Merge tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Fix error output of boot instance creation failure

   Currently if a boot instance creation fails, instead of printing out
   the name of the instance that failed, it prints "(null)". That is
   because it prints "cur_str" that had already been processed by
   strsep(). Print the saved name instead.

   While at it, print the error code of the failure.

 - Fix use-after-free for same named historgrams

   Histograms can be named so that they can be used in multiple events.
   But if the named histogram has a variable attached, the second event
   that uses the named histogram which duplicates it and needs to free
   the original after duplication leaves the old variable in place and
   still visible. If another histogram uses than variable, it will use
   the stale one which will try to reference the freed duplicate
   histogram and crash the kernel.

   Free the duplicate variables along with the duplicated histogram
   data.

 - Check return value of kthread_run() in event self test

   The events self tests uses a kthread for testing but does not check
   if it succeeded in creating a kthread. If the kthread creation were
   to fail, the code will still try to call kthread_stop() on the error
   returned.

 - Fix race between reading trace_pipe and updating subbuffer size

   If a user is reading the trace_pipe file at the same time they update
   the ring buffer sub-buffer size, can cause the trace_pipe read to
   read stale data. Add trace_access_lock() around updating the ring
   buffer sub-buffer size.

 - Fix eventfs_inode on failure path in creation of the events directory

   In the creation of the "events" directory, if after allocating the
   eventfs_inode a failure is detected, it calls cleanup_ei() which
   calls free_ei(). The free_ei() will test if eventfs_inode being freed
   has no children. It is a bug if it does. But on the failure case of
   the creation of the "events" directory, the children lists have not
   yet been initialized and the free will trigger a warning because
   list_empty() on an uninitialized list returns false.

   Move the initialization into init_ei() where it makes more sense and
   makes sure that a created eventfs_inode has its lists initialized
   upon creation.

 - Check return value of kthread_run() in ftrace direct sample code

   The sample code that shows how to use the ftrace direct calls does
   not test the return of kthread_run() to see if it succeeds. Return a
   failure if the kthread_run() doesn't succeed.

 - Clear user events state on fork in case of alloc failure

   On fork, the child gets a pointer to the parent's user events state.
   It makes a copy of it then updates the child's pointer to it. But if
   the allocation fails, the duplication function leaves the child with
   a pointer to its parent's descriptor. When the child cleans up its
   data, it will free the parent's descriptor while the parent is still
   using it.

   In the duplication function, set the child's user_event_mm to NULL
   before testing if the allocation succeeded, and when it exits it will
   not free the parent's descriptor.

 - Fix retry exhaustion in simple ring buffer reader swap

   simple_ring_buffer_swap_reader_page() starts with retry set to 8 and
   post-decrements it only after a failed link replacement. On the final
   attempt, a successful replacement leaves retry at zero, while a
   failed replacement leaves it at -1.

   But the check for success expects the retry value to be non-zero and
   exits with an error on zero. This is the opposite result. Fix it.

 - Fail nicely when the remote swap_reader_page() returns an error

   Currently, if the swap_reader_page() of a remote buffer fails, it
   triggers a WARN_ON_ONCE() and continues normally. Instead, have it
   exit with an error and a pr_warn() print instead of a full WARNING.

* tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  ring-buffer: Stop remote reader update when page swap fails
  tracing: Fix retry exhaustion in simple ring buffer reader swap
  tracing/user_events: Clear copied tracing state before fork duplication
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-multi-modify
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-modify
  eventfs: Initialize ei->children and ei->list in init_ei()
  tracing: Fix use-after-free in trace_pipe read on sub-buffer order change
  tracing: Fix crash passing ERR_PTR to kthread_stop()
  tracing: Fix use-after-free with same-name named triggers
  tracing: Fix logged instance name on creation failure

22 files changed, 26881 insertions, 0 deletions

diff --git a/drivers/s390/block/Kconfig b/drivers/s390/block/Kconfig
new file mode 100644
index 000000000..877a9bc7f
--- /dev/null
+++ b/drivers/s390/block/Kconfig
@@ -0,0 +1,73 @@
+# SPDX-License-Identifier: GPL-2.0
+comment "S/390 block device drivers"
+	depends on S390 && BLOCK
+
+config DCSSBLK
+	def_tristate m
+	prompt "DCSSBLK support"
+	depends on S390 && BLOCK && ZONE_DEVICE
+	select FS_DAX
+	help
+	  Support for dcss block device
+
+config DASD
+	def_tristate y
+	prompt "Support for DASD devices"
+	depends on CCW && BLOCK
+	help
+	  Enable this option if you want to access DASDs directly utilizing
+	  S/390s channel subsystem commands. This is necessary for running
+	  natively on a single image or an LPAR.
+
+config DASD_PROFILE
+	def_bool y
+	prompt "Profiling support for dasd devices"
+	depends on DASD
+	help
+	  Enable this option if you want to see profiling information
+	  in /proc/dasd/statistics.
+
+config DASD_ECKD
+	def_tristate y
+	prompt "Support for ECKD Disks"
+	depends on DASD
+	help
+	  ECKD devices are the most commonly used devices. You should enable
+	  this option unless you are very sure to have no ECKD device.
+
+config DASD_FBA
+	def_tristate y
+	prompt "Support for FBA  Disks"
+	depends on DASD
+	help
+	  Select this option to be able to access FBA devices. It is safe to
+	  say "Y".
+
+config DASD_DIAG
+	def_tristate y
+	prompt "Support for DIAG access to Disks"
+	depends on DASD
+	help
+	  Select this option if you want to use Diagnose250 command to access
+	  Disks under VM.  If you are not running under VM or unsure what it is,
+	  say "N".
+
+config DASD_EER
+	def_bool y
+	prompt "Extended error reporting (EER)"
+	depends on DASD
+	help
+	  This driver provides a character device interface to the
+	  DASD extended error reporting. This is only needed if you want to
+	  use applications written for the EER facility.
+
+config SCM_BLOCK
+	def_tristate m
+	prompt "Support for Storage Class Memory"
+	depends on S390 && BLOCK && EADM_SCH && SCM_BUS
+	help
+	  Block device driver for Storage Class Memory (SCM). This driver
+	  provides a block device interface for each available SCM increment.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called scm_block.
diff --git a/drivers/s390/block/Makefile b/drivers/s390/block/Makefile
new file mode 100644
index 000000000..3dd1a3ce7
--- /dev/null
+++ b/drivers/s390/block/Makefile
@@ -0,0 +1,24 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# S/390 block devices
+#
+
+CONTEXT_ANALYSIS := y
+
+dasd_eckd_mod-objs := dasd_eckd.o dasd_3990_erp.o dasd_alias.o
+dasd_fba_mod-objs  := dasd_fba.o
+dasd_diag_mod-objs := dasd_diag.o
+dasd_mod-objs      := dasd.o dasd_ioctl.o dasd_proc.o dasd_devmap.o \
+			dasd_genhd.o dasd_erp.o
+ifdef CONFIG_DASD_EER
+dasd_mod-objs      += dasd_eer.o
+endif
+
+obj-$(CONFIG_DASD) += dasd_mod.o
+obj-$(CONFIG_DASD_DIAG) += dasd_diag_mod.o
+obj-$(CONFIG_DASD_ECKD) += dasd_eckd_mod.o
+obj-$(CONFIG_DASD_FBA)  += dasd_fba_mod.o
+obj-$(CONFIG_DCSSBLK) += dcssblk.o
+
+scm_block-objs := scm_drv.o scm_blk.o
+obj-$(CONFIG_SCM_BLOCK) += scm_block.o
diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
new file mode 100644
index 000000000..597975831
--- /dev/null
+++ b/drivers/s390/block/dasd.c
@@ -0,0 +1,4228 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Carsten Otte <Cotte@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2009
+ */
+
+#include <linux/export.h>
+#include <linux/kmod.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/ctype.h>
+#include <linux/major.h>
+#include <linux/slab.h>
+#include <linux/hdreg.h>
+#include <linux/async.h>
+#include <linux/mutex.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <linux/vmalloc.h>
+#include <linux/delay.h>
+
+#include <asm/machine.h>
+#include <asm/ccwdev.h>
+#include <asm/ebcdic.h>
+#include <asm/idals.h>
+#include <asm/itcw.h>
+#include <asm/diag.h>
+
+#include "dasd_int.h"
+/*
+ * SECTION: Constant definitions to be used within this file
+ */
+#define DASD_CHANQ_MAX_SIZE 4
+
+#define DASD_DIAG_MOD		"dasd_diag_mod"
+
+/*
+ * SECTION: exported variables of dasd.c
+ */
+debug_info_t *dasd_debug_area;
+EXPORT_SYMBOL(dasd_debug_area);
+static struct dentry *dasd_debugfs_root_entry;
+struct dasd_discipline *dasd_diag_discipline_pointer;
+EXPORT_SYMBOL(dasd_diag_discipline_pointer);
+void dasd_int_handler(struct ccw_device *, unsigned long, struct irb *);
+
+MODULE_AUTHOR("Holger Smolinski <Holger.Smolinski@de.ibm.com>");
+MODULE_DESCRIPTION("Linux on S/390 DASD device driver,"
+		   " Copyright IBM Corp. 2000");
+MODULE_LICENSE("GPL");
+
+/*
+ * SECTION: prototypes for static functions of dasd.c
+ */
+static int dasd_flush_block_queue(struct dasd_block *);
+static void dasd_device_tasklet(unsigned long);
+static void dasd_block_tasklet(unsigned long);
+static void do_kick_device(struct work_struct *);
+static void do_reload_device(struct work_struct *);
+static void do_requeue_requests(struct work_struct *);
+static void dasd_return_cqr_cb(struct dasd_ccw_req *, void *);
+static void dasd_device_timeout(struct timer_list *);
+static void dasd_block_timeout(struct timer_list *);
+static void __dasd_process_erp(struct dasd_device *, struct dasd_ccw_req *);
+static void dasd_profile_init(struct dasd_profile *, struct dentry *);
+static void dasd_profile_exit(struct dasd_profile *);
+static void dasd_hosts_init(struct dentry *, struct dasd_device *);
+static void dasd_hosts_exit(struct dasd_device *);
+static int dasd_handle_autoquiesce(struct dasd_device *, struct dasd_ccw_req *,
+				   unsigned int);
+/*
+ * SECTION: Operations on the device structure.
+ */
+static wait_queue_head_t dasd_init_waitq;
+static wait_queue_head_t dasd_flush_wq;
+static wait_queue_head_t generic_waitq;
+static wait_queue_head_t shutdown_waitq;
+
+/*
+ * Allocate memory for a new device structure.
+ */
+struct dasd_device *dasd_alloc_device(void)
+{
+	struct dasd_device *device;
+
+	device = kzalloc_obj(struct dasd_device, GFP_KERNEL);
+	if (!device)
+		return ERR_PTR(-ENOMEM);
+
+	/*
+	 * Four pages: a full-track ITCW is roughly twice the size of a plain
+	 * track-mode one, so this keeps two maximum-size requests in flight.
+	 */
+	device->ccw_mem = (void *)__get_free_pages(GFP_KERNEL | GFP_DMA, 2);
+	if (!device->ccw_mem) {
+		kfree(device);
+		return ERR_PTR(-ENOMEM);
+	}
+	/* per-request track-filler buffers (R0 + count records) */
+	device->fill_mem = (void *)__get_free_pages(GFP_KERNEL | GFP_DMA, 1);
+	if (!device->fill_mem) {
+		free_pages((unsigned long)device->ccw_mem, 2);
+		kfree(device);
+		return ERR_PTR(-ENOMEM);
+	}
+	/* Get one page for error recovery. */
+	device->erp_mem = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!device->erp_mem) {
+		free_pages((unsigned long)device->fill_mem, 1);
+		free_pages((unsigned long)device->ccw_mem, 2);
+		kfree(device);
+		return ERR_PTR(-ENOMEM);
+	}
+	/* sized like ccw_chunks: two max-size NRF format requests in flight */
+	device->ese_mem = (void *)__get_free_pages(GFP_KERNEL | GFP_DMA, 2);
+	if (!device->ese_mem) {
+		free_page((unsigned long)device->erp_mem);
+		free_pages((unsigned long)device->fill_mem, 1);
+		free_pages((unsigned long)device->ccw_mem, 2);
+		kfree(device);
+		return ERR_PTR(-ENOMEM);
+	}
+	device->nulldata = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!device->nulldata) {
+		free_page((unsigned long)device->erp_mem);
+		free_pages((unsigned long)device->fill_mem, 1);
+		free_pages((unsigned long)device->ccw_mem, 2);
+		free_pages((unsigned long)device->ese_mem, 2);
+		kfree(device);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	dasd_init_chunklist(&device->ccw_chunks, device->ccw_mem, PAGE_SIZE * 4);
+	dasd_init_chunklist(&device->fill_chunks, device->fill_mem, PAGE_SIZE * 2);
+	dasd_init_chunklist(&device->erp_chunks, device->erp_mem, PAGE_SIZE);
+	dasd_init_chunklist(&device->ese_chunks, device->ese_mem, PAGE_SIZE * 4);
+	spin_lock_init(&device->mem_lock);
+	atomic_set(&device->tasklet_scheduled, 0);
+	tasklet_init(&device->tasklet, dasd_device_tasklet,
+		     (unsigned long) device);
+	INIT_LIST_HEAD(&device->ccw_queue);
+	timer_setup(&device->timer, dasd_device_timeout, 0);
+	INIT_WORK(&device->kick_work, do_kick_device);
+	INIT_WORK(&device->reload_device, do_reload_device);
+	INIT_WORK(&device->requeue_requests, do_requeue_requests);
+	device->state = DASD_STATE_NEW;
+	device->target = DASD_STATE_NEW;
+	mutex_init(&device->state_mutex);
+	spin_lock_init(&device->profile.lock);
+	return device;
+}
+
+/*
+ * Free memory of a device structure.
+ */
+void dasd_free_device(struct dasd_device *device)
+{
+	kfree(device->private);
+	free_pages((unsigned long)device->ese_mem, 2);
+	free_page((unsigned long)device->erp_mem);
+	free_pages((unsigned long)device->fill_mem, 1);
+	free_pages((unsigned long)device->ccw_mem, 2);
+	free_page((unsigned long)device->nulldata);
+	kfree(device);
+}
+
+/*
+ * Allocate memory for a new device structure.
+ */
+struct dasd_block *dasd_alloc_block(void)
+{
+	struct dasd_block *block;
+
+	block = kzalloc_obj(*block, GFP_ATOMIC);
+	if (!block)
+		return ERR_PTR(-ENOMEM);
+	/* open_count = 0 means device online but not in use */
+	atomic_set(&block->open_count, -1);
+
+	atomic_set(&block->tasklet_scheduled, 0);
+	tasklet_init(&block->tasklet, dasd_block_tasklet,
+		     (unsigned long) block);
+	INIT_LIST_HEAD(&block->ccw_queue);
+	spin_lock_init(&block->queue_lock);
+	INIT_LIST_HEAD(&block->format_list);
+	spin_lock_init(&block->format_lock);
+	INIT_LIST_HEAD(&block->ese_staging);
+	spin_lock_init(&block->ese_lock);
+	timer_setup(&block->timer, dasd_block_timeout, 0);
+	spin_lock_init(&block->profile.lock);
+
+	return block;
+}
+EXPORT_SYMBOL_GPL(dasd_alloc_block);
+
+/*
+ * Free memory of a device structure.
+ */
+void dasd_free_block(struct dasd_block *block)
+{
+	kfree(block);
+}
+EXPORT_SYMBOL_GPL(dasd_free_block);
+
+/*
+ * Make a new device known to the system.
+ */
+static int dasd_state_new_to_known(struct dasd_device *device)
+{
+	/*
+	 * As long as the device is not in state DASD_STATE_NEW we want to
+	 * keep the reference count > 0.
+	 */
+	dasd_get_device(device);
+	device->state = DASD_STATE_KNOWN;
+	return 0;
+}
+
+/*
+ * Let the system forget about a device.
+ */
+static int dasd_state_known_to_new(struct dasd_device *device)
+{
+	/* Disable extended error reporting for this device. */
+	dasd_eer_disable(device);
+	device->state = DASD_STATE_NEW;
+
+	/* Give up reference we took in dasd_state_new_to_known. */
+	dasd_put_device(device);
+	return 0;
+}
+
+/*
+ * Request the irq line for the device.
+ */
+static int dasd_state_known_to_basic(struct dasd_device *device)
+{
+	struct dasd_block *block = device->block;
+	int rc = 0;
+
+	/* Allocate and register gendisk structure. */
+	if (block) {
+		rc = dasd_gendisk_alloc(block);
+		if (rc)
+			return rc;
+		block->debugfs_dentry =
+			debugfs_create_dir(block->gdp->disk_name,
+					   dasd_debugfs_root_entry);
+		dasd_profile_init(&block->profile, block->debugfs_dentry);
+		if (dasd_global_profile_level == DASD_PROFILE_ON)
+			dasd_profile_on(&device->block->profile);
+	}
+	device->debugfs_dentry =
+		debugfs_create_dir(dev_name(&device->cdev->dev),
+				   dasd_debugfs_root_entry);
+	dasd_profile_init(&device->profile, device->debugfs_dentry);
+	dasd_hosts_init(device->debugfs_dentry, device);
+
+	/* register 'device' debug area, used for all DBF_DEV_XXX calls */
+	device->debug_area = debug_register(dev_name(&device->cdev->dev), 4, 1,
+					    8 * sizeof(long));
+	debug_register_view(device->debug_area, &debug_sprintf_view);
+	debug_set_level(device->debug_area, DBF_WARNING);
+	DBF_DEV_EVENT(DBF_EMERG, device, "%s", "debug area created");
+
+	device->state = DASD_STATE_BASIC;
+
+	return rc;
+}
+
+/*
+ * Release the irq line for the device. Terminate any running i/o.
+ */
+static int dasd_state_basic_to_known(struct dasd_device *device)
+{
+	int rc;
+
+	if (device->discipline->basic_to_known) {
+		rc = device->discipline->basic_to_known(device);
+		if (rc)
+			return rc;
+	}
+
+	if (device->block) {
+		dasd_profile_exit(&device->block->profile);
+		debugfs_remove(device->block->debugfs_dentry);
+		dasd_gendisk_free(device->block);
+		dasd_block_clear_timer(device->block);
+	}
+	rc = dasd_flush_device_queue(device);
+	if (rc)
+		return rc;
+	dasd_device_clear_timer(device);
+	dasd_profile_exit(&device->profile);
+	dasd_hosts_exit(device);
+	debugfs_remove(device->debugfs_dentry);
+	DBF_DEV_EVENT(DBF_EMERG, device, "%p debug area deleted", device);
+	if (device->debug_area != NULL) {
+		debug_unregister(device->debug_area);
+		device->debug_area = NULL;
+	}
+	device->state = DASD_STATE_KNOWN;
+	return 0;
+}
+
+/*
+ * Do the initial analysis. The do_analysis function may return
+ * -EAGAIN in which case the device keeps the state DASD_STATE_BASIC
+ * until the discipline decides to continue the startup sequence
+ * by calling the function dasd_change_state. The eckd disciplines
+ * uses this to start a ccw that detects the format. The completion
+ * interrupt for this detection ccw uses the kernel event daemon to
+ * trigger the call to dasd_change_state. All this is done in the
+ * discipline code, see dasd_eckd.c.
+ * After the analysis ccw is done (do_analysis returned 0) the block
+ * device is setup.
+ * In case the analysis returns an error, the device setup is stopped
+ * (a fake disk was already added to allow formatting).
+ */
+static int dasd_state_basic_to_ready(struct dasd_device *device)
+{
+	struct dasd_block *block = device->block;
+	struct queue_limits lim;
+	int rc = 0;
+
+	/* make disk known with correct capacity */
+	if (!block) {
+		device->state = DASD_STATE_READY;
+		goto out;
+	}
+
+	if (block->base->discipline->do_analysis != NULL)
+		rc = block->base->discipline->do_analysis(block);
+	if (rc) {
+		if (rc == -EAGAIN)
+			return rc;
+		device->state = DASD_STATE_UNFMT;
+		kobject_uevent(&disk_to_dev(device->block->gdp)->kobj,
+			       KOBJ_CHANGE);
+		goto out;
+	}
+
+	lim = queue_limits_start_update(block->gdp->queue);
+	lim.max_dev_sectors = device->discipline->max_sectors(block);
+	lim.max_hw_sectors = lim.max_dev_sectors;
+	lim.logical_block_size = block->bp_block;
+	/*
+	 * Adjust dma_alignment to match block_size - 1
+	 * to ensure proper buffer alignment checks in the block layer.
+	 */
+	lim.dma_alignment = lim.logical_block_size - 1;
+
+	if (device->features & DASD_FEATURE_DISCARD) {
+		unsigned int max_bytes;
+
+		if (device->discipline->disc_limits) {
+			device->discipline->disc_limits(block, &lim);
+		} else {
+			lim.discard_granularity = block->bp_block;
+			/* Calculate max_discard_sectors and make it PAGE aligned */
+			max_bytes = USHRT_MAX * block->bp_block;
+			max_bytes = ALIGN_DOWN(max_bytes, PAGE_SIZE);
+			lim.max_hw_discard_sectors = max_bytes / block->bp_block;
+			lim.max_write_zeroes_sectors = lim.max_hw_discard_sectors;
+		}
+	}
+	rc = queue_limits_commit_update(block->gdp->queue, &lim);
+	if (rc)
+		return rc;
+
+	set_capacity(block->gdp, block->blocks << block->s2b_shift);
+	device->state = DASD_STATE_READY;
+
+	rc = dasd_scan_partitions(block);
+	if (rc) {
+		device->state = DASD_STATE_BASIC;
+		return rc;
+	}
+
+out:
+	if (device->discipline->basic_to_ready)
+		rc = device->discipline->basic_to_ready(device);
+	return rc;
+}
+
+static inline
+int _wait_for_empty_queues(struct dasd_device *device)
+{
+	if (device->block)
+		return list_empty(&device->ccw_queue) &&
+			list_empty(&device->block->ccw_queue) &&
+			list_empty(&device->block->ese_staging);
+	else
+		return list_empty(&device->ccw_queue);
+}
+
+/*
+ * Remove device from block device layer. Destroy dirty buffers.
+ * Forget format information. Check if the target level is basic
+ * and if it is create fake disk for formatting.
+ */
+static int dasd_state_ready_to_basic(struct dasd_device *device)
+{
+	int rc;
+
+	device->state = DASD_STATE_BASIC;
+	if (device->block) {
+		struct dasd_block *block = device->block;
+		rc = dasd_flush_block_queue(block);
+		if (rc) {
+			device->state = DASD_STATE_READY;
+			return rc;
+		}
+		dasd_destroy_partitions(block);
+		block->blocks = 0;
+		block->bp_block = 0;
+		block->s2b_shift = 0;
+	}
+	return 0;
+}
+
+/*
+ * Back to basic.
+ */
+static int dasd_state_unfmt_to_basic(struct dasd_device *device)
+{
+	device->state = DASD_STATE_BASIC;
+	return 0;
+}
+
+/*
+ * Make the device online and schedule the bottom half to start
+ * the requeueing of requests from the linux request queue to the
+ * ccw queue.
+ */
+static int
+dasd_state_ready_to_online(struct dasd_device * device)
+{
+	device->state = DASD_STATE_ONLINE;
+	if (device->block) {
+		dasd_schedule_block_bh(device->block);
+		if ((device->features & DASD_FEATURE_USERAW)) {
+			kobject_uevent(&disk_to_dev(device->block->gdp)->kobj,
+					KOBJ_CHANGE);
+			return 0;
+		}
+		disk_uevent(file_bdev(device->block->bdev_file)->bd_disk,
+			    KOBJ_CHANGE);
+	}
+	return 0;
+}
+
+/*
+ * Stop the requeueing of requests again.
+ */
+static int dasd_state_online_to_ready(struct dasd_device *device)
+{
+	int rc;
+
+	if (device->discipline->online_to_ready) {
+		rc = device->discipline->online_to_ready(device);
+		if (rc)
+			return rc;
+	}
+
+	device->state = DASD_STATE_READY;
+	if (device->block && !(device->features & DASD_FEATURE_USERAW))
+		disk_uevent(file_bdev(device->block->bdev_file)->bd_disk,
+			    KOBJ_CHANGE);
+	return 0;
+}
+
+/*
+ * Device startup state changes.
+ */
+static int dasd_increase_state(struct dasd_device *device)
+{
+	int rc;
+
+	rc = 0;
+	if (device->state == DASD_STATE_NEW &&
+	    device->target >= DASD_STATE_KNOWN)
+		rc = dasd_state_new_to_known(device);
+
+	if (!rc &&
+	    device->state == DASD_STATE_KNOWN &&
+	    device->target >= DASD_STATE_BASIC)
+		rc = dasd_state_known_to_basic(device);
+
+	if (!rc &&
+	    device->state == DASD_STATE_BASIC &&
+	    device->target >= DASD_STATE_READY)
+		rc = dasd_state_basic_to_ready(device);
+
+	if (!rc &&
+	    device->state == DASD_STATE_UNFMT &&
+	    device->target > DASD_STATE_UNFMT)
+		rc = -EPERM;
+
+	if (!rc &&
+	    device->state == DASD_STATE_READY &&
+	    device->target >= DASD_STATE_ONLINE)
+		rc = dasd_state_ready_to_online(device);
+
+	return rc;
+}
+
+/*
+ * Device shutdown state changes.
+ */
+static int dasd_decrease_state(struct dasd_device *device)
+{
+	int rc;
+
+	rc = 0;
+	if (device->state == DASD_STATE_ONLINE &&
+	    device->target <= DASD_STATE_READY)
+		rc = dasd_state_online_to_ready(device);
+
+	if (!rc &&
+	    device->state == DASD_STATE_READY &&
+	    device->target <= DASD_STATE_BASIC)
+		rc = dasd_state_ready_to_basic(device);
+
+	if (!rc &&
+	    device->state == DASD_STATE_UNFMT &&
+	    device->target <= DASD_STATE_BASIC)
+		rc = dasd_state_unfmt_to_basic(device);
+
+	if (!rc &&
+	    device->state == DASD_STATE_BASIC &&
+	    device->target <= DASD_STATE_KNOWN)
+		rc = dasd_state_basic_to_known(device);
+
+	if (!rc &&
+	    device->state == DASD_STATE_KNOWN &&
+	    device->target <= DASD_STATE_NEW)
+		rc = dasd_state_known_to_new(device);
+
+	return rc;
+}
+
+/*
+ * This is the main startup/shutdown routine.
+ */
+static void dasd_change_state(struct dasd_device *device)
+{
+	int rc;
+
+	if (device->state == device->target)
+		/* Already where we want to go today... */
+		return;
+	if (device->state < device->target)
+		rc = dasd_increase_state(device);
+	else
+		rc = dasd_decrease_state(device);
+	if (rc == -EAGAIN)
+		return;
+	if (rc)
+		device->target = device->state;
+
+	/* let user-space know that the device status changed */
+	kobject_uevent(&device->cdev->dev.kobj, KOBJ_CHANGE);
+
+	if (device->state == device->target)
+		wake_up(&dasd_init_waitq);
+}
+
+/*
+ * Kick starter for devices that did not complete the startup/shutdown
+ * procedure or were sleeping because of a pending state.
+ * dasd_kick_device will schedule a call do do_kick_device to the kernel
+ * event daemon.
+ */
+static void do_kick_device(struct work_struct *work)
+{
+	struct dasd_device *device = container_of(work, struct dasd_device, kick_work);
+	mutex_lock(&device->state_mutex);
+	dasd_change_state(device);
+	mutex_unlock(&device->state_mutex);
+	dasd_schedule_device_bh(device);
+	dasd_put_device(device);
+}
+
+void dasd_kick_device(struct dasd_device *device)
+{
+	dasd_get_device(device);
+	/* queue call to dasd_kick_device to the kernel event daemon. */
+	if (!schedule_work(&device->kick_work))
+		dasd_put_device(device);
+}
+EXPORT_SYMBOL(dasd_kick_device);
+
+/*
+ * dasd_reload_device will schedule a call do do_reload_device to the kernel
+ * event daemon.
+ */
+static void do_reload_device(struct work_struct *work)
+{
+	struct dasd_device *device = container_of(work, struct dasd_device,
+						  reload_device);
+	device->discipline->reload(device);
+	dasd_put_device(device);
+}
+
+void dasd_reload_device(struct dasd_device *device)
+{
+	dasd_get_device(device);
+	/* queue call to dasd_reload_device to the kernel event daemon. */
+	if (!schedule_work(&device->reload_device))
+		dasd_put_device(device);
+}
+EXPORT_SYMBOL(dasd_reload_device);
+
+/*
+ * Set the target state for a device and starts the state change.
+ */
+void dasd_set_target_state(struct dasd_device *device, int target)
+{
+	dasd_get_device(device);
+	mutex_lock(&device->state_mutex);
+	/* If we are in probeonly mode stop at DASD_STATE_READY. */
+	if (dasd_probeonly && target > DASD_STATE_READY)
+		target = DASD_STATE_READY;
+	if (device->target != target) {
+		if (device->state == target)
+			wake_up(&dasd_init_waitq);
+		device->target = target;
+	}
+	if (device->state != device->target)
+		dasd_change_state(device);
+	mutex_unlock(&device->state_mutex);
+	dasd_put_device(device);
+}
+
+/*
+ * Enable devices with device numbers in [from..to].
+ */
+static inline int _wait_for_device(struct dasd_device *device)
+{
+	return (device->state == device->target);
+}
+
+void dasd_enable_device(struct dasd_device *device)
+{
+	dasd_set_target_state(device, DASD_STATE_ONLINE);
+	if (device->state <= DASD_STATE_KNOWN)
+		/* No discipline for device found. */
+		dasd_set_target_state(device, DASD_STATE_NEW);
+	/* Now wait for the devices to come up. */
+	wait_event(dasd_init_waitq, _wait_for_device(device));
+
+	dasd_reload_device(device);
+	if (device->discipline->kick_validate)
+		device->discipline->kick_validate(device);
+}
+EXPORT_SYMBOL(dasd_enable_device);
+
+/*
+ * SECTION: device operation (interrupt handler, start i/o, term i/o ...)
+ */
+
+unsigned int dasd_global_profile_level = DASD_PROFILE_OFF;
+
+#ifdef CONFIG_DASD_PROFILE
+struct dasd_profile dasd_global_profile = {
+	.lock = __SPIN_LOCK_UNLOCKED(dasd_global_profile.lock),
+};
+static struct dentry *dasd_debugfs_global_entry;
+
+/*
+ * Add profiling information for cqr before execution.
+ */
+static void dasd_profile_start(struct dasd_block *block,
+			       struct dasd_ccw_req *cqr,
+			       struct request *req)
+{
+	struct list_head *l;
+	unsigned int counter;
+	struct dasd_device *device;
+
+	/* count the length of the chanq for statistics */
+	counter = 0;
+	if (dasd_global_profile_level || block->profile.data)
+		list_for_each(l, &block->ccw_queue)
+			if (++counter >= 31)
+				break;
+
+	spin_lock(&dasd_global_profile.lock);
+	if (dasd_global_profile.data) {
+		dasd_global_profile.data->dasd_io_nr_req[counter]++;
+		if (rq_data_dir(req) == READ)
+			dasd_global_profile.data->dasd_read_nr_req[counter]++;
+	}
+	spin_unlock(&dasd_global_profile.lock);
+
+	spin_lock(&block->profile.lock);
+	if (block->profile.data) {
+		block->profile.data->dasd_io_nr_req[counter]++;
+		if (rq_data_dir(req) == READ)
+			block->profile.data->dasd_read_nr_req[counter]++;
+	}
+	spin_unlock(&block->profile.lock);
+
+	/*
+	 * We count the request for the start device, even though it may run on
+	 * some other device due to error recovery. This way we make sure that
+	 * we count each request only once.
+	 */
+	device = cqr->startdev;
+	if (!device->profile.data)
+		return;
+
+	spin_lock(get_ccwdev_lock(device->cdev));
+	counter = 1; /* request is not yet queued on the start device */
+	list_for_each(l, &device->ccw_queue)
+		if (++counter >= 31)
+			break;
+	spin_unlock(get_ccwdev_lock(device->cdev));
+
+	spin_lock(&device->profile.lock);
+	device->profile.data->dasd_io_nr_req[counter]++;
+	if (rq_data_dir(req) == READ)
+		device->profile.data->dasd_read_nr_req[counter]++;
+	spin_unlock(&device->profile.lock);
+}
+
+/*
+ * Add profiling information for cqr after execution.
+ */
+
+#define dasd_profile_counter(value, index)			   \
+{								   \
+	for (index = 0; index < 31 && value >> (2+index); index++) \
+		;						   \
+}
+
+static void dasd_profile_end_add_data(struct dasd_profile_info *data,
+				      int is_alias,
+				      int is_tpm,
+				      int is_read,
+				      long sectors,
+				      int sectors_ind,
+				      int tottime_ind,
+				      int tottimeps_ind,
+				      int strtime_ind,
+				      int irqtime_ind,
+				      int irqtimeps_ind,
+				      int endtime_ind)
+{
+	/* in case of an overflow, reset the whole profile */
+	if (data->dasd_io_reqs == UINT_MAX) {
+			memset(data, 0, sizeof(*data));
+			ktime_get_real_ts64(&data->starttod);
+	}
+	data->dasd_io_reqs++;
+	data->dasd_io_sects += sectors;
+	if (is_alias)
+		data->dasd_io_alias++;
+	if (is_tpm)
+		data->dasd_io_tpm++;
+
+	data->dasd_io_secs[sectors_ind]++;
+	data->dasd_io_times[tottime_ind]++;
+	data->dasd_io_timps[tottimeps_ind]++;
+	data->dasd_io_time1[strtime_ind]++;
+	data->dasd_io_time2[irqtime_ind]++;
+	data->dasd_io_time2ps[irqtimeps_ind]++;
+	data->dasd_io_time3[endtime_ind]++;
+
+	if (is_read) {
+		data->dasd_read_reqs++;
+		data->dasd_read_sects += sectors;
+		if (is_alias)
+			data->dasd_read_alias++;
+		if (is_tpm)
+			data->dasd_read_tpm++;
+		data->dasd_read_secs[sectors_ind]++;
+		data->dasd_read_times[tottime_ind]++;
+		data->dasd_read_time1[strtime_ind]++;
+		data->dasd_read_time2[irqtime_ind]++;
+		data->dasd_read_time3[endtime_ind]++;
+	}
+}
+
+static void dasd_profile_end(struct dasd_block *block,
+			     struct dasd_ccw_req *cqr,
+			     struct request *req)
+{
+	unsigned long strtime, irqtime, endtime, tottime;
+	unsigned long tottimeps, sectors;
+	struct dasd_device *device;
+	int sectors_ind, tottime_ind, tottimeps_ind, strtime_ind;
+	int irqtime_ind, irqtimeps_ind, endtime_ind;
+	struct dasd_profile_info *data;
+
+	device = cqr->startdev;
+	if (!(dasd_global_profile_level ||
+	      block->profile.data ||
+	      device->profile.data))
+		return;
+
+	sectors = blk_rq_sectors(req);
+	if (!cqr->buildclk || !cqr->startclk ||
+	    !cqr->stopclk || !cqr->endclk ||
+	    !sectors)
+		return;
+
+	strtime = ((cqr->startclk - cqr->buildclk) >> 12);
+	irqtime = ((cqr->stopclk - cqr->startclk) >> 12);
+	endtime = ((cqr->endclk - cqr->stopclk) >> 12);
+	tottime = ((cqr->endclk - cqr->buildclk) >> 12);
+	tottimeps = tottime / sectors;
+
+	dasd_profile_counter(sectors, sectors_ind);
+	dasd_profile_counter(tottime, tottime_ind);
+	dasd_profile_counter(tottimeps, tottimeps_ind);
+	dasd_profile_counter(strtime, strtime_ind);
+	dasd_profile_counter(irqtime, irqtime_ind);
+	dasd_profile_counter(irqtime / sectors, irqtimeps_ind);
+	dasd_profile_counter(endtime, endtime_ind);
+
+	spin_lock(&dasd_global_profile.lock);
+	if (dasd_global_profile.data) {
+		data = dasd_global_profile.data;
+		data->dasd_sum_times += tottime;
+		data->dasd_sum_time_str += strtime;
+		data->dasd_sum_time_irq += irqtime;
+		data->dasd_sum_time_end += endtime;
+		dasd_profile_end_add_data(dasd_global_profile.data,
+					  cqr->startdev != block->base,
+					  cqr->cpmode == 1,
+					  rq_data_dir(req) == READ,
+					  sectors, sectors_ind, tottime_ind,
+					  tottimeps_ind, strtime_ind,
+					  irqtime_ind, irqtimeps_ind,
+					  endtime_ind);
+	}
+	spin_unlock(&dasd_global_profile.lock);
+
+	spin_lock(&block->profile.lock);
+	if (block->profile.data) {
+		data = block->profile.data;
+		data->dasd_sum_times += tottime;
+		data->dasd_sum_time_str += strtime;
+		data->dasd_sum_time_irq += irqtime;
+		data->dasd_sum_time_end += endtime;
+		dasd_profile_end_add_data(block->profile.data,
+					  cqr->startdev != block->base,
+					  cqr->cpmode == 1,
+					  rq_data_dir(req) == READ,
+					  sectors, sectors_ind, tottime_ind,
+					  tottimeps_ind, strtime_ind,
+					  irqtime_ind, irqtimeps_ind,
+					  endtime_ind);
+	}
+	spin_unlock(&block->profile.lock);
+
+	spin_lock(&device->profile.lock);
+	if (device->profile.data) {
+		data = device->profile.data;
+		data->dasd_sum_times += tottime;
+		data->dasd_sum_time_str += strtime;
+		data->dasd_sum_time_irq += irqtime;
+		data->dasd_sum_time_end += endtime;
+		dasd_profile_end_add_data(device->profile.data,
+					  cqr->startdev != block->base,
+					  cqr->cpmode == 1,
+					  rq_data_dir(req) == READ,
+					  sectors, sectors_ind, tottime_ind,
+					  tottimeps_ind, strtime_ind,
+					  irqtime_ind, irqtimeps_ind,
+					  endtime_ind);
+	}
+	spin_unlock(&device->profile.lock);
+}
+
+void dasd_profile_reset(struct dasd_profile *profile)
+{
+	struct dasd_profile_info *data;
+
+	spin_lock_bh(&profile->lock);
+	data = profile->data;
+	if (!data) {
+		spin_unlock_bh(&profile->lock);
+		return;
+	}
+	memset(data, 0, sizeof(*data));
+	ktime_get_real_ts64(&data->starttod);
+	spin_unlock_bh(&profile->lock);
+}
+
+int dasd_profile_on(struct dasd_profile *profile)
+{
+	struct dasd_profile_info *data;
+
+	data = kzalloc_obj(*data);
+	if (!data)
+		return -ENOMEM;
+	spin_lock_bh(&profile->lock);
+	if (profile->data) {
+		spin_unlock_bh(&profile->lock);
+		kfree(data);
+		return 0;
+	}
+	ktime_get_real_ts64(&data->starttod);
+	profile->data = data;
+	spin_unlock_bh(&profile->lock);
+	return 0;
+}
+
+void dasd_profile_off(struct dasd_profile *profile)
+{
+	spin_lock_bh(&profile->lock);
+	kfree(profile->data);
+	profile->data = NULL;
+	spin_unlock_bh(&profile->lock);
+}
+
+char *dasd_get_user_string(const char __user *user_buf, size_t user_len)
+{
+	char *buffer;
+
+	buffer = vmalloc(user_len + 1);
+	if (buffer == NULL)
+		return ERR_PTR(-ENOMEM);
+	if (copy_from_user(buffer, user_buf, user_len) != 0) {
+		vfree(buffer);
+		return ERR_PTR(-EFAULT);
+	}
+	/* got the string, now strip linefeed. */
+	if (buffer[user_len - 1] == '\n')
+		buffer[user_len - 1] = 0;
+	else
+		buffer[user_len] = 0;
+	return buffer;
+}
+
+static ssize_t dasd_stats_write(struct file *file,
+				const char __user *user_buf,
+				size_t user_len, loff_t *pos)
+{
+	char *buffer, *str;
+	int rc;
+	struct seq_file *m = (struct seq_file *)file->private_data;
+	struct dasd_profile *prof = m->private;
+
+	if (user_len > 65536)
+		user_len = 65536;
+	buffer = dasd_get_user_string(user_buf, user_len);
+	if (IS_ERR(buffer))
+		return PTR_ERR(buffer);
+
+	str = skip_spaces(buffer);
+	rc = user_len;
+	if (strncmp(str, "reset", 5) == 0) {
+		dasd_profile_reset(prof);
+	} else if (strncmp(str, "on", 2) == 0) {
+		rc = dasd_profile_on(prof);
+		if (rc)
+			goto out;
+		rc = user_len;
+		if (prof == &dasd_global_profile) {
+			dasd_profile_reset(prof);
+			dasd_global_profile_level = DASD_PROFILE_GLOBAL_ONLY;
+		}
+	} else if (strncmp(str, "off", 3) == 0) {
+		if (prof == &dasd_global_profile)
+			dasd_global_profile_level = DASD_PROFILE_OFF;
+		dasd_profile_off(prof);
+	} else
+		rc = -EINVAL;
+out:
+	vfree(buffer);
+	return rc;
+}
+
+static void dasd_stats_array(struct seq_file *m, unsigned int *array)
+{
+	int i;
+
+	for (i = 0; i < 32; i++)
+		seq_printf(m, "%u ", array[i]);
+	seq_putc(m, '\n');
+}
+
+static void dasd_stats_seq_print(struct seq_file *m,
+				 struct dasd_profile_info *data)
+{
+	seq_printf(m, "start_time %ptSp\n", &data->starttod);
+	seq_printf(m, "total_requests %u\n", data->dasd_io_reqs);
+	seq_printf(m, "total_sectors %u\n", data->dasd_io_sects);
+	seq_printf(m, "total_pav %u\n", data->dasd_io_alias);
+	seq_printf(m, "total_hpf %u\n", data->dasd_io_tpm);
+	seq_printf(m, "avg_total %lu\n", data->dasd_io_reqs ?
+		   data->dasd_sum_times / data->dasd_io_reqs : 0UL);
+	seq_printf(m, "avg_build_to_ssch %lu\n", data->dasd_io_reqs ?
+		   data->dasd_sum_time_str / data->dasd_io_reqs : 0UL);
+	seq_printf(m, "avg_ssch_to_irq %lu\n", data->dasd_io_reqs ?
+		   data->dasd_sum_time_irq / data->dasd_io_reqs : 0UL);
+	seq_printf(m, "avg_irq_to_end %lu\n", data->dasd_io_reqs ?
+		   data->dasd_sum_time_end / data->dasd_io_reqs : 0UL);
+	seq_puts(m, "histogram_sectors ");
+	dasd_stats_array(m, data->dasd_io_secs);
+	seq_puts(m, "histogram_io_times ");
+	dasd_stats_array(m, data->dasd_io_times);
+	seq_puts(m, "histogram_io_times_weighted ");
+	dasd_stats_array(m, data->dasd_io_timps);
+	seq_puts(m, "histogram_time_build_to_ssch ");
+	dasd_stats_array(m, data->dasd_io_time1);
+	seq_puts(m, "histogram_time_ssch_to_irq ");
+	dasd_stats_array(m, data->dasd_io_time2);
+	seq_puts(m, "histogram_time_ssch_to_irq_weighted ");
+	dasd_stats_array(m, data->dasd_io_time2ps);
+	seq_puts(m, "histogram_time_irq_to_end ");
+	dasd_stats_array(m, data->dasd_io_time3);
+	seq_puts(m, "histogram_ccw_queue_length ");
+	dasd_stats_array(m, data->dasd_io_nr_req);
+	seq_printf(m, "total_read_requests %u\n", data->dasd_read_reqs);
+	seq_printf(m, "total_read_sectors %u\n", data->dasd_read_sects);
+	seq_printf(m, "total_read_pav %u\n", data->dasd_read_alias);
+	seq_printf(m, "total_read_hpf %u\n", data->dasd_read_tpm);
+	seq_puts(m, "histogram_read_sectors ");
+	dasd_stats_array(m, data->dasd_read_secs);
+	seq_puts(m, "histogram_read_times ");
+	dasd_stats_array(m, data->dasd_read_times);
+	seq_puts(m, "histogram_read_time_build_to_ssch ");
+	dasd_stats_array(m, data->dasd_read_time1);
+	seq_puts(m, "histogram_read_time_ssch_to_irq ");
+	dasd_stats_array(m, data->dasd_read_time2);
+	seq_puts(m, "histogram_read_time_irq_to_end ");
+	dasd_stats_array(m, data->dasd_read_time3);
+	seq_puts(m, "histogram_read_ccw_queue_length ");
+	dasd_stats_array(m, data->dasd_read_nr_req);
+}
+
+static int dasd_stats_show(struct seq_file *m, void *v)
+{
+	struct dasd_profile *profile;
+	struct dasd_profile_info *data;
+
+	profile = m->private;
+	spin_lock_bh(&profile->lock);
+	data = profile->data;
+	if (!data) {
+		spin_unlock_bh(&profile->lock);
+		seq_puts(m, "disabled\n");
+		return 0;
+	}
+	dasd_stats_seq_print(m, data);
+	spin_unlock_bh(&profile->lock);
+	return 0;
+}
+
+static int dasd_stats_open(struct inode *inode, struct file *file)
+{
+	struct dasd_profile *profile = inode->i_private;
+	return single_open(file, dasd_stats_show, profile);
+}
+
+static const struct file_operations dasd_stats_raw_fops = {
+	.owner		= THIS_MODULE,
+	.open		= dasd_stats_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.write		= dasd_stats_write,
+};
+
+static void dasd_profile_init(struct dasd_profile *profile,
+			      struct dentry *base_dentry)
+{
+	profile->data = NULL;
+	profile->dentry = debugfs_create_file("statistics", 0600, base_dentry,
+					      profile, &dasd_stats_raw_fops);
+}
+
+static void dasd_profile_exit(struct dasd_profile *profile)
+{
+	dasd_profile_off(profile);
+	debugfs_remove(profile->dentry);
+	profile->dentry = NULL;
+}
+
+static void dasd_statistics_removeroot(void)
+{
+	dasd_global_profile_level = DASD_PROFILE_OFF;
+	dasd_profile_exit(&dasd_global_profile);
+	debugfs_remove(dasd_debugfs_global_entry);
+	debugfs_remove(dasd_debugfs_root_entry);
+}
+
+static void dasd_statistics_createroot(void)
+{
+	dasd_debugfs_root_entry = debugfs_create_dir("dasd", NULL);
+	dasd_debugfs_global_entry = debugfs_create_dir("global", dasd_debugfs_root_entry);
+	dasd_profile_init(&dasd_global_profile, dasd_debugfs_global_entry);
+}
+
+#else
+#define dasd_profile_start(block, cqr, req) do {} while (0)
+#define dasd_profile_end(block, cqr, req) do {} while (0)
+
+static void dasd_statistics_createroot(void)
+{
+	return;
+}
+
+static void dasd_statistics_removeroot(void)
+{
+	return;
+}
+
+static void dasd_profile_init(struct dasd_profile *profile,
+			      struct dentry *base_dentry)
+{
+	return;
+}
+
+static void dasd_profile_exit(struct dasd_profile *profile)
+{
+	return;
+}
+
+int dasd_profile_on(struct dasd_profile *profile)
+{
+	return 0;
+}
+
+#endif				/* CONFIG_DASD_PROFILE */
+
+static int dasd_hosts_show(struct seq_file *m, void *v)
+{
+	struct dasd_device *device;
+	int rc = -EOPNOTSUPP;
+
+	device = m->private;
+	dasd_get_device(device);
+
+	if (device->discipline->hosts_print)
+		rc = device->discipline->hosts_print(device, m);
+
+	dasd_put_device(device);
+	return rc;
+}
+
+DEFINE_SHOW_ATTRIBUTE(dasd_hosts);
+
+static void dasd_hosts_exit(struct dasd_device *device)
+{
+	debugfs_remove(device->hosts_dentry);
+	device->hosts_dentry = NULL;
+}
+
+static void dasd_hosts_init(struct dentry *base_dentry,
+			    struct dasd_device *device)
+{
+	device->hosts_dentry = debugfs_create_file("host_access_list", 0400, base_dentry,
+						   device, &dasd_hosts_fops);
+}
+
+struct dasd_ccw_req *dasd_smalloc_request(int magic, int cplength, int datasize,
+					  struct dasd_device *device,
+					  struct dasd_ccw_req *cqr)
+{
+	unsigned long flags;
+	char *data, *chunk;
+	int size = 0;
+
+	if (cplength > 0)
+		size += cplength * sizeof(struct ccw1);
+	if (datasize > 0)
+		size += datasize;
+	if (!cqr)
+		size += (sizeof(*cqr) + 7L) & -8L;
+
+	spin_lock_irqsave(&device->mem_lock, flags);
+	data = chunk = dasd_alloc_chunk(&device->ccw_chunks, size);
+	spin_unlock_irqrestore(&device->mem_lock, flags);
+	if (!chunk)
+		return ERR_PTR(-ENOMEM);
+	if (!cqr) {
+		cqr = (void *) data;
+		data += (sizeof(*cqr) + 7L) & -8L;
+	}
+	memset(cqr, 0, sizeof(*cqr));
+	cqr->mem_chunk = chunk;
+	if (cplength > 0) {
+		cqr->cpaddr = data;
+		data += cplength * sizeof(struct ccw1);
+		memset(cqr->cpaddr, 0, cplength * sizeof(struct ccw1));
+	}
+	if (datasize > 0) {
+		cqr->data = data;
+ 		memset(cqr->data, 0, datasize);
+	}
+	cqr->magic = magic;
+	set_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	dasd_get_device(device);
+	return cqr;
+}
+EXPORT_SYMBOL(dasd_smalloc_request);
+
+struct dasd_ccw_req *dasd_fmalloc_request(int magic, int cplength,
+					  int datasize,
+					  struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr;
+	unsigned long flags;
+	int size, cqr_size;
+	char *data;
+
+	cqr_size = (sizeof(*cqr) + 7L) & -8L;
+	size = cqr_size;
+	if (cplength > 0)
+		size += cplength * sizeof(struct ccw1);
+	if (datasize > 0)
+		size += datasize;
+
+	spin_lock_irqsave(&device->mem_lock, flags);
+	cqr = dasd_alloc_chunk(&device->ese_chunks, size);
+	spin_unlock_irqrestore(&device->mem_lock, flags);
+	if (!cqr)
+		return ERR_PTR(-ENOMEM);
+	memset(cqr, 0, sizeof(*cqr));
+	data = (char *)cqr + cqr_size;
+	cqr->cpaddr = NULL;
+	if (cplength > 0) {
+		cqr->cpaddr = data;
+		data += cplength * sizeof(struct ccw1);
+		memset(cqr->cpaddr, 0, cplength * sizeof(struct ccw1));
+	}
+	cqr->data = NULL;
+	if (datasize > 0) {
+		cqr->data = data;
+		memset(cqr->data, 0, datasize);
+	}
+
+	cqr->magic = magic;
+	set_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	dasd_get_device(device);
+
+	return cqr;
+}
+EXPORT_SYMBOL(dasd_fmalloc_request);
+
+void dasd_sfree_request(struct dasd_ccw_req *cqr, struct dasd_device *device)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&device->mem_lock, flags);
+	/*
+	 * Free the request block from the pool it came from: smalloc() sets
+	 * mem_chunk (ccw_chunks), fmalloc() leaves it NULL (ese_chunks). A
+	 * full-track request also frees its track-filler buffer.
+	 */
+	if (cqr->filldata)
+		dasd_free_chunk(&device->fill_chunks, cqr->filldata);
+	if (cqr->mem_chunk)
+		dasd_free_chunk(&device->ccw_chunks, cqr->mem_chunk);
+	else
+		dasd_free_chunk(&device->ese_chunks, cqr);
+
+	spin_unlock_irqrestore(&device->mem_lock, flags);
+	dasd_put_device(device);
+}
+EXPORT_SYMBOL(dasd_sfree_request);
+
+void dasd_ffree_request(struct dasd_ccw_req *cqr, struct dasd_device *device)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&device->mem_lock, flags);
+	if (cqr->filldata)
+		dasd_free_chunk(&device->fill_chunks, cqr->filldata);
+	dasd_free_chunk(&device->ese_chunks, cqr);
+	spin_unlock_irqrestore(&device->mem_lock, flags);
+	dasd_put_device(device);
+}
+EXPORT_SYMBOL(dasd_ffree_request);
+
+/*
+ * Check discipline magic in cqr.
+ */
+static inline int dasd_check_cqr(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device;
+
+	if (cqr == NULL)
+		return -EINVAL;
+	device = cqr->startdev;
+	if (strncmp((char *) &cqr->magic, device->discipline->ebcname, 4)) {
+		DBF_DEV_EVENT(DBF_WARNING, device,
+			    " dasd_ccw_req 0x%08x magic doesn't match"
+			    " discipline 0x%08x",
+			    cqr->magic,
+			    *(unsigned int *) device->discipline->name);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/*
+ * Terminate the current i/o and set the request to clear_pending.
+ * Timer keeps device runnig.
+ * ccw_device_clear can fail if the i/o subsystem
+ * is in a bad mood.
+ */
+int dasd_term_IO(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device;
+	int retries, rc;
+
+	/* Check the cqr */
+	rc = dasd_check_cqr(cqr);
+	if (rc)
+		return rc;
+	retries = 0;
+	device = (struct dasd_device *) cqr->startdev;
+	while ((retries < 5) && (cqr->status == DASD_CQR_IN_IO)) {
+		rc = ccw_device_clear(device->cdev, (long) cqr);
+		switch (rc) {
+		case 0:	/* termination successful */
+			cqr->status = DASD_CQR_CLEAR_PENDING;
+			cqr->stopclk = get_tod_clock();
+			cqr->starttime = 0;
+			DBF_DEV_EVENT(DBF_DEBUG, device,
+				      "terminate cqr %p successful",
+				      cqr);
+			break;
+		case -ENODEV:
+			DBF_DEV_EVENT(DBF_ERR, device, "%s",
+				      "device gone, retry");
+			break;
+		case -EINVAL:
+			/*
+			 * device not valid so no I/O could be running
+			 * handle CQR as termination successful
+			 */
+			cqr->status = DASD_CQR_CLEARED;
+			cqr->stopclk = get_tod_clock();
+			cqr->starttime = 0;
+			/* no retries for invalid devices */
+			cqr->retries = -1;
+			DBF_DEV_EVENT(DBF_ERR, device, "%s",
+				      "EINVAL, handle as terminated");
+			/* fake rc to success */
+			rc = 0;
+			break;
+		default:
+			dev_err(&device->cdev->dev,
+				"Unexpected error during request termination %d\n", rc);
+			BUG();
+			break;
+		}
+		retries++;
+	}
+	dasd_schedule_device_bh(device);
+	return rc;
+}
+EXPORT_SYMBOL(dasd_term_IO);
+
+/*
+ * Start the i/o. This start_IO can fail if the channel is really busy.
+ * In that case set up a timer to start the request later.
+ */
+int dasd_start_IO(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device;
+	int rc;
+
+	/* Check the cqr */
+	rc = dasd_check_cqr(cqr);
+	if (rc) {
+		cqr->intrc = rc;
+		return rc;
+	}
+	device = (struct dasd_device *) cqr->startdev;
+	if (((cqr->block &&
+	      test_bit(DASD_FLAG_LOCK_STOLEN, &cqr->block->base->flags)) ||
+	     test_bit(DASD_FLAG_LOCK_STOLEN, &device->flags)) &&
+	    !test_bit(DASD_CQR_ALLOW_SLOCK, &cqr->flags)) {
+		DBF_DEV_EVENT(DBF_DEBUG, device, "start_IO: return request %p "
+			      "because of stolen lock", cqr);
+		cqr->status = DASD_CQR_ERROR;
+		cqr->intrc = -EPERM;
+		return -EPERM;
+	}
+	if (cqr->retries < 0) {
+		dev_err(&device->cdev->dev,
+			"Start I/O ran out of retries\n");
+		cqr->status = DASD_CQR_ERROR;
+		return -EIO;
+	}
+	cqr->startclk = get_tod_clock();
+	cqr->starttime = jiffies;
+	cqr->retries--;
+	if (!test_bit(DASD_CQR_VERIFY_PATH, &cqr->flags)) {
+		cqr->lpm &= dasd_path_get_opm(device);
+		if (!cqr->lpm)
+			cqr->lpm = dasd_path_get_opm(device);
+	}
+	if (cqr->cpmode == 1) {
+		rc = ccw_device_tm_start(device->cdev, cqr->cpaddr,
+					 (long) cqr, cqr->lpm);
+	} else {
+		rc = ccw_device_start(device->cdev, cqr->cpaddr,
+				      (long) cqr, cqr->lpm, 0);
+	}
+	switch (rc) {
+	case 0:
+		cqr->status = DASD_CQR_IN_IO;
+		break;
+	case -EBUSY:
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "start_IO: device busy, retry later");
+		break;
+	case -EACCES:
+		/* -EACCES indicates that the request used only a subset of the
+		 * available paths and all these paths are gone. If the lpm of
+		 * this request was only a subset of the opm (e.g. the ppm) then
+		 * we just do a retry with all available paths.
+		 * If we already use the full opm, something is amiss, and we
+		 * need a full path verification.
+		 */
+		if (test_bit(DASD_CQR_VERIFY_PATH, &cqr->flags)) {
+			DBF_DEV_EVENT(DBF_WARNING, device,
+				      "start_IO: selected paths gone (%x)",
+				      cqr->lpm);
+		} else if (cqr->lpm != dasd_path_get_opm(device)) {
+			cqr->lpm = dasd_path_get_opm(device);
+			DBF_DEV_EVENT(DBF_DEBUG, device, "%s",
+				      "start_IO: selected paths gone,"
+				      " retry on all paths");
+		} else {
+			DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+				      "start_IO: all paths in opm gone,"
+				      " do path verification");
+			dasd_generic_last_path_gone(device);
+			dasd_path_no_path(device);
+			dasd_path_set_tbvpm(device,
+					  ccw_device_get_path_mask(
+						  device->cdev));
+		}
+		break;
+	case -ENODEV:
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "start_IO: -ENODEV device gone, retry");
+		/* this is equivalent to CC=3 for SSCH report this to EER */
+		dasd_handle_autoquiesce(device, cqr, DASD_EER_STARTIO);
+		break;
+	case -EIO:
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "start_IO: -EIO device gone, retry");
+		break;
+	case -EINVAL:
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "start_IO: -EINVAL device currently "
+			      "not accessible");
+		break;
+	default:
+		dev_err(&device->cdev->dev,
+			"Unexpected error during request start %d", rc);
+		BUG();
+		break;
+	}
+	cqr->intrc = rc;
+	return rc;
+}
+EXPORT_SYMBOL(dasd_start_IO);
+
+/*
+ * Timeout function for dasd devices. This is used for different purposes
+ *  1) missing interrupt handler for normal operation
+ *  2) delayed start of request where start_IO failed with -EBUSY
+ *  3) timeout for missing state change interrupts
+ * The head of the ccw queue will have status DASD_CQR_IN_IO for 1),
+ * DASD_CQR_QUEUED for 2) and 3).
+ */
+static void dasd_device_timeout(struct timer_list *t)
+{
+	unsigned long flags;
+	struct dasd_device *device;
+
+	device = timer_container_of(device, t, timer);
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	/* re-activate request queue */
+	dasd_device_remove_stop_bits(device, DASD_STOPPED_PENDING);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	dasd_schedule_device_bh(device);
+}
+
+/*
+ * Setup timeout for a device in jiffies.
+ */
+void dasd_device_set_timer(struct dasd_device *device, int expires)
+{
+	if (expires == 0)
+		timer_delete(&device->timer);
+	else
+		mod_timer(&device->timer, jiffies + expires);
+}
+EXPORT_SYMBOL(dasd_device_set_timer);
+
+/*
+ * Clear timeout for a device.
+ */
+void dasd_device_clear_timer(struct dasd_device *device)
+{
+	timer_delete(&device->timer);
+}
+EXPORT_SYMBOL(dasd_device_clear_timer);
+
+static void dasd_handle_killed_request(struct ccw_device *cdev,
+				       unsigned long intparm)
+{
+	struct dasd_ccw_req *cqr;
+	struct dasd_device *device;
+
+	if (!intparm)
+		return;
+	cqr = (struct dasd_ccw_req *) intparm;
+	if (cqr->status != DASD_CQR_IN_IO) {
+		DBF_EVENT_DEVID(DBF_DEBUG, cdev,
+				"invalid status in handle_killed_request: "
+				"%02x", cqr->status);
+		return;
+	}
+
+	device = dasd_device_from_cdev_locked(cdev);
+	if (IS_ERR(device)) {
+		DBF_EVENT_DEVID(DBF_DEBUG, cdev, "%s",
+				"unable to get device from cdev");
+		return;
+	}
+
+	if (!cqr->startdev ||
+	    device != cqr->startdev ||
+	    strncmp(cqr->startdev->discipline->ebcname,
+		    (char *) &cqr->magic, 4)) {
+		DBF_EVENT_DEVID(DBF_DEBUG, cdev, "%s",
+				"invalid device in request");
+		dasd_put_device(device);
+		return;
+	}
+
+	/* Schedule request to be retried. */
+	cqr->status = DASD_CQR_QUEUED;
+
+	dasd_device_clear_timer(device);
+	dasd_schedule_device_bh(device);
+	dasd_put_device(device);
+}
+
+void dasd_generic_handle_state_change(struct dasd_device *device)
+{
+	/* First of all start sense subsystem status request. */
+	dasd_eer_snss(device);
+
+	dasd_device_remove_stop_bits(device, DASD_STOPPED_PENDING);
+	dasd_schedule_device_bh(device);
+	if (device->block) {
+		dasd_schedule_block_bh(device->block);
+		if (device->block->gdp)
+			blk_mq_run_hw_queues(device->block->gdp->queue, true);
+	}
+}
+EXPORT_SYMBOL_GPL(dasd_generic_handle_state_change);
+
+static int dasd_check_hpf_error(struct irb *irb)
+{
+	return (scsw_tm_is_valid_schxs(&irb->scsw) &&
+	    (irb->scsw.tm.sesq == SCSW_SESQ_DEV_NOFCX ||
+	     irb->scsw.tm.sesq == SCSW_SESQ_PATH_NOFCX));
+}
+
+static int dasd_ese_needs_format(struct dasd_block *block, struct irb *irb)
+{
+	struct dasd_device *device = NULL;
+	u8 *sense = NULL;
+
+	if (!block)
+		return 0;
+	device = block->base;
+	if (!device || !device->discipline->is_ese)
+		return 0;
+	if (!device->discipline->is_ese(device))
+		return 0;
+
+	sense = dasd_get_sense(irb);
+	if (!sense)
+		return 0;
+
+	if (sense[1] & SNS1_NO_REC_FOUND)
+		return 1;
+
+	if ((sense[1] & SNS1_INV_TRACK_FORMAT) &&
+	    scsw_is_tm(&irb->scsw) &&
+	    !(sense[2] & SNS2_ENV_DATA_PRESENT))
+		return 1;
+
+	return 0;
+}
+
+static int dasd_ese_oos_cond(u8 *sense)
+{
+	return sense[0] & SNS0_EQUIPMENT_CHECK &&
+		sense[1] & SNS1_PERM_ERR &&
+		sense[1] & SNS1_WRITE_INHIBITED &&
+		sense[25] == 0x01;
+}
+
+/*
+ * Interrupt handler for "normal" ssch-io based dasd devices.
+ */
+void dasd_int_handler(struct ccw_device *cdev, unsigned long intparm,
+		      struct irb *irb)
+{
+	struct dasd_ccw_req *cqr, *next;
+	struct dasd_device *device;
+	unsigned long now;
+	int nrf_suppressed = 0;
+	int it_suppressed = 0;
+	struct request *req;
+	u8 *sense = NULL;
+	int expires;
+
+	cqr = (struct dasd_ccw_req *) intparm;
+	if (IS_ERR(irb)) {
+		switch (PTR_ERR(irb)) {
+		case -EIO:
+			if (cqr && cqr->status == DASD_CQR_CLEAR_PENDING) {
+				device = cqr->startdev;
+				cqr->status = DASD_CQR_CLEARED;
+				dasd_device_clear_timer(device);
+				wake_up(&dasd_flush_wq);
+				dasd_schedule_device_bh(device);
+				return;
+			}
+			break;
+		case -ETIMEDOUT:
+			DBF_EVENT_DEVID(DBF_WARNING, cdev, "%s: "
+					"request timed out\n", __func__);
+			break;
+		default:
+			DBF_EVENT_DEVID(DBF_WARNING, cdev, "%s: "
+					"unknown error %ld\n", __func__,
+					PTR_ERR(irb));
+		}
+		dasd_handle_killed_request(cdev, intparm);
+		return;
+	}
+
+	now = get_tod_clock();
+	/* check for conditions that should be handled immediately */
+	if (!cqr ||
+	    !(scsw_dstat(&irb->scsw) == (DEV_STAT_CHN_END | DEV_STAT_DEV_END) &&
+	      scsw_cstat(&irb->scsw) == 0)) {
+		if (cqr)
+			memcpy(&cqr->irb, irb, sizeof(*irb));
+		device = dasd_device_from_cdev_locked(cdev);
+		if (IS_ERR(device))
+			return;
+		/* ignore unsolicited interrupts for DIAG discipline */
+		if (device->discipline == dasd_diag_discipline_pointer) {
+			dasd_put_device(device);
+			return;
+		}
+
+		/*
+		 * In some cases 'File Protected' or 'No Record Found' errors
+		 * might be expected and debug log messages for the
+		 * corresponding interrupts shouldn't be written then.
+		 * Check if either of the according suppress bits is set.
+		 */
+		sense = dasd_get_sense(irb);
+		if (sense) {
+			it_suppressed =	(sense[1] & SNS1_INV_TRACK_FORMAT) &&
+				!(sense[2] & SNS2_ENV_DATA_PRESENT) &&
+				test_bit(DASD_CQR_SUPPRESS_IT, &cqr->flags);
+			nrf_suppressed = (sense[1] & SNS1_NO_REC_FOUND) &&
+				test_bit(DASD_CQR_SUPPRESS_NRF, &cqr->flags);
+
+			/*
+			 * Extent pool probably out-of-space.
+			 * Stop device and check exhaust level.
+			 */
+			if (dasd_ese_oos_cond(sense)) {
+				dasd_generic_space_exhaust(device, cqr);
+				device->discipline->ext_pool_exhaust(device, cqr);
+				dasd_put_device(device);
+				return;
+			}
+		}
+		if (!(it_suppressed || nrf_suppressed))
+			device->discipline->dump_sense_dbf(device, irb, "int");
+
+		if (device->features & DASD_FEATURE_ERPLOG)
+			device->discipline->dump_sense(device, cqr, irb);
+		device->discipline->check_for_device_change(device, cqr, irb);
+		dasd_put_device(device);
+	}
+
+	/* check for attention message */
+	if (scsw_dstat(&irb->scsw) & DEV_STAT_ATTENTION) {
+		device = dasd_device_from_cdev_locked(cdev);
+		if (!IS_ERR(device)) {
+			device->discipline->check_attention(device,
+							    irb->esw.esw1.lpum);
+			dasd_put_device(device);
+		}
+	}
+
+	if (!cqr)
+		return;
+
+	device = (struct dasd_device *) cqr->startdev;
+	if (!device ||
+	    strncmp(device->discipline->ebcname, (char *) &cqr->magic, 4)) {
+		DBF_EVENT_DEVID(DBF_DEBUG, cdev, "%s",
+				"invalid device in request");
+		return;
+	}
+
+	if (dasd_ese_needs_format(cqr->block, irb)) {
+		req = dasd_get_callback_data(cqr);
+		if (!req) {
+			cqr->status = DASD_CQR_ERROR;
+			return;
+		}
+		if (rq_data_dir(req) == READ) {
+			if (device->discipline->ese_read(cqr, irb))
+				cqr->status = DASD_CQR_ERROR;
+			else
+				cqr->status = DASD_CQR_SUCCESS;
+			cqr->stopclk = now;
+			dasd_device_clear_timer(device);
+			dasd_schedule_device_bh(device);
+			return;
+		}
+		if (cqr->filldata) {
+			/*
+			 * A WRITE_FULL_TRACK cqr carries the complete
+			 * track image; INV_TRACK_FORMAT here means the
+			 * generated image or the media itself is bad, not
+			 * that the track still needs formatting - retrying
+			 * via ese_format() would just resubmit the same
+			 * write. Let it fail instead.
+			 */
+			cqr->status = DASD_CQR_ERROR;
+			cqr->stopclk = now;
+			dasd_device_clear_timer(device);
+			dasd_schedule_device_bh(device);
+			return;
+		}
+		device->discipline->ese_format(device, cqr, irb);
+		return;
+	}
+
+	/* Check for clear pending */
+	if (cqr->status == DASD_CQR_CLEAR_PENDING &&
+	    scsw_fctl(&irb->scsw) & SCSW_FCTL_CLEAR_FUNC) {
+		cqr->status = DASD_CQR_CLEARED;
+		dasd_device_clear_timer(device);
+		wake_up(&dasd_flush_wq);
+		dasd_schedule_device_bh(device);
+		return;
+	}
+
+	/* check status - the request might have been killed by dyn detach */
+	if (cqr->status != DASD_CQR_IN_IO) {
+		DBF_DEV_EVENT(DBF_DEBUG, device, "invalid status: bus_id %s, "
+			      "status %02x", dev_name(&cdev->dev), cqr->status);
+		return;
+	}
+
+	next = NULL;
+	expires = 0;
+	if (scsw_dstat(&irb->scsw) == (DEV_STAT_CHN_END | DEV_STAT_DEV_END) &&
+	    scsw_cstat(&irb->scsw) == 0) {
+		/* request was completed successfully */
+		cqr->status = DASD_CQR_SUCCESS;
+		cqr->stopclk = now;
+		/* Start first request on queue if possible -> fast_io. */
+		if (cqr->devlist.next != &device->ccw_queue) {
+			next = list_entry(cqr->devlist.next,
+					  struct dasd_ccw_req, devlist);
+		}
+	} else {  /* error */
+		/* check for HPF error
+		 * call discipline function to requeue all requests
+		 * and disable HPF accordingly
+		 */
+		if (cqr->cpmode && dasd_check_hpf_error(irb) &&
+		    device->discipline->handle_hpf_error)
+			device->discipline->handle_hpf_error(device, irb);
+		/*
+		 * If we don't want complex ERP for this request, then just
+		 * reset this and retry it in the fastpath
+		 */
+		if (!test_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags) &&
+		    cqr->retries > 0) {
+			if (cqr->lpm == dasd_path_get_opm(device))
+				DBF_DEV_EVENT(DBF_DEBUG, device,
+					      "default ERP in fastpath "
+					      "(%i retries left)",
+					      cqr->retries);
+			if (!test_bit(DASD_CQR_VERIFY_PATH, &cqr->flags))
+				cqr->lpm = dasd_path_get_opm(device);
+			cqr->status = DASD_CQR_QUEUED;
+			next = cqr;
+		} else
+			cqr->status = DASD_CQR_ERROR;
+	}
+	if (next && (next->status == DASD_CQR_QUEUED) &&
+	    (!device->stopped)) {
+		if (device->discipline->start_IO(next) == 0)
+			expires = next->expires;
+	}
+	if (expires != 0)
+		dasd_device_set_timer(device, expires);
+	else
+		dasd_device_clear_timer(device);
+	dasd_schedule_device_bh(device);
+}
+EXPORT_SYMBOL(dasd_int_handler);
+
+enum uc_todo dasd_generic_uc_handler(struct ccw_device *cdev, struct irb *irb)
+{
+	struct dasd_device *device;
+
+	device = dasd_device_from_cdev_locked(cdev);
+
+	if (IS_ERR(device))
+		goto out;
+	if (test_bit(DASD_FLAG_OFFLINE, &device->flags) ||
+	   device->state != device->target ||
+	   !device->discipline->check_for_device_change){
+		dasd_put_device(device);
+		goto out;
+	}
+	if (device->discipline->dump_sense_dbf)
+		device->discipline->dump_sense_dbf(device, irb, "uc");
+	device->discipline->check_for_device_change(device, NULL, irb);
+	dasd_put_device(device);
+out:
+	return UC_TODO_RETRY;
+}
+EXPORT_SYMBOL_GPL(dasd_generic_uc_handler);
+
+/*
+ * If we have an error on a dasd_block layer request then we cancel
+ * and return all further requests from the same dasd_block as well.
+ */
+static void __dasd_device_recovery(struct dasd_device *device,
+				   struct dasd_ccw_req *ref_cqr)
+{
+	struct list_head *l, *n;
+	struct dasd_ccw_req *cqr;
+
+	/*
+	 * only requeue request that came from the dasd_block layer
+	 */
+	if (!ref_cqr->block)
+		return;
+
+	list_for_each_safe(l, n, &device->ccw_queue) {
+		cqr = list_entry(l, struct dasd_ccw_req, devlist);
+		if (cqr->status == DASD_CQR_QUEUED &&
+		    ref_cqr->block == cqr->block) {
+			cqr->status = DASD_CQR_CLEARED;
+		}
+	}
+};
+
+/*
+ * Remove those ccw requests from the queue that need to be returned
+ * to the upper layer.
+ */
+static void __dasd_device_process_ccw_queue(struct dasd_device *device,
+					    struct list_head *final_queue)
+{
+	struct list_head *l, *n;
+	struct dasd_ccw_req *cqr;
+
+	/* Process request with final status. */
+	list_for_each_safe(l, n, &device->ccw_queue) {
+		cqr = list_entry(l, struct dasd_ccw_req, devlist);
+
+		/* Skip any non-final request. */
+		if (cqr->status == DASD_CQR_QUEUED ||
+		    cqr->status == DASD_CQR_IN_IO ||
+		    cqr->status == DASD_CQR_CLEAR_PENDING)
+			continue;
+		if (cqr->status == DASD_CQR_ERROR) {
+			__dasd_device_recovery(device, cqr);
+		}
+		/* Rechain finished requests to final queue */
+		list_move_tail(&cqr->devlist, final_queue);
+	}
+}
+
+static void __dasd_process_cqr(struct dasd_device *device,
+			       struct dasd_ccw_req *cqr)
+{
+	switch (cqr->status) {
+	case DASD_CQR_SUCCESS:
+		cqr->status = DASD_CQR_DONE;
+		break;
+	case DASD_CQR_ERROR:
+		cqr->status = DASD_CQR_NEED_ERP;
+		break;
+	case DASD_CQR_CLEARED:
+		cqr->status = DASD_CQR_TERMINATED;
+		break;
+	case DASD_CQR_ABORT:
+		cqr->status = DASD_CQR_ABORTED;
+		/*
+		 * ABORT is only set on the block-layer origin write that a
+		 * full-track format replaces. Clear the callback so the request
+		 * is not completed here - the replacement completes it. Internal
+		 * requests never take this path, so no sleep_on waiter is left
+		 * without its wakeup.
+		 */
+		cqr->callback = NULL;
+		break;
+	default:
+		dev_err(&device->cdev->dev,
+			"Unexpected CQR status %02x", cqr->status);
+		BUG();
+	}
+	if (cqr->callback)
+		cqr->callback(cqr, cqr->callback_data);
+}
+
+/*
+ * the cqrs from the final queue are returned to the upper layer
+ * by setting a dasd_block state and calling the callback function
+ */
+static void __dasd_device_process_final_queue(struct dasd_device *device,
+					      struct list_head *final_queue)
+{
+	struct list_head *l, *n;
+	struct dasd_ccw_req *cqr;
+	struct dasd_block *block;
+
+	list_for_each_safe(l, n, final_queue) {
+		cqr = list_entry(l, struct dasd_ccw_req, devlist);
+		list_del_init(&cqr->devlist);
+		block = cqr->block;
+		if (!block) {
+			__dasd_process_cqr(device, cqr);
+		} else {
+			spin_lock_bh(&block->queue_lock);
+			__dasd_process_cqr(device, cqr);
+			spin_unlock_bh(&block->queue_lock);
+		}
+	}
+}
+
+/*
+ * check if device should be autoquiesced due to too many timeouts
+ */
+static void __dasd_device_check_autoquiesce_timeout(struct dasd_device *device,
+						    struct dasd_ccw_req *cqr)
+{
+	if ((device->default_retries - cqr->retries) >= device->aq_timeouts)
+		dasd_handle_autoquiesce(device, cqr, DASD_EER_TIMEOUTS);
+}
+
+/*
+ * Take a look at the first request on the ccw queue and check
+ * if it reached its expire time. If so, terminate the IO.
+ */
+static void __dasd_device_check_expire(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr;
+
+	if (list_empty(&device->ccw_queue))
+		return;
+	cqr = list_entry(device->ccw_queue.next, struct dasd_ccw_req, devlist);
+	if ((cqr->status == DASD_CQR_IN_IO && cqr->expires != 0) &&
+	    (time_after_eq(jiffies, cqr->expires + cqr->starttime))) {
+		if (test_bit(DASD_FLAG_SAFE_OFFLINE_RUNNING, &device->flags)) {
+			/*
+			 * IO in safe offline processing should not
+			 * run out of retries
+			 */
+			cqr->retries++;
+		}
+		if (device->discipline->term_IO(cqr) != 0) {
+			/* Hmpf, try again in 5 sec */
+			dev_err(&device->cdev->dev,
+				"CQR timed out (%lus) but cannot be ended, retrying in 5s\n",
+				(cqr->expires / HZ));
+			cqr->expires += 5*HZ;
+			dasd_device_set_timer(device, 5*HZ);
+		} else {
+			dev_err(&device->cdev->dev,
+				"CQR timed out (%lus), %i retries remaining\n",
+				(cqr->expires / HZ), cqr->retries);
+		}
+		__dasd_device_check_autoquiesce_timeout(device, cqr);
+	}
+}
+
+/*
+ * return 1 when device is not eligible for IO
+ */
+static int __dasd_device_is_unusable(struct dasd_device *device,
+				     struct dasd_ccw_req *cqr)
+{
+	int mask = ~(DASD_STOPPED_DC_WAIT | DASD_STOPPED_NOSPC);
+
+	if (test_bit(DASD_FLAG_OFFLINE, &device->flags) &&
+	    !test_bit(DASD_FLAG_SAFE_OFFLINE_RUNNING, &device->flags)) {
+		/*
+		 * dasd is being set offline
+		 * but it is no safe offline where we have to allow I/O
+		 */
+		return 1;
+	}
+	if (device->stopped) {
+		if (device->stopped & mask) {
+			/* stopped and CQR will not change that. */
+			return 1;
+		}
+		if (!test_bit(DASD_CQR_VERIFY_PATH, &cqr->flags)) {
+			/* CQR is not able to change device to
+			 * operational. */
+			return 1;
+		}
+		/* CQR required to get device operational. */
+	}
+	return 0;
+}
+
+/*
+ * Take a look at the first request on the ccw queue and check
+ * if it needs to be started.
+ */
+static void __dasd_device_start_head(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr;
+	int rc;
+
+	if (list_empty(&device->ccw_queue))
+		return;
+	cqr = list_entry(device->ccw_queue.next, struct dasd_ccw_req, devlist);
+	if (cqr->status != DASD_CQR_QUEUED)
+		return;
+	/* if device is not usable return request to upper layer */
+	if (__dasd_device_is_unusable(device, cqr)) {
+		cqr->intrc = -EAGAIN;
+		cqr->status = DASD_CQR_CLEARED;
+		dasd_schedule_device_bh(device);
+		return;
+	}
+
+	rc = device->discipline->start_IO(cqr);
+	if (rc == 0)
+		dasd_device_set_timer(device, cqr->expires);
+	else if (rc == -EACCES) {
+		dasd_schedule_device_bh(device);
+	} else
+		/* Hmpf, try again in 1/2 sec */
+		dasd_device_set_timer(device, 50);
+}
+
+static void __dasd_device_check_path_events(struct dasd_device *device)
+{
+	__u8 tbvpm, fcsecpm;
+	int rc;
+
+	tbvpm = dasd_path_get_tbvpm(device);
+	fcsecpm = dasd_path_get_fcsecpm(device);
+
+	if (!tbvpm && !fcsecpm)
+		return;
+
+	if (device->stopped & ~(DASD_STOPPED_DC_WAIT))
+		return;
+
+	dasd_path_clear_all_verify(device);
+	dasd_path_clear_all_fcsec(device);
+
+	rc = device->discipline->pe_handler(device, tbvpm, fcsecpm);
+	if (rc) {
+		dasd_path_add_tbvpm(device, tbvpm);
+		dasd_path_add_fcsecpm(device, fcsecpm);
+		dasd_device_set_timer(device, 50);
+	}
+};
+
+/*
+ * Go through all request on the dasd_device request queue,
+ * terminate them on the cdev if necessary, and return them to the
+ * submitting layer via callback.
+ * Note:
+ * Make sure that all 'submitting layers' still exist when
+ * this function is called!. In other words, when 'device' is a base
+ * device then all block layer requests must have been removed before
+ * via dasd_flush_block_queue.
+ */
+int dasd_flush_device_queue(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr, *n;
+	int rc;
+	struct list_head flush_queue;
+
+	INIT_LIST_HEAD(&flush_queue);
+	spin_lock_irq(get_ccwdev_lock(device->cdev));
+	rc = 0;
+	list_for_each_entry_safe(cqr, n, &device->ccw_queue, devlist) {
+		/* Check status and move request to flush_queue */
+		switch (cqr->status) {
+		case DASD_CQR_IN_IO:
+			rc = device->discipline->term_IO(cqr);
+			if (rc) {
+				/* unable to terminate request */
+				dev_err(&device->cdev->dev,
+					"Flushing the DASD request queue failed\n");
+				/* stop flush processing */
+				goto finished;
+			}
+			break;
+		case DASD_CQR_QUEUED:
+			cqr->stopclk = get_tod_clock();
+			cqr->status = DASD_CQR_CLEARED;
+			break;
+		default: /* no need to modify the others */
+			break;
+		}
+		list_move_tail(&cqr->devlist, &flush_queue);
+	}
+finished:
+	spin_unlock_irq(get_ccwdev_lock(device->cdev));
+	/*
+	 * After this point all requests must be in state CLEAR_PENDING,
+	 * CLEARED, SUCCESS or ERROR. Now wait for CLEAR_PENDING to become
+	 * one of the others.
+	 */
+	list_for_each_entry_safe(cqr, n, &flush_queue, devlist)
+		wait_event(dasd_flush_wq,
+			   (cqr->status != DASD_CQR_CLEAR_PENDING));
+	/*
+	 * Now set each request back to TERMINATED, DONE or NEED_ERP
+	 * and call the callback function of flushed requests
+	 */
+	__dasd_device_process_final_queue(device, &flush_queue);
+	return rc;
+}
+EXPORT_SYMBOL_GPL(dasd_flush_device_queue);
+
+/*
+ * Acquire the device lock and process queues for the device.
+ */
+static void dasd_device_tasklet(unsigned long data)
+{
+	struct dasd_device *device = (struct dasd_device *) data;
+	struct list_head final_queue;
+
+	atomic_set (&device->tasklet_scheduled, 0);
+	INIT_LIST_HEAD(&final_queue);
+	spin_lock_irq(get_ccwdev_lock(device->cdev));
+	/* Check expire time of first request on the ccw queue. */
+	__dasd_device_check_expire(device);
+	/* find final requests on ccw queue */
+	__dasd_device_process_ccw_queue(device, &final_queue);
+	__dasd_device_check_path_events(device);
+	spin_unlock_irq(get_ccwdev_lock(device->cdev));
+	/* Now call the callback function of requests with final status */
+	__dasd_device_process_final_queue(device, &final_queue);
+	spin_lock_irq(get_ccwdev_lock(device->cdev));
+	/* Now check if the head of the ccw queue needs to be started. */
+	__dasd_device_start_head(device);
+	spin_unlock_irq(get_ccwdev_lock(device->cdev));
+	if (waitqueue_active(&shutdown_waitq))
+		wake_up(&shutdown_waitq);
+	dasd_put_device(device);
+}
+
+/*
+ * Schedules a call to dasd_tasklet over the device tasklet.
+ */
+void dasd_schedule_device_bh(struct dasd_device *device)
+{
+	/* Protect against rescheduling. */
+	if (atomic_cmpxchg (&device->tasklet_scheduled, 0, 1) != 0)
+		return;
+	dasd_get_device(device);
+	tasklet_hi_schedule(&device->tasklet);
+}
+EXPORT_SYMBOL(dasd_schedule_device_bh);
+
+void dasd_device_set_stop_bits(struct dasd_device *device, int bits)
+{
+	device->stopped |= bits;
+}
+EXPORT_SYMBOL_GPL(dasd_device_set_stop_bits);
+
+void dasd_device_remove_stop_bits(struct dasd_device *device, int bits)
+{
+	device->stopped &= ~bits;
+	if (!device->stopped)
+		wake_up(&generic_waitq);
+}
+EXPORT_SYMBOL_GPL(dasd_device_remove_stop_bits);
+
+/*
+ * Queue a request to the head of the device ccw_queue.
+ * Start the I/O if possible.
+ */
+void dasd_add_request_head(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device;
+	unsigned long flags;
+
+	device = cqr->startdev;
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	cqr->status = DASD_CQR_QUEUED;
+	list_add(&cqr->devlist, &device->ccw_queue);
+	/* let the bh start the request to keep them in order */
+	dasd_schedule_device_bh(device);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+}
+EXPORT_SYMBOL(dasd_add_request_head);
+
+/*
+ * Queue a request to the tail of the device ccw_queue.
+ * Start the I/O if possible.
+ */
+void dasd_add_request_tail(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device;
+	unsigned long flags;
+
+	device = cqr->startdev;
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	cqr->status = DASD_CQR_QUEUED;
+	list_add_tail(&cqr->devlist, &device->ccw_queue);
+	/* let the bh start the request to keep them in order */
+	dasd_schedule_device_bh(device);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+}
+EXPORT_SYMBOL(dasd_add_request_tail);
+
+/*
+ * Wakeup helper for the 'sleep_on' functions.
+ */
+void dasd_wakeup_cb(struct dasd_ccw_req *cqr, void *data)
+{
+	spin_lock_irq(get_ccwdev_lock(cqr->startdev->cdev));
+	cqr->endclk = get_tod_clock();
+	cqr->callback_data = DASD_SLEEPON_END_TAG;
+	spin_unlock_irq(get_ccwdev_lock(cqr->startdev->cdev));
+	wake_up(&generic_waitq);
+}
+EXPORT_SYMBOL_GPL(dasd_wakeup_cb);
+
+static inline int _wait_for_wakeup(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device;
+	int rc;
+
+	device = cqr->startdev;
+	spin_lock_irq(get_ccwdev_lock(device->cdev));
+	rc = (cqr->callback_data == DASD_SLEEPON_END_TAG);
+	spin_unlock_irq(get_ccwdev_lock(device->cdev));
+	return rc;
+}
+
+/*
+ * checks if error recovery is necessary, returns 1 if yes, 0 otherwise.
+ */
+static int __dasd_sleep_on_erp(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device;
+	dasd_erp_fn_t erp_fn;
+
+	if (cqr->status == DASD_CQR_FILLED)
+		return 0;
+	device = cqr->startdev;
+	if (test_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags)) {
+		if (cqr->status == DASD_CQR_TERMINATED) {
+			device->discipline->handle_terminated_request(cqr);
+			return 1;
+		}
+		if (cqr->status == DASD_CQR_NEED_ERP) {
+			erp_fn = device->discipline->erp_action(cqr);
+			erp_fn(cqr);
+			return 1;
+		}
+		if (cqr->status == DASD_CQR_FAILED)
+			dasd_log_sense(cqr, &cqr->irb);
+		if (cqr->refers) {
+			__dasd_process_erp(device, cqr);
+			return 1;
+		}
+	}
+	return 0;
+}
+
+static int __dasd_sleep_on_loop_condition(struct dasd_ccw_req *cqr)
+{
+	if (test_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags)) {
+		if (cqr->refers) /* erp is not done yet */
+			return 1;
+		return ((cqr->status != DASD_CQR_DONE) &&
+			(cqr->status != DASD_CQR_FAILED));
+	} else
+		return (cqr->status == DASD_CQR_FILLED);
+}
+
+static int _dasd_sleep_on(struct dasd_ccw_req *maincqr, int interruptible)
+{
+	struct dasd_device *device;
+	int rc;
+	struct list_head ccw_queue;
+	struct dasd_ccw_req *cqr;
+
+	INIT_LIST_HEAD(&ccw_queue);
+	maincqr->status = DASD_CQR_FILLED;
+	device = maincqr->startdev;
+	list_add(&maincqr->blocklist, &ccw_queue);
+	for (cqr = maincqr;  __dasd_sleep_on_loop_condition(cqr);
+	     cqr = list_first_entry(&ccw_queue,
+				    struct dasd_ccw_req, blocklist)) {
+
+		if (__dasd_sleep_on_erp(cqr))
+			continue;
+		if (cqr->status != DASD_CQR_FILLED) /* could be failed */
+			continue;
+		if (test_bit(DASD_FLAG_LOCK_STOLEN, &device->flags) &&
+		    !test_bit(DASD_CQR_ALLOW_SLOCK, &cqr->flags)) {
+			cqr->status = DASD_CQR_FAILED;
+			cqr->intrc = -EPERM;
+			continue;
+		}
+		/* Non-temporary stop condition will trigger fail fast */
+		if (device->stopped & ~DASD_STOPPED_PENDING &&
+		    test_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags) &&
+		    !dasd_eer_enabled(device) && device->aq_mask == 0) {
+			cqr->status = DASD_CQR_FAILED;
+			cqr->intrc = -ENOLINK;
+			continue;
+		}
+		/*
+		 * Don't try to start requests if device is in
+		 * offline processing, it might wait forever
+		 */
+		if (test_bit(DASD_FLAG_OFFLINE, &device->flags)) {
+			cqr->status = DASD_CQR_FAILED;
+			cqr->intrc = -ENODEV;
+			continue;
+		}
+		/*
+		 * Don't try to start requests if device is stopped
+		 * except path verification requests
+		 */
+		if (!test_bit(DASD_CQR_VERIFY_PATH, &cqr->flags)) {
+			if (interruptible) {
+				rc = wait_event_interruptible(
+					generic_waitq, !(device->stopped));
+				if (rc == -ERESTARTSYS) {
+					cqr->status = DASD_CQR_FAILED;
+					maincqr->intrc = rc;
+					continue;
+				}
+			} else
+				wait_event(generic_waitq, !(device->stopped));
+		}
+		if (!cqr->callback)
+			cqr->callback = dasd_wakeup_cb;
+
+		cqr->callback_data = DASD_SLEEPON_START_TAG;
+		dasd_add_request_tail(cqr);
+		if (interruptible) {
+			rc = wait_event_interruptible(
+				generic_waitq, _wait_for_wakeup(cqr));
+			if (rc == -ERESTARTSYS) {
+				dasd_cancel_req(cqr);
+				/* wait (non-interruptible) for final status */
+				wait_event(generic_waitq,
+					   _wait_for_wakeup(cqr));
+				cqr->status = DASD_CQR_FAILED;
+				maincqr->intrc = rc;
+				continue;
+			}
+		} else
+			wait_event(generic_waitq, _wait_for_wakeup(cqr));
+	}
+
+	maincqr->endclk = get_tod_clock();
+	if ((maincqr->status != DASD_CQR_DONE) &&
+	    (maincqr->intrc != -ERESTARTSYS))
+		dasd_log_sense(maincqr, &maincqr->irb);
+	if (maincqr->status == DASD_CQR_DONE)
+		rc = 0;
+	else if (maincqr->intrc)
+		rc = maincqr->intrc;
+	else
+		rc = -EIO;
+	return rc;
+}
+
+static inline int _wait_for_wakeup_queue(struct list_head *ccw_queue)
+{
+	struct dasd_ccw_req *cqr;
+
+	list_for_each_entry(cqr, ccw_queue, blocklist) {
+		if (cqr->callback_data != DASD_SLEEPON_END_TAG)
+			return 0;
+	}
+
+	return 1;
+}
+
+static int _dasd_sleep_on_queue(struct list_head *ccw_queue, int interruptible)
+{
+	struct dasd_device *device;
+	struct dasd_ccw_req *cqr, *n;
+	u8 *sense = NULL;
+	int rc;
+
+retry:
+	list_for_each_entry_safe(cqr, n, ccw_queue, blocklist) {
+		device = cqr->startdev;
+		if (cqr->status != DASD_CQR_FILLED) /*could be failed*/
+			continue;
+
+		if (test_bit(DASD_FLAG_LOCK_STOLEN, &device->flags) &&
+		    !test_bit(DASD_CQR_ALLOW_SLOCK, &cqr->flags)) {
+			cqr->status = DASD_CQR_FAILED;
+			cqr->intrc = -EPERM;
+			continue;
+		}
+		/*Non-temporary stop condition will trigger fail fast*/
+		if (device->stopped & ~DASD_STOPPED_PENDING &&
+		    test_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags) &&
+		    !dasd_eer_enabled(device)) {
+			cqr->status = DASD_CQR_FAILED;
+			cqr->intrc = -EAGAIN;
+			continue;
+		}
+
+		/*Don't try to start requests if device is stopped*/
+		if (interruptible) {
+			rc = wait_event_interruptible(
+				generic_waitq, !device->stopped);
+			if (rc == -ERESTARTSYS) {
+				cqr->status = DASD_CQR_FAILED;
+				cqr->intrc = rc;
+				continue;
+			}
+		} else
+			wait_event(generic_waitq, !(device->stopped));
+
+		if (!cqr->callback)
+			cqr->callback = dasd_wakeup_cb;
+		cqr->callback_data = DASD_SLEEPON_START_TAG;
+		dasd_add_request_tail(cqr);
+	}
+
+	wait_event(generic_waitq, _wait_for_wakeup_queue(ccw_queue));
+
+	rc = 0;
+	list_for_each_entry_safe(cqr, n, ccw_queue, blocklist) {
+		/*
+		 * In some cases certain errors might be expected and
+		 * error recovery would be unnecessary in these cases.
+		 * Check if the according suppress bit is set.
+		 */
+		sense = dasd_get_sense(&cqr->irb);
+		if (sense && (sense[1] & SNS1_INV_TRACK_FORMAT) &&
+		    !(sense[2] & SNS2_ENV_DATA_PRESENT) &&
+		    test_bit(DASD_CQR_SUPPRESS_IT, &cqr->flags))
+			continue;
+		if (sense && (sense[1] & SNS1_NO_REC_FOUND) &&
+		    test_bit(DASD_CQR_SUPPRESS_NRF, &cqr->flags))
+			continue;
+		if (scsw_cstat(&cqr->irb.scsw) == 0x40 &&
+		    test_bit(DASD_CQR_SUPPRESS_IL, &cqr->flags))
+			continue;
+
+		/*
+		 * for alias devices simplify error recovery and
+		 * return to upper layer
+		 * do not skip ERP requests
+		 */
+		if (cqr->startdev != cqr->basedev && !cqr->refers &&
+		    (cqr->status == DASD_CQR_TERMINATED ||
+		     cqr->status == DASD_CQR_NEED_ERP))
+			return -EAGAIN;
+
+		/* normal recovery for basedev IO */
+		if (__dasd_sleep_on_erp(cqr))
+			/* handle erp first */
+			goto retry;
+	}
+
+	return 0;
+}
+
+/*
+ * Queue a request to the tail of the device ccw_queue and wait for
+ * it's completion.
+ */
+int dasd_sleep_on(struct dasd_ccw_req *cqr)
+{
+	return _dasd_sleep_on(cqr, 0);
+}
+EXPORT_SYMBOL(dasd_sleep_on);
+
+/*
+ * Start requests from a ccw_queue and wait for their completion.
+ */
+int dasd_sleep_on_queue(struct list_head *ccw_queue)
+{
+	return _dasd_sleep_on_queue(ccw_queue, 0);
+}
+EXPORT_SYMBOL(dasd_sleep_on_queue);
+
+/*
+ * Start requests from a ccw_queue and wait interruptible for their completion.
+ */
+int dasd_sleep_on_queue_interruptible(struct list_head *ccw_queue)
+{
+	return _dasd_sleep_on_queue(ccw_queue, 1);
+}
+EXPORT_SYMBOL(dasd_sleep_on_queue_interruptible);
+
+/*
+ * Queue a request to the tail of the device ccw_queue and wait
+ * interruptible for it's completion.
+ */
+int dasd_sleep_on_interruptible(struct dasd_ccw_req *cqr)
+{
+	return _dasd_sleep_on(cqr, 1);
+}
+EXPORT_SYMBOL(dasd_sleep_on_interruptible);
+
+/*
+ * Whoa nelly now it gets really hairy. For some functions (e.g. steal lock
+ * for eckd devices) the currently running request has to be terminated
+ * and be put back to status queued, before the special request is added
+ * to the head of the queue. Then the special request is waited on normally.
+ */
+static inline int _dasd_term_running_cqr(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr;
+	int rc;
+
+	if (list_empty(&device->ccw_queue))
+		return 0;
+	cqr = list_entry(device->ccw_queue.next, struct dasd_ccw_req, devlist);
+	/*
+	 * Path verification requests must not be terminated. They are critical
+	 * for bringing paths back online. Terminating them would cause rc=-EIO
+	 * because CLEARED requests skip the retry path.
+	 */
+	if (test_bit(DASD_CQR_VERIFY_PATH, &cqr->flags))
+		return -EAGAIN;
+	rc = device->discipline->term_IO(cqr);
+	if (!rc)
+		/*
+		 * CQR terminated because a more important request is pending.
+		 * Undo decreasing of retry counter because this is
+		 * not an error case.
+		 */
+		cqr->retries++;
+	return rc;
+}
+
+int dasd_sleep_on_immediatly(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device;
+	int rc;
+
+	device = cqr->startdev;
+	if (test_bit(DASD_FLAG_LOCK_STOLEN, &device->flags) &&
+	    !test_bit(DASD_CQR_ALLOW_SLOCK, &cqr->flags)) {
+		cqr->status = DASD_CQR_FAILED;
+		cqr->intrc = -EPERM;
+		return -EIO;
+	}
+	spin_lock_irq(get_ccwdev_lock(device->cdev));
+	while ((rc = _dasd_term_running_cqr(device)) == -EAGAIN) {
+		spin_unlock_irq(get_ccwdev_lock(device->cdev));
+		msleep(1);
+		spin_lock_irq(get_ccwdev_lock(device->cdev));
+	}
+	if (rc) {
+		spin_unlock_irq(get_ccwdev_lock(device->cdev));
+		return rc;
+	}
+	cqr->callback = dasd_wakeup_cb;
+	cqr->callback_data = DASD_SLEEPON_START_TAG;
+	cqr->status = DASD_CQR_QUEUED;
+	/*
+	 * add new request as second
+	 * first the terminated cqr needs to be finished
+	 */
+	list_add(&cqr->devlist, device->ccw_queue.next);
+
+	/* let the bh start the request to keep them in order */
+	dasd_schedule_device_bh(device);
+
+	spin_unlock_irq(get_ccwdev_lock(device->cdev));
+
+	wait_event(generic_waitq, _wait_for_wakeup(cqr));
+
+	if (cqr->status == DASD_CQR_DONE)
+		rc = 0;
+	else if (cqr->intrc)
+		rc = cqr->intrc;
+	else
+		rc = -EIO;
+
+	/* kick tasklets */
+	dasd_schedule_device_bh(device);
+	if (device->block)
+		dasd_schedule_block_bh(device->block);
+
+	return rc;
+}
+EXPORT_SYMBOL(dasd_sleep_on_immediatly);
+
+/*
+ * Cancels a request that was started with dasd_sleep_on_req.
+ * This is useful to timeout requests. The request will be
+ * terminated if it is currently in i/o.
+ * Returns 0 if request termination was successful
+ *	   negative error code if termination failed
+ * Cancellation of a request is an asynchronous operation! The calling
+ * function has to wait until the request is properly returned via callback.
+ */
+static int __dasd_cancel_req(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device = cqr->startdev;
+	int rc = 0;
+
+	switch (cqr->status) {
+	case DASD_CQR_QUEUED:
+		/* request was not started - just set to cleared */
+		cqr->status = DASD_CQR_CLEARED;
+		break;
+	case DASD_CQR_IN_IO:
+		/* request in IO - terminate IO and release again */
+		rc = device->discipline->term_IO(cqr);
+		if (rc) {
+			dev_err(&device->cdev->dev,
+				"Cancelling request failed with rc=%d\n", rc);
+		} else {
+			cqr->stopclk = get_tod_clock();
+		}
+		break;
+	default: /* already finished or clear pending - do nothing */
+		break;
+	}
+	dasd_schedule_device_bh(device);
+	return rc;
+}
+
+int dasd_cancel_req(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device = cqr->startdev;
+	unsigned long flags;
+	int rc;
+
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	rc = __dasd_cancel_req(cqr);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	return rc;
+}
+
+/*
+ * SECTION: Operations of the dasd_block layer.
+ */
+
+/*
+ * Timeout function for dasd_block. This is used when the block layer
+ * is waiting for something that may not come reliably, (e.g. a state
+ * change interrupt)
+ */
+static void dasd_block_timeout(struct timer_list *t)
+{
+	unsigned long flags;
+	struct dasd_block *block;
+
+	block = timer_container_of(block, t, timer);
+	spin_lock_irqsave(get_ccwdev_lock(block->base->cdev), flags);
+	/* re-activate request queue */
+	dasd_device_remove_stop_bits(block->base, DASD_STOPPED_PENDING);
+	spin_unlock_irqrestore(get_ccwdev_lock(block->base->cdev), flags);
+	dasd_schedule_block_bh(block);
+	blk_mq_run_hw_queues(block->gdp->queue, true);
+}
+
+/*
+ * Setup timeout for a dasd_block in jiffies.
+ */
+void dasd_block_set_timer(struct dasd_block *block, int expires)
+{
+	if (expires == 0)
+		timer_delete(&block->timer);
+	else
+		mod_timer(&block->timer, jiffies + expires);
+}
+EXPORT_SYMBOL(dasd_block_set_timer);
+
+/*
+ * Clear timeout for a dasd_block.
+ */
+void dasd_block_clear_timer(struct dasd_block *block)
+{
+	timer_delete(&block->timer);
+}
+EXPORT_SYMBOL(dasd_block_clear_timer);
+
+/*
+ * Process finished error recovery ccw.
+ */
+static void __dasd_process_erp(struct dasd_device *device,
+			       struct dasd_ccw_req *cqr)
+{
+	dasd_erp_fn_t erp_fn;
+
+	if (cqr->status == DASD_CQR_DONE)
+		DBF_DEV_EVENT(DBF_NOTICE, device, "%s", "ERP successful");
+	else if (cqr->status == DASD_CQR_ABORTED)
+		/*
+		 * ESE format aborts the request and replaces it with a format
+		 * CQR - this is not an ERP failure.
+		 */
+		DBF_DEV_EVENT(DBF_NOTICE, device, "%s",
+			      "ERP request aborted, replaced by ESE format");
+	else
+		dev_err(&device->cdev->dev, "ERP failed for the DASD\n");
+	erp_fn = device->discipline->erp_postaction(cqr);
+	erp_fn(cqr);
+}
+
+static void __dasd_cleanup_cqr(struct dasd_ccw_req *cqr)
+{
+	struct request *req;
+	blk_status_t error = BLK_STS_OK;
+	unsigned int proc_bytes;
+	int status, intrc;
+
+	req = (struct request *) cqr->callback_data;
+	dasd_profile_end(cqr->block, cqr, req);
+
+	/*
+	 * free_cp() returns the request block to its memory pool, so snapshot
+	 * everything still needed from cqr before calling it - another CPU can
+	 * reallocate and overwrite the block right after.
+	 */
+	proc_bytes = cqr->proc_bytes;
+	intrc = cqr->intrc;
+	status = cqr->block->base->discipline->free_cp(cqr, req);
+	if (status < 0)
+		error = errno_to_blk_status(status);
+	else if (status == 0) {
+		switch (intrc) {
+		case -EPERM:
+			/*
+			 * DASD doesn't implement SCSI/NVMe reservations, but it
+			 * implements a locking scheme similar to them. We
+			 * return this error when we no longer have the lock.
+			 */
+			error = BLK_STS_RESV_CONFLICT;
+			break;
+		case -ENOLINK:
+			error = BLK_STS_TRANSPORT;
+			break;
+		case -ETIMEDOUT:
+			error = BLK_STS_TIMEOUT;
+			break;
+		default:
+			error = BLK_STS_IOERR;
+			break;
+		}
+	} else if (status == DASD_CQR_ABORTED) {
+		/* aborted requests are replaced with a new one so do not complete this */
+		return;
+	}
+
+	/*
+	 * We need to take care for ETIMEDOUT errors here since the
+	 * complete callback does not get called in this case.
+	 * Take care of all errors here and avoid additional code to
+	 * transfer the error value to the complete callback.
+	 */
+	if (error) {
+		blk_mq_end_request(req, error);
+		blk_mq_run_hw_queues(req->q, true);
+	} else {
+		/*
+		 * Partial completed requests can happen with ESE devices.
+		 * During read we might have gotten a NRF error and have to
+		 * complete a request partially.
+		 */
+		if (proc_bytes) {
+			blk_update_request(req, BLK_STS_OK, proc_bytes);
+			blk_mq_requeue_request(req, true);
+		} else if (likely(!blk_should_fake_timeout(req->q))) {
+			blk_mq_complete_request(req);
+		}
+	}
+}
+
+/*
+ * Process ccw request queue.
+ */
+static void __dasd_process_block_ccw_queue(struct dasd_block *block,
+					   struct list_head *final_queue)
+{
+	struct list_head *l, *n;
+	struct dasd_ccw_req *cqr;
+	dasd_erp_fn_t erp_fn;
+	unsigned long flags;
+	struct dasd_device *base = block->base;
+
+restart:
+	/* Process request with final status. */
+	list_for_each_safe(l, n, &block->ccw_queue) {
+		cqr = list_entry(l, struct dasd_ccw_req, blocklist);
+		if (cqr->status != DASD_CQR_DONE &&
+		    cqr->status != DASD_CQR_FAILED &&
+		    cqr->status != DASD_CQR_NEED_ERP &&
+		    cqr->status != DASD_CQR_TERMINATED &&
+		    cqr->status != DASD_CQR_ABORTED)
+			continue;
+
+		if (cqr->status == DASD_CQR_TERMINATED) {
+			base->discipline->handle_terminated_request(cqr);
+			goto restart;
+		}
+
+		/*  Process requests that may be recovered */
+		if (cqr->status == DASD_CQR_NEED_ERP) {
+			erp_fn = base->discipline->erp_action(cqr);
+			if (IS_ERR(erp_fn(cqr)))
+				continue;
+			goto restart;
+		}
+
+		/* log sense for fatal error */
+		if (cqr->status == DASD_CQR_FAILED) {
+			dasd_log_sense(cqr, &cqr->irb);
+		}
+
+		/*
+		 * First call extended error reporting and check for autoquiesce
+		 */
+		spin_lock_irqsave(get_ccwdev_lock(base->cdev), flags);
+		if (cqr->status == DASD_CQR_FAILED &&
+		    dasd_handle_autoquiesce(base, cqr, DASD_EER_FATALERROR)) {
+			cqr->status = DASD_CQR_FILLED;
+			cqr->retries = 255;
+			spin_unlock_irqrestore(get_ccwdev_lock(base->cdev), flags);
+			goto restart;
+		}
+		spin_unlock_irqrestore(get_ccwdev_lock(base->cdev), flags);
+
+		/* Process finished ERP request. */
+		if (cqr->refers) {
+			__dasd_process_erp(base, cqr);
+			goto restart;
+		}
+
+		/* Rechain finished requests to final queue */
+		cqr->endclk = get_tod_clock();
+		list_move_tail(&cqr->blocklist, final_queue);
+	}
+}
+
+static void dasd_return_cqr_cb(struct dasd_ccw_req *cqr, void *data)
+{
+	struct dasd_ccw_req *temp_cqr;
+	struct dasd_block *block;
+
+	/* only format CQRs are candidates */
+	if (!cqr->block || unlikely(!cqr->format))
+		goto out;
+
+	block = cqr->block;
+	/*
+	 * Mark in-flight (IN_IO) CQRs that overlap this just-completed format
+	 * range so they re-check in test_and_set_format on completion; FILLED
+	 * or QUEUED CQRs re-check the format_list on their next round anyway.
+	 */
+	list_for_each_entry(temp_cqr, &block->ccw_queue, blocklist) {
+		if (temp_cqr != cqr &&
+		    temp_cqr->status != DASD_CQR_FILLED &&
+		    temp_cqr->status != DASD_CQR_QUEUED &&
+		    dasd_req_conflict(cqr, temp_cqr)) {
+			WRITE_ONCE(temp_cqr->collision, true);
+		}
+	}
+out:
+	dasd_schedule_block_bh(cqr->block);
+}
+
+static void __dasd_block_start_head(struct dasd_block *block)
+{
+	struct dasd_ccw_req *cqr;
+
+	if (list_empty(&block->ccw_queue))
+		return;
+	/* We allways begin with the first requests on the queue, as some
+	 * of previously started requests have to be enqueued on a
+	 * dasd_device again for error recovery.
+	 */
+	list_for_each_entry(cqr, &block->ccw_queue, blocklist) {
+		if (cqr->status != DASD_CQR_FILLED)
+			continue;
+		if (test_bit(DASD_FLAG_LOCK_STOLEN, &block->base->flags) &&
+		    !test_bit(DASD_CQR_ALLOW_SLOCK, &cqr->flags)) {
+			cqr->status = DASD_CQR_FAILED;
+			cqr->intrc = -EPERM;
+			dasd_schedule_block_bh(block);
+			continue;
+		}
+		/* Non-temporary stop condition will trigger fail fast */
+		if (block->base->stopped & ~DASD_STOPPED_PENDING &&
+		    test_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags) &&
+		    !dasd_eer_enabled(block->base) && block->base->aq_mask == 0) {
+			cqr->status = DASD_CQR_FAILED;
+			cqr->intrc = -ENOLINK;
+			dasd_schedule_block_bh(block);
+			continue;
+		}
+		/* Don't try to start requests if device is stopped */
+		if (block->base->stopped)
+			return;
+
+		/* just a fail safe check, should not happen */
+		if (!cqr->startdev)
+			cqr->startdev = block->base;
+
+		/* make sure that the requests we submit find their way back */
+		cqr->callback = dasd_return_cqr_cb;
+
+		dasd_add_request_tail(cqr);
+	}
+}
+
+/*
+ * Central dasd_block layer routine. Takes requests from the generic
+ * block layer request queue, creates ccw requests, enqueues them on
+ * a dasd_device and processes ccw requests that have been returned.
+ */
+static void dasd_block_tasklet(unsigned long data)
+{
+	struct dasd_block *block = (struct dasd_block *) data;
+	struct list_head final_queue;
+	struct list_head *l, *n;
+	struct dasd_ccw_req *cqr;
+	struct dasd_queue *dq;
+
+	atomic_set(&block->tasklet_scheduled, 0);
+	INIT_LIST_HEAD(&final_queue);
+	spin_lock_irq(&block->queue_lock);
+	/*
+	 * Splice the hardirq-staged ESE format CQRs onto ccw_queue. Splice to
+	 * the tail so an aborted origin request (already on ccw_queue) is
+	 * retired before its format-CQR replacement completes and requeues it.
+	 */
+	spin_lock(&block->ese_lock);
+	list_splice_tail_init(&block->ese_staging, &block->ccw_queue);
+	spin_unlock(&block->ese_lock);
+	/* Finish off requests on ccw queue */
+	__dasd_process_block_ccw_queue(block, &final_queue);
+	spin_unlock_irq(&block->queue_lock);
+
+	/* Now call the callback function of requests with final status */
+	list_for_each_safe(l, n, &final_queue) {
+		cqr = list_entry(l, struct dasd_ccw_req, blocklist);
+		dq = cqr->dq;
+		spin_lock_irq(&dq->lock);
+		list_del_init(&cqr->blocklist);
+		__dasd_cleanup_cqr(cqr);
+		spin_unlock_irq(&dq->lock);
+	}
+
+	spin_lock_irq(&block->queue_lock);
+	/* Now check if the head of the ccw queue needs to be started. */
+	__dasd_block_start_head(block);
+	spin_unlock_irq(&block->queue_lock);
+
+	if (waitqueue_active(&shutdown_waitq))
+		wake_up(&shutdown_waitq);
+	dasd_put_device(block->base);
+}
+
+static void _dasd_wake_block_flush_cb(struct dasd_ccw_req *cqr, void *data)
+{
+	wake_up(&dasd_flush_wq);
+}
+
+/*
+ * Requeue a request back to the block request queue
+ * only works for block requests
+ */
+static void _dasd_requeue_request(struct dasd_ccw_req *cqr)
+{
+	struct request *req;
+
+	/*
+	 * If the request is an ERP request there is nothing to requeue.
+	 * This will be done with the remaining original request.
+	 */
+	if (cqr->refers)
+		return;
+	spin_lock_irq(&cqr->dq->lock);
+	req = (struct request *) cqr->callback_data;
+	blk_mq_requeue_request(req, true);
+	spin_unlock_irq(&cqr->dq->lock);
+
+	return;
+}
+
+static int _dasd_requests_to_flushqueue(struct dasd_block *block,
+					struct list_head *flush_queue)
+{
+	struct dasd_ccw_req *cqr, *n;
+	unsigned long flags;
+	int rc, i;
+
+	spin_lock_irqsave(&block->queue_lock, flags);
+	/*
+	 * Splice any hardirq-staged ESE format CQRs onto ccw_queue first so
+	 * they are seen and canceled by the walk below instead of being
+	 * orphaned across this flush / state transition. Mirrors the splice
+	 * in dasd_block_tasklet().
+	 */
+	spin_lock(&block->ese_lock);
+	list_splice_tail_init(&block->ese_staging, &block->ccw_queue);
+	spin_unlock(&block->ese_lock);
+	rc = 0;
+restart:
+	list_for_each_entry_safe(cqr, n, &block->ccw_queue, blocklist) {
+		/* if this request currently owned by a dasd_device cancel it */
+		if (cqr->status >= DASD_CQR_QUEUED)
+			rc = dasd_cancel_req(cqr);
+		if (rc < 0)
+			break;
+		/* Rechain request (including erp chain) so it won't be
+		 * touched by the dasd_block_tasklet anymore.
+		 * Replace the callback so we notice when the request
+		 * is returned from the dasd_device layer.
+		 */
+		cqr->callback = _dasd_wake_block_flush_cb;
+		for (i = 0; cqr; cqr = cqr->refers, i++)
+			list_move_tail(&cqr->blocklist, flush_queue);
+		if (i > 1)
+			/* moved more than one request - need to restart */
+			goto restart;
+	}
+	spin_unlock_irqrestore(&block->queue_lock, flags);
+
+	return rc;
+}
+
+/*
+ * Go through all request on the dasd_block request queue, cancel them
+ * on the respective dasd_device, and return them to the generic
+ * block layer.
+ */
+static int dasd_flush_block_queue(struct dasd_block *block)
+{
+	struct dasd_ccw_req *cqr, *n;
+	struct list_head flush_queue;
+	unsigned long flags;
+	int rc;
+
+	INIT_LIST_HEAD(&flush_queue);
+	rc = _dasd_requests_to_flushqueue(block, &flush_queue);
+
+	/* Now call the callback function of flushed requests */
+restart_cb:
+	list_for_each_entry_safe(cqr, n, &flush_queue, blocklist) {
+		wait_event(dasd_flush_wq, (cqr->status < DASD_CQR_QUEUED));
+		/* Process finished ERP request. */
+		if (cqr->refers) {
+			spin_lock_bh(&block->queue_lock);
+			__dasd_process_erp(block->base, cqr);
+			spin_unlock_bh(&block->queue_lock);
+			/* restart list_for_xx loop since dasd_process_erp
+			 * might remove multiple elements */
+			goto restart_cb;
+		}
+		/* call the callback function */
+		spin_lock_irqsave(&cqr->dq->lock, flags);
+		cqr->endclk = get_tod_clock();
+		list_del_init(&cqr->blocklist);
+		__dasd_cleanup_cqr(cqr);
+		spin_unlock_irqrestore(&cqr->dq->lock, flags);
+	}
+	return rc;
+}
+
+/*
+ * Schedules a call to dasd_tasklet over the device tasklet.
+ */
+void dasd_schedule_block_bh(struct dasd_block *block)
+{
+	/* Protect against rescheduling. */
+	if (atomic_cmpxchg(&block->tasklet_scheduled, 0, 1) != 0)
+		return;
+	/* life cycle of block is bound to it's base device */
+	dasd_get_device(block->base);
+	tasklet_hi_schedule(&block->tasklet);
+}
+EXPORT_SYMBOL(dasd_schedule_block_bh);
+
+
+/*
+ * SECTION: external block device operations
+ * (request queue handling, open, release, etc.)
+ */
+
+/*
+ * Dasd request queue function. Called from ll_rw_blk.c
+ */
+static blk_status_t do_dasd_request(struct blk_mq_hw_ctx *hctx,
+				    const struct blk_mq_queue_data *qd)
+{
+	struct dasd_block *block = hctx->queue->queuedata;
+	struct dasd_queue *dq = hctx->driver_data;
+	struct request *req = qd->rq;
+	struct dasd_device *basedev;
+	struct dasd_ccw_req *cqr;
+	blk_status_t rc = BLK_STS_OK;
+	bool complete_noop = false;
+
+	basedev = block->base;
+	spin_lock_irq(&dq->lock);
+	if (basedev->state < DASD_STATE_READY ||
+	    test_bit(DASD_FLAG_OFFLINE, &basedev->flags)) {
+		DBF_DEV_EVENT(DBF_ERR, basedev,
+			      "device not ready for request %p", req);
+		rc = BLK_STS_IOERR;
+		goto out;
+	}
+
+	/*
+	 * if device is stopped do not fetch new requests
+	 * except failfast is active which will let requests fail
+	 * immediately in __dasd_block_start_head()
+	 */
+	if (basedev->stopped && !(basedev->features & DASD_FEATURE_FAILFAST)) {
+		DBF_DEV_EVENT(DBF_ERR, basedev,
+			      "device stopped request %p", req);
+		rc = BLK_STS_RESOURCE;
+		goto out;
+	}
+
+	if (basedev->features & DASD_FEATURE_READONLY &&
+	    rq_data_dir(req) == WRITE) {
+		DBF_DEV_EVENT(DBF_ERR, basedev,
+			      "Rejecting write request %p", req);
+		rc = BLK_STS_IOERR;
+		goto out;
+	}
+
+	if (test_bit(DASD_FLAG_ABORTALL, &basedev->flags) &&
+	    (basedev->features & DASD_FEATURE_FAILFAST ||
+	     blk_noretry_request(req))) {
+		DBF_DEV_EVENT(DBF_ERR, basedev,
+			      "Rejecting failfast request %p", req);
+		rc = BLK_STS_IOERR;
+		goto out;
+	}
+
+	cqr = basedev->discipline->build_cp(basedev, block, req);
+	if (IS_ERR(cqr)) {
+		if (PTR_ERR(cqr) == -EBUSY ||
+		    PTR_ERR(cqr) == -ENOMEM ||
+		    PTR_ERR(cqr) == -EAGAIN) {
+			rc = BLK_STS_RESOURCE;
+		} else if (PTR_ERR(cqr) == -EINVAL) {
+			rc = BLK_STS_INVAL;
+		} else if (PTR_ERR(cqr) == -EOPNOTSUPP) {
+			/*
+			 * A discard that covers no whole extent releases
+			 * nothing. Discard is advisory, so complete it as a
+			 * benign no-op: the device does support discard, this
+			 * range just does not align to the large ESE extent
+			 * granularity.
+			 * Completed after the lock is dropped.
+			 */
+			rc = BLK_STS_OK;
+			complete_noop = true;
+		} else {
+			DBF_DEV_EVENT(DBF_ERR, basedev,
+				      "CCW creation failed (rc=%ld) on request %p",
+				      PTR_ERR(cqr), req);
+			rc = BLK_STS_IOERR;
+		}
+		goto out;
+	}
+	if (!cqr) {
+		/* build_cp may collapse a non-transient build error to NULL */
+		DBF_DEV_EVENT(DBF_ERR, basedev,
+			      "CCW creation returned NULL on request %p", req);
+		rc = BLK_STS_IOERR;
+		goto out;
+	}
+	/*
+	 *  Note: callback is set to dasd_return_cqr_cb in
+	 * __dasd_block_start_head to cover erp requests as well
+	 */
+	cqr->callback_data = req;
+	cqr->status = DASD_CQR_FILLED;
+	cqr->dq = dq;
+
+	blk_mq_start_request(req);
+	spin_lock(&block->queue_lock);
+	list_add_tail(&cqr->blocklist, &block->ccw_queue);
+	INIT_LIST_HEAD(&cqr->devlist);
+	dasd_profile_start(block, cqr, req);
+	dasd_schedule_block_bh(block);
+	spin_unlock(&block->queue_lock);
+
+out:
+	spin_unlock_irq(&dq->lock);
+	if (complete_noop)
+		blk_mq_end_request(req, BLK_STS_OK);
+	return rc;
+}
+
+/*
+ * Block timeout callback, called from the block layer
+ *
+ * Return values:
+ * BLK_EH_RESET_TIMER if the request should be left running
+ * BLK_EH_DONE if the request is handled or terminated
+ *		      by the driver.
+ */
+enum blk_eh_timer_return dasd_times_out(struct request *req)
+{
+	struct dasd_block *block = req->q->queuedata;
+	struct dasd_device *device;
+	struct dasd_ccw_req *cqr;
+	unsigned long flags;
+	int rc = 0;
+
+	cqr = blk_mq_rq_to_pdu(req);
+	if (!cqr)
+		return BLK_EH_DONE;
+
+	spin_lock_irqsave(&cqr->dq->lock, flags);
+	device = cqr->startdev ? cqr->startdev : block->base;
+	if (!device->blk_timeout) {
+		spin_unlock_irqrestore(&cqr->dq->lock, flags);
+		return BLK_EH_RESET_TIMER;
+	}
+	DBF_DEV_EVENT(DBF_WARNING, device,
+		      " dasd_times_out cqr %p status %x",
+		      cqr, cqr->status);
+
+	spin_lock(&block->queue_lock);
+	spin_lock(get_ccwdev_lock(device->cdev));
+	cqr->retries = -1;
+	cqr->intrc = -ETIMEDOUT;
+	if (cqr->status >= DASD_CQR_QUEUED) {
+		rc = __dasd_cancel_req(cqr);
+	} else if (cqr->status == DASD_CQR_FILLED ||
+		   cqr->status == DASD_CQR_NEED_ERP) {
+		cqr->status = DASD_CQR_TERMINATED;
+	} else if (cqr->status == DASD_CQR_IN_ERP) {
+		struct dasd_ccw_req *searchcqr, *nextcqr, *tmpcqr;
+
+		list_for_each_entry_safe(searchcqr, nextcqr,
+					 &block->ccw_queue, blocklist) {
+			tmpcqr = searchcqr;
+			while (tmpcqr->refers)
+				tmpcqr = tmpcqr->refers;
+			if (tmpcqr != cqr)
+				continue;
+			/* searchcqr is an ERP request for cqr */
+			searchcqr->retries = -1;
+			searchcqr->intrc = -ETIMEDOUT;
+			if (searchcqr->status >= DASD_CQR_QUEUED) {
+				rc = __dasd_cancel_req(searchcqr);
+			} else if ((searchcqr->status == DASD_CQR_FILLED) ||
+				   (searchcqr->status == DASD_CQR_NEED_ERP)) {
+				searchcqr->status = DASD_CQR_TERMINATED;
+				rc = 0;
+			} else if (searchcqr->status == DASD_CQR_IN_ERP) {
+				/*
+				 * Shouldn't happen; most recent ERP
+				 * request is at the front of queue
+				 */
+				continue;
+			}
+			break;
+		}
+	}
+	spin_unlock(get_ccwdev_lock(device->cdev));
+	dasd_schedule_block_bh(block);
+	spin_unlock(&block->queue_lock);
+	spin_unlock_irqrestore(&cqr->dq->lock, flags);
+
+	return rc ? BLK_EH_RESET_TIMER : BLK_EH_DONE;
+}
+
+static int dasd_init_hctx(struct blk_mq_hw_ctx *hctx, void *data,
+			  unsigned int idx)
+{
+	struct dasd_queue *dq = kzalloc_obj(*dq);
+
+	if (!dq)
+		return -ENOMEM;
+
+	spin_lock_init(&dq->lock);
+	hctx->driver_data = dq;
+
+	return 0;
+}
+
+static void dasd_exit_hctx(struct blk_mq_hw_ctx *hctx, unsigned int idx)
+{
+	kfree(hctx->driver_data);
+	hctx->driver_data = NULL;
+}
+
+static void dasd_request_done(struct request *req)
+{
+	blk_mq_end_request(req, 0);
+	blk_mq_run_hw_queues(req->q, true);
+}
+
+struct blk_mq_ops dasd_mq_ops = {
+	.queue_rq = do_dasd_request,
+	.complete = dasd_request_done,
+	.timeout = dasd_times_out,
+	.init_hctx = dasd_init_hctx,
+	.exit_hctx = dasd_exit_hctx,
+};
+
+static int dasd_open(struct gendisk *disk, blk_mode_t mode)
+{
+	struct dasd_device *base;
+	int rc;
+
+	base = dasd_device_from_gendisk(disk);
+	if (!base)
+		return -ENODEV;
+
+	atomic_inc(&base->block->open_count);
+	if (test_bit(DASD_FLAG_OFFLINE, &base->flags)) {
+		rc = -ENODEV;
+		goto unlock;
+	}
+
+	if (!try_module_get(base->discipline->owner)) {
+		rc = -EINVAL;
+		goto unlock;
+	}
+
+	if (dasd_probeonly) {
+		dev_info(&base->cdev->dev,
+			 "Accessing the DASD failed because it is in "
+			 "probeonly mode\n");
+		rc = -EPERM;
+		goto out;
+	}
+
+	if (base->state <= DASD_STATE_BASIC) {
+		DBF_DEV_EVENT(DBF_ERR, base, " %s",
+			      " Cannot open unrecognized device");
+		rc = -ENODEV;
+		goto out;
+	}
+	if ((mode & BLK_OPEN_WRITE) &&
+	    (test_bit(DASD_FLAG_DEVICE_RO, &base->flags) ||
+	     (base->features & DASD_FEATURE_READONLY))) {
+		rc = -EROFS;
+		goto out;
+	}
+	dasd_put_device(base);
+	return 0;
+
+out:
+	module_put(base->discipline->owner);
+unlock:
+	atomic_dec(&base->block->open_count);
+	dasd_put_device(base);
+	return rc;
+}
+
+static void dasd_release(struct gendisk *disk)
+{
+	struct dasd_device *base = dasd_device_from_gendisk(disk);
+	if (base) {
+		atomic_dec(&base->block->open_count);
+		module_put(base->discipline->owner);
+		dasd_put_device(base);
+	}
+}
+
+/*
+ * Return disk geometry.
+ */
+static int dasd_getgeo(struct gendisk *disk, struct hd_geometry *geo)
+{
+	struct dasd_device *base;
+
+	base = dasd_device_from_gendisk(disk);
+	if (!base)
+		return -ENODEV;
+
+	if (!base->discipline ||
+	    !base->discipline->fill_geometry) {
+		dasd_put_device(base);
+		return -EINVAL;
+	}
+	base->discipline->fill_geometry(base->block, geo);
+	// geo->start is left unchanged by the above
+	geo->start >>= base->block->s2b_shift;
+	dasd_put_device(base);
+	return 0;
+}
+
+const struct block_device_operations
+dasd_device_operations = {
+	.owner		= THIS_MODULE,
+	.open		= dasd_open,
+	.release	= dasd_release,
+	.ioctl		= dasd_ioctl,
+	.getgeo		= dasd_getgeo,
+	.set_read_only	= dasd_set_read_only,
+};
+
+/*******************************************************************************
+ * end of block device operations
+ */
+
+static void
+dasd_exit(void)
+{
+#ifdef CONFIG_PROC_FS
+	dasd_proc_exit();
+#endif
+	dasd_eer_exit();
+	kmem_cache_destroy(dasd_page_cache);
+	dasd_page_cache = NULL;
+	dasd_gendisk_exit();
+	dasd_devmap_exit();
+	if (dasd_debug_area != NULL) {
+		debug_unregister(dasd_debug_area);
+		dasd_debug_area = NULL;
+	}
+	dasd_statistics_removeroot();
+}
+
+/*
+ * SECTION: common functions for ccw_driver use
+ */
+
+/*
+ * Is the device read-only?
+ * Note that this function does not report the setting of the
+ * readonly device attribute, but how it is configured in z/VM.
+ */
+int dasd_device_is_ro(struct dasd_device *device)
+{
+	struct ccw_dev_id dev_id;
+	struct diag210 diag_data;
+	int rc;
+
+	if (!machine_is_vm())
+		return 0;
+	ccw_device_get_id(device->cdev, &dev_id);
+	memset(&diag_data, 0, sizeof(diag_data));
+	diag_data.vrdcdvno = dev_id.devno;
+	diag_data.vrdclen = sizeof(diag_data);
+	rc = diag210(&diag_data);
+	if (rc == 0 || rc == 2) {
+		return diag_data.vrdcvfla & 0x80;
+	} else {
+		DBF_EVENT(DBF_WARNING, "diag210 failed for dev=%04x with rc=%d",
+			  dev_id.devno, rc);
+		return 0;
+	}
+}
+EXPORT_SYMBOL_GPL(dasd_device_is_ro);
+
+static void dasd_generic_auto_online(void *data, async_cookie_t cookie)
+{
+	struct ccw_device *cdev = data;
+	int ret;
+
+	ret = ccw_device_set_online(cdev);
+	if (ret)
+		dev_warn(&cdev->dev, "Setting the DASD online failed with rc=%d\n", ret);
+}
+
+/*
+ * Initial attempt at a probe function. this can be simplified once
+ * the other detection code is gone.
+ */
+int dasd_generic_probe(struct ccw_device *cdev)
+{
+	cdev->handler = &dasd_int_handler;
+
+	/*
+	 * Automatically online either all dasd devices (dasd_autodetect)
+	 * or all devices specified with dasd= parameters during
+	 * initial probe.
+	 */
+	if ((dasd_get_feature(cdev, DASD_FEATURE_INITIAL_ONLINE) > 0 ) ||
+	    (dasd_autodetect && dasd_busid_known(dev_name(&cdev->dev)) != 0))
+		async_schedule(dasd_generic_auto_online, cdev);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(dasd_generic_probe);
+
+void dasd_generic_free_discipline(struct dasd_device *device)
+{
+	/* Forget the discipline information. */
+	if (device->discipline) {
+		if (device->discipline->uncheck_device)
+			device->discipline->uncheck_device(device);
+		module_put(device->discipline->owner);
+		device->discipline = NULL;
+	}
+	if (device->base_discipline) {
+		module_put(device->base_discipline->owner);
+		device->base_discipline = NULL;
+	}
+}
+EXPORT_SYMBOL_GPL(dasd_generic_free_discipline);
+
+/*
+ * This will one day be called from a global not_oper handler.
+ * It is also used by driver_unregister during module unload.
+ */
+void dasd_generic_remove(struct ccw_device *cdev)
+{
+	struct dasd_device *device;
+	struct dasd_block *block;
+
+	device = dasd_device_from_cdev(cdev);
+	if (IS_ERR(device))
+		return;
+
+	if (test_and_set_bit(DASD_FLAG_OFFLINE, &device->flags) &&
+	    !test_bit(DASD_FLAG_SAFE_OFFLINE_RUNNING, &device->flags)) {
+		/* Already doing offline processing */
+		dasd_put_device(device);
+		return;
+	}
+	/*
+	 * This device is removed unconditionally. Set offline
+	 * flag to prevent dasd_open from opening it while it is
+	 * no quite down yet.
+	 */
+	dasd_set_target_state(device, DASD_STATE_NEW);
+	cdev->handler = NULL;
+	/* dasd_delete_device destroys the device reference. */
+	block = device->block;
+	dasd_delete_device(device);
+	/*
+	 * life cycle of block is bound to device, so delete it after
+	 * device was safely removed
+	 */
+	if (block)
+		dasd_free_block(block);
+}
+EXPORT_SYMBOL_GPL(dasd_generic_remove);
+
+/*
+ * Activate a device. This is called from dasd_{eckd,fba}_probe() when either
+ * the device is detected for the first time and is supposed to be used
+ * or the user has started activation through sysfs.
+ */
+int dasd_generic_set_online(struct ccw_device *cdev,
+			    struct dasd_discipline *base_discipline)
+{
+	struct dasd_discipline *discipline;
+	struct dasd_device *device;
+	struct device *dev;
+	int rc;
+
+	dev = &cdev->dev;
+
+	/* first online clears initial online feature flag */
+	dasd_set_feature(cdev, DASD_FEATURE_INITIAL_ONLINE, 0);
+	device = dasd_create_device(cdev);
+	if (IS_ERR(device))
+		return PTR_ERR(device);
+
+	discipline = base_discipline;
+	if (device->features & DASD_FEATURE_USEDIAG) {
+	  	if (!dasd_diag_discipline_pointer) {
+			/* Try to load the required module. */
+			rc = request_module(DASD_DIAG_MOD);
+			if (rc) {
+				dev_warn(dev, "Setting the DASD online failed "
+					 "because the required module %s "
+					 "could not be loaded (rc=%d)\n",
+					 DASD_DIAG_MOD, rc);
+				dasd_delete_device(device);
+				return -ENODEV;
+			}
+		}
+		/* Module init could have failed, so check again here after
+		 * request_module(). */
+		if (!dasd_diag_discipline_pointer) {
+			dev_warn(dev, "Setting the DASD online failed because of missing DIAG discipline\n");
+			dasd_delete_device(device);
+			return -ENODEV;
+		}
+		discipline = dasd_diag_discipline_pointer;
+	}
+	if (!try_module_get(base_discipline->owner)) {
+		dasd_delete_device(device);
+		return -EINVAL;
+	}
+	device->base_discipline = base_discipline;
+	if (!try_module_get(discipline->owner)) {
+		dasd_delete_device(device);
+		return -EINVAL;
+	}
+	device->discipline = discipline;
+
+	/* check_device will allocate block device if necessary */
+	rc = discipline->check_device(device);
+	if (rc) {
+		dev_warn(dev, "Setting the DASD online with discipline %s failed with rc=%i\n",
+			 discipline->name, rc);
+		dasd_delete_device(device);
+		return rc;
+	}
+
+	dasd_set_target_state(device, DASD_STATE_ONLINE);
+	if (device->state <= DASD_STATE_KNOWN) {
+		dev_warn(dev, "Setting the DASD online failed because of a missing discipline\n");
+		rc = -ENODEV;
+		dasd_set_target_state(device, DASD_STATE_NEW);
+		if (device->block)
+			dasd_free_block(device->block);
+		dasd_delete_device(device);
+	} else {
+		dev_dbg(dev, "dasd_generic device found\n");
+	}
+
+	wait_event(dasd_init_waitq, _wait_for_device(device));
+
+	dasd_put_device(device);
+	return rc;
+}
+EXPORT_SYMBOL_GPL(dasd_generic_set_online);
+
+int dasd_generic_set_offline(struct ccw_device *cdev)
+{
+	int max_count, open_count, rc;
+	struct dasd_device *device;
+	struct dasd_block *block;
+	unsigned long flags;
+	struct device *dev;
+
+	dev = &cdev->dev;
+
+	rc = 0;
+	spin_lock_irqsave(get_ccwdev_lock(cdev), flags);
+	device = dasd_device_from_cdev_locked(cdev);
+	if (IS_ERR(device)) {
+		spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+		return PTR_ERR(device);
+	}
+
+	/*
+	 * We must make sure that this device is currently not in use.
+	 * The open_count is increased for every opener, that includes
+	 * the blkdev_get in dasd_scan_partitions. We are only interested
+	 * in the other openers.
+	 */
+	if (device->block) {
+		max_count = device->block->bdev_file ? 0 : -1;
+		open_count = atomic_read(&device->block->open_count);
+		if (open_count > max_count) {
+			if (open_count > 0)
+				dev_warn(dev, "The DASD cannot be set offline with open count %i\n",
+					 open_count);
+			else
+				dev_warn(dev, "The DASD cannot be set offline while it is in use\n");
+			rc = -EBUSY;
+			goto out_err;
+		}
+	}
+
+	/*
+	 * Test if the offline processing is already running and exit if so.
+	 * If a safe offline is being processed this could only be a normal
+	 * offline that should be able to overtake the safe offline and
+	 * cancel any I/O we do not want to wait for any longer
+	 */
+	if (test_bit(DASD_FLAG_OFFLINE, &device->flags)) {
+		if (test_bit(DASD_FLAG_SAFE_OFFLINE_RUNNING, &device->flags)) {
+			clear_bit(DASD_FLAG_SAFE_OFFLINE_RUNNING,
+				  &device->flags);
+		} else {
+			rc = -EBUSY;
+			goto out_err;
+		}
+	}
+	set_bit(DASD_FLAG_OFFLINE, &device->flags);
+
+	/*
+	 * if safe_offline is called set safe_offline_running flag and
+	 * clear safe_offline so that a call to normal offline
+	 * can overrun safe_offline processing
+	 */
+	if (test_and_clear_bit(DASD_FLAG_SAFE_OFFLINE, &device->flags) &&
+	    !test_and_set_bit(DASD_FLAG_SAFE_OFFLINE_RUNNING, &device->flags)) {
+		/* need to unlock here to wait for outstanding I/O */
+		spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+		/*
+		 * If we want to set the device safe offline all IO operations
+		 * should be finished before continuing the offline process
+		 * so sync bdev first and then wait for our queues to become
+		 * empty
+		 */
+		if (device->block && device->block->bdev_file)
+			bdev_mark_dead(file_bdev(device->block->bdev_file), false);
+		dasd_schedule_device_bh(device);
+		rc = wait_event_interruptible(shutdown_waitq,
+					      _wait_for_empty_queues(device));
+		if (rc != 0)
+			goto interrupted;
+
+		/*
+		 * check if a normal offline process overtook the offline
+		 * processing in this case simply do nothing beside returning
+		 * that we got interrupted
+		 * otherwise mark safe offline as not running any longer and
+		 * continue with normal offline
+		 */
+		spin_lock_irqsave(get_ccwdev_lock(cdev), flags);
+		if (!test_bit(DASD_FLAG_SAFE_OFFLINE_RUNNING, &device->flags)) {
+			rc = -ERESTARTSYS;
+			goto out_err;
+		}
+		clear_bit(DASD_FLAG_SAFE_OFFLINE_RUNNING, &device->flags);
+	}
+	spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+
+	dasd_set_target_state(device, DASD_STATE_NEW);
+	/* dasd_delete_device destroys the device reference. */
+	block = device->block;
+	dasd_delete_device(device);
+	/*
+	 * life cycle of block is bound to device, so delete it after
+	 * device was safely removed
+	 */
+	if (block)
+		dasd_free_block(block);
+
+	return 0;
+
+interrupted:
+	/* interrupted by signal */
+	spin_lock_irqsave(get_ccwdev_lock(cdev), flags);
+	clear_bit(DASD_FLAG_SAFE_OFFLINE_RUNNING, &device->flags);
+	clear_bit(DASD_FLAG_OFFLINE, &device->flags);
+out_err:
+	dasd_put_device(device);
+	spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+	return rc;
+}
+EXPORT_SYMBOL_GPL(dasd_generic_set_offline);
+
+int dasd_generic_last_path_gone(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr;
+
+	dev_warn(&device->cdev->dev, "No operational channel path is left "
+		 "for the device\n");
+	DBF_DEV_EVENT(DBF_WARNING, device, "%s", "last path gone");
+	/* First call extended error reporting and check for autoquiesce. */
+	dasd_handle_autoquiesce(device, NULL, DASD_EER_NOPATH);
+
+	if (device->state < DASD_STATE_BASIC)
+		return 0;
+	/* Device is active. We want to keep it. */
+	list_for_each_entry(cqr, &device->ccw_queue, devlist)
+		if ((cqr->status == DASD_CQR_IN_IO) ||
+		    (cqr->status == DASD_CQR_CLEAR_PENDING)) {
+			cqr->status = DASD_CQR_QUEUED;
+			cqr->retries++;
+		}
+	dasd_device_set_stop_bits(device, DASD_STOPPED_DC_WAIT);
+	dasd_device_clear_timer(device);
+	dasd_schedule_device_bh(device);
+	return 1;
+}
+EXPORT_SYMBOL_GPL(dasd_generic_last_path_gone);
+
+int dasd_generic_path_operational(struct dasd_device *device)
+{
+	dev_info(&device->cdev->dev, "A channel path to the device has become "
+		 "operational\n");
+	DBF_DEV_EVENT(DBF_WARNING, device, "%s", "path operational");
+	dasd_device_remove_stop_bits(device, DASD_STOPPED_DC_WAIT);
+	dasd_schedule_device_bh(device);
+	if (device->block) {
+		dasd_schedule_block_bh(device->block);
+		if (device->block->gdp)
+			blk_mq_run_hw_queues(device->block->gdp->queue, true);
+	}
+
+	if (!device->stopped)
+		wake_up(&generic_waitq);
+
+	return 1;
+}
+EXPORT_SYMBOL_GPL(dasd_generic_path_operational);
+
+int dasd_generic_notify(struct ccw_device *cdev, int event)
+{
+	struct dasd_device *device;
+	int ret;
+
+	device = dasd_device_from_cdev_locked(cdev);
+	if (IS_ERR(device))
+		return 0;
+	ret = 0;
+	switch (event) {
+	case CIO_GONE:
+	case CIO_BOXED:
+	case CIO_NO_PATH:
+		dasd_path_no_path(device);
+		ret = dasd_generic_last_path_gone(device);
+		break;
+	case CIO_OPER:
+		ret = 1;
+		if (dasd_path_get_opm(device))
+			ret = dasd_generic_path_operational(device);
+		break;
+	}
+	dasd_put_device(device);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(dasd_generic_notify);
+
+void dasd_generic_path_event(struct ccw_device *cdev, int *path_event)
+{
+	struct dasd_device *device;
+	int chp, oldopm, hpfpm, ifccpm;
+
+	device = dasd_device_from_cdev_locked(cdev);
+	if (IS_ERR(device))
+		return;
+
+	oldopm = dasd_path_get_opm(device);
+	for (chp = 0; chp < 8; chp++) {
+		if (path_event[chp] & PE_PATH_GONE) {
+			dasd_path_notoper(device, chp);
+		}
+		if (path_event[chp] & PE_PATH_AVAILABLE) {
+			dasd_path_available(device, chp);
+			dasd_schedule_device_bh(device);
+		}
+		if (path_event[chp] & PE_PATHGROUP_ESTABLISHED) {
+			if (!dasd_path_is_operational(device, chp) &&
+			    !dasd_path_need_verify(device, chp)) {
+				/*
+				 * we can not establish a pathgroup on an
+				 * unavailable path, so trigger a path
+				 * verification first
+				 */
+			dasd_path_available(device, chp);
+			dasd_schedule_device_bh(device);
+			}
+			DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+				      "Pathgroup re-established\n");
+			if (device->discipline->kick_validate)
+				device->discipline->kick_validate(device);
+		}
+		if (path_event[chp] & PE_PATH_FCES_EVENT) {
+			dasd_path_fcsec_update(device, chp);
+			dasd_schedule_device_bh(device);
+		}
+	}
+	hpfpm = dasd_path_get_hpfpm(device);
+	ifccpm = dasd_path_get_ifccpm(device);
+	if (!dasd_path_get_opm(device) && hpfpm) {
+		/*
+		 * device has no operational paths but at least one path is
+		 * disabled due to HPF errors
+		 * disable HPF at all and use the path(s) again
+		 */
+		if (device->discipline->disable_hpf)
+			device->discipline->disable_hpf(device);
+		dasd_device_set_stop_bits(device, DASD_STOPPED_NOT_ACC);
+		dasd_path_set_tbvpm(device, hpfpm);
+		dasd_schedule_device_bh(device);
+		dasd_schedule_requeue(device);
+	} else if (!dasd_path_get_opm(device) && ifccpm) {
+		/*
+		 * device has no operational paths but at least one path is
+		 * disabled due to IFCC errors
+		 * trigger path verification on paths with IFCC errors
+		 */
+		dasd_path_set_tbvpm(device, ifccpm);
+		dasd_schedule_device_bh(device);
+	}
+	if (oldopm && !dasd_path_get_opm(device) && !hpfpm && !ifccpm) {
+		dev_warn(&device->cdev->dev,
+			 "No verified channel paths remain for the device\n");
+		DBF_DEV_EVENT(DBF_WARNING, device,
+			      "%s", "last verified path gone");
+		/* First call extended error reporting and check for autoquiesce. */
+		dasd_handle_autoquiesce(device, NULL, DASD_EER_NOPATH);
+		dasd_device_set_stop_bits(device,
+					  DASD_STOPPED_DC_WAIT);
+	}
+	dasd_put_device(device);
+}
+EXPORT_SYMBOL_GPL(dasd_generic_path_event);
+
+int dasd_generic_verify_path(struct dasd_device *device, __u8 lpm)
+{
+	if (!dasd_path_get_opm(device) && lpm) {
+		dasd_path_set_opm(device, lpm);
+		dasd_generic_path_operational(device);
+	} else
+		dasd_path_add_opm(device, lpm);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(dasd_generic_verify_path);
+
+void dasd_generic_space_exhaust(struct dasd_device *device,
+				struct dasd_ccw_req *cqr)
+{
+	/* First call extended error reporting and check for autoquiesce. */
+	dasd_handle_autoquiesce(device, NULL, DASD_EER_NOSPC);
+
+	if (device->state < DASD_STATE_BASIC)
+		return;
+
+	if (cqr->status == DASD_CQR_IN_IO ||
+	    cqr->status == DASD_CQR_CLEAR_PENDING) {
+		cqr->status = DASD_CQR_QUEUED;
+		cqr->retries++;
+	}
+	dasd_device_set_stop_bits(device, DASD_STOPPED_NOSPC);
+	dasd_device_clear_timer(device);
+	dasd_schedule_device_bh(device);
+}
+EXPORT_SYMBOL_GPL(dasd_generic_space_exhaust);
+
+void dasd_generic_space_avail(struct dasd_device *device)
+{
+	dev_info(&device->cdev->dev, "Extent pool space is available\n");
+	DBF_DEV_EVENT(DBF_WARNING, device, "%s", "space available");
+
+	dasd_device_remove_stop_bits(device, DASD_STOPPED_NOSPC);
+	dasd_schedule_device_bh(device);
+
+	if (device->block) {
+		dasd_schedule_block_bh(device->block);
+		if (device->block->gdp)
+			blk_mq_run_hw_queues(device->block->gdp->queue, true);
+	}
+	if (!device->stopped)
+		wake_up(&generic_waitq);
+}
+EXPORT_SYMBOL_GPL(dasd_generic_space_avail);
+
+/*
+ * clear active requests and requeue them to block layer if possible
+ */
+int dasd_generic_requeue_all_requests(struct dasd_device *device)
+{
+	struct dasd_block *block = device->block;
+	struct list_head requeue_queue;
+	struct dasd_ccw_req *cqr, *n;
+	int rc;
+
+	if (!block)
+		return 0;
+
+	INIT_LIST_HEAD(&requeue_queue);
+	rc = _dasd_requests_to_flushqueue(block, &requeue_queue);
+
+	/* Now call the callback function of flushed requests */
+restart_cb:
+	list_for_each_entry_safe(cqr, n, &requeue_queue, blocklist) {
+		wait_event(dasd_flush_wq, (cqr->status < DASD_CQR_QUEUED));
+		/* Process finished ERP request. */
+		if (cqr->refers) {
+			spin_lock_bh(&block->queue_lock);
+			__dasd_process_erp(block->base, cqr);
+			spin_unlock_bh(&block->queue_lock);
+			/* restart list_for_xx loop since dasd_process_erp
+			 * might remove multiple elements
+			 */
+			goto restart_cb;
+		}
+		/*
+		 * An aborted request was replaced by a full-track write and is
+		 * retired by that replacement; do not requeue it, just release
+		 * it (mirrors the DASD_CQR_ABORTED handling in
+		 * __dasd_cleanup_cqr()).
+		 */
+		if (cqr->status == DASD_CQR_ABORTED) {
+			struct request *req = cqr->callback_data;
+
+			list_del_init(&cqr->blocklist);
+			cqr->block->base->discipline->free_cp(cqr, req);
+			continue;
+		}
+		_dasd_requeue_request(cqr);
+		list_del_init(&cqr->blocklist);
+		cqr->block->base->discipline->free_cp(
+			cqr, (struct request *) cqr->callback_data);
+	}
+	dasd_schedule_device_bh(device);
+	return rc;
+}
+EXPORT_SYMBOL_GPL(dasd_generic_requeue_all_requests);
+
+static void do_requeue_requests(struct work_struct *work)
+{
+	struct dasd_device *device = container_of(work, struct dasd_device,
+						  requeue_requests);
+	dasd_generic_requeue_all_requests(device);
+	dasd_device_remove_stop_bits(device, DASD_STOPPED_NOT_ACC);
+	if (device->block)
+		dasd_schedule_block_bh(device->block);
+	dasd_put_device(device);
+}
+
+void dasd_schedule_requeue(struct dasd_device *device)
+{
+	dasd_get_device(device);
+	/* queue call to dasd_reload_device to the kernel event daemon. */
+	if (!schedule_work(&device->requeue_requests))
+		dasd_put_device(device);
+}
+EXPORT_SYMBOL(dasd_schedule_requeue);
+
+static int dasd_handle_autoquiesce(struct dasd_device *device,
+				   struct dasd_ccw_req *cqr,
+				   unsigned int reason)
+{
+	/* in any case write eer message with reason */
+	if (dasd_eer_enabled(device))
+		dasd_eer_write(device, cqr, reason);
+
+	if (!test_bit(reason, &device->aq_mask))
+		return 0;
+
+	/* notify eer about autoquiesce */
+	if (dasd_eer_enabled(device))
+		dasd_eer_write(device, NULL, DASD_EER_AUTOQUIESCE);
+
+	dev_info(&device->cdev->dev,
+		 "The DASD has been put in the quiesce state\n");
+	dasd_device_set_stop_bits(device, DASD_STOPPED_QUIESCE);
+
+	if (device->features & DASD_FEATURE_REQUEUEQUIESCE)
+		dasd_schedule_requeue(device);
+
+	return 1;
+}
+
+static struct dasd_ccw_req *dasd_generic_build_rdc(struct dasd_device *device,
+						   int rdc_buffer_size,
+						   int magic)
+{
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+
+	cqr = dasd_smalloc_request(magic, 1 /* RDC */, rdc_buffer_size, device,
+				   NULL);
+
+	if (IS_ERR(cqr)) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+				"Could not allocate RDC request");
+		return cqr;
+	}
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = CCW_CMD_RDC;
+	ccw->cda = virt_to_dma32(cqr->data);
+	ccw->flags = 0;
+	ccw->count = rdc_buffer_size;
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->expires = 10*HZ;
+	cqr->retries = 256;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	return cqr;
+}
+
+
+int dasd_generic_read_dev_chars(struct dasd_device *device, int magic,
+				void *rdc_buffer, int rdc_buffer_size)
+{
+	int ret;
+	struct dasd_ccw_req *cqr;
+
+	cqr = dasd_generic_build_rdc(device, rdc_buffer_size, magic);
+	if (IS_ERR(cqr))
+		return PTR_ERR(cqr);
+
+	ret = dasd_sleep_on(cqr);
+	if (ret == 0)
+		memcpy(rdc_buffer, cqr->data, rdc_buffer_size);
+	dasd_sfree_request(cqr, cqr->memdev);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(dasd_generic_read_dev_chars);
+
+/*
+ *   In command mode and transport mode we need to look for sense
+ *   data in different places. The sense data itself is allways
+ *   an array of 32 bytes, so we can unify the sense data access
+ *   for both modes.
+ */
+char *dasd_get_sense(struct irb *irb)
+{
+	struct tsb *tsb = NULL;
+	char *sense = NULL;
+
+	if (scsw_is_tm(&irb->scsw) && (irb->scsw.tm.fcxs == 0x01)) {
+		if (irb->scsw.tm.tcw)
+			tsb = tcw_get_tsb(dma32_to_virt(irb->scsw.tm.tcw));
+		if (tsb && tsb->length == 64 && tsb->flags)
+			switch (tsb->flags & 0x07) {
+			case 1:	/* tsa_iostat */
+				sense = tsb->tsa.iostat.sense;
+				break;
+			case 2: /* tsa_ddpc */
+				sense = tsb->tsa.ddpc.sense;
+				break;
+			default:
+				/* currently we don't use interrogate data */
+				break;
+			}
+	} else if (irb->esw.esw0.erw.cons) {
+		sense = irb->ecw;
+	}
+	return sense;
+}
+EXPORT_SYMBOL_GPL(dasd_get_sense);
+
+void dasd_generic_shutdown(struct ccw_device *cdev)
+{
+	struct dasd_device *device;
+
+	device = dasd_device_from_cdev(cdev);
+	if (IS_ERR(device))
+		return;
+
+	if (device->block)
+		dasd_schedule_block_bh(device->block);
+
+	dasd_schedule_device_bh(device);
+
+	wait_event(shutdown_waitq, _wait_for_empty_queues(device));
+}
+EXPORT_SYMBOL_GPL(dasd_generic_shutdown);
+
+static int __init dasd_init(void)
+{
+	int rc;
+
+	init_waitqueue_head(&dasd_init_waitq);
+	init_waitqueue_head(&dasd_flush_wq);
+	init_waitqueue_head(&generic_waitq);
+	init_waitqueue_head(&shutdown_waitq);
+
+	/* register 'common' DASD debug area, used for all DBF_XXX calls */
+	dasd_debug_area = debug_register("dasd", 1, 1, 8 * sizeof(long));
+	if (dasd_debug_area == NULL) {
+		rc = -ENOMEM;
+		goto failed;
+	}
+	debug_register_view(dasd_debug_area, &debug_sprintf_view);
+	debug_set_level(dasd_debug_area, DBF_WARNING);
+
+	DBF_EVENT(DBF_EMERG, "%s", "debug area created");
+
+	dasd_diag_discipline_pointer = NULL;
+
+	dasd_statistics_createroot();
+
+	rc = dasd_devmap_init();
+	if (rc)
+		goto failed;
+	rc = dasd_gendisk_init();
+	if (rc)
+		goto failed;
+	rc = dasd_parse();
+	if (rc)
+		goto failed;
+	rc = dasd_eer_init();
+	if (rc)
+		goto failed;
+#ifdef CONFIG_PROC_FS
+	rc = dasd_proc_init();
+	if (rc)
+		goto failed;
+#endif
+
+	return 0;
+failed:
+	pr_info("The DASD device driver could not be initialized\n");
+	dasd_exit();
+	return rc;
+}
+
+module_init(dasd_init);
+module_exit(dasd_exit);
diff --git a/drivers/s390/block/dasd_3990_erp.c b/drivers/s390/block/dasd_3990_erp.c
new file mode 100644
index 000000000..736459477
--- /dev/null
+++ b/drivers/s390/block/dasd_3990_erp.c
@@ -0,0 +1,2816 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Author(s)......: Horst  Hummel    <Horst.Hummel@de.ibm.com>
+ *		    Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 2000, 2001
+ *
+ */
+
+#include <linux/timer.h>
+#include <asm/idals.h>
+
+#include "dasd_int.h"
+#include "dasd_eckd.h"
+
+
+struct DCTL_data {
+	unsigned char subcommand;  /* e.g Inhibit Write, Enable Write,... */
+	unsigned char modifier;	   /* Subcommand modifier */
+	unsigned short res;	   /* reserved */
+} __attribute__ ((packed));
+
+/*
+ *****************************************************************************
+ * SECTION ERP HANDLING
+ *****************************************************************************
+ */
+/*
+ *****************************************************************************
+ * 24 and 32 byte sense ERP functions
+ *****************************************************************************
+ */
+
+/*
+ * DASD_3990_ERP_CLEANUP
+ *
+ * DESCRIPTION
+ *   Removes the already build but not necessary ERP request and sets
+ *   the status of the original cqr / erp to the given (final) status
+ *
+ *  PARAMETER
+ *   erp		request to be blocked
+ *   final_status	either DASD_CQR_DONE or DASD_CQR_FAILED
+ *
+ * RETURN VALUES
+ *   cqr		original cqr
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_cleanup(struct dasd_ccw_req * erp, char final_status)
+{
+	struct dasd_ccw_req *cqr = erp->refers;
+
+	dasd_free_erp_request(erp, erp->memdev);
+	cqr->status = final_status;
+	return cqr;
+
+}				/* end dasd_3990_erp_cleanup */
+
+/*
+ * DASD_3990_ERP_BLOCK_QUEUE
+ *
+ * DESCRIPTION
+ *   Block the given device request queue to prevent from further
+ *   processing until the started timer has expired or an related
+ *   interrupt was received.
+ */
+static void dasd_3990_erp_block_queue(struct dasd_ccw_req *erp, int expires)
+{
+
+	struct dasd_device *device = erp->startdev;
+	unsigned long flags;
+
+	DBF_DEV_EVENT(DBF_INFO, device,
+		    "blocking request queue for %is", expires/HZ);
+
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	dasd_device_set_stop_bits(device, DASD_STOPPED_PENDING);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	erp->status = DASD_CQR_FILLED;
+	if (erp->block)
+		dasd_block_set_timer(erp->block, expires);
+	else
+		dasd_device_set_timer(device, expires);
+}
+
+/*
+ * DASD_3990_ERP_INT_REQ
+ *
+ * DESCRIPTION
+ *   Handles 'Intervention Required' error.
+ *   This means either device offline or not installed.
+ *
+ * PARAMETER
+ *   erp		current erp
+ * RETURN VALUES
+ *   erp		modified erp
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_int_req(struct dasd_ccw_req * erp)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	/* first time set initial retry counter and erp_function */
+	/* and retry once without blocking queue		 */
+	/* (this enables easier enqueing of the cqr)		 */
+	if (erp->function != dasd_3990_erp_int_req) {
+
+		erp->retries = 256;
+		erp->function = dasd_3990_erp_int_req;
+
+	} else {
+
+		/* issue a message and wait for 'device ready' interrupt */
+		dev_err(&device->cdev->dev,
+			    "is offline or not installed - "
+			    "INTERVENTION REQUIRED!!\n");
+
+		dasd_3990_erp_block_queue(erp, 60*HZ);
+	}
+
+	return erp;
+
+}				/* end dasd_3990_erp_int_req */
+
+/*
+ * DASD_3990_ERP_ALTERNATE_PATH
+ *
+ * DESCRIPTION
+ *   Repeat the operation on a different channel path.
+ *   If all alternate paths have been tried, the request is posted with a
+ *   permanent error.
+ *
+ *  PARAMETER
+ *   erp		pointer to the current ERP
+ *
+ * RETURN VALUES
+ *   erp		modified pointer to the ERP
+ */
+static void
+dasd_3990_erp_alternate_path(struct dasd_ccw_req * erp)
+{
+	struct dasd_device *device = erp->startdev;
+	__u8 opm;
+	unsigned long flags;
+
+	/* try alternate valid path */
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	opm = ccw_device_get_path_mask(device->cdev);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	if (erp->lpm == 0)
+		erp->lpm = dasd_path_get_opm(device) &
+			~(erp->irb.esw.esw0.sublog.lpum);
+	else
+		erp->lpm &= ~(erp->irb.esw.esw0.sublog.lpum);
+
+	if ((erp->lpm & opm) != 0x00) {
+
+		DBF_DEV_EVENT(DBF_WARNING, device,
+			    "try alternate lpm=%x (lpum=%x / opm=%x)",
+			    erp->lpm, erp->irb.esw.esw0.sublog.lpum, opm);
+
+		/* reset status to submit the request again... */
+		erp->status = DASD_CQR_FILLED;
+		erp->retries = 10;
+	} else {
+		dev_err(&device->cdev->dev,
+			"The DASD cannot be reached on any path (lpum=%x"
+			"/opm=%x)\n", erp->irb.esw.esw0.sublog.lpum, opm);
+
+		/* post request with permanent error */
+		erp->status = DASD_CQR_FAILED;
+	}
+}				/* end dasd_3990_erp_alternate_path */
+
+/*
+ * DASD_3990_ERP_DCTL
+ *
+ * DESCRIPTION
+ *   Setup cqr to do the Diagnostic Control (DCTL) command with an
+ *   Inhibit Write subcommand (0x20) and the given modifier.
+ *
+ *  PARAMETER
+ *   erp		pointer to the current (failed) ERP
+ *   modifier		subcommand modifier
+ *
+ * RETURN VALUES
+ *   dctl_cqr		pointer to NEW dctl_cqr
+ *
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_DCTL(struct dasd_ccw_req * erp, char modifier)
+{
+
+	struct dasd_device *device = erp->startdev;
+	struct DCTL_data *DCTL_data;
+	struct ccw1 *ccw;
+	struct dasd_ccw_req *dctl_cqr;
+
+	dctl_cqr = dasd_alloc_erp_request(erp->magic, 1,
+					  sizeof(struct DCTL_data),
+					  device);
+	if (IS_ERR(dctl_cqr)) {
+		dev_err(&device->cdev->dev,
+			    "Unable to allocate DCTL-CQR\n");
+		erp->status = DASD_CQR_FAILED;
+		return erp;
+	}
+
+	DCTL_data = dctl_cqr->data;
+
+	DCTL_data->subcommand = 0x02;	/* Inhibit Write */
+	DCTL_data->modifier = modifier;
+
+	ccw = dctl_cqr->cpaddr;
+	memset(ccw, 0, sizeof(struct ccw1));
+	ccw->cmd_code = CCW_CMD_DCTL;
+	ccw->count = 4;
+	ccw->cda = virt_to_dma32(DCTL_data);
+	dctl_cqr->flags = erp->flags;
+	dctl_cqr->function = dasd_3990_erp_DCTL;
+	dctl_cqr->refers = erp;
+	dctl_cqr->startdev = device;
+	dctl_cqr->memdev = device;
+	dctl_cqr->magic = erp->magic;
+	dctl_cqr->expires = 5 * 60 * HZ;
+	dctl_cqr->retries = 2;
+
+	dctl_cqr->buildclk = get_tod_clock();
+
+	dctl_cqr->status = DASD_CQR_FILLED;
+
+	return dctl_cqr;
+
+}				/* end dasd_3990_erp_DCTL */
+
+/*
+ * DASD_3990_ERP_ACTION_1
+ *
+ * DESCRIPTION
+ *   Setup ERP to do the ERP action 1 (see Reference manual).
+ *   Repeat the operation on a different channel path.
+ *   As deviation from the recommended recovery action, we reset the path mask
+ *   after we have tried each path and go through all paths a second time.
+ *   This will cover situations where only one path at a time is actually down,
+ *   but all paths fail and recover just with the same sequence and timing as
+ *   we try to use them (flapping links).
+ *   If all alternate paths have been tried twice, the request is posted with
+ *   a permanent error.
+ *
+ *  PARAMETER
+ *   erp		pointer to the current ERP
+ *
+ * RETURN VALUES
+ *   erp		pointer to the ERP
+ *
+ */
+static struct dasd_ccw_req *dasd_3990_erp_action_1_sec(struct dasd_ccw_req *erp)
+{
+	erp->function = dasd_3990_erp_action_1_sec;
+	dasd_3990_erp_alternate_path(erp);
+	return erp;
+}
+
+static struct dasd_ccw_req *dasd_3990_erp_action_1(struct dasd_ccw_req *erp)
+{
+	erp->function = dasd_3990_erp_action_1;
+	dasd_3990_erp_alternate_path(erp);
+	if (erp->status == DASD_CQR_FAILED &&
+	    !test_bit(DASD_CQR_VERIFY_PATH, &erp->flags)) {
+		erp->status = DASD_CQR_FILLED;
+		erp->retries = 10;
+		erp->lpm = dasd_path_get_opm(erp->startdev);
+		erp->function = dasd_3990_erp_action_1_sec;
+	}
+	return erp;
+}				/* end dasd_3990_erp_action_1(b) */
+
+/*
+ * DASD_3990_ERP_ACTION_4
+ *
+ * DESCRIPTION
+ *   Setup ERP to do the ERP action 4 (see Reference manual).
+ *   Set the current request to PENDING to block the CQR queue for that device
+ *   until the state change interrupt appears.
+ *   Use a timer (20 seconds) to retry the cqr if the interrupt is still
+ *   missing.
+ *
+ *  PARAMETER
+ *   sense		sense data of the actual error
+ *   erp		pointer to the current ERP
+ *
+ * RETURN VALUES
+ *   erp		pointer to the ERP
+ *
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_action_4(struct dasd_ccw_req * erp, char *sense)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	/* first time set initial retry counter and erp_function    */
+	/* and retry once without waiting for state change pending  */
+	/* interrupt (this enables easier enqueing of the cqr)	    */
+	if (erp->function != dasd_3990_erp_action_4) {
+
+		DBF_DEV_EVENT(DBF_INFO, device, "%s",
+			    "dasd_3990_erp_action_4: first time retry");
+
+		erp->retries = 256;
+		erp->function = dasd_3990_erp_action_4;
+
+	} else {
+		if (sense && (sense[25] == 0x1D)) { /* state change pending */
+
+			DBF_DEV_EVENT(DBF_INFO, device,
+				    "waiting for state change pending "
+				    "interrupt, %d retries left",
+				    erp->retries);
+
+			dasd_3990_erp_block_queue(erp, 30*HZ);
+
+		} else if (sense && (sense[25] == 0x1E)) {	/* busy */
+			DBF_DEV_EVENT(DBF_INFO, device,
+				    "busy - redriving request later, "
+				    "%d retries left",
+				    erp->retries);
+                        dasd_3990_erp_block_queue(erp, HZ);
+		} else {
+			/* no state change pending - retry */
+			DBF_DEV_EVENT(DBF_INFO, device,
+				     "redriving request immediately, "
+				     "%d retries left",
+				     erp->retries);
+			erp->status = DASD_CQR_FILLED;
+		}
+	}
+
+	return erp;
+
+}				/* end dasd_3990_erp_action_4 */
+
+/*
+ *****************************************************************************
+ * 24 byte sense ERP functions (only)
+ *****************************************************************************
+ */
+
+/*
+ * DASD_3990_ERP_ACTION_5
+ *
+ * DESCRIPTION
+ *   Setup ERP to do the ERP action 5 (see Reference manual).
+ *   NOTE: Further handling is done in xxx_further_erp after the retries.
+ *
+ *  PARAMETER
+ *   erp		pointer to the current ERP
+ *
+ * RETURN VALUES
+ *   erp		pointer to the ERP
+ *
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_action_5(struct dasd_ccw_req * erp)
+{
+
+	/* first of all retry */
+	erp->retries = 10;
+	erp->function = dasd_3990_erp_action_5;
+
+	return erp;
+
+}				/* end dasd_3990_erp_action_5 */
+
+/*
+ * DASD_3990_HANDLE_ENV_DATA
+ *
+ * DESCRIPTION
+ *   Handles 24 byte 'Environmental data present'.
+ *   Does a analysis of the sense data (message Format)
+ *   and prints the error messages.
+ *
+ * PARAMETER
+ *   sense		current sense data
+ *
+ * RETURN VALUES
+ *   void
+ */
+static void
+dasd_3990_handle_env_data(struct dasd_ccw_req * erp, char *sense)
+{
+
+	struct dasd_device *device = erp->startdev;
+	char msg_format = (sense[7] & 0xF0);
+	char msg_no = (sense[7] & 0x0F);
+
+	switch (msg_format) {
+	case 0x00:		/* Format 0 - Program or System Checks */
+
+		if (sense[1] & 0x10) {	/* check message to operator bit */
+
+			switch (msg_no) {
+			case 0x00:	/* No Message */
+				break;
+			case 0x01:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Invalid Command\n");
+				break;
+			case 0x02:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Invalid Command "
+					    "Sequence\n");
+				break;
+			case 0x03:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - CCW Count less than "
+					    "required\n");
+				break;
+			case 0x04:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Invalid Parameter\n");
+				break;
+			case 0x05:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Diagnostic of Special"
+					    " Command Violates File Mask\n");
+				break;
+			case 0x07:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Channel Returned with "
+					    "Incorrect retry CCW\n");
+				break;
+			case 0x08:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Reset Notification\n");
+				break;
+			case 0x09:
+				dev_warn(&device->cdev->dev,
+					 "FORMAT 0 - Storage Path Restart\n");
+				break;
+			case 0x0A:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Channel requested "
+					    "... %02x\n", sense[8]);
+				break;
+			case 0x0B:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Invalid Defective/"
+					    "Alternate Track Pointer\n");
+				break;
+			case 0x0C:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - DPS Installation "
+					    "Check\n");
+				break;
+			case 0x0E:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Command Invalid on "
+					    "Secondary Address\n");
+				break;
+			case 0x0F:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Status Not As "
+					    "Required: reason %02x\n",
+					 sense[8]);
+				break;
+			default:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Reserved\n");
+			}
+		} else {
+			switch (msg_no) {
+			case 0x00:	/* No Message */
+				break;
+			case 0x01:
+				dev_warn(&device->cdev->dev,
+					 "FORMAT 0 - Device Error "
+					 "Source\n");
+				break;
+			case 0x02:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Reserved\n");
+				break;
+			case 0x03:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Device Fenced - "
+					    "device = %02x\n", sense[4]);
+				break;
+			case 0x04:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Data Pinned for "
+					    "Device\n");
+				break;
+			default:
+				dev_warn(&device->cdev->dev,
+					    "FORMAT 0 - Reserved\n");
+			}
+		}
+		break;
+
+	case 0x10:		/* Format 1 - Device Equipment Checks */
+		switch (msg_no) {
+		case 0x00:	/* No Message */
+			break;
+		case 0x01:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Device Status 1 not as "
+				    "expected\n");
+			break;
+		case 0x03:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Index missing\n");
+			break;
+		case 0x04:
+			dev_warn(&device->cdev->dev,
+				 "FORMAT 1 - Interruption cannot be "
+				 "reset\n");
+			break;
+		case 0x05:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Device did not respond to "
+				    "selection\n");
+			break;
+		case 0x06:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Device check-2 error or Set "
+				    "Sector is not complete\n");
+			break;
+		case 0x07:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Head address does not "
+				    "compare\n");
+			break;
+		case 0x08:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Device status 1 not valid\n");
+			break;
+		case 0x09:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Device not ready\n");
+			break;
+		case 0x0A:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Track physical address did "
+				    "not compare\n");
+			break;
+		case 0x0B:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Missing device address bit\n");
+			break;
+		case 0x0C:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Drive motor switch is off\n");
+			break;
+		case 0x0D:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Seek incomplete\n");
+			break;
+		case 0x0E:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Cylinder address did not "
+				    "compare\n");
+			break;
+		case 0x0F:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Offset active cannot be "
+				    "reset\n");
+			break;
+		default:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 1 - Reserved\n");
+		}
+		break;
+
+	case 0x20:		/* Format 2 - 3990 Equipment Checks */
+		switch (msg_no) {
+		case 0x08:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 2 - 3990 check-2 error\n");
+			break;
+		case 0x0E:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 2 - Support facility errors\n");
+			break;
+		case 0x0F:
+			dev_warn(&device->cdev->dev,
+				 "FORMAT 2 - Microcode detected error "
+				 "%02x\n",
+				 sense[8]);
+			break;
+		default:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 2 - Reserved\n");
+		}
+		break;
+
+	case 0x30:		/* Format 3 - 3990 Control Checks */
+		switch (msg_no) {
+		case 0x0F:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 3 - Allegiance terminated\n");
+			break;
+		default:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 3 - Reserved\n");
+		}
+		break;
+
+	case 0x40:		/* Format 4 - Data Checks */
+		switch (msg_no) {
+		case 0x00:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - Home address area error\n");
+			break;
+		case 0x01:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - Count area error\n");
+			break;
+		case 0x02:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - Key area error\n");
+			break;
+		case 0x03:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - Data area error\n");
+			break;
+		case 0x04:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - No sync byte in home address "
+				    "area\n");
+			break;
+		case 0x05:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - No sync byte in count address "
+				    "area\n");
+			break;
+		case 0x06:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - No sync byte in key area\n");
+			break;
+		case 0x07:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - No sync byte in data area\n");
+			break;
+		case 0x08:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - Home address area error; "
+				    "offset active\n");
+			break;
+		case 0x09:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - Count area error; offset "
+				    "active\n");
+			break;
+		case 0x0A:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - Key area error; offset "
+				    "active\n");
+			break;
+		case 0x0B:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - Data area error; "
+				    "offset active\n");
+			break;
+		case 0x0C:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - No sync byte in home "
+				    "address area; offset active\n");
+			break;
+		case 0x0D:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - No sync byte in count "
+				    "address area; offset active\n");
+			break;
+		case 0x0E:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - No sync byte in key area; "
+				    "offset active\n");
+			break;
+		case 0x0F:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - No sync byte in data area; "
+				    "offset active\n");
+			break;
+		default:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 4 - Reserved\n");
+		}
+		break;
+
+	case 0x50:  /* Format 5 - Data Check with displacement information */
+		switch (msg_no) {
+		case 0x00:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 5 - Data Check in the "
+				    "home address area\n");
+			break;
+		case 0x01:
+			dev_warn(&device->cdev->dev,
+				 "FORMAT 5 - Data Check in the count "
+				 "area\n");
+			break;
+		case 0x02:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 5 - Data Check in the key area\n");
+			break;
+		case 0x03:
+			dev_warn(&device->cdev->dev,
+				 "FORMAT 5 - Data Check in the data "
+				 "area\n");
+			break;
+		case 0x08:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 5 - Data Check in the "
+				    "home address area; offset active\n");
+			break;
+		case 0x09:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 5 - Data Check in the count area; "
+				    "offset active\n");
+			break;
+		case 0x0A:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 5 - Data Check in the key area; "
+				    "offset active\n");
+			break;
+		case 0x0B:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 5 - Data Check in the data area; "
+				    "offset active\n");
+			break;
+		default:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 5 - Reserved\n");
+		}
+		break;
+
+	case 0x60:  /* Format 6 - Usage Statistics/Overrun Errors */
+		switch (msg_no) {
+		case 0x00:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 6 - Overrun on channel A\n");
+			break;
+		case 0x01:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 6 - Overrun on channel B\n");
+			break;
+		case 0x02:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 6 - Overrun on channel C\n");
+			break;
+		case 0x03:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 6 - Overrun on channel D\n");
+			break;
+		case 0x04:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 6 - Overrun on channel E\n");
+			break;
+		case 0x05:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 6 - Overrun on channel F\n");
+			break;
+		case 0x06:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 6 - Overrun on channel G\n");
+			break;
+		case 0x07:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 6 - Overrun on channel H\n");
+			break;
+		default:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 6 - Reserved\n");
+		}
+		break;
+
+	case 0x70:  /* Format 7 - Device Connection Control Checks */
+		switch (msg_no) {
+		case 0x00:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - RCC initiated by a connection "
+				    "check alert\n");
+			break;
+		case 0x01:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - RCC 1 sequence not "
+				    "successful\n");
+			break;
+		case 0x02:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - RCC 1 and RCC 2 sequences not "
+				    "successful\n");
+			break;
+		case 0x03:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - Invalid tag-in during "
+				    "selection sequence\n");
+			break;
+		case 0x04:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - extra RCC required\n");
+			break;
+		case 0x05:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - Invalid DCC selection "
+				    "response or timeout\n");
+			break;
+		case 0x06:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - Missing end operation; device "
+				    "transfer complete\n");
+			break;
+		case 0x07:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - Missing end operation; device "
+				    "transfer incomplete\n");
+			break;
+		case 0x08:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - Invalid tag-in for an "
+				    "immediate command sequence\n");
+			break;
+		case 0x09:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - Invalid tag-in for an "
+				    "extended command sequence\n");
+			break;
+		case 0x0A:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - 3990 microcode time out when "
+				    "stopping selection\n");
+			break;
+		case 0x0B:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - No response to selection "
+				    "after a poll interruption\n");
+			break;
+		case 0x0C:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - Permanent path error (DASD "
+				    "controller not available)\n");
+			break;
+		case 0x0D:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - DASD controller not available"
+				    " on disconnected command chain\n");
+			break;
+		default:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 7 - Reserved\n");
+		}
+		break;
+
+	case 0x80:  /* Format 8 - Additional Device Equipment Checks */
+		switch (msg_no) {
+		case 0x00:	/* No Message */
+		case 0x01:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 8 - Error correction code "
+				    "hardware fault\n");
+			break;
+		case 0x03:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 8 - Unexpected end operation "
+				    "response code\n");
+			break;
+		case 0x04:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 8 - End operation with transfer "
+				    "count not zero\n");
+			break;
+		case 0x05:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 8 - End operation with transfer "
+				    "count zero\n");
+			break;
+		case 0x06:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 8 - DPS checks after a system "
+				    "reset or selective reset\n");
+			break;
+		case 0x07:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 8 - DPS cannot be filled\n");
+			break;
+		case 0x08:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 8 - Short busy time-out during "
+				    "device selection\n");
+			break;
+		case 0x09:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 8 - DASD controller failed to "
+				    "set or reset the long busy latch\n");
+			break;
+		case 0x0A:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 8 - No interruption from device "
+				    "during a command chain\n");
+			break;
+		default:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 8 - Reserved\n");
+		}
+		break;
+
+	case 0x90:  /* Format 9 - Device Read, Write, and Seek Checks */
+		switch (msg_no) {
+		case 0x00:
+			break;	/* No Message */
+		case 0x06:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 9 - Device check-2 error\n");
+			break;
+		case 0x07:
+			dev_warn(&device->cdev->dev,
+				 "FORMAT 9 - Head address did not "
+				 "compare\n");
+			break;
+		case 0x0A:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 9 - Track physical address did "
+				    "not compare while oriented\n");
+			break;
+		case 0x0E:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 9 - Cylinder address did not "
+				    "compare\n");
+			break;
+		default:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT 9 - Reserved\n");
+		}
+		break;
+
+	case 0xF0:		/* Format F - Cache Storage Checks */
+		switch (msg_no) {
+		case 0x00:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Operation Terminated\n");
+			break;
+		case 0x01:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Subsystem Processing Error\n");
+			break;
+		case 0x02:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Cache or nonvolatile storage "
+				    "equipment failure\n");
+			break;
+		case 0x04:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Caching terminated\n");
+			break;
+		case 0x06:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Cache fast write access not "
+				    "authorized\n");
+			break;
+		case 0x07:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Track format incorrect\n");
+			break;
+		case 0x09:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Caching reinitiated\n");
+			break;
+		case 0x0A:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Nonvolatile storage "
+				    "terminated\n");
+			break;
+		case 0x0B:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Volume is suspended duplex\n");
+			/* call extended error reporting (EER) */
+			dasd_eer_write(device, erp->refers,
+				       DASD_EER_PPRCSUSPEND);
+			break;
+		case 0x0C:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Subsystem status cannot be "
+				    "determined\n");
+			break;
+		case 0x0D:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Caching status reset to "
+				    "default\n");
+			break;
+		case 0x0E:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - DASD Fast Write inhibited\n");
+			break;
+		default:
+			dev_warn(&device->cdev->dev,
+				    "FORMAT F - Reserved\n");
+		}
+		break;
+
+	default:
+		dev_err(&device->cdev->dev,
+			"Unknown message format %02x", msg_format);
+		break;
+	}			/* end switch message format */
+
+}				/* end dasd_3990_handle_env_data */
+
+/*
+ * DASD_3990_ERP_COM_REJ
+ *
+ * DESCRIPTION
+ *   Handles 24 byte 'Command Reject' error.
+ *
+ * PARAMETER
+ *   erp		current erp_head
+ *   sense		current sense data
+ *
+ * RETURN VALUES
+ *   erp		'new' erp_head - pointer to new ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_com_rej(struct dasd_ccw_req * erp, char *sense)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	erp->function = dasd_3990_erp_com_rej;
+
+	/* env data present (ACTION 10 - retry should work) */
+	if (sense[2] & SNS2_ENV_DATA_PRESENT) {
+
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Command Reject - environmental data present");
+
+		dasd_3990_handle_env_data(erp, sense);
+
+		erp->retries = 5;
+
+	} else if (sense[1] & SNS1_WRITE_INHIBITED) {
+		dev_err(&device->cdev->dev, "An I/O request was rejected"
+			" because writing is inhibited\n");
+		erp = dasd_3990_erp_cleanup(erp, DASD_CQR_FAILED);
+	} else if (sense[7] == SNS7_INVALID_ON_SEC) {
+		dev_err(&device->cdev->dev, "An I/O request was rejected on a copy pair secondary device\n");
+		/* suppress dump of sense data for this error */
+		set_bit(DASD_CQR_SUPPRESS_CR, &erp->refers->flags);
+		erp = dasd_3990_erp_cleanup(erp, DASD_CQR_FAILED);
+	} else {
+		if (!test_bit(DASD_CQR_SUPPRESS_CR, &erp->flags))
+			dev_err(&device->cdev->dev,
+				"An I/O command request was rejected\n");
+
+		erp = dasd_3990_erp_cleanup(erp, DASD_CQR_FAILED);
+	}
+
+	return erp;
+
+}				/* end dasd_3990_erp_com_rej */
+
+/*
+ * DASD_3990_ERP_BUS_OUT
+ *
+ * DESCRIPTION
+ *   Handles 24 byte 'Bus Out Parity Check' error.
+ *
+ * PARAMETER
+ *   erp		current erp_head
+ * RETURN VALUES
+ *   erp		new erp_head - pointer to new ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_bus_out(struct dasd_ccw_req * erp)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	/* first time set initial retry counter and erp_function */
+	/* and retry once without blocking queue		 */
+	/* (this enables easier enqueing of the cqr)		 */
+	if (erp->function != dasd_3990_erp_bus_out) {
+		erp->retries = 256;
+		erp->function = dasd_3990_erp_bus_out;
+
+	} else {
+
+		/* issue a message and wait for 'device ready' interrupt */
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "bus out parity error or BOPC requested by "
+			    "channel");
+
+		dasd_3990_erp_block_queue(erp, 60*HZ);
+
+	}
+
+	return erp;
+
+}				/* end dasd_3990_erp_bus_out */
+
+/*
+ * DASD_3990_ERP_EQUIP_CHECK
+ *
+ * DESCRIPTION
+ *   Handles 24 byte 'Equipment Check' error.
+ *
+ * PARAMETER
+ *   erp		current erp_head
+ * RETURN VALUES
+ *   erp		new erp_head - pointer to new ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_equip_check(struct dasd_ccw_req * erp, char *sense)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	erp->function = dasd_3990_erp_equip_check;
+
+	if (sense[1] & SNS1_WRITE_INHIBITED) {
+		dev_err(&device->cdev->dev, "Write inhibited path encountered\n");
+
+		erp = dasd_3990_erp_action_1(erp);
+
+	} else if (sense[2] & SNS2_ENV_DATA_PRESENT) {
+
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Equipment Check - " "environmental data present");
+
+		dasd_3990_handle_env_data(erp, sense);
+
+		erp = dasd_3990_erp_action_4(erp, sense);
+
+	} else if (sense[1] & SNS1_PERM_ERR) {
+
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Equipment Check - retry exhausted or "
+			    "undesirable");
+
+		erp = dasd_3990_erp_action_1(erp);
+
+	} else {
+		/* all other equipment checks - Action 5 */
+		/* rest is done when retries == 0 */
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Equipment check or processing error");
+
+		erp = dasd_3990_erp_action_5(erp);
+	}
+	return erp;
+
+}				/* end dasd_3990_erp_equip_check */
+
+/*
+ * DASD_3990_ERP_DATA_CHECK
+ *
+ * DESCRIPTION
+ *   Handles 24 byte 'Data Check' error.
+ *
+ * PARAMETER
+ *   erp		current erp_head
+ * RETURN VALUES
+ *   erp		new erp_head - pointer to new ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_data_check(struct dasd_ccw_req * erp, char *sense)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	erp->function = dasd_3990_erp_data_check;
+
+	if (sense[2] & SNS2_CORRECTABLE) {	/* correctable data check */
+
+		/* issue message that the data has been corrected */
+		dev_emerg(&device->cdev->dev,
+			    "Data recovered during retry with PCI "
+			    "fetch mode active\n");
+
+		/* not possible to handle this situation in Linux */
+		panic("No way to inform application about the possibly "
+		      "incorrect data");
+
+	} else if (sense[2] & SNS2_ENV_DATA_PRESENT) {
+
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Uncorrectable data check recovered secondary "
+			    "addr of duplex pair");
+
+		erp = dasd_3990_erp_action_4(erp, sense);
+
+	} else if (sense[1] & SNS1_PERM_ERR) {
+
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Uncorrectable data check with internal "
+			    "retry exhausted");
+
+		erp = dasd_3990_erp_action_1(erp);
+
+	} else {
+		/* all other data checks */
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Uncorrectable data check with retry count "
+			    "exhausted...");
+
+		erp = dasd_3990_erp_action_5(erp);
+	}
+
+	return erp;
+
+}				/* end dasd_3990_erp_data_check */
+
+/*
+ * DASD_3990_ERP_OVERRUN
+ *
+ * DESCRIPTION
+ *   Handles 24 byte 'Overrun' error.
+ *
+ * PARAMETER
+ *   erp		current erp_head
+ * RETURN VALUES
+ *   erp		new erp_head - pointer to new ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_overrun(struct dasd_ccw_req * erp, char *sense)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	erp->function = dasd_3990_erp_overrun;
+
+	DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+		    "Overrun - service overrun or overrun"
+		    " error requested by channel");
+
+	erp = dasd_3990_erp_action_5(erp);
+
+	return erp;
+
+}				/* end dasd_3990_erp_overrun */
+
+/*
+ * DASD_3990_ERP_INV_FORMAT
+ *
+ * DESCRIPTION
+ *   Handles 24 byte 'Invalid Track Format' error.
+ *
+ * PARAMETER
+ *   erp		current erp_head
+ * RETURN VALUES
+ *   erp		new erp_head - pointer to new ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_inv_format(struct dasd_ccw_req * erp, char *sense)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	erp->function = dasd_3990_erp_inv_format;
+
+	if (sense[2] & SNS2_ENV_DATA_PRESENT) {
+
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Track format error when destaging or "
+			    "staging data");
+
+		dasd_3990_handle_env_data(erp, sense);
+
+		erp = dasd_3990_erp_action_4(erp, sense);
+
+	} else {
+		dev_err(&device->cdev->dev, "Track format is not valid\n");
+		erp = dasd_3990_erp_cleanup(erp, DASD_CQR_FAILED);
+	}
+
+	return erp;
+
+}				/* end dasd_3990_erp_inv_format */
+
+/*
+ * DASD_3990_ERP_EOC
+ *
+ * DESCRIPTION
+ *   Handles 24 byte 'End-of-Cylinder' error.
+ *
+ * PARAMETER
+ *   erp		already added default erp
+ * RETURN VALUES
+ *   erp		pointer to original (failed) cqr.
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_EOC(struct dasd_ccw_req * default_erp, char *sense)
+{
+
+	struct dasd_device *device = default_erp->startdev;
+
+	dev_err(&device->cdev->dev,
+		"The cylinder data for accessing the DASD is inconsistent\n");
+
+	/* implement action 7 - BUG */
+	return dasd_3990_erp_cleanup(default_erp, DASD_CQR_FAILED);
+
+}				/* end dasd_3990_erp_EOC */
+
+/*
+ * DASD_3990_ERP_ENV_DATA
+ *
+ * DESCRIPTION
+ *   Handles 24 byte 'Environmental-Data Present' error.
+ *
+ * PARAMETER
+ *   erp		current erp_head
+ * RETURN VALUES
+ *   erp		new erp_head - pointer to new ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_env_data(struct dasd_ccw_req * erp, char *sense)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	erp->function = dasd_3990_erp_env_data;
+
+	DBF_DEV_EVENT(DBF_WARNING, device, "%s", "Environmental data present");
+
+	dasd_3990_handle_env_data(erp, sense);
+
+	/* don't retry on disabled interface */
+	if (sense[7] != 0x0F) {
+		erp = dasd_3990_erp_action_4(erp, sense);
+	} else {
+		erp->status = DASD_CQR_FILLED;
+	}
+
+	return erp;
+
+}				/* end dasd_3990_erp_env_data */
+
+/*
+ * DASD_3990_ERP_NO_REC
+ *
+ * DESCRIPTION
+ *   Handles 24 byte 'No Record Found' error.
+ *
+ * PARAMETER
+ *   erp		already added default ERP
+ *
+ * RETURN VALUES
+ *   erp		new erp_head - pointer to new ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_no_rec(struct dasd_ccw_req * default_erp, char *sense)
+{
+
+	struct dasd_device *device = default_erp->startdev;
+
+	/*
+	 * In some cases the 'No Record Found' error might be expected and
+	 * log messages shouldn't be written then.
+	 * Check if the according suppress bit is set.
+	 */
+	if (!test_bit(DASD_CQR_SUPPRESS_NRF, &default_erp->flags))
+		dev_err(&device->cdev->dev,
+			"The specified record was not found\n");
+
+	return dasd_3990_erp_cleanup(default_erp, DASD_CQR_FAILED);
+
+}				/* end dasd_3990_erp_no_rec */
+
+/*
+ * DASD_3990_ERP_FILE_PROT
+ *
+ * DESCRIPTION
+ *   Handles 24 byte 'File Protected' error.
+ *   Note: Seek related recovery is not implemented because
+ *	   wee don't use the seek command yet.
+ *
+ * PARAMETER
+ *   erp		current erp_head
+ * RETURN VALUES
+ *   erp		new erp_head - pointer to new ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_file_prot(struct dasd_ccw_req * erp)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	dev_err(&device->cdev->dev,
+		"Accessing the DASD failed because of a hardware error\n");
+
+	return dasd_3990_erp_cleanup(erp, DASD_CQR_FAILED);
+
+}				/* end dasd_3990_erp_file_prot */
+
+/*
+ * DASD_3990_ERP_INSPECT_ALIAS
+ *
+ * DESCRIPTION
+ *   Checks if the original request was started on an alias device.
+ *   If yes, it modifies the original and the erp request so that
+ *   the erp request can be started on a base device.
+ *
+ * PARAMETER
+ *   erp		pointer to the currently created default ERP
+ *
+ * RETURN VALUES
+ *   erp		pointer to the modified ERP, or NULL
+ */
+
+static struct dasd_ccw_req *dasd_3990_erp_inspect_alias(
+						struct dasd_ccw_req *erp)
+{
+	struct dasd_ccw_req *cqr = erp->refers;
+	char *sense;
+
+	if (cqr->block &&
+	    (cqr->block->base != cqr->startdev)) {
+
+		sense = dasd_get_sense(&erp->refers->irb);
+		/*
+		 * dynamic pav may have changed base alias mapping
+		 */
+		if (!test_bit(DASD_FLAG_OFFLINE, &cqr->startdev->flags) && sense
+		    && (sense[0] == 0x10) && (sense[7] == 0x0F)
+		    && (sense[8] == 0x67)) {
+			/*
+			 * remove device from alias handling to prevent new
+			 * requests from being scheduled on the
+			 * wrong alias device
+			 */
+			dasd_alias_remove_device(cqr->startdev);
+
+			/* schedule worker to reload device */
+			dasd_reload_device(cqr->startdev);
+		}
+
+		if (cqr->startdev->features & DASD_FEATURE_ERPLOG) {
+			DBF_DEV_EVENT(DBF_ERR, cqr->startdev,
+				    "ERP on alias device for request %p,"
+				    " recover on base device %s", cqr,
+				    dev_name(&cqr->block->base->cdev->dev));
+		}
+		dasd_eckd_reset_ccw_to_base_io(cqr);
+		erp->startdev = cqr->block->base;
+		erp->function = dasd_3990_erp_inspect_alias;
+		return erp;
+	} else
+		return NULL;
+}
+
+
+/*
+ * DASD_3990_ERP_INSPECT_24
+ *
+ * DESCRIPTION
+ *   Does a detailed inspection of the 24 byte sense data
+ *   and sets up a related error recovery action.
+ *
+ * PARAMETER
+ *   sense		sense data of the actual error
+ *   erp		pointer to the currently created default ERP
+ *
+ * RETURN VALUES
+ *   erp		pointer to the (addtitional) ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_inspect_24(struct dasd_ccw_req * erp, char *sense)
+{
+
+	struct dasd_ccw_req *erp_filled = NULL;
+
+	/* Check sense for ....	   */
+	/* 'Command Reject'	   */
+	if ((erp_filled == NULL) && (sense[0] & SNS0_CMD_REJECT)) {
+		erp_filled = dasd_3990_erp_com_rej(erp, sense);
+	}
+	/* 'Intervention Required' */
+	if ((erp_filled == NULL) && (sense[0] & SNS0_INTERVENTION_REQ)) {
+		erp_filled = dasd_3990_erp_int_req(erp);
+	}
+	/* 'Bus Out Parity Check'  */
+	if ((erp_filled == NULL) && (sense[0] & SNS0_BUS_OUT_CHECK)) {
+		erp_filled = dasd_3990_erp_bus_out(erp);
+	}
+	/* 'Equipment Check'	   */
+	if ((erp_filled == NULL) && (sense[0] & SNS0_EQUIPMENT_CHECK)) {
+		erp_filled = dasd_3990_erp_equip_check(erp, sense);
+	}
+	/* 'Data Check'		   */
+	if ((erp_filled == NULL) && (sense[0] & SNS0_DATA_CHECK)) {
+		erp_filled = dasd_3990_erp_data_check(erp, sense);
+	}
+	/* 'Overrun'		   */
+	if ((erp_filled == NULL) && (sense[0] & SNS0_OVERRUN)) {
+		erp_filled = dasd_3990_erp_overrun(erp, sense);
+	}
+	/* 'Invalid Track Format'  */
+	if ((erp_filled == NULL) && (sense[1] & SNS1_INV_TRACK_FORMAT)) {
+		erp_filled = dasd_3990_erp_inv_format(erp, sense);
+	}
+	/* 'End-of-Cylinder'	   */
+	if ((erp_filled == NULL) && (sense[1] & SNS1_EOC)) {
+		erp_filled = dasd_3990_erp_EOC(erp, sense);
+	}
+	/* 'Environmental Data'	   */
+	if ((erp_filled == NULL) && (sense[2] & SNS2_ENV_DATA_PRESENT)) {
+		erp_filled = dasd_3990_erp_env_data(erp, sense);
+	}
+	/* 'No Record Found'	   */
+	if ((erp_filled == NULL) && (sense[1] & SNS1_NO_REC_FOUND)) {
+		erp_filled = dasd_3990_erp_no_rec(erp, sense);
+	}
+	/* 'File Protected'	   */
+	if ((erp_filled == NULL) && (sense[1] & SNS1_FILE_PROTECTED)) {
+		erp_filled = dasd_3990_erp_file_prot(erp);
+	}
+	/* other (unknown) error - do default ERP */
+	if (erp_filled == NULL) {
+
+		erp_filled = erp;
+	}
+
+	return erp_filled;
+
+}				/* END dasd_3990_erp_inspect_24 */
+
+/*
+ *****************************************************************************
+ * 32 byte sense ERP functions (only)
+ *****************************************************************************
+ */
+
+/*
+ * DASD_3990_ERPACTION_10_32
+ *
+ * DESCRIPTION
+ *   Handles 32 byte 'Action 10' of Single Program Action Codes.
+ *   Just retry and if retry doesn't work, return with error.
+ *
+ * PARAMETER
+ *   erp		current erp_head
+ *   sense		current sense data
+ * RETURN VALUES
+ *   erp		modified erp_head
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_action_10_32(struct dasd_ccw_req * erp, char *sense)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	erp->retries = 256;
+	erp->function = dasd_3990_erp_action_10_32;
+
+	DBF_DEV_EVENT(DBF_WARNING, device, "%s", "Perform logging requested");
+
+	return erp;
+
+}				/* end dasd_3990_erp_action_10_32 */
+
+/*
+ * DASD_3990_ERP_ACTION_1B_32
+ *
+ * DESCRIPTION
+ *   Handles 32 byte 'Action 1B' of Single Program Action Codes.
+ *   A write operation could not be finished because of an unexpected
+ *   condition.
+ *   The already created 'default erp' is used to get the link to
+ *   the erp chain, but it can not be used for this recovery
+ *   action because it contains no DE/LO data space.
+ *
+ * PARAMETER
+ *   default_erp	already added default erp.
+ *   sense		current sense data
+ *
+ * RETURN VALUES
+ *   erp		new erp or
+ *			default_erp in case of imprecise ending or error
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_action_1B_32(struct dasd_ccw_req * default_erp, char *sense)
+{
+
+	struct dasd_device *device = default_erp->startdev;
+	dma32_t cpa = 0;
+	struct dasd_ccw_req *cqr;
+	struct dasd_ccw_req *erp;
+	struct DE_eckd_data *DE_data;
+	struct PFX_eckd_data *PFX_data;
+	char *LO_data;		/* LO_eckd_data_t */
+	struct ccw1 *ccw, *oldccw;
+
+	DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+		    "Write not finished because of unexpected condition");
+
+	default_erp->function = dasd_3990_erp_action_1B_32;
+
+	/* determine the original cqr */
+	cqr = default_erp;
+
+	while (cqr->refers != NULL) {
+		cqr = cqr->refers;
+	}
+
+	if (scsw_is_tm(&cqr->irb.scsw)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "32 bit sense, action 1B is not defined"
+			      " in transport mode - just retry");
+		return default_erp;
+	}
+
+	/* for imprecise ending just do default erp */
+	if (sense[1] & 0x01) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Imprecise ending is set - just retry");
+
+		return default_erp;
+	}
+
+	/* determine the address of the CCW to be restarted */
+	/* Imprecise ending is not set -> addr from IRB-SCSW */
+	cpa = default_erp->refers->irb.scsw.cmd.cpa;
+
+	if (cpa == 0) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Unable to determine address of the CCW "
+			    "to be restarted");
+
+		return dasd_3990_erp_cleanup(default_erp, DASD_CQR_FAILED);
+	}
+
+	/* Build new ERP request including DE/LO */
+	erp = dasd_alloc_erp_request(cqr->magic,
+				     2 + 1,/* DE/LO + TIC */
+				     sizeof(struct DE_eckd_data) +
+				     sizeof(struct LO_eckd_data), device);
+
+	if (IS_ERR(erp)) {
+		DBF_DEV_EVENT(DBF_ERR, device, "%s",
+			      "Unable to allocate ERP request (1B 32)");
+		return dasd_3990_erp_cleanup(default_erp, DASD_CQR_FAILED);
+	}
+
+	/* use original DE */
+	DE_data = erp->data;
+	oldccw = cqr->cpaddr;
+	if (oldccw->cmd_code == DASD_ECKD_CCW_PFX) {
+		PFX_data = cqr->data;
+		memcpy(DE_data, &PFX_data->define_extent,
+		       sizeof(struct DE_eckd_data));
+	} else
+		memcpy(DE_data, cqr->data, sizeof(struct DE_eckd_data));
+
+	/* create LO */
+	LO_data = erp->data + sizeof(struct DE_eckd_data);
+
+	if ((sense[3] == 0x01) && (LO_data[1] & 0x01)) {
+		/* should not */
+		return dasd_3990_erp_cleanup(default_erp, DASD_CQR_FAILED);
+	}
+
+	if ((sense[7] & 0x3F) == 0x01) {
+		/* operation code is WRITE DATA -> data area orientation */
+		LO_data[0] = 0x81;
+
+	} else if ((sense[7] & 0x3F) == 0x03) {
+		/* operation code is FORMAT WRITE -> index orientation */
+		LO_data[0] = 0xC3;
+
+	} else {
+		LO_data[0] = sense[7];	/* operation */
+	}
+
+	LO_data[1] = sense[8];	/* auxiliary */
+	LO_data[2] = sense[9];
+	LO_data[3] = sense[3];	/* count */
+	LO_data[4] = sense[29];	/* seek_addr.cyl */
+	LO_data[5] = sense[30];	/* seek_addr.cyl 2nd byte */
+	LO_data[7] = sense[31];	/* seek_addr.head 2nd byte */
+
+	memcpy(&(LO_data[8]), &(sense[11]), 8);
+
+	/* create DE ccw */
+	ccw = erp->cpaddr;
+	memset(ccw, 0, sizeof(struct ccw1));
+	ccw->cmd_code = DASD_ECKD_CCW_DEFINE_EXTENT;
+	ccw->flags = CCW_FLAG_CC;
+	ccw->count = 16;
+	ccw->cda = virt_to_dma32(DE_data);
+
+	/* create LO ccw */
+	ccw++;
+	memset(ccw, 0, sizeof(struct ccw1));
+	ccw->cmd_code = DASD_ECKD_CCW_LOCATE_RECORD;
+	ccw->flags = CCW_FLAG_CC;
+	ccw->count = 16;
+	ccw->cda = virt_to_dma32(LO_data);
+
+	/* TIC to the failed ccw */
+	ccw++;
+	ccw->cmd_code = CCW_CMD_TIC;
+	ccw->cda = cpa;
+
+	/* fill erp related fields */
+	erp->flags = default_erp->flags;
+	erp->function = dasd_3990_erp_action_1B_32;
+	erp->refers = default_erp->refers;
+	erp->startdev = device;
+	erp->memdev = device;
+	erp->magic = default_erp->magic;
+	erp->expires = default_erp->expires;
+	erp->retries = 256;
+	erp->buildclk = get_tod_clock();
+	erp->status = DASD_CQR_FILLED;
+
+	/* remove the default erp */
+	dasd_free_erp_request(default_erp, device);
+
+	return erp;
+
+}				/* end dasd_3990_erp_action_1B_32 */
+
+/*
+ * DASD_3990_UPDATE_1B
+ *
+ * DESCRIPTION
+ *   Handles the update to the 32 byte 'Action 1B' of Single Program
+ *   Action Codes in case the first action was not successful.
+ *   The already created 'previous_erp' is the currently not successful
+ *   ERP.
+ *
+ * PARAMETER
+ *   previous_erp	already created previous erp.
+ *   sense		current sense data
+ * RETURN VALUES
+ *   erp		modified erp
+ */
+static struct dasd_ccw_req *
+dasd_3990_update_1B(struct dasd_ccw_req * previous_erp, char *sense)
+{
+
+	struct dasd_device *device = previous_erp->startdev;
+	dma32_t cpa = 0;
+	struct dasd_ccw_req *cqr;
+	struct dasd_ccw_req *erp;
+	char *LO_data;		/* struct LO_eckd_data */
+	struct ccw1 *ccw;
+
+	DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+		    "Write not finished because of unexpected condition"
+		    " - follow on");
+
+	/* determine the original cqr */
+	cqr = previous_erp;
+
+	while (cqr->refers != NULL) {
+		cqr = cqr->refers;
+	}
+
+	if (scsw_is_tm(&cqr->irb.scsw)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "32 bit sense, action 1B, update,"
+			      " in transport mode - just retry");
+		return previous_erp;
+	}
+
+	/* for imprecise ending just do default erp */
+	if (sense[1] & 0x01) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Imprecise ending is set - just retry");
+
+		previous_erp->status = DASD_CQR_FILLED;
+
+		return previous_erp;
+	}
+
+	/* determine the address of the CCW to be restarted */
+	/* Imprecise ending is not set -> addr from IRB-SCSW */
+	cpa = previous_erp->irb.scsw.cmd.cpa;
+
+	if (cpa == 0) {
+		dev_err(&device->cdev->dev,
+			"Unable to determine address of to be restarted CCW\n");
+
+		previous_erp->status = DASD_CQR_FAILED;
+
+		return previous_erp;
+	}
+
+	erp = previous_erp;
+
+	/* update the LO with the new returned sense data  */
+	LO_data = erp->data + sizeof(struct DE_eckd_data);
+
+	if ((sense[3] == 0x01) && (LO_data[1] & 0x01)) {
+		/* should not happen */
+		previous_erp->status = DASD_CQR_FAILED;
+
+		return previous_erp;
+	}
+
+	if ((sense[7] & 0x3F) == 0x01) {
+		/* operation code is WRITE DATA -> data area orientation */
+		LO_data[0] = 0x81;
+
+	} else if ((sense[7] & 0x3F) == 0x03) {
+		/* operation code is FORMAT WRITE -> index orientation */
+		LO_data[0] = 0xC3;
+
+	} else {
+		LO_data[0] = sense[7];	/* operation */
+	}
+
+	LO_data[1] = sense[8];	/* auxiliary */
+	LO_data[2] = sense[9];
+	LO_data[3] = sense[3];	/* count */
+	LO_data[4] = sense[29];	/* seek_addr.cyl */
+	LO_data[5] = sense[30];	/* seek_addr.cyl 2nd byte */
+	LO_data[7] = sense[31];	/* seek_addr.head 2nd byte */
+
+	memcpy(&(LO_data[8]), &(sense[11]), 8);
+
+	/* TIC to the failed ccw */
+	ccw = erp->cpaddr;	/* addr of DE ccw */
+	ccw++;			/* addr of LE ccw */
+	ccw++;			/* addr of TIC ccw */
+	ccw->cda = cpa;
+
+	erp->status = DASD_CQR_FILLED;
+
+	return erp;
+
+}				/* end dasd_3990_update_1B */
+
+/*
+ * DASD_3990_ERP_COMPOUND_RETRY
+ *
+ * DESCRIPTION
+ *   Handles the compound ERP action retry code.
+ *   NOTE: At least one retry is done even if zero is specified
+ *	   by the sense data. This makes enqueueing of the request
+ *	   easier.
+ *
+ * PARAMETER
+ *   sense		sense data of the actual error
+ *   erp		pointer to the currently created ERP
+ *
+ * RETURN VALUES
+ *   erp		modified ERP pointer
+ *
+ */
+static void
+dasd_3990_erp_compound_retry(struct dasd_ccw_req * erp, char *sense)
+{
+
+	switch (sense[25] & 0x03) {
+	case 0x00:		/* no not retry */
+		erp->retries = 1;
+		break;
+
+	case 0x01:		/* retry 2 times */
+		erp->retries = 2;
+		break;
+
+	case 0x02:		/* retry 10 times */
+		erp->retries = 10;
+		break;
+
+	case 0x03:		/* retry 256 times */
+		erp->retries = 256;
+		break;
+
+	default:
+		BUG();
+	}
+
+	erp->function = dasd_3990_erp_compound_retry;
+
+}				/* end dasd_3990_erp_compound_retry */
+
+/*
+ * DASD_3990_ERP_COMPOUND_PATH
+ *
+ * DESCRIPTION
+ *   Handles the compound ERP action for retry on alternate
+ *   channel path.
+ *
+ * PARAMETER
+ *   sense		sense data of the actual error
+ *   erp		pointer to the currently created ERP
+ *
+ * RETURN VALUES
+ *   erp		modified ERP pointer
+ *
+ */
+static void
+dasd_3990_erp_compound_path(struct dasd_ccw_req * erp, char *sense)
+{
+	if (sense[25] & DASD_SENSE_BIT_3) {
+		dasd_3990_erp_alternate_path(erp);
+
+		if (erp->status == DASD_CQR_FAILED &&
+		    !test_bit(DASD_CQR_VERIFY_PATH, &erp->flags)) {
+			/* reset the lpm and the status to be able to
+			 * try further actions. */
+			erp->lpm = dasd_path_get_opm(erp->startdev);
+			erp->status = DASD_CQR_NEED_ERP;
+		}
+	}
+
+	erp->function = dasd_3990_erp_compound_path;
+
+}				/* end dasd_3990_erp_compound_path */
+
+/*
+ * DASD_3990_ERP_COMPOUND_CODE
+ *
+ * DESCRIPTION
+ *   Handles the compound ERP action for retry code.
+ *
+ * PARAMETER
+ *   sense		sense data of the actual error
+ *   erp		pointer to the currently created ERP
+ *
+ * RETURN VALUES
+ *   erp		NEW ERP pointer
+ *
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_compound_code(struct dasd_ccw_req * erp, char *sense)
+{
+
+	if (sense[25] & DASD_SENSE_BIT_2) {
+
+		switch (sense[28]) {
+		case 0x17:
+			/* issue a Diagnostic Control command with an
+			 * Inhibit Write subcommand and controller modifier */
+			erp = dasd_3990_erp_DCTL(erp, 0x20);
+			break;
+
+		case 0x25:
+			/* wait for 5 seconds and retry again */
+			erp->retries = 1;
+
+			dasd_3990_erp_block_queue (erp, 5*HZ);
+			break;
+
+		default:
+			/* should not happen - continue */
+			break;
+		}
+	}
+
+	erp->function = dasd_3990_erp_compound_code;
+
+	return erp;
+
+}				/* end dasd_3990_erp_compound_code */
+
+/*
+ * DASD_3990_ERP_COMPOUND_CONFIG
+ *
+ * DESCRIPTION
+ *   Handles the compound ERP action for configuration
+ *   dependent error.
+ *   Note: duplex handling is not implemented (yet).
+ *
+ * PARAMETER
+ *   sense		sense data of the actual error
+ *   erp		pointer to the currently created ERP
+ *
+ * RETURN VALUES
+ *   erp		modified ERP pointer
+ *
+ */
+static void
+dasd_3990_erp_compound_config(struct dasd_ccw_req * erp, char *sense)
+{
+
+	if ((sense[25] & DASD_SENSE_BIT_1) && (sense[26] & DASD_SENSE_BIT_2)) {
+		struct dasd_device *device = erp->startdev;
+		dev_err(&device->cdev->dev,
+			"Compound configuration error occurred\n");
+	}
+
+	erp->function = dasd_3990_erp_compound_config;
+
+}				/* end dasd_3990_erp_compound_config */
+
+/*
+ * DASD_3990_ERP_COMPOUND
+ *
+ * DESCRIPTION
+ *   Does the further compound program action if
+ *   compound retry was not successful.
+ *
+ * PARAMETER
+ *   sense		sense data of the actual error
+ *   erp		pointer to the current (failed) ERP
+ *
+ * RETURN VALUES
+ *   erp		(additional) ERP pointer
+ *
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_compound(struct dasd_ccw_req * erp, char *sense)
+{
+
+	if ((erp->function == dasd_3990_erp_compound_retry) &&
+	    (erp->status == DASD_CQR_NEED_ERP)) {
+
+		dasd_3990_erp_compound_path(erp, sense);
+	}
+
+	if ((erp->function == dasd_3990_erp_compound_path) &&
+	    (erp->status == DASD_CQR_NEED_ERP)) {
+
+		erp = dasd_3990_erp_compound_code(erp, sense);
+	}
+
+	if ((erp->function == dasd_3990_erp_compound_code) &&
+	    (erp->status == DASD_CQR_NEED_ERP)) {
+
+		dasd_3990_erp_compound_config(erp, sense);
+	}
+
+	/* if no compound action ERP specified, the request failed */
+	if (erp->status == DASD_CQR_NEED_ERP)
+		erp->status = DASD_CQR_FAILED;
+
+	return erp;
+
+}				/* end dasd_3990_erp_compound */
+
+/*
+ *DASD_3990_ERP_HANDLE_SIM
+ *
+ *DESCRIPTION
+ *  inspects the SIM SENSE data and starts an appropriate action
+ *
+ * PARAMETER
+ *   sense	   sense data of the actual error
+ *
+ * RETURN VALUES
+ *   none
+ */
+void
+dasd_3990_erp_handle_sim(struct dasd_device *device, char *sense)
+{
+	/* print message according to log or message to operator mode */
+	if ((sense[24] & DASD_SIM_MSG_TO_OP) || (sense[1] & 0x10)) {
+		/* print SIM SRC from RefCode */
+		dev_err(&device->cdev->dev, "SIM - SRC: "
+			    "%02x%02x%02x%02x\n", sense[22],
+			    sense[23], sense[11], sense[12]);
+	} else if (sense[24] & DASD_SIM_LOG) {
+		/* print SIM SRC Refcode */
+		dev_warn(&device->cdev->dev, "log SIM - SRC: "
+			    "%02x%02x%02x%02x\n", sense[22],
+			    sense[23], sense[11], sense[12]);
+	}
+}
+
+/*
+ * DASD_3990_ERP_INSPECT_32
+ *
+ * DESCRIPTION
+ *   Does a detailed inspection of the 32 byte sense data
+ *   and sets up a related error recovery action.
+ *
+ * PARAMETER
+ *   sense		sense data of the actual error
+ *   erp		pointer to the currently created default ERP
+ *
+ * RETURN VALUES
+ *   erp_filled		pointer to the ERP
+ *
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_inspect_32(struct dasd_ccw_req * erp, char *sense)
+{
+
+	struct dasd_device *device = erp->startdev;
+
+	erp->function = dasd_3990_erp_inspect_32;
+
+	/* check for SIM sense data */
+	if ((sense[6] & DASD_SIM_SENSE) == DASD_SIM_SENSE)
+		dasd_3990_erp_handle_sim(device, sense);
+
+	if (sense[25] & DASD_SENSE_BIT_0) {
+
+		/* compound program action codes (byte25 bit 0 == '1') */
+		dasd_3990_erp_compound_retry(erp, sense);
+
+	} else {
+
+		/* single program action codes (byte25 bit 0 == '0') */
+		switch (sense[25]) {
+
+		case 0x00:	/* success - use default ERP for retries */
+			DBF_DEV_EVENT(DBF_DEBUG, device, "%s",
+				    "ERP called for successful request"
+				    " - just retry");
+			break;
+
+		case 0x01:	/* fatal error */
+			dev_err(&device->cdev->dev,
+				    "ERP failed for the DASD\n");
+
+			erp = dasd_3990_erp_cleanup(erp, DASD_CQR_FAILED);
+			break;
+
+		case 0x02:	/* intervention required */
+		case 0x03:	/* intervention required during dual copy */
+			erp = dasd_3990_erp_int_req(erp);
+			break;
+
+		case 0x0F:
+			dev_err(&device->cdev->dev,
+				"Update write command error occurred\n");
+
+			erp = dasd_3990_erp_cleanup(erp, DASD_CQR_FAILED);
+			break;
+
+		case 0x10:  /* logging required for other channel program */
+			erp = dasd_3990_erp_action_10_32(erp, sense);
+			break;
+
+		case 0x15:
+			dev_err(&device->cdev->dev,
+				"Track outside defined extent error occurred\n");
+
+			erp = dasd_3990_erp_cleanup(erp, DASD_CQR_FAILED);
+			break;
+
+		case 0x1B:	/* unexpected condition during write */
+
+			erp = dasd_3990_erp_action_1B_32(erp, sense);
+			break;
+
+		case 0x1C:	/* invalid data */
+			dev_emerg(&device->cdev->dev,
+				    "Data recovered during retry with PCI "
+				    "fetch mode active\n");
+
+			/* not possible to handle this situation in Linux */
+			panic
+			    ("Invalid data - No way to inform application "
+			     "about the possibly incorrect data");
+			break;
+
+		case 0x1D:	/* state-change pending */
+			DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+				    "A State change pending condition exists "
+				    "for the subsystem or device");
+
+			erp = dasd_3990_erp_action_4(erp, sense);
+			break;
+
+		case 0x1E:	/* busy */
+			DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+				    "Busy condition exists "
+				    "for the subsystem or device");
+                        erp = dasd_3990_erp_action_4(erp, sense);
+			break;
+
+		default:	/* all others errors - default erp  */
+			break;
+		}
+	}
+
+	return erp;
+
+}				/* end dasd_3990_erp_inspect_32 */
+
+static void dasd_3990_erp_disable_path(struct dasd_device *device, __u8 lpum)
+{
+	int pos = pathmask_to_pos(lpum);
+
+	if (!(device->features & DASD_FEATURE_PATH_AUTODISABLE)) {
+		dev_err(&device->cdev->dev,
+			"Path %x.%02x (pathmask %02x) is operational despite excessive IFCCs\n",
+			device->path[pos].cssid, device->path[pos].chpid, lpum);
+		goto out;
+	}
+
+	/* no remaining path, cannot disable */
+	if (!(dasd_path_get_opm(device) & ~lpum)) {
+		dev_err(&device->cdev->dev,
+			"Last path %x.%02x (pathmask %02x) is operational despite excessive IFCCs\n",
+			device->path[pos].cssid, device->path[pos].chpid, lpum);
+		goto out;
+	}
+
+	dev_err(&device->cdev->dev,
+		"Path %x.%02x (pathmask %02x) is disabled - IFCC threshold exceeded\n",
+		device->path[pos].cssid, device->path[pos].chpid, lpum);
+	dasd_path_remove_opm(device, lpum);
+	dasd_path_add_ifccpm(device, lpum);
+
+out:
+	device->path[pos].errorclk = 0;
+	atomic_set(&device->path[pos].error_count, 0);
+}
+
+static void dasd_3990_erp_account_error(struct dasd_ccw_req *erp)
+{
+	struct dasd_device *device = erp->startdev;
+	__u8 lpum = erp->refers->irb.esw.esw1.lpum;
+	int pos = pathmask_to_pos(lpum);
+	unsigned long clk;
+
+	if (!device->path_thrhld)
+		return;
+
+	clk = get_tod_clock();
+	/*
+	 * check if the last error is longer ago than the timeout,
+	 * if so reset error state
+	 */
+	if ((tod_to_ns(clk - device->path[pos].errorclk) / NSEC_PER_SEC)
+	    >= device->path_interval) {
+		atomic_set(&device->path[pos].error_count, 0);
+		device->path[pos].errorclk = 0;
+	}
+	atomic_inc(&device->path[pos].error_count);
+	device->path[pos].errorclk = clk;
+	/* threshold exceeded disable path if possible */
+	if (atomic_read(&device->path[pos].error_count) >=
+	    device->path_thrhld)
+		dasd_3990_erp_disable_path(device, lpum);
+}
+
+/*
+ *****************************************************************************
+ * main ERP control functions (24 and 32 byte sense)
+ *****************************************************************************
+ */
+
+/*
+ * DASD_3990_ERP_CONTROL_CHECK
+ *
+ * DESCRIPTION
+ *   Does a generic inspection if a control check occurred and sets up
+ *   the related error recovery procedure
+ *
+ * PARAMETER
+ *   erp		pointer to the currently created default ERP
+ *
+ * RETURN VALUES
+ *   erp_filled		pointer to the erp
+ */
+
+static struct dasd_ccw_req *
+dasd_3990_erp_control_check(struct dasd_ccw_req *erp)
+{
+	struct dasd_device *device = erp->startdev;
+
+	if (scsw_cstat(&erp->refers->irb.scsw) & (SCHN_STAT_INTF_CTRL_CHK
+					   | SCHN_STAT_CHN_CTRL_CHK)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "channel or interface control check");
+		dasd_3990_erp_account_error(erp);
+		erp = dasd_3990_erp_action_4(erp, NULL);
+	}
+	return erp;
+}
+
+/*
+ * DASD_3990_ERP_INSPECT
+ *
+ * DESCRIPTION
+ *   Does a detailed inspection for sense data by calling either
+ *   the 24-byte or the 32-byte inspection routine.
+ *
+ * PARAMETER
+ *   erp		pointer to the currently created default ERP
+ * RETURN VALUES
+ *   erp_new		contens was possibly modified
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_inspect(struct dasd_ccw_req *erp)
+{
+
+	struct dasd_ccw_req *erp_new = NULL;
+	char *sense;
+
+	/* if this problem occurred on an alias retry on base */
+	erp_new = dasd_3990_erp_inspect_alias(erp);
+	if (erp_new)
+		return erp_new;
+
+	/* sense data are located in the refers record of the
+	 * already set up new ERP !
+	 * check if concurrent sens is available
+	 */
+	sense = dasd_get_sense(&erp->refers->irb);
+	if (!sense)
+		erp_new = dasd_3990_erp_control_check(erp);
+	/* distinguish between 24 and 32 byte sense data */
+	else if (sense[27] & DASD_SENSE_BIT_0) {
+
+		/* inspect the 24 byte sense data */
+		erp_new = dasd_3990_erp_inspect_24(erp, sense);
+
+	} else {
+
+		/* inspect the 32 byte sense data */
+		erp_new = dasd_3990_erp_inspect_32(erp, sense);
+
+	}	/* end distinguish between 24 and 32 byte sense data */
+
+	return erp_new;
+}
+
+/*
+ * DASD_3990_ERP_ADD_ERP
+ *
+ * DESCRIPTION
+ *   This function adds an additional request block (ERP) to the head of
+ *   the given cqr (or erp).
+ *   For a command mode cqr the erp is initialized as an default erp
+ *   (retry TIC).
+ *   For transport mode we make a copy of the original TCW (points to
+ *   the original TCCB, TIDALs, etc.) but give it a fresh
+ *   TSB so the original sense data will not be changed.
+ *
+ * PARAMETER
+ *   cqr		head of the current ERP-chain (or single cqr if
+ *			first error)
+ * RETURN VALUES
+ *   erp		pointer to new ERP-chain head
+ */
+static struct dasd_ccw_req *dasd_3990_erp_add_erp(struct dasd_ccw_req *cqr)
+{
+
+	struct dasd_device *device = cqr->startdev;
+	struct ccw1 *ccw;
+	struct dasd_ccw_req *erp;
+	int cplength, datasize;
+	struct tcw *tcw;
+	struct tsb *tsb;
+
+	if (cqr->cpmode == 1) {
+		cplength = 0;
+		/* TCW needs to be 64 byte aligned, so leave enough room */
+		datasize = 64 + sizeof(struct tcw) + sizeof(struct tsb);
+	} else {
+		cplength = 2;
+		datasize = 0;
+	}
+
+	/* allocate additional request block */
+	erp = dasd_alloc_erp_request(cqr->magic,
+				     cplength, datasize, device);
+	if (IS_ERR(erp)) {
+                if (cqr->retries <= 0) {
+			DBF_DEV_EVENT(DBF_ERR, device, "%s",
+				    "Unable to allocate ERP request");
+			cqr->status = DASD_CQR_FAILED;
+			cqr->stopclk = get_tod_clock();
+		} else {
+			DBF_DEV_EVENT(DBF_ERR, device,
+                                     "Unable to allocate ERP request "
+				     "(%i retries left)",
+                                     cqr->retries);
+			dasd_block_set_timer(device->block, (HZ << 3));
+                }
+		return erp;
+	}
+
+	ccw = cqr->cpaddr;
+	if (cqr->cpmode == 1) {
+		/* make a shallow copy of the original tcw but set new tsb */
+		erp->cpmode = 1;
+		erp->cpaddr = PTR_ALIGN(erp->data, 64);
+		tcw = erp->cpaddr;
+		tsb = (struct tsb *) &tcw[1];
+		*tcw = *((struct tcw *)cqr->cpaddr);
+		tcw->tsb = virt_to_dma64(tsb);
+	} else if (ccw->cmd_code == DASD_ECKD_CCW_PSF) {
+		/* PSF cannot be chained from NOOP/TIC */
+		erp->cpaddr = cqr->cpaddr;
+	} else {
+		/* initialize request with default TIC to current ERP/CQR */
+		ccw = erp->cpaddr;
+		ccw->cmd_code = CCW_CMD_NOOP;
+		ccw->flags = CCW_FLAG_CC;
+		ccw++;
+		ccw->cmd_code = CCW_CMD_TIC;
+		ccw->cda      = virt_to_dma32(cqr->cpaddr);
+	}
+
+	erp->flags = cqr->flags;
+	erp->function = dasd_3990_erp_add_erp;
+	erp->refers   = cqr;
+	erp->startdev = device;
+	erp->memdev   = device;
+	erp->block    = cqr->block;
+	erp->filldata = cqr->filldata;
+	erp->magic    = cqr->magic;
+	erp->expires  = cqr->expires;
+	erp->retries  = device->default_retries;
+	erp->buildclk = get_tod_clock();
+	erp->status = DASD_CQR_FILLED;
+
+	return erp;
+}
+
+/*
+ * DASD_3990_ERP_ADDITIONAL_ERP
+ *
+ * DESCRIPTION
+ *   An additional ERP is needed to handle the current error.
+ *   Add ERP to the head of the ERP-chain containing the ERP processing
+ *   determined based on the sense data.
+ *
+ * PARAMETER
+ *   cqr		head of the current ERP-chain (or single cqr if
+ *			first error)
+ *
+ * RETURN VALUES
+ *   erp		pointer to new ERP-chain head
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_additional_erp(struct dasd_ccw_req * cqr)
+{
+
+	struct dasd_ccw_req *erp = NULL;
+
+	/* add erp and initialize with default TIC */
+	erp = dasd_3990_erp_add_erp(cqr);
+
+	if (IS_ERR(erp))
+		return erp;
+
+	/* inspect sense, determine specific ERP if possible */
+	if (erp != cqr) {
+
+		erp = dasd_3990_erp_inspect(erp);
+	}
+
+	return erp;
+
+}				/* end dasd_3990_erp_additional_erp */
+
+/*
+ * DASD_3990_ERP_ERROR_MATCH
+ *
+ * DESCRIPTION
+ *   Check if the device status of the given cqr is the same.
+ *   This means that the failed CCW and the relevant sense data
+ *   must match.
+ *   I don't distinguish between 24 and 32 byte sense because in case of
+ *   24 byte sense byte 25 and 27 is set as well.
+ *
+ * PARAMETER
+ *   cqr1		first cqr, which will be compared with the
+ *   cqr2		second cqr.
+ *
+ * RETURN VALUES
+ *   match		'boolean' for match found
+ *			returns 1 if match found, otherwise 0.
+ */
+static int dasd_3990_erp_error_match(struct dasd_ccw_req *cqr1,
+				     struct dasd_ccw_req *cqr2)
+{
+	char *sense1, *sense2;
+
+	if (cqr1->startdev != cqr2->startdev)
+		return 0;
+
+	sense1 = dasd_get_sense(&cqr1->irb);
+	sense2 = dasd_get_sense(&cqr2->irb);
+
+	/* one request has sense data, the other not -> no match, return 0 */
+	if (!sense1 != !sense2)
+		return 0;
+	/* no sense data in both cases -> check cstat for IFCC */
+	if (!sense1 && !sense2)	{
+		if ((scsw_cstat(&cqr1->irb.scsw) & (SCHN_STAT_INTF_CTRL_CHK |
+						    SCHN_STAT_CHN_CTRL_CHK)) ==
+		    (scsw_cstat(&cqr2->irb.scsw) & (SCHN_STAT_INTF_CTRL_CHK |
+						    SCHN_STAT_CHN_CTRL_CHK)))
+			return 1; /* match with ifcc*/
+	}
+	/* check sense data; byte 0-2,25,27 */
+	if (!(sense1 && sense2 &&
+	      (memcmp(sense1, sense2, 3) == 0) &&
+	      (sense1[27] == sense2[27]) &&
+	      (sense1[25] == sense2[25]))) {
+
+		return 0;	/* sense doesn't match */
+	}
+
+	return 1;		/* match */
+
+}				/* end dasd_3990_erp_error_match */
+
+/*
+ * DASD_3990_ERP_IN_ERP
+ *
+ * DESCRIPTION
+ *   check if the current error already happened before.
+ *   quick exit if current cqr is not an ERP (cqr->refers=NULL)
+ *
+ * PARAMETER
+ *   cqr		failed cqr (either original cqr or already an erp)
+ *
+ * RETURN VALUES
+ *   erp		erp-pointer to the already defined error
+ *			recovery procedure OR
+ *			NULL if a 'new' error occurred.
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_in_erp(struct dasd_ccw_req *cqr)
+{
+
+	struct dasd_ccw_req *erp_head = cqr,	/* save erp chain head */
+	*erp_match = NULL;	/* save erp chain head */
+	int match = 0;		/* 'boolean' for matching error found */
+
+	if (cqr->refers == NULL) {	/* return if not in erp */
+		return NULL;
+	}
+
+	/* check the erp/cqr chain for current error */
+	do {
+		match = dasd_3990_erp_error_match(erp_head, cqr->refers);
+		erp_match = cqr;	/* save possible matching erp  */
+		cqr = cqr->refers;	/* check next erp/cqr in queue */
+
+	} while ((cqr->refers != NULL) && (!match));
+
+	if (!match) {
+		return NULL;	/* no match was found */
+	}
+
+	return erp_match;	/* return address of matching erp */
+
+}				/* END dasd_3990_erp_in_erp */
+
+/*
+ * DASD_3990_ERP_FURTHER_ERP (24 & 32 byte sense)
+ *
+ * DESCRIPTION
+ *   No retry is left for the current ERP. Check what has to be done
+ *   with the ERP.
+ *     - do further defined ERP action or
+ *     - wait for interrupt or
+ *     - exit with permanent error
+ *
+ * PARAMETER
+ *   erp		ERP which is in progress with no retry left
+ *
+ * RETURN VALUES
+ *   erp		modified/additional ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_further_erp(struct dasd_ccw_req *erp)
+{
+
+	struct dasd_device *device = erp->startdev;
+	char *sense = dasd_get_sense(&erp->irb);
+
+	/* check for 24 byte sense ERP */
+	if ((erp->function == dasd_3990_erp_bus_out) ||
+	    (erp->function == dasd_3990_erp_action_1) ||
+	    (erp->function == dasd_3990_erp_action_4)) {
+
+		erp = dasd_3990_erp_action_1(erp);
+
+	} else if (erp->function == dasd_3990_erp_action_1_sec) {
+		erp = dasd_3990_erp_action_1_sec(erp);
+	} else if (erp->function == dasd_3990_erp_action_5) {
+
+		/* retries have not been successful */
+		/* prepare erp for retry on different channel path */
+		erp = dasd_3990_erp_action_1(erp);
+
+		if (sense && !(sense[2] & DASD_SENSE_BIT_0)) {
+
+			/* issue a Diagnostic Control command with an
+			 * Inhibit Write subcommand */
+
+			switch (sense[25]) {
+			case 0x17:
+			case 0x57:{	/* controller */
+					erp = dasd_3990_erp_DCTL(erp, 0x20);
+					break;
+				}
+			case 0x18:
+			case 0x58:{	/* channel path */
+					erp = dasd_3990_erp_DCTL(erp, 0x40);
+					break;
+				}
+			case 0x19:
+			case 0x59:{	/* storage director */
+					erp = dasd_3990_erp_DCTL(erp, 0x80);
+					break;
+				}
+			default:
+				DBF_DEV_EVENT(DBF_WARNING, device,
+					    "invalid subcommand modifier 0x%x "
+					    "for Diagnostic Control Command",
+					    sense[25]);
+			}
+		}
+
+		/* check for 32 byte sense ERP */
+	} else if (sense &&
+		   ((erp->function == dasd_3990_erp_compound_retry) ||
+		    (erp->function == dasd_3990_erp_compound_path) ||
+		    (erp->function == dasd_3990_erp_compound_code) ||
+		    (erp->function == dasd_3990_erp_compound_config))) {
+
+		erp = dasd_3990_erp_compound(erp, sense);
+
+	} else {
+		/*
+		 * No retry left and no additional special handling
+		 * necessary
+		 */
+		dev_err(&device->cdev->dev,
+			"ERP %px has run out of retries and failed\n", erp);
+
+		erp->status = DASD_CQR_FAILED;
+	}
+
+	return erp;
+
+}				/* end dasd_3990_erp_further_erp */
+
+/*
+ * DASD_3990_ERP_HANDLE_MATCH_ERP
+ *
+ * DESCRIPTION
+ *   An error occurred again and an ERP has been detected which is already
+ *   used to handle this error (e.g. retries).
+ *   All prior ERP's are asumed to be successful and therefore removed
+ *   from queue.
+ *   If retry counter of matching erp is already 0, it is checked if further
+ *   action is needed (besides retry) or if the ERP has failed.
+ *
+ * PARAMETER
+ *   erp_head		first ERP in ERP-chain
+ *   erp		ERP that handles the actual error.
+ *			(matching erp)
+ *
+ * RETURN VALUES
+ *   erp		modified/additional ERP
+ */
+static struct dasd_ccw_req *
+dasd_3990_erp_handle_match_erp(struct dasd_ccw_req *erp_head,
+			       struct dasd_ccw_req *erp)
+{
+
+	struct dasd_device *device = erp_head->startdev;
+	struct dasd_ccw_req *erp_done = erp_head;	/* finished req */
+	struct dasd_ccw_req *erp_free = NULL;	/* req to be freed */
+
+	/* loop over successful ERPs and remove them from chanq */
+	while (erp_done != erp) {
+
+		if (erp_done == NULL)	/* end of chain reached */
+			panic("Programming error in ERP! The original request was lost\n");
+
+		/* remove the request from the device queue */
+		list_del(&erp_done->blocklist);
+
+		erp_free = erp_done;
+		erp_done = erp_done->refers;
+
+		/* free the finished erp request */
+		dasd_free_erp_request(erp_free, erp_free->memdev);
+
+	}			/* end while */
+
+	if (erp->retries > 0) {
+
+		char *sense = dasd_get_sense(&erp->refers->irb);
+
+		/* check for special retries */
+		if (sense && erp->function == dasd_3990_erp_action_4) {
+
+			erp = dasd_3990_erp_action_4(erp, sense);
+
+		} else if (sense &&
+			   erp->function == dasd_3990_erp_action_1B_32) {
+
+			erp = dasd_3990_update_1B(erp, sense);
+
+		} else if (sense && erp->function == dasd_3990_erp_int_req) {
+
+			erp = dasd_3990_erp_int_req(erp);
+
+		} else {
+			/* simple retry	  */
+			DBF_DEV_EVENT(DBF_DEBUG, device,
+				    "%i retries left for erp %p",
+				    erp->retries, erp);
+
+			/* handle the request again... */
+			erp->status = DASD_CQR_FILLED;
+		}
+
+	} else {
+		/* no retry left - check for further necessary action	 */
+		/* if no further actions, handle rest as permanent error */
+		erp = dasd_3990_erp_further_erp(erp);
+	}
+
+	return erp;
+
+}				/* end dasd_3990_erp_handle_match_erp */
+
+/*
+ * DASD_3990_ERP_ACTION
+ *
+ * DESCRIPTION
+ *   control routine for 3990 erp actions.
+ *   Has to be called with the queue lock (namely the s390_irq_lock) acquired.
+ *
+ * PARAMETER
+ *   cqr		failed cqr (either original cqr or already an erp)
+ *
+ * RETURN VALUES
+ *   erp		erp-pointer to the head of the ERP action chain.
+ *			This means:
+ *			 - either a ptr to an additional ERP cqr or
+ *			 - the original given cqr (which's status might
+ *			   be modified)
+ */
+struct dasd_ccw_req *
+dasd_3990_erp_action(struct dasd_ccw_req * cqr)
+{
+	struct dasd_ccw_req *erp = NULL;
+	struct dasd_device *device = cqr->startdev;
+	struct dasd_ccw_req *temp_erp = NULL;
+
+	if (device->features & DASD_FEATURE_ERPLOG) {
+		/* print current erp_chain */
+		dev_err(&device->cdev->dev,
+			    "ERP chain at BEGINNING of ERP-ACTION\n");
+		for (temp_erp = cqr;
+		     temp_erp != NULL; temp_erp = temp_erp->refers) {
+			dev_err(&device->cdev->dev,
+				"ERP %px (%02x) refers to %px\n",
+				temp_erp, temp_erp->status, temp_erp->refers);
+		}
+	}
+
+	/* double-check if current erp/cqr was successful */
+	if ((scsw_cstat(&cqr->irb.scsw) == 0x00) &&
+	    (scsw_dstat(&cqr->irb.scsw) ==
+	     (DEV_STAT_CHN_END | DEV_STAT_DEV_END))) {
+
+		DBF_DEV_EVENT(DBF_DEBUG, device,
+			    "ERP called for successful request %p"
+			    " - NO ERP necessary", cqr);
+
+		cqr->status = DASD_CQR_DONE;
+
+		return cqr;
+	}
+
+	/* check if error happened before */
+	erp = dasd_3990_erp_in_erp(cqr);
+
+	if (erp == NULL) {
+		/* no matching erp found - set up erp */
+		erp = dasd_3990_erp_additional_erp(cqr);
+		if (IS_ERR(erp))
+			return erp;
+	} else {
+		/* matching erp found - set all leading erp's to DONE */
+		erp = dasd_3990_erp_handle_match_erp(cqr, erp);
+	}
+
+
+	/*
+	 * For path verification work we need to stick with the path that was
+	 * originally chosen so that the per path configuration data is
+	 * assigned correctly.
+	 */
+	if (test_bit(DASD_CQR_VERIFY_PATH, &erp->flags) && cqr->lpm) {
+		erp->lpm = cqr->lpm;
+	}
+
+	if (device->features & DASD_FEATURE_ERPLOG) {
+		/* print current erp_chain */
+		dev_err(&device->cdev->dev,
+			    "ERP chain at END of ERP-ACTION\n");
+		for (temp_erp = erp;
+		     temp_erp != NULL; temp_erp = temp_erp->refers) {
+			dev_err(&device->cdev->dev,
+				"ERP %px (%02x) refers to %px\n",
+				temp_erp, temp_erp->status, temp_erp->refers);
+		}
+	}
+
+	/* enqueue ERP request if it's a new one */
+	if (list_empty(&erp->blocklist)) {
+		cqr->status = DASD_CQR_IN_ERP;
+		/* add erp request before the cqr */
+		list_add_tail(&erp->blocklist, &cqr->blocklist);
+	}
+
+
+
+	return erp;
+
+}				/* end dasd_3990_erp_action */
diff --git a/drivers/s390/block/dasd_alias.c b/drivers/s390/block/dasd_alias.c
new file mode 100644
index 000000000..8d6c16d35
--- /dev/null
+++ b/drivers/s390/block/dasd_alias.c
@@ -0,0 +1,972 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * PAV alias management for the DASD ECKD discipline
+ *
+ * Copyright IBM Corp. 2007
+ * Author(s): Stefan Weinhuber <wein@de.ibm.com>
+ */
+
+#include <linux/list.h>
+#include <linux/slab.h>
+#include <asm/ebcdic.h>
+#include "dasd_int.h"
+#include "dasd_eckd.h"
+
+/*
+ * General concept of alias management:
+ * - PAV and DASD alias management is specific to the eckd discipline.
+ * - A device is connected to an lcu as long as the device exists.
+ *   dasd_alias_make_device_known_to_lcu will be called wenn the
+ *   device is checked by the eckd discipline and
+ *   dasd_alias_disconnect_device_from_lcu will be called
+ *   before the device is deleted.
+ * - The dasd_alias_add_device / dasd_alias_remove_device
+ *   functions mark the point when a device is 'ready for service'.
+ * - A summary unit check is a rare occasion, but it is mandatory to
+ *   support it. It requires some complex recovery actions before the
+ *   devices can be used again (see dasd_alias_handle_summary_unit_check).
+ * - dasd_alias_get_start_dev will find an alias device that can be used
+ *   instead of the base device and does some (very simple) load balancing.
+ *   This is the function that gets called for each I/O, so when improving
+ *   something, this function should get faster or better, the rest has just
+ *   to be correct.
+ */
+
+
+static void summary_unit_check_handling_work(struct work_struct *);
+static void lcu_update_work(struct work_struct *);
+static int _schedule_lcu_update(struct alias_lcu *, struct dasd_device *);
+
+static struct alias_root aliastree = {
+	.serverlist = LIST_HEAD_INIT(aliastree.serverlist),
+	.lock = __SPIN_LOCK_UNLOCKED(aliastree.lock),
+};
+
+static struct alias_server *_find_server(struct dasd_uid *uid)
+{
+	struct alias_server *pos;
+	list_for_each_entry(pos, &aliastree.serverlist, server) {
+		if (!strncmp(pos->uid.vendor, uid->vendor,
+			     sizeof(uid->vendor))
+		    && !strncmp(pos->uid.serial, uid->serial,
+				sizeof(uid->serial)))
+			return pos;
+	}
+	return NULL;
+}
+
+static struct alias_lcu *_find_lcu(struct alias_server *server,
+				   struct dasd_uid *uid)
+{
+	struct alias_lcu *pos;
+	list_for_each_entry(pos, &server->lculist, lcu) {
+		if (pos->uid.ssid == uid->ssid)
+			return pos;
+	}
+	return NULL;
+}
+
+static struct alias_pav_group *_find_group(struct alias_lcu *lcu,
+					   struct dasd_uid *uid)
+{
+	struct alias_pav_group *pos;
+	__u8 search_unit_addr;
+
+	/* for hyper pav there is only one group */
+	if (lcu->pav == HYPER_PAV) {
+		if (list_empty(&lcu->grouplist))
+			return NULL;
+		else
+			return list_first_entry(&lcu->grouplist,
+						struct alias_pav_group, group);
+	}
+
+	/* for base pav we have to find the group that matches the base */
+	if (uid->type == UA_BASE_DEVICE)
+		search_unit_addr = uid->real_unit_addr;
+	else
+		search_unit_addr = uid->base_unit_addr;
+	list_for_each_entry(pos, &lcu->grouplist, group) {
+		if (pos->uid.base_unit_addr == search_unit_addr &&
+		    !strncmp(pos->uid.vduit, uid->vduit, sizeof(uid->vduit)))
+			return pos;
+	}
+	return NULL;
+}
+
+static struct alias_server *_allocate_server(struct dasd_uid *uid)
+{
+	struct alias_server *server;
+
+	server = kzalloc_obj(*server);
+	if (!server)
+		return ERR_PTR(-ENOMEM);
+	memcpy(server->uid.vendor, uid->vendor, sizeof(uid->vendor));
+	memcpy(server->uid.serial, uid->serial, sizeof(uid->serial));
+	INIT_LIST_HEAD(&server->server);
+	INIT_LIST_HEAD(&server->lculist);
+	return server;
+}
+
+static void _free_server(struct alias_server *server)
+{
+	kfree(server);
+}
+
+static struct alias_lcu *_allocate_lcu(struct dasd_uid *uid)
+{
+	struct alias_lcu *lcu;
+
+	lcu = kzalloc_obj(*lcu);
+	if (!lcu)
+		return ERR_PTR(-ENOMEM);
+	lcu->uac = kzalloc_obj(*(lcu->uac), GFP_KERNEL | GFP_DMA);
+	if (!lcu->uac)
+		goto out_err1;
+	lcu->rsu_cqr = kzalloc_obj(*lcu->rsu_cqr, GFP_KERNEL | GFP_DMA);
+	if (!lcu->rsu_cqr)
+		goto out_err2;
+	lcu->rsu_cqr->cpaddr = kzalloc_obj(struct ccw1, GFP_KERNEL | GFP_DMA);
+	if (!lcu->rsu_cqr->cpaddr)
+		goto out_err3;
+	lcu->rsu_cqr->data = kzalloc(16, GFP_KERNEL | GFP_DMA);
+	if (!lcu->rsu_cqr->data)
+		goto out_err4;
+
+	memcpy(lcu->uid.vendor, uid->vendor, sizeof(uid->vendor));
+	memcpy(lcu->uid.serial, uid->serial, sizeof(uid->serial));
+	lcu->uid.ssid = uid->ssid;
+	lcu->pav = NO_PAV;
+	lcu->flags = NEED_UAC_UPDATE | UPDATE_PENDING;
+	INIT_LIST_HEAD(&lcu->lcu);
+	INIT_LIST_HEAD(&lcu->inactive_devices);
+	INIT_LIST_HEAD(&lcu->active_devices);
+	INIT_LIST_HEAD(&lcu->grouplist);
+	INIT_WORK(&lcu->suc_data.worker, summary_unit_check_handling_work);
+	INIT_DELAYED_WORK(&lcu->ruac_data.dwork, lcu_update_work);
+	spin_lock_init(&lcu->lock);
+	init_completion(&lcu->lcu_setup);
+	return lcu;
+
+out_err4:
+	kfree(lcu->rsu_cqr->cpaddr);
+out_err3:
+	kfree(lcu->rsu_cqr);
+out_err2:
+	kfree(lcu->uac);
+out_err1:
+	kfree(lcu);
+	return ERR_PTR(-ENOMEM);
+}
+
+static void _free_lcu(struct alias_lcu *lcu)
+{
+	kfree(lcu->rsu_cqr->data);
+	kfree(lcu->rsu_cqr->cpaddr);
+	kfree(lcu->rsu_cqr);
+	kfree(lcu->uac);
+	kfree(lcu);
+}
+
+/*
+ * This is the function that will allocate all the server and lcu data,
+ * so this function must be called first for a new device.
+ * If the return value is 1, the lcu was already known before, if it
+ * is 0, this is a new lcu.
+ * Negative return code indicates that something went wrong (e.g. -ENOMEM)
+ */
+int dasd_alias_make_device_known_to_lcu(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	unsigned long flags;
+	struct alias_server *server, *newserver;
+	struct alias_lcu *lcu, *newlcu;
+	struct dasd_uid uid;
+
+	device->discipline->get_uid(device, &uid);
+	spin_lock_irqsave(&aliastree.lock, flags);
+	server = _find_server(&uid);
+	if (!server) {
+		spin_unlock_irqrestore(&aliastree.lock, flags);
+		newserver = _allocate_server(&uid);
+		if (IS_ERR(newserver))
+			return PTR_ERR(newserver);
+		spin_lock_irqsave(&aliastree.lock, flags);
+		server = _find_server(&uid);
+		if (!server) {
+			list_add(&newserver->server, &aliastree.serverlist);
+			server = newserver;
+		} else {
+			/* someone was faster */
+			_free_server(newserver);
+		}
+	}
+
+	lcu = _find_lcu(server, &uid);
+	if (!lcu) {
+		spin_unlock_irqrestore(&aliastree.lock, flags);
+		newlcu = _allocate_lcu(&uid);
+		if (IS_ERR(newlcu))
+			return PTR_ERR(newlcu);
+		spin_lock_irqsave(&aliastree.lock, flags);
+		lcu = _find_lcu(server, &uid);
+		if (!lcu) {
+			list_add(&newlcu->lcu, &server->lculist);
+			lcu = newlcu;
+		} else {
+			/* someone was faster */
+			_free_lcu(newlcu);
+		}
+	}
+	spin_lock(&lcu->lock);
+	list_add(&device->alias_list, &lcu->inactive_devices);
+	private->lcu = lcu;
+	spin_unlock(&lcu->lock);
+	spin_unlock_irqrestore(&aliastree.lock, flags);
+
+	return 0;
+}
+
+/*
+ * This function removes a device from the scope of alias management.
+ * The complicated part is to make sure that it is not in use by
+ * any of the workers. If necessary cancel the work.
+ */
+void dasd_alias_disconnect_device_from_lcu(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	unsigned long flags;
+	struct alias_lcu *lcu;
+	struct alias_server *server;
+	int was_pending;
+	struct dasd_uid uid;
+
+	lcu = private->lcu;
+	/* nothing to do if already disconnected */
+	if (!lcu)
+		return;
+	device->discipline->get_uid(device, &uid);
+	spin_lock_irqsave(&lcu->lock, flags);
+	/* make sure that the workers don't use this device */
+	if (device == lcu->suc_data.device) {
+		spin_unlock_irqrestore(&lcu->lock, flags);
+		cancel_work_sync(&lcu->suc_data.worker);
+		spin_lock_irqsave(&lcu->lock, flags);
+		if (device == lcu->suc_data.device) {
+			dasd_put_device(device);
+			lcu->suc_data.device = NULL;
+		}
+	}
+	was_pending = 0;
+	if (device == lcu->ruac_data.device) {
+		spin_unlock_irqrestore(&lcu->lock, flags);
+		was_pending = 1;
+		cancel_delayed_work_sync(&lcu->ruac_data.dwork);
+		spin_lock_irqsave(&lcu->lock, flags);
+		if (device == lcu->ruac_data.device) {
+			dasd_put_device(device);
+			lcu->ruac_data.device = NULL;
+		}
+	}
+	private->lcu = NULL;
+	spin_unlock_irqrestore(&lcu->lock, flags);
+
+	spin_lock_irqsave(&aliastree.lock, flags);
+	spin_lock(&lcu->lock);
+	list_del_init(&device->alias_list);
+	if (list_empty(&lcu->grouplist) &&
+	    list_empty(&lcu->active_devices) &&
+	    list_empty(&lcu->inactive_devices)) {
+		list_del(&lcu->lcu);
+		spin_unlock(&lcu->lock);
+		_free_lcu(lcu);
+		lcu = NULL;
+	} else {
+		if (was_pending)
+			_schedule_lcu_update(lcu, NULL);
+		spin_unlock(&lcu->lock);
+	}
+	server = _find_server(&uid);
+	if (server && list_empty(&server->lculist)) {
+		list_del(&server->server);
+		_free_server(server);
+	}
+	spin_unlock_irqrestore(&aliastree.lock, flags);
+}
+
+/*
+ * This function assumes that the unit address configuration stored
+ * in the lcu is up to date and will update the device uid before
+ * adding it to a pav group.
+ */
+
+static int _add_device_to_lcu(struct alias_lcu *lcu,
+			      struct dasd_device *device,
+			      struct dasd_device *pos)
+{
+
+	struct dasd_eckd_private *private = device->private;
+	struct alias_pav_group *group;
+	struct dasd_uid uid;
+
+	spin_lock(get_ccwdev_lock(device->cdev));
+	private->uid.type = lcu->uac->unit[private->uid.real_unit_addr].ua_type;
+	private->uid.base_unit_addr =
+		lcu->uac->unit[private->uid.real_unit_addr].base_ua;
+	uid = private->uid;
+	spin_unlock(get_ccwdev_lock(device->cdev));
+	/* if we have no PAV anyway, we don't need to bother with PAV groups */
+	if (lcu->pav == NO_PAV) {
+		list_move(&device->alias_list, &lcu->active_devices);
+		return 0;
+	}
+	group = _find_group(lcu, &uid);
+	if (!group) {
+		group = kzalloc_obj(*group, GFP_ATOMIC);
+		if (!group)
+			return -ENOMEM;
+		memcpy(group->uid.vendor, uid.vendor, sizeof(uid.vendor));
+		memcpy(group->uid.serial, uid.serial, sizeof(uid.serial));
+		group->uid.ssid = uid.ssid;
+		if (uid.type == UA_BASE_DEVICE)
+			group->uid.base_unit_addr = uid.real_unit_addr;
+		else
+			group->uid.base_unit_addr = uid.base_unit_addr;
+		memcpy(group->uid.vduit, uid.vduit, sizeof(uid.vduit));
+		INIT_LIST_HEAD(&group->group);
+		INIT_LIST_HEAD(&group->baselist);
+		INIT_LIST_HEAD(&group->aliaslist);
+		list_add(&group->group, &lcu->grouplist);
+	}
+	if (uid.type == UA_BASE_DEVICE)
+		list_move(&device->alias_list, &group->baselist);
+	else
+		list_move(&device->alias_list, &group->aliaslist);
+	private->pavgroup = group;
+	return 0;
+};
+
+static void _remove_device_from_lcu(struct alias_lcu *lcu,
+				    struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct alias_pav_group *group;
+
+	list_move(&device->alias_list, &lcu->inactive_devices);
+	group = private->pavgroup;
+	if (!group)
+		return;
+	private->pavgroup = NULL;
+	if (list_empty(&group->baselist) && list_empty(&group->aliaslist)) {
+		list_del(&group->group);
+		kfree(group);
+		return;
+	}
+	if (group->next == device)
+		group->next = NULL;
+};
+
+static int
+suborder_not_supported(struct dasd_ccw_req *cqr)
+{
+	char *sense;
+	char reason;
+	char msg_format;
+	char msg_no;
+
+	/*
+	 * intrc values ENODEV, ENOLINK and EPERM
+	 * will be optained from sleep_on to indicate that no
+	 * IO operation can be started
+	 */
+	if (cqr->intrc == -ENODEV)
+		return 1;
+
+	if (cqr->intrc == -ENOLINK)
+		return 1;
+
+	if (cqr->intrc == -EPERM)
+		return 1;
+
+	sense = dasd_get_sense(&cqr->irb);
+	if (!sense)
+		return 0;
+
+	reason = sense[0];
+	msg_format = (sense[7] & 0xF0);
+	msg_no = (sense[7] & 0x0F);
+
+	/* command reject, Format 0 MSG 4 - invalid parameter */
+	if ((reason == 0x80) && (msg_format == 0x00) && (msg_no == 0x04))
+		return 1;
+
+	return 0;
+}
+
+static int read_unit_address_configuration(struct dasd_device *device,
+					   struct alias_lcu *lcu)
+{
+	struct dasd_psf_prssd_data *prssdp;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int rc;
+	unsigned long flags;
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */	+ 1 /* RSSD */,
+				   (sizeof(struct dasd_psf_prssd_data)),
+				   device, NULL);
+	if (IS_ERR(cqr))
+		return PTR_ERR(cqr);
+	cqr->startdev = device;
+	cqr->memdev = device;
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	cqr->retries = 10;
+	cqr->expires = 20 * HZ;
+
+	/* Prepare for Read Subsystem Data */
+	prssdp = (struct dasd_psf_prssd_data *) cqr->data;
+	memset(prssdp, 0, sizeof(struct dasd_psf_prssd_data));
+	prssdp->order = PSF_ORDER_PRSSD;
+	prssdp->suborder = 0x0e;	/* Read unit address configuration */
+	/* all other bytes of prssdp must be zero */
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->count = sizeof(struct dasd_psf_prssd_data);
+	ccw->flags |= CCW_FLAG_CC;
+	ccw->cda = virt_to_dma32(prssdp);
+
+	/* Read Subsystem Data - feature codes */
+	memset(lcu->uac, 0, sizeof(*(lcu->uac)));
+
+	ccw++;
+	ccw->cmd_code = DASD_ECKD_CCW_RSSD;
+	ccw->count = sizeof(*(lcu->uac));
+	ccw->cda = virt_to_dma32(lcu->uac);
+
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	/* need to unset flag here to detect race with summary unit check */
+	spin_lock_irqsave(&lcu->lock, flags);
+	lcu->flags &= ~NEED_UAC_UPDATE;
+	spin_unlock_irqrestore(&lcu->lock, flags);
+
+	rc = dasd_sleep_on(cqr);
+	if (!rc)
+		goto out;
+
+	if (suborder_not_supported(cqr)) {
+		/* suborder not supported or device unusable for IO */
+		rc = -EOPNOTSUPP;
+	} else {
+		/* IO failed but should be retried */
+		spin_lock_irqsave(&lcu->lock, flags);
+		lcu->flags |= NEED_UAC_UPDATE;
+		spin_unlock_irqrestore(&lcu->lock, flags);
+	}
+out:
+	dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+static int _lcu_update(struct dasd_device *refdev, struct alias_lcu *lcu)
+{
+	unsigned long flags;
+	struct alias_pav_group *pavgroup, *tempgroup;
+	struct dasd_device *device, *tempdev;
+	int i, rc;
+	struct dasd_eckd_private *private;
+
+	spin_lock_irqsave(&lcu->lock, flags);
+	list_for_each_entry_safe(pavgroup, tempgroup, &lcu->grouplist, group) {
+		list_for_each_entry_safe(device, tempdev, &pavgroup->baselist,
+					 alias_list) {
+			list_move(&device->alias_list, &lcu->active_devices);
+			private = device->private;
+			private->pavgroup = NULL;
+		}
+		list_for_each_entry_safe(device, tempdev, &pavgroup->aliaslist,
+					 alias_list) {
+			list_move(&device->alias_list, &lcu->active_devices);
+			private = device->private;
+			private->pavgroup = NULL;
+		}
+		list_del(&pavgroup->group);
+		kfree(pavgroup);
+	}
+	spin_unlock_irqrestore(&lcu->lock, flags);
+
+	rc = read_unit_address_configuration(refdev, lcu);
+	if (rc)
+		return rc;
+
+	spin_lock_irqsave(&lcu->lock, flags);
+	/*
+	 * there is another update needed skip the remaining handling
+	 * the data might already be outdated
+	 * but especially do not add the device to an LCU with pending
+	 * update
+	 */
+	if (lcu->flags & NEED_UAC_UPDATE)
+		goto out;
+	lcu->pav = NO_PAV;
+	for (i = 0; i < MAX_DEVICES_PER_LCU; ++i) {
+		switch (lcu->uac->unit[i].ua_type) {
+		case UA_BASE_PAV_ALIAS:
+			lcu->pav = BASE_PAV;
+			break;
+		case UA_HYPER_PAV_ALIAS:
+			lcu->pav = HYPER_PAV;
+			break;
+		}
+		if (lcu->pav != NO_PAV)
+			break;
+	}
+
+	list_for_each_entry_safe(device, tempdev, &lcu->active_devices,
+				 alias_list) {
+		_add_device_to_lcu(lcu, device, refdev);
+	}
+out:
+	spin_unlock_irqrestore(&lcu->lock, flags);
+	return 0;
+}
+
+static void lcu_update_work(struct work_struct *work)
+{
+	struct alias_lcu *lcu;
+	struct read_uac_work_data *ruac_data;
+	struct dasd_device *device;
+	unsigned long flags;
+	int rc;
+
+	ruac_data = container_of(work, struct read_uac_work_data, dwork.work);
+	lcu = container_of(ruac_data, struct alias_lcu, ruac_data);
+	device = ruac_data->device;
+	rc = _lcu_update(device, lcu);
+	/*
+	 * Need to check flags again, as there could have been another
+	 * prepare_update or a new device a new device while we were still
+	 * processing the data
+	 */
+	spin_lock_irqsave(&lcu->lock, flags);
+	if ((rc && (rc != -EOPNOTSUPP)) || (lcu->flags & NEED_UAC_UPDATE)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "could not update"
+			    " alias data in lcu (rc = %d), retry later", rc);
+		if (!schedule_delayed_work(&lcu->ruac_data.dwork, 30*HZ))
+			dasd_put_device(device);
+	} else {
+		dasd_put_device(device);
+		lcu->ruac_data.device = NULL;
+		lcu->flags &= ~UPDATE_PENDING;
+	}
+	spin_unlock_irqrestore(&lcu->lock, flags);
+}
+
+static int _schedule_lcu_update(struct alias_lcu *lcu,
+				struct dasd_device *device)
+{
+	struct dasd_device *usedev = NULL;
+	struct alias_pav_group *group;
+
+	lcu->flags |= NEED_UAC_UPDATE;
+	if (lcu->ruac_data.device) {
+		/* already scheduled or running */
+		return 0;
+	}
+	if (device && !list_empty(&device->alias_list))
+		usedev = device;
+
+	if (!usedev && !list_empty(&lcu->grouplist)) {
+		group = list_first_entry(&lcu->grouplist,
+					 struct alias_pav_group, group);
+		if (!list_empty(&group->baselist))
+			usedev = list_first_entry(&group->baselist,
+						  struct dasd_device,
+						  alias_list);
+		else if (!list_empty(&group->aliaslist))
+			usedev = list_first_entry(&group->aliaslist,
+						  struct dasd_device,
+						  alias_list);
+	}
+	if (!usedev && !list_empty(&lcu->active_devices)) {
+		usedev = list_first_entry(&lcu->active_devices,
+					  struct dasd_device, alias_list);
+	}
+	/*
+	 * if we haven't found a proper device yet, give up for now, the next
+	 * device that will be set active will trigger an lcu update
+	 */
+	if (!usedev)
+		return -EINVAL;
+	dasd_get_device(usedev);
+	lcu->ruac_data.device = usedev;
+	if (!schedule_delayed_work(&lcu->ruac_data.dwork, 0))
+		dasd_put_device(usedev);
+	return 0;
+}
+
+int dasd_alias_add_device(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	__u8 uaddr = private->uid.real_unit_addr;
+	struct alias_lcu *lcu = private->lcu;
+	unsigned long flags;
+	int rc;
+
+	rc = 0;
+	spin_lock_irqsave(&lcu->lock, flags);
+	/*
+	 * Check if device and lcu type differ. If so, the uac data may be
+	 * outdated and needs to be updated.
+	 */
+	if (private->uid.type !=  lcu->uac->unit[uaddr].ua_type) {
+		lcu->flags |= UPDATE_PENDING;
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "uid type mismatch - trigger rescan");
+	}
+	if (!(lcu->flags & UPDATE_PENDING)) {
+		rc = _add_device_to_lcu(lcu, device, device);
+		if (rc)
+			lcu->flags |= UPDATE_PENDING;
+	}
+	if (lcu->flags & UPDATE_PENDING) {
+		list_move(&device->alias_list, &lcu->active_devices);
+		private->pavgroup = NULL;
+		_schedule_lcu_update(lcu, device);
+	}
+	spin_unlock_irqrestore(&lcu->lock, flags);
+	return rc;
+}
+
+int dasd_alias_update_add_device(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	private->lcu->flags |= UPDATE_PENDING;
+	return dasd_alias_add_device(device);
+}
+
+int dasd_alias_remove_device(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct alias_lcu *lcu = private->lcu;
+	unsigned long flags;
+
+	/* nothing to do if already removed */
+	if (!lcu)
+		return 0;
+	spin_lock_irqsave(&lcu->lock, flags);
+	_remove_device_from_lcu(lcu, device);
+	spin_unlock_irqrestore(&lcu->lock, flags);
+	return 0;
+}
+
+struct dasd_device *dasd_alias_get_start_dev(struct dasd_device *base_device)
+{
+	struct dasd_eckd_private *alias_priv, *private = base_device->private;
+	struct alias_lcu *lcu = private->lcu;
+	struct dasd_device *alias_device;
+	struct alias_pav_group *group;
+	unsigned long flags;
+
+	if (!lcu)
+		return NULL;
+	if (lcu->pav == NO_PAV ||
+	    lcu->flags & (NEED_UAC_UPDATE | UPDATE_PENDING))
+		return NULL;
+	if (unlikely(!(private->features.feature[8] & 0x01))) {
+		/*
+		 * PAV enabled but prefix not, very unlikely
+		 * seems to be a lost pathgroup
+		 * use base device to do IO
+		 */
+		DBF_DEV_EVENT(DBF_ERR, base_device, "%s",
+			      "Prefix not enabled with PAV enabled\n");
+		return NULL;
+	}
+
+	spin_lock_irqsave(&lcu->lock, flags);
+	group = private->pavgroup;
+	if (!group) {
+		spin_unlock_irqrestore(&lcu->lock, flags);
+		return NULL;
+	}
+	alias_device = group->next;
+	if (!alias_device) {
+		if (list_empty(&group->aliaslist)) {
+			spin_unlock_irqrestore(&lcu->lock, flags);
+			return NULL;
+		} else {
+			alias_device = list_first_entry(&group->aliaslist,
+							struct dasd_device,
+							alias_list);
+		}
+	}
+	if (list_is_last(&alias_device->alias_list, &group->aliaslist))
+		group->next = list_first_entry(&group->aliaslist,
+					       struct dasd_device, alias_list);
+	else
+		group->next = list_first_entry(&alias_device->alias_list,
+					       struct dasd_device, alias_list);
+	spin_unlock_irqrestore(&lcu->lock, flags);
+	alias_priv = alias_device->private;
+	if ((alias_priv->count < private->count) && !alias_device->stopped &&
+	    !test_bit(DASD_FLAG_OFFLINE, &alias_device->flags))
+		return alias_device;
+	else
+		return NULL;
+}
+
+/*
+ * Summary unit check handling depends on the way alias devices
+ * are handled so it is done here rather then in dasd_eckd.c
+ */
+static int reset_summary_unit_check(struct alias_lcu *lcu,
+				    struct dasd_device *device,
+				    char reason)
+{
+	struct dasd_ccw_req *cqr;
+	int rc = 0;
+	struct ccw1 *ccw;
+
+	cqr = lcu->rsu_cqr;
+	memcpy((char *) &cqr->magic, "ECKD", 4);
+	ASCEBC((char *) &cqr->magic, 4);
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_RSCK;
+	ccw->flags = CCW_FLAG_SLI;
+	ccw->count = 16;
+	ccw->cda = virt_to_dma32(cqr->data);
+	((char *)cqr->data)[0] = reason;
+
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	cqr->retries = 255;	/* set retry counter to enable basic ERP */
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->expires = 5 * HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	rc = dasd_sleep_on_immediatly(cqr);
+	return rc;
+}
+
+static void _restart_all_base_devices_on_lcu(struct alias_lcu *lcu)
+{
+	struct alias_pav_group *pavgroup;
+	struct dasd_device *device;
+	struct dasd_eckd_private *private;
+
+	/* active and inactive list can contain alias as well as base devices */
+	list_for_each_entry(device, &lcu->active_devices, alias_list) {
+		private = device->private;
+		if (private->uid.type != UA_BASE_DEVICE)
+			continue;
+		dasd_schedule_block_bh(device->block);
+		dasd_schedule_device_bh(device);
+	}
+	list_for_each_entry(device, &lcu->inactive_devices, alias_list) {
+		private = device->private;
+		if (private->uid.type != UA_BASE_DEVICE)
+			continue;
+		dasd_schedule_block_bh(device->block);
+		dasd_schedule_device_bh(device);
+	}
+	list_for_each_entry(pavgroup, &lcu->grouplist, group) {
+		list_for_each_entry(device, &pavgroup->baselist, alias_list) {
+			dasd_schedule_block_bh(device->block);
+			dasd_schedule_device_bh(device);
+		}
+	}
+}
+
+static void flush_all_alias_devices_on_lcu(struct alias_lcu *lcu)
+{
+	struct alias_pav_group *pavgroup;
+	struct dasd_device *device, *temp;
+	struct dasd_eckd_private *private;
+	unsigned long flags;
+	LIST_HEAD(active);
+
+	/*
+	 * Problem here ist that dasd_flush_device_queue may wait
+	 * for termination of a request to complete. We can't keep
+	 * the lcu lock during that time, so we must assume that
+	 * the lists may have changed.
+	 * Idea: first gather all active alias devices in a separate list,
+	 * then flush the first element of this list unlocked, and afterwards
+	 * check if it is still on the list before moving it to the
+	 * active_devices list.
+	 */
+
+	spin_lock_irqsave(&lcu->lock, flags);
+	list_for_each_entry_safe(device, temp, &lcu->active_devices,
+				 alias_list) {
+		private = device->private;
+		if (private->uid.type == UA_BASE_DEVICE)
+			continue;
+		list_move(&device->alias_list, &active);
+	}
+
+	list_for_each_entry(pavgroup, &lcu->grouplist, group) {
+		list_splice_init(&pavgroup->aliaslist, &active);
+	}
+	while (!list_empty(&active)) {
+		device = list_first_entry(&active, struct dasd_device,
+					  alias_list);
+		spin_unlock_irqrestore(&lcu->lock, flags);
+		dasd_flush_device_queue(device);
+		spin_lock_irqsave(&lcu->lock, flags);
+		/*
+		 * only move device around if it wasn't moved away while we
+		 * were waiting for the flush
+		 */
+		if (device == list_first_entry(&active,
+					       struct dasd_device, alias_list)) {
+			list_move(&device->alias_list, &lcu->active_devices);
+			private = device->private;
+			private->pavgroup = NULL;
+		}
+	}
+	spin_unlock_irqrestore(&lcu->lock, flags);
+}
+
+static void _stop_all_devices_on_lcu(struct alias_lcu *lcu)
+{
+	struct alias_pav_group *pavgroup;
+	struct dasd_device *device;
+
+	list_for_each_entry(device, &lcu->active_devices, alias_list) {
+		spin_lock(get_ccwdev_lock(device->cdev));
+		dasd_device_set_stop_bits(device, DASD_STOPPED_SU);
+		spin_unlock(get_ccwdev_lock(device->cdev));
+	}
+	list_for_each_entry(device, &lcu->inactive_devices, alias_list) {
+		spin_lock(get_ccwdev_lock(device->cdev));
+		dasd_device_set_stop_bits(device, DASD_STOPPED_SU);
+		spin_unlock(get_ccwdev_lock(device->cdev));
+	}
+	list_for_each_entry(pavgroup, &lcu->grouplist, group) {
+		list_for_each_entry(device, &pavgroup->baselist, alias_list) {
+			spin_lock(get_ccwdev_lock(device->cdev));
+			dasd_device_set_stop_bits(device, DASD_STOPPED_SU);
+			spin_unlock(get_ccwdev_lock(device->cdev));
+		}
+		list_for_each_entry(device, &pavgroup->aliaslist, alias_list) {
+			spin_lock(get_ccwdev_lock(device->cdev));
+			dasd_device_set_stop_bits(device, DASD_STOPPED_SU);
+			spin_unlock(get_ccwdev_lock(device->cdev));
+		}
+	}
+}
+
+static void _unstop_all_devices_on_lcu(struct alias_lcu *lcu)
+{
+	struct alias_pav_group *pavgroup;
+	struct dasd_device *device;
+
+	list_for_each_entry(device, &lcu->active_devices, alias_list) {
+		spin_lock(get_ccwdev_lock(device->cdev));
+		dasd_device_remove_stop_bits(device, DASD_STOPPED_SU);
+		spin_unlock(get_ccwdev_lock(device->cdev));
+	}
+	list_for_each_entry(device, &lcu->inactive_devices, alias_list) {
+		spin_lock(get_ccwdev_lock(device->cdev));
+		dasd_device_remove_stop_bits(device, DASD_STOPPED_SU);
+		spin_unlock(get_ccwdev_lock(device->cdev));
+	}
+	list_for_each_entry(pavgroup, &lcu->grouplist, group) {
+		list_for_each_entry(device, &pavgroup->baselist, alias_list) {
+			spin_lock(get_ccwdev_lock(device->cdev));
+			dasd_device_remove_stop_bits(device, DASD_STOPPED_SU);
+			spin_unlock(get_ccwdev_lock(device->cdev));
+		}
+		list_for_each_entry(device, &pavgroup->aliaslist, alias_list) {
+			spin_lock(get_ccwdev_lock(device->cdev));
+			dasd_device_remove_stop_bits(device, DASD_STOPPED_SU);
+			spin_unlock(get_ccwdev_lock(device->cdev));
+		}
+	}
+}
+
+static void summary_unit_check_handling_work(struct work_struct *work)
+{
+	struct alias_lcu *lcu;
+	struct summary_unit_check_work_data *suc_data;
+	unsigned long flags;
+	struct dasd_device *device;
+
+	suc_data = container_of(work, struct summary_unit_check_work_data,
+				worker);
+	lcu = container_of(suc_data, struct alias_lcu, suc_data);
+	device = suc_data->device;
+
+	/* 1. flush alias devices */
+	flush_all_alias_devices_on_lcu(lcu);
+
+	/* 2. reset summary unit check */
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	dasd_device_remove_stop_bits(device,
+				     (DASD_STOPPED_SU | DASD_STOPPED_PENDING));
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	reset_summary_unit_check(lcu, device, suc_data->reason);
+
+	spin_lock_irqsave(&lcu->lock, flags);
+	_unstop_all_devices_on_lcu(lcu);
+	_restart_all_base_devices_on_lcu(lcu);
+	/* 3. read new alias configuration */
+	_schedule_lcu_update(lcu, device);
+	lcu->suc_data.device = NULL;
+	dasd_put_device(device);
+	spin_unlock_irqrestore(&lcu->lock, flags);
+}
+
+void dasd_alias_handle_summary_unit_check(struct work_struct *work)
+{
+	struct dasd_device *device = container_of(work, struct dasd_device,
+						  suc_work);
+	struct dasd_eckd_private *private = device->private;
+	struct alias_lcu *lcu;
+	unsigned long flags;
+
+	lcu = private->lcu;
+	if (!lcu) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "device not ready to handle summary"
+			    " unit check (no lcu structure)");
+		goto out;
+	}
+	spin_lock_irqsave(&lcu->lock, flags);
+	/* If this device is about to be removed just return and wait for
+	 * the next interrupt on a different device
+	 */
+	if (list_empty(&device->alias_list)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "device is in offline processing,"
+			    " don't do summary unit check handling");
+		goto out_unlock;
+	}
+	if (lcu->suc_data.device) {
+		/* already scheduled or running */
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "previous instance of summary unit check worker"
+			    " still pending");
+		goto out_unlock;
+	}
+	_stop_all_devices_on_lcu(lcu);
+	/* prepare for lcu_update */
+	lcu->flags |= NEED_UAC_UPDATE | UPDATE_PENDING;
+	lcu->suc_data.reason = private->suc_reason;
+	lcu->suc_data.device = device;
+	dasd_get_device(device);
+	if (!schedule_work(&lcu->suc_data.worker))
+		dasd_put_device(device);
+out_unlock:
+	spin_unlock_irqrestore(&lcu->lock, flags);
+out:
+	clear_bit(DASD_FLAG_SUC, &device->flags);
+	dasd_put_device(device);
+};
diff --git a/drivers/s390/block/dasd_devmap.c b/drivers/s390/block/dasd_devmap.c
new file mode 100644
index 000000000..d07d384a0
--- /dev/null
+++ b/drivers/s390/block/dasd_devmap.c
@@ -0,0 +1,2708 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Carsten Otte <Cotte@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999,2001
+ *
+ * Device mapping and dasd= parameter parsing functions. All devmap
+ * functions may not be called from interrupt context. In particular
+ * dasd_get_device is a no-no from interrupt context.
+ *
+ */
+
+#include <linux/export.h>
+#include <linux/ctype.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+
+#include <asm/machine.h>
+#include <asm/debug.h>
+#include <linux/uaccess.h>
+#include <asm/ipl.h>
+
+#define DASD_MAX_PARAMS 256
+
+#include "dasd_int.h"
+
+struct kmem_cache *dasd_page_cache;
+EXPORT_SYMBOL_GPL(dasd_page_cache);
+
+/*
+ * dasd_devmap_t is used to store the features and the relation
+ * between device number and device index. To find a dasd_devmap_t
+ * that corresponds to a device number of a device index each
+ * dasd_devmap_t is added to two linked lists, one to search by
+ * the device number and one to search by the device index. As
+ * soon as big minor numbers are available the device index list
+ * can be removed since the device number will then be identical
+ * to the device index.
+ */
+struct dasd_devmap {
+	struct list_head list;
+	char bus_id[DASD_BUS_ID_SIZE];
+        unsigned int devindex;
+        unsigned short features;
+	struct dasd_device *device;
+	struct dasd_copy_relation *copy;
+	unsigned int aq_mask;
+};
+
+/*
+ * Parameter parsing functions for dasd= parameter. The syntax is:
+ *   <devno>		: (0x)?[0-9a-fA-F]+
+ *   <busid>		: [0-0a-f]\.[0-9a-f]\.(0x)?[0-9a-fA-F]+
+ *   <feature>		: ro
+ *   <feature_list>	: \(<feature>(:<feature>)*\)
+ *   <devno-range>	: <devno>(-<devno>)?<feature_list>?
+ *   <busid-range>	: <busid>(-<busid>)?<feature_list>?
+ *   <devices>		: <devno-range>|<busid-range>
+ *   <dasd_module>	: dasd_diag_mod|dasd_eckd_mod|dasd_fba_mod
+ *
+ *   <dasd>		: autodetect|probeonly|<devices>(,<devices>)*
+ */
+
+int dasd_probeonly =  0;	/* is true, when probeonly mode is active */
+int dasd_autodetect = 0;	/* is true, when autodetection is active */
+int dasd_nopav = 0;		/* is true, when PAV is disabled */
+EXPORT_SYMBOL_GPL(dasd_nopav);
+int dasd_nofcx;			/* disable High Performance Ficon */
+EXPORT_SYMBOL_GPL(dasd_nofcx);
+
+/*
+ * char *dasd[] is intended to hold the ranges supplied by the dasd= statement
+ * it is named 'dasd' to directly be filled by insmod with the comma separated
+ * strings when running as a module.
+ */
+static char *dasd[DASD_MAX_PARAMS];
+module_param_array(dasd, charp, NULL, S_IRUGO);
+
+/*
+ * Single spinlock to protect devmap and servermap structures and lists.
+ */
+static DEFINE_SPINLOCK(dasd_devmap_lock);
+
+/*
+ * Hash lists for devmap structures.
+ */
+static struct list_head dasd_hashlists[256];
+int dasd_max_devindex;
+
+static struct dasd_devmap *dasd_add_busid(const char *, int);
+
+static inline int
+dasd_hash_busid(const char *bus_id)
+{
+	int hash, i;
+
+	hash = 0;
+	for (i = 0; (i < DASD_BUS_ID_SIZE) && *bus_id; i++, bus_id++)
+		hash += *bus_id;
+	return hash & 0xff;
+}
+
+#ifndef MODULE
+static int __init dasd_call_setup(char *opt)
+{
+	static int i __initdata;
+	char *tmp;
+
+	while (i < DASD_MAX_PARAMS) {
+		tmp = strsep(&opt, ",");
+		if (!tmp)
+			break;
+
+		dasd[i++] = tmp;
+	}
+
+	return 1;
+}
+
+__setup ("dasd=", dasd_call_setup);
+#endif	/* #ifndef MODULE */
+
+#define	DASD_IPLDEV	"ipldev"
+
+/*
+ * Read a device busid/devno from a string.
+ */
+static int dasd_busid(char *str, int *id0, int *id1, int *devno)
+{
+	unsigned int val;
+	char *tok;
+
+	/* Interpret ipldev busid */
+	if (strncmp(DASD_IPLDEV, str, strlen(DASD_IPLDEV)) == 0) {
+		if (ipl_info.type != IPL_TYPE_CCW) {
+			pr_err("The IPL device is not a CCW device\n");
+			return -EINVAL;
+		}
+		*id0 = 0;
+		*id1 = ipl_info.data.ccw.dev_id.ssid;
+		*devno = ipl_info.data.ccw.dev_id.devno;
+
+		return 0;
+	}
+
+	/* Old style 0xXXXX or XXXX */
+	if (!kstrtouint(str, 16, &val)) {
+		*id0 = *id1 = 0;
+		if (val > 0xffff)
+			return -EINVAL;
+		*devno = val;
+		return 0;
+	}
+
+	/* New style x.y.z busid */
+	tok = strsep(&str, ".");
+	if (kstrtouint(tok, 16, &val) || val > 0xff)
+		return -EINVAL;
+	*id0 = val;
+
+	tok = strsep(&str, ".");
+	if (kstrtouint(tok, 16, &val) || val > 0xff)
+		return -EINVAL;
+	*id1 = val;
+
+	tok = strsep(&str, ".");
+	if (kstrtouint(tok, 16, &val) || val > 0xffff)
+		return -EINVAL;
+	*devno = val;
+
+	return 0;
+}
+
+/*
+ * Read colon separated list of dasd features.
+ */
+static int __init dasd_feature_list(char *str)
+{
+	int features, len, rc;
+
+	features = 0;
+	rc = 0;
+
+	if (!str)
+		return DASD_FEATURE_DEFAULT;
+
+	while (1) {
+		for (len = 0;
+		     str[len] && str[len] != ':' && str[len] != ')'; len++);
+		if (len == 2 && !strncmp(str, "ro", 2))
+			features |= DASD_FEATURE_READONLY;
+		else if (len == 4 && !strncmp(str, "diag", 4))
+			features |= DASD_FEATURE_USEDIAG;
+		else if (len == 3 && !strncmp(str, "raw", 3))
+			features |= DASD_FEATURE_USERAW;
+		else if (len == 6 && !strncmp(str, "erplog", 6))
+			features |= DASD_FEATURE_ERPLOG;
+		else if (len == 8 && !strncmp(str, "failfast", 8))
+			features |= DASD_FEATURE_FAILFAST;
+		else {
+			pr_warn("%.*s is not a supported device option\n",
+				len, str);
+			rc = -EINVAL;
+		}
+		str += len;
+		if (*str != ':')
+			break;
+		str++;
+	}
+
+	return rc ? : features;
+}
+
+/*
+ * Try to match the first element on the comma separated parse string
+ * with one of the known keywords. If a keyword is found, take the approprate
+ * action and return a pointer to the residual string. If the first element
+ * could not be matched to any keyword then return an error code.
+ */
+static int __init dasd_parse_keyword(char *keyword)
+{
+	int length = strlen(keyword);
+
+	if (strncmp("autodetect", keyword, length) == 0) {
+		dasd_autodetect = 1;
+		pr_info("The autodetection mode has been activated\n");
+		return 0;
+        }
+	if (strncmp("probeonly", keyword, length) == 0) {
+		dasd_probeonly = 1;
+		pr_info("The probeonly mode has been activated\n");
+		return 0;
+        }
+	if (strncmp("nopav", keyword, length) == 0) {
+		if (machine_is_vm())
+			pr_info("'nopav' is not supported on z/VM\n");
+		else {
+			dasd_nopav = 1;
+			pr_info("PAV support has be deactivated\n");
+		}
+		return 0;
+	}
+	if (strncmp("nofcx", keyword, length) == 0) {
+		dasd_nofcx = 1;
+		pr_info("High Performance FICON support has been "
+			"deactivated\n");
+		return 0;
+	}
+	if (strncmp("fixedbuffers", keyword, length) == 0) {
+		if (dasd_page_cache)
+			return 0;
+		dasd_page_cache =
+			kmem_cache_create("dasd_page_cache", PAGE_SIZE,
+					  PAGE_SIZE, SLAB_CACHE_DMA,
+					  NULL);
+		if (!dasd_page_cache)
+			DBF_EVENT(DBF_WARNING, "%s", "Failed to create slab, "
+				"fixed buffer mode disabled.");
+		else
+			DBF_EVENT(DBF_INFO, "%s",
+				 "turning on fixed buffer mode");
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+/*
+ * Split a string of a device range into its pieces and return the from, to, and
+ * feature parts separately.
+ * e.g.:
+ * 0.0.1234-0.0.5678(ro:erplog) -> from: 0.0.1234 to: 0.0.5678 features: ro:erplog
+ * 0.0.8765(raw) -> from: 0.0.8765 to: null features: raw
+ * 0x4321 -> from: 0x4321 to: null features: null
+ */
+static int __init dasd_evaluate_range_param(char *range, char **from_str,
+					    char **to_str, char **features_str)
+{
+	int rc = 0;
+
+	/* Do we have a range or a single device? */
+	if (strchr(range, '-')) {
+		*from_str = strsep(&range, "-");
+		*to_str = strsep(&range, "(");
+		*features_str = strsep(&range, ")");
+	} else {
+		*from_str = strsep(&range, "(");
+		*features_str = strsep(&range, ")");
+	}
+
+	if (*features_str && !range) {
+		pr_warn("A closing parenthesis ')' is missing in the dasd= parameter\n");
+		rc = -EINVAL;
+	}
+
+	return rc;
+}
+
+/*
+ * Try to interprete the range string as a device number or a range of devices.
+ * If the interpretation is successful, create the matching dasd_devmap entries.
+ * If interpretation fails or in case of an error, return an error code.
+ */
+static int __init dasd_parse_range(const char *range)
+{
+	struct dasd_devmap *devmap;
+	int from, from_id0, from_id1;
+	int to, to_id0, to_id1;
+	int features;
+	char bus_id[DASD_BUS_ID_SIZE + 1];
+	char *features_str = NULL;
+	char *from_str = NULL;
+	char *to_str = NULL;
+	int rc = 0;
+	char *tmp;
+
+	tmp = kstrdup(range, GFP_KERNEL);
+	if (!tmp)
+		return -ENOMEM;
+
+	if (dasd_evaluate_range_param(tmp, &from_str, &to_str, &features_str)) {
+		rc = -EINVAL;
+		goto out;
+	}
+
+	if (dasd_busid(from_str, &from_id0, &from_id1, &from)) {
+		rc = -EINVAL;
+		goto out;
+	}
+
+	to = from;
+	to_id0 = from_id0;
+	to_id1 = from_id1;
+	if (to_str) {
+		if (dasd_busid(to_str, &to_id0, &to_id1, &to)) {
+			rc = -EINVAL;
+			goto out;
+		}
+		if (from_id0 != to_id0 || from_id1 != to_id1 || from > to) {
+			pr_err("%s is not a valid device range\n", range);
+			rc = -EINVAL;
+			goto out;
+		}
+	}
+
+	features = dasd_feature_list(features_str);
+	if (features < 0) {
+		rc = -EINVAL;
+		goto out;
+	}
+	/* each device in dasd= parameter should be set initially online */
+	features |= DASD_FEATURE_INITIAL_ONLINE;
+	while (from <= to) {
+		scnprintf(bus_id, sizeof(bus_id),
+			  "%01x.%01x.%04x", from_id0, from_id1, from++);
+		devmap = dasd_add_busid(bus_id, features);
+		if (IS_ERR(devmap)) {
+			rc = PTR_ERR(devmap);
+			goto out;
+		}
+	}
+
+out:
+	kfree(tmp);
+
+	return rc;
+}
+
+/*
+ * Parse parameters stored in dasd[]
+ * The 'dasd=...' parameter allows to specify a comma separated list of
+ * keywords and device ranges. The parameters in that list will be stored as
+ * separate elementes in dasd[].
+ */
+int __init dasd_parse(void)
+{
+	int rc, i;
+	char *cur;
+
+	rc = 0;
+	for (i = 0; i < DASD_MAX_PARAMS; i++) {
+		cur = dasd[i];
+		if (!cur)
+			break;
+		if (*cur == '\0')
+			continue;
+
+		rc = dasd_parse_keyword(cur);
+		if (rc)
+			rc = dasd_parse_range(cur);
+
+		if (rc)
+			break;
+	}
+
+	return rc;
+}
+
+/*
+ * Add a devmap for the device specified by busid. It is possible that
+ * the devmap already exists (dasd= parameter). The order of the devices
+ * added through this function will define the kdevs for the individual
+ * devices.
+ */
+static struct dasd_devmap *
+dasd_add_busid(const char *bus_id, int features)
+{
+	struct dasd_devmap *devmap, *new, *tmp;
+	int hash;
+
+	new = kzalloc_obj(struct dasd_devmap);
+	if (!new)
+		return ERR_PTR(-ENOMEM);
+	spin_lock(&dasd_devmap_lock);
+	devmap = NULL;
+	hash = dasd_hash_busid(bus_id);
+	list_for_each_entry(tmp, &dasd_hashlists[hash], list)
+		if (strncmp(tmp->bus_id, bus_id, DASD_BUS_ID_SIZE) == 0) {
+			devmap = tmp;
+			break;
+		}
+	if (!devmap) {
+		/* This bus_id is new. */
+		new->devindex = dasd_max_devindex++;
+		strscpy(new->bus_id, bus_id, DASD_BUS_ID_SIZE);
+		new->features = features;
+		new->device = NULL;
+		list_add(&new->list, &dasd_hashlists[hash]);
+		devmap = new;
+		new = NULL;
+	}
+	spin_unlock(&dasd_devmap_lock);
+	kfree(new);
+	return devmap;
+}
+
+static struct dasd_devmap *
+dasd_find_busid_locked(const char *bus_id)
+{
+	struct dasd_devmap *devmap, *tmp;
+	int hash;
+
+	devmap = ERR_PTR(-ENODEV);
+	hash = dasd_hash_busid(bus_id);
+	list_for_each_entry(tmp, &dasd_hashlists[hash], list) {
+		if (strncmp(tmp->bus_id, bus_id, DASD_BUS_ID_SIZE) == 0) {
+			devmap = tmp;
+			break;
+		}
+	}
+	return devmap;
+}
+
+/*
+ * Find devmap for device with given bus_id.
+ */
+static struct dasd_devmap *
+dasd_find_busid(const char *bus_id)
+{
+	struct dasd_devmap *devmap;
+
+	spin_lock(&dasd_devmap_lock);
+	devmap = dasd_find_busid_locked(bus_id);
+	spin_unlock(&dasd_devmap_lock);
+	return devmap;
+}
+
+/*
+ * Check if busid has been added to the list of dasd ranges.
+ */
+int
+dasd_busid_known(const char *bus_id)
+{
+	return IS_ERR(dasd_find_busid(bus_id)) ? -ENOENT : 0;
+}
+
+/*
+ * Forget all about the device numbers added so far.
+ * This may only be called at module unload or system shutdown.
+ */
+static void
+dasd_forget_ranges(void)
+{
+	struct dasd_devmap *devmap, *n;
+	int i;
+
+	spin_lock(&dasd_devmap_lock);
+	for (i = 0; i < 256; i++) {
+		list_for_each_entry_safe(devmap, n, &dasd_hashlists[i], list) {
+			BUG_ON(devmap->device != NULL);
+			list_del(&devmap->list);
+			kfree(devmap);
+		}
+	}
+	spin_unlock(&dasd_devmap_lock);
+}
+
+/*
+ * Find the device struct by its device index.
+ */
+struct dasd_device *
+dasd_device_from_devindex(int devindex)
+{
+	struct dasd_devmap *devmap, *tmp;
+	struct dasd_device *device;
+	int i;
+
+	spin_lock(&dasd_devmap_lock);
+	devmap = NULL;
+	for (i = 0; (i < 256) && !devmap; i++)
+		list_for_each_entry(tmp, &dasd_hashlists[i], list)
+			if (tmp->devindex == devindex) {
+				/* Found the devmap for the device. */
+				devmap = tmp;
+				break;
+			}
+	if (devmap && devmap->device) {
+		device = devmap->device;
+		dasd_get_device(device);
+	} else
+		device = ERR_PTR(-ENODEV);
+	spin_unlock(&dasd_devmap_lock);
+	return device;
+}
+
+/*
+ * Return devmap for cdev. If no devmap exists yet, create one and
+ * connect it to the cdev.
+ */
+static struct dasd_devmap *
+dasd_devmap_from_cdev(struct ccw_device *cdev)
+{
+	struct dasd_devmap *devmap;
+
+	devmap = dasd_find_busid(dev_name(&cdev->dev));
+	if (IS_ERR(devmap))
+		devmap = dasd_add_busid(dev_name(&cdev->dev),
+					DASD_FEATURE_DEFAULT);
+	return devmap;
+}
+
+/*
+ * Create a dasd device structure for cdev.
+ */
+struct dasd_device *
+dasd_create_device(struct ccw_device *cdev)
+{
+	struct dasd_devmap *devmap;
+	struct dasd_device *device;
+	unsigned long flags;
+	int rc;
+
+	devmap = dasd_devmap_from_cdev(cdev);
+	if (IS_ERR(devmap))
+		return (void *) devmap;
+
+	device = dasd_alloc_device();
+	if (IS_ERR(device))
+		return device;
+	atomic_set(&device->ref_count, 3);
+
+	spin_lock(&dasd_devmap_lock);
+	if (!devmap->device) {
+		devmap->device = device;
+		device->devindex = devmap->devindex;
+		device->features = devmap->features;
+		get_device(&cdev->dev);
+		device->cdev = cdev;
+		rc = 0;
+	} else
+		/* Someone else was faster. */
+		rc = -EBUSY;
+	spin_unlock(&dasd_devmap_lock);
+
+	if (rc) {
+		dasd_free_device(device);
+		return ERR_PTR(rc);
+	}
+
+	spin_lock_irqsave(get_ccwdev_lock(cdev), flags);
+	dev_set_drvdata(&cdev->dev, device);
+	spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+
+	device->paths_info = kset_create_and_add("paths_info", NULL,
+						 &device->cdev->dev.kobj);
+	if (!device->paths_info)
+		dev_warn(&cdev->dev, "Could not create paths_info kset\n");
+
+	return device;
+}
+
+/*
+ * allocate a PPRC data structure and call the discipline function to fill
+ */
+static int dasd_devmap_get_pprc_status(struct dasd_device *device,
+				       struct dasd_pprc_data_sc4 **data)
+{
+	struct dasd_pprc_data_sc4 *temp;
+
+	if (!device->discipline || !device->discipline->pprc_status) {
+		dev_warn(&device->cdev->dev, "Unable to query copy relation status\n");
+		return -EOPNOTSUPP;
+	}
+	temp = kzalloc_obj(*temp);
+	if (!temp)
+		return -ENOMEM;
+
+	/* get PPRC information from storage */
+	if (device->discipline->pprc_status(device, temp)) {
+		dev_warn(&device->cdev->dev, "Error during copy relation status query\n");
+		kfree(temp);
+		return -EINVAL;
+	}
+	*data = temp;
+
+	return 0;
+}
+
+/*
+ * find an entry in a PPRC device_info array by a given UID
+ * depending on the primary/secondary state of the device it has to be
+ * matched with the respective fields
+ */
+static int dasd_devmap_entry_from_pprc_data(struct dasd_pprc_data_sc4 *data,
+					    struct dasd_uid uid,
+					    bool primary)
+{
+	int i;
+
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		if (primary) {
+			if (data->dev_info[i].prim_cu_ssid == uid.ssid &&
+			    data->dev_info[i].primary == uid.real_unit_addr)
+				return i;
+		} else {
+			if (data->dev_info[i].sec_cu_ssid == uid.ssid &&
+			    data->dev_info[i].secondary == uid.real_unit_addr)
+				return i;
+		}
+	}
+	return -1;
+}
+
+/*
+ * check the consistency of a specified copy relation by checking
+ * the following things:
+ *
+ *   - is the given device part of a copy pair setup
+ *   - does the state of the device match the state in the PPRC status data
+ *   - does the device UID match with the UID in the PPRC status data
+ *   - to prevent misrouted IO check if the given device is present in all
+ *     related PPRC status data
+ */
+static int dasd_devmap_check_copy_relation(struct dasd_device *device,
+					   struct dasd_copy_entry *entry,
+					   struct dasd_pprc_data_sc4 *data,
+					   struct dasd_copy_relation *copy)
+{
+	struct dasd_pprc_data_sc4 *tmp_dat;
+	struct dasd_device *tmp_dev;
+	struct dasd_uid uid;
+	int i, j;
+
+	if (!device->discipline || !device->discipline->get_uid ||
+	    device->discipline->get_uid(device, &uid))
+		return 1;
+
+	i = dasd_devmap_entry_from_pprc_data(data, uid, entry->primary);
+	if (i < 0) {
+		dev_warn(&device->cdev->dev, "Device not part of a copy relation\n");
+		return 1;
+	}
+
+	/* double check which role the current device has */
+	if (entry->primary) {
+		if (data->dev_info[i].flags & 0x80) {
+			dev_warn(&device->cdev->dev, "Copy pair secondary is setup as primary\n");
+			return 1;
+		}
+		if (data->dev_info[i].prim_cu_ssid != uid.ssid ||
+		    data->dev_info[i].primary != uid.real_unit_addr) {
+			dev_warn(&device->cdev->dev,
+				 "Primary device %s does not match copy pair status primary device %04x\n",
+				 dev_name(&device->cdev->dev),
+				 data->dev_info[i].prim_cu_ssid |
+				 data->dev_info[i].primary);
+			return 1;
+		}
+	} else {
+		if (!(data->dev_info[i].flags & 0x80)) {
+			dev_warn(&device->cdev->dev, "Copy pair primary is setup as secondary\n");
+			return 1;
+		}
+		if (data->dev_info[i].sec_cu_ssid != uid.ssid ||
+		    data->dev_info[i].secondary != uid.real_unit_addr) {
+			dev_warn(&device->cdev->dev,
+				 "Secondary device %s does not match copy pair status secondary device %04x\n",
+				 dev_name(&device->cdev->dev),
+				 data->dev_info[i].sec_cu_ssid |
+				 data->dev_info[i].secondary);
+			return 1;
+		}
+	}
+
+	/*
+	 * the current device has to be part of the copy relation of all
+	 * entries to prevent misrouted IO to another copy pair
+	 */
+	for (j = 0; j < DASD_CP_ENTRIES; j++) {
+		if (entry == &copy->entry[j])
+			tmp_dev = device;
+		else
+			tmp_dev = copy->entry[j].device;
+
+		if (!tmp_dev)
+			continue;
+
+		if (dasd_devmap_get_pprc_status(tmp_dev, &tmp_dat))
+			return 1;
+
+		if (dasd_devmap_entry_from_pprc_data(tmp_dat, uid, entry->primary) < 0) {
+			dev_warn(&tmp_dev->cdev->dev,
+				 "Copy pair relation does not contain device: %s\n",
+				 dev_name(&device->cdev->dev));
+			kfree(tmp_dat);
+			return 1;
+		}
+		kfree(tmp_dat);
+	}
+	return 0;
+}
+
+/* delete device from copy relation entry */
+static void dasd_devmap_delete_copy_relation_device(struct dasd_device *device)
+{
+	struct dasd_copy_relation *copy;
+	int i;
+
+	if (!device->copy)
+		return;
+
+	copy = device->copy;
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		if (copy->entry[i].device == device)
+			copy->entry[i].device = NULL;
+	}
+	dasd_put_device(device);
+	device->copy = NULL;
+}
+
+/*
+ * read all required information for a copy relation setup and setup the device
+ * accordingly
+ */
+int dasd_devmap_set_device_copy_relation(struct ccw_device *cdev,
+					 bool pprc_enabled)
+{
+	struct dasd_pprc_data_sc4 *data = NULL;
+	struct dasd_copy_entry *entry = NULL;
+	struct dasd_copy_relation *copy;
+	struct dasd_devmap *devmap;
+	struct dasd_device *device;
+	int i, rc = 0;
+
+	devmap = dasd_devmap_from_cdev(cdev);
+	if (IS_ERR(devmap))
+		return PTR_ERR(devmap);
+
+	device = devmap->device;
+	if (!device)
+		return -ENODEV;
+
+	copy = devmap->copy;
+	/* no copy pair setup for this device */
+	if (!copy)
+		goto out;
+
+	rc = dasd_devmap_get_pprc_status(device, &data);
+	if (rc)
+		return rc;
+
+	/* print error if PPRC is requested but not enabled on storage server */
+	if (!pprc_enabled) {
+		dev_err(&cdev->dev, "Copy relation not enabled on storage server\n");
+		rc = -EINVAL;
+		goto out;
+	}
+
+	if (!data->dev_info[0].state) {
+		dev_warn(&device->cdev->dev, "Copy pair setup requested for device not in copy relation\n");
+		rc = -EINVAL;
+		goto out;
+	}
+	/* find entry */
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		if (copy->entry[i].configured &&
+		    strncmp(dev_name(&cdev->dev),
+			    copy->entry[i].busid, DASD_BUS_ID_SIZE) == 0) {
+			entry = &copy->entry[i];
+			break;
+		}
+	}
+	if (!entry) {
+		dev_warn(&device->cdev->dev, "Copy relation entry not found\n");
+		rc = -EINVAL;
+		goto out;
+	}
+	/* check if the copy relation is valid */
+	if (dasd_devmap_check_copy_relation(device, entry, data, copy)) {
+		dev_warn(&device->cdev->dev, "Copy relation faulty\n");
+		rc = -EINVAL;
+		goto out;
+	}
+
+	dasd_get_device(device);
+	copy->entry[i].device = device;
+	device->copy = copy;
+out:
+	kfree(data);
+	return rc;
+}
+EXPORT_SYMBOL_GPL(dasd_devmap_set_device_copy_relation);
+
+/*
+ * Wait queue for dasd_delete_device waits.
+ */
+static DECLARE_WAIT_QUEUE_HEAD(dasd_delete_wq);
+
+/*
+ * Remove a dasd device structure. The passed referenced
+ * is destroyed.
+ */
+void
+dasd_delete_device(struct dasd_device *device)
+{
+	struct ccw_device *cdev;
+	struct dasd_devmap *devmap;
+	unsigned long flags;
+
+	/* First remove device pointer from devmap. */
+	devmap = dasd_find_busid(dev_name(&device->cdev->dev));
+	BUG_ON(IS_ERR(devmap));
+	spin_lock(&dasd_devmap_lock);
+	if (devmap->device != device) {
+		spin_unlock(&dasd_devmap_lock);
+		dasd_put_device(device);
+		return;
+	}
+	devmap->device = NULL;
+	spin_unlock(&dasd_devmap_lock);
+
+	/* Disconnect dasd_device structure from ccw_device structure. */
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	dev_set_drvdata(&device->cdev->dev, NULL);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+
+	/* Remove copy relation */
+	dasd_devmap_delete_copy_relation_device(device);
+	/*
+	 * Drop ref_count by 3, one for the devmap reference, one for
+	 * the cdev reference and one for the passed reference.
+	 */
+	atomic_sub(3, &device->ref_count);
+
+	/* Wait for reference counter to drop to zero. */
+	wait_event(dasd_delete_wq, atomic_read(&device->ref_count) == 0);
+
+	dasd_generic_free_discipline(device);
+
+	kset_unregister(device->paths_info);
+
+	/* Disconnect dasd_device structure from ccw_device structure. */
+	cdev = device->cdev;
+	device->cdev = NULL;
+
+	/* Put ccw_device structure. */
+	put_device(&cdev->dev);
+
+	/* Now the device structure can be freed. */
+	dasd_free_device(device);
+}
+
+/*
+ * Reference counter dropped to zero. Wake up waiter
+ * in dasd_delete_device.
+ */
+void
+dasd_put_device_wake(struct dasd_device *device)
+{
+	wake_up(&dasd_delete_wq);
+}
+EXPORT_SYMBOL_GPL(dasd_put_device_wake);
+
+/*
+ * Return dasd_device structure associated with cdev.
+ * This function needs to be called with the ccw device
+ * lock held. It can be used from interrupt context.
+ */
+struct dasd_device *
+dasd_device_from_cdev_locked(struct ccw_device *cdev)
+{
+	struct dasd_device *device = dev_get_drvdata(&cdev->dev);
+
+	if (!device)
+		return ERR_PTR(-ENODEV);
+	dasd_get_device(device);
+	return device;
+}
+
+/*
+ * Return dasd_device structure associated with cdev.
+ */
+struct dasd_device *
+dasd_device_from_cdev(struct ccw_device *cdev)
+{
+	struct dasd_device *device;
+	unsigned long flags;
+
+	spin_lock_irqsave(get_ccwdev_lock(cdev), flags);
+	device = dasd_device_from_cdev_locked(cdev);
+	spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+	return device;
+}
+
+void dasd_add_link_to_gendisk(struct gendisk *gdp, struct dasd_device *device)
+{
+	struct dasd_devmap *devmap;
+
+	devmap = dasd_find_busid(dev_name(&device->cdev->dev));
+	if (IS_ERR(devmap))
+		return;
+	spin_lock(&dasd_devmap_lock);
+	gdp->private_data = devmap;
+	spin_unlock(&dasd_devmap_lock);
+}
+EXPORT_SYMBOL(dasd_add_link_to_gendisk);
+
+struct dasd_device *dasd_device_from_gendisk(struct gendisk *gdp)
+{
+	struct dasd_device *device;
+	struct dasd_devmap *devmap;
+
+	if (!gdp->private_data)
+		return NULL;
+	device = NULL;
+	spin_lock(&dasd_devmap_lock);
+	devmap = gdp->private_data;
+	if (devmap && devmap->device) {
+		device = devmap->device;
+		dasd_get_device(device);
+	}
+	spin_unlock(&dasd_devmap_lock);
+	return device;
+}
+
+/*
+ * SECTION: files in sysfs
+ */
+
+/*
+ * failfast controls the behaviour, if no path is available
+ */
+static ssize_t dasd_ff_show(struct device *dev, struct device_attribute *attr,
+			    char *buf)
+{
+	struct dasd_devmap *devmap;
+	int ff_flag;
+
+	devmap = dasd_find_busid(dev_name(dev));
+	if (!IS_ERR(devmap))
+		ff_flag = (devmap->features & DASD_FEATURE_FAILFAST) != 0;
+	else
+		ff_flag = (DASD_FEATURE_DEFAULT & DASD_FEATURE_FAILFAST) != 0;
+	return sysfs_emit(buf, ff_flag ? "1\n" : "0\n");
+}
+
+static ssize_t dasd_ff_store(struct device *dev, struct device_attribute *attr,
+	      const char *buf, size_t count)
+{
+	unsigned int val;
+	int rc;
+
+	if (kstrtouint(buf, 0, &val) || val > 1)
+		return -EINVAL;
+
+	rc = dasd_set_feature(to_ccwdev(dev), DASD_FEATURE_FAILFAST, val);
+
+	return rc ? : count;
+}
+
+static DEVICE_ATTR(failfast, 0644, dasd_ff_show, dasd_ff_store);
+
+/*
+ * readonly controls the readonly status of a dasd
+ */
+static ssize_t
+dasd_ro_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct dasd_devmap *devmap;
+	struct dasd_device *device;
+	int ro_flag = 0;
+
+	devmap = dasd_find_busid(dev_name(dev));
+	if (IS_ERR(devmap))
+		goto out;
+
+	ro_flag = !!(devmap->features & DASD_FEATURE_READONLY);
+
+	spin_lock(&dasd_devmap_lock);
+	device = devmap->device;
+	if (device)
+		ro_flag |= test_bit(DASD_FLAG_DEVICE_RO, &device->flags);
+	spin_unlock(&dasd_devmap_lock);
+
+out:
+	return sysfs_emit(buf, ro_flag ? "1\n" : "0\n");
+}
+
+static ssize_t
+dasd_ro_store(struct device *dev, struct device_attribute *attr,
+	      const char *buf, size_t count)
+{
+	struct ccw_device *cdev = to_ccwdev(dev);
+	struct dasd_device *device;
+	unsigned long flags;
+	unsigned int val;
+	int rc;
+
+	if (kstrtouint(buf, 0, &val) || val > 1)
+		return -EINVAL;
+
+	rc = dasd_set_feature(cdev, DASD_FEATURE_READONLY, val);
+	if (rc)
+		return rc;
+
+	device = dasd_device_from_cdev(cdev);
+	if (IS_ERR(device))
+		return count;
+
+	spin_lock_irqsave(get_ccwdev_lock(cdev), flags);
+	val = val || test_bit(DASD_FLAG_DEVICE_RO, &device->flags);
+
+	if (!device->block || !device->block->gdp ||
+	    test_bit(DASD_FLAG_OFFLINE, &device->flags)) {
+		spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+		goto out;
+	}
+	/* Increase open_count to avoid losing the block device */
+	atomic_inc(&device->block->open_count);
+	spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+
+	set_disk_ro(device->block->gdp, val);
+	atomic_dec(&device->block->open_count);
+
+out:
+	dasd_put_device(device);
+
+	return count;
+}
+
+static DEVICE_ATTR(readonly, 0644, dasd_ro_show, dasd_ro_store);
+/*
+ * erplog controls the logging of ERP related data
+ * (e.g. failing channel programs).
+ */
+static ssize_t
+dasd_erplog_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct dasd_devmap *devmap;
+	int erplog;
+
+	devmap = dasd_find_busid(dev_name(dev));
+	if (!IS_ERR(devmap))
+		erplog = (devmap->features & DASD_FEATURE_ERPLOG) != 0;
+	else
+		erplog = (DASD_FEATURE_DEFAULT & DASD_FEATURE_ERPLOG) != 0;
+	return sysfs_emit(buf, erplog ? "1\n" : "0\n");
+}
+
+static ssize_t
+dasd_erplog_store(struct device *dev, struct device_attribute *attr,
+	      const char *buf, size_t count)
+{
+	unsigned int val;
+	int rc;
+
+	if (kstrtouint(buf, 0, &val) || val > 1)
+		return -EINVAL;
+
+	rc = dasd_set_feature(to_ccwdev(dev), DASD_FEATURE_ERPLOG, val);
+
+	return rc ? : count;
+}
+
+static DEVICE_ATTR(erplog, 0644, dasd_erplog_show, dasd_erplog_store);
+
+/*
+ * use_diag controls whether the driver should use diag rather than ssch
+ * to talk to the device
+ */
+static ssize_t
+dasd_use_diag_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct dasd_devmap *devmap;
+	int use_diag;
+
+	devmap = dasd_find_busid(dev_name(dev));
+	if (!IS_ERR(devmap))
+		use_diag = (devmap->features & DASD_FEATURE_USEDIAG) != 0;
+	else
+		use_diag = (DASD_FEATURE_DEFAULT & DASD_FEATURE_USEDIAG) != 0;
+	return sysfs_emit(buf, use_diag ? "1\n" : "0\n");
+}
+
+static ssize_t
+dasd_use_diag_store(struct device *dev, struct device_attribute *attr,
+		    const char *buf, size_t count)
+{
+	struct dasd_devmap *devmap;
+	unsigned int val;
+	ssize_t rc;
+
+	devmap = dasd_devmap_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(devmap))
+		return PTR_ERR(devmap);
+
+	if (kstrtouint(buf, 0, &val) || val > 1)
+		return -EINVAL;
+
+	spin_lock(&dasd_devmap_lock);
+	/* Changing diag discipline flag is only allowed in offline state. */
+	rc = count;
+	if (!devmap->device && !(devmap->features & DASD_FEATURE_USERAW)) {
+		if (val)
+			devmap->features |= DASD_FEATURE_USEDIAG;
+		else
+			devmap->features &= ~DASD_FEATURE_USEDIAG;
+	} else
+		rc = -EPERM;
+	spin_unlock(&dasd_devmap_lock);
+	return rc;
+}
+
+static DEVICE_ATTR(use_diag, 0644, dasd_use_diag_show, dasd_use_diag_store);
+
+/*
+ * use_raw controls whether the driver should give access to raw eckd data or
+ * operate in standard mode
+ */
+static ssize_t
+dasd_use_raw_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct dasd_devmap *devmap;
+	int use_raw;
+
+	devmap = dasd_find_busid(dev_name(dev));
+	if (!IS_ERR(devmap))
+		use_raw = (devmap->features & DASD_FEATURE_USERAW) != 0;
+	else
+		use_raw = (DASD_FEATURE_DEFAULT & DASD_FEATURE_USERAW) != 0;
+	return sysfs_emit(buf, use_raw ? "1\n" : "0\n");
+}
+
+static ssize_t
+dasd_use_raw_store(struct device *dev, struct device_attribute *attr,
+		    const char *buf, size_t count)
+{
+	struct dasd_devmap *devmap;
+	ssize_t rc;
+	unsigned long val;
+
+	devmap = dasd_devmap_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(devmap))
+		return PTR_ERR(devmap);
+
+	if ((kstrtoul(buf, 10, &val) != 0) || val > 1)
+		return -EINVAL;
+
+	spin_lock(&dasd_devmap_lock);
+	/* Changing diag discipline flag is only allowed in offline state. */
+	rc = count;
+	if (!devmap->device && !(devmap->features & DASD_FEATURE_USEDIAG)) {
+		if (val)
+			devmap->features |= DASD_FEATURE_USERAW;
+		else
+			devmap->features &= ~DASD_FEATURE_USERAW;
+	} else
+		rc = -EPERM;
+	spin_unlock(&dasd_devmap_lock);
+	return rc;
+}
+
+static DEVICE_ATTR(raw_track_access, 0644, dasd_use_raw_show,
+		   dasd_use_raw_store);
+
+static ssize_t
+dasd_safe_offline_store(struct device *dev, struct device_attribute *attr,
+			const char *buf, size_t count)
+{
+	struct ccw_device *cdev = to_ccwdev(dev);
+	struct dasd_device *device;
+	unsigned long flags;
+	int rc;
+
+	spin_lock_irqsave(get_ccwdev_lock(cdev), flags);
+	device = dasd_device_from_cdev_locked(cdev);
+	if (IS_ERR(device)) {
+		rc = PTR_ERR(device);
+		spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+		goto out;
+	}
+
+	if (test_bit(DASD_FLAG_OFFLINE, &device->flags) ||
+	    test_bit(DASD_FLAG_SAFE_OFFLINE_RUNNING, &device->flags)) {
+		/* Already doing offline processing */
+		dasd_put_device(device);
+		spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+		rc = -EBUSY;
+		goto out;
+	}
+
+	set_bit(DASD_FLAG_SAFE_OFFLINE, &device->flags);
+	dasd_put_device(device);
+	spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags);
+
+	rc = ccw_device_set_offline(cdev);
+
+out:
+	return rc ? rc : count;
+}
+
+static DEVICE_ATTR(safe_offline, 0200, NULL, dasd_safe_offline_store);
+
+static ssize_t
+dasd_access_show(struct device *dev, struct device_attribute *attr,
+		 char *buf)
+{
+	struct ccw_device *cdev = to_ccwdev(dev);
+	struct dasd_device *device;
+	int count;
+
+	device = dasd_device_from_cdev(cdev);
+	if (IS_ERR(device))
+		return PTR_ERR(device);
+
+	if (!device->discipline)
+		count = -ENODEV;
+	else if (!device->discipline->host_access_count)
+		count = -EOPNOTSUPP;
+	else
+		count = device->discipline->host_access_count(device);
+
+	dasd_put_device(device);
+	if (count < 0)
+		return count;
+
+	return sysfs_emit(buf, "%d\n", count);
+}
+
+static DEVICE_ATTR(host_access_count, 0444, dasd_access_show, NULL);
+
+static ssize_t
+dasd_discipline_show(struct device *dev, struct device_attribute *attr,
+		     char *buf)
+{
+	struct dasd_device *device;
+	ssize_t len;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		goto out;
+	else if (!device->discipline) {
+		dasd_put_device(device);
+		goto out;
+	} else {
+		len = sysfs_emit(buf, "%s\n",
+				 device->discipline->name);
+		dasd_put_device(device);
+		return len;
+	}
+out:
+	len = sysfs_emit(buf, "none\n");
+	return len;
+}
+
+static DEVICE_ATTR(discipline, 0444, dasd_discipline_show, NULL);
+
+static ssize_t
+dasd_device_status_show(struct device *dev, struct device_attribute *attr,
+		     char *buf)
+{
+	struct dasd_device *device;
+	ssize_t len;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (!IS_ERR(device)) {
+		switch (device->state) {
+		case DASD_STATE_NEW:
+			len = sysfs_emit(buf, "new\n");
+			break;
+		case DASD_STATE_KNOWN:
+			len = sysfs_emit(buf, "detected\n");
+			break;
+		case DASD_STATE_BASIC:
+			len = sysfs_emit(buf, "basic\n");
+			break;
+		case DASD_STATE_UNFMT:
+			len = sysfs_emit(buf, "unformatted\n");
+			break;
+		case DASD_STATE_READY:
+			len = sysfs_emit(buf, "ready\n");
+			break;
+		case DASD_STATE_ONLINE:
+			len = sysfs_emit(buf, "online\n");
+			break;
+		default:
+			len = sysfs_emit(buf, "no stat\n");
+			break;
+		}
+		dasd_put_device(device);
+	} else
+		len = sysfs_emit(buf, "unknown\n");
+	return len;
+}
+
+static DEVICE_ATTR(status, 0444, dasd_device_status_show, NULL);
+
+static ssize_t dasd_alias_show(struct device *dev,
+			       struct device_attribute *attr, char *buf)
+{
+	struct dasd_device *device;
+	struct dasd_uid uid;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return sysfs_emit(buf, "0\n");
+
+	if (device->discipline && device->discipline->get_uid &&
+	    !device->discipline->get_uid(device, &uid)) {
+		if (uid.type == UA_BASE_PAV_ALIAS ||
+		    uid.type == UA_HYPER_PAV_ALIAS) {
+			dasd_put_device(device);
+			return sysfs_emit(buf, "1\n");
+		}
+	}
+	dasd_put_device(device);
+
+	return sysfs_emit(buf, "0\n");
+}
+
+static DEVICE_ATTR(alias, 0444, dasd_alias_show, NULL);
+
+static ssize_t dasd_vendor_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct dasd_device *device;
+	struct dasd_uid uid;
+	char *vendor;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	vendor = "";
+	if (IS_ERR(device))
+		return sysfs_emit(buf, "%s\n", vendor);
+
+	if (device->discipline && device->discipline->get_uid &&
+	    !device->discipline->get_uid(device, &uid))
+			vendor = uid.vendor;
+
+	dasd_put_device(device);
+
+	return sysfs_emit(buf, "%s\n", vendor);
+}
+
+static DEVICE_ATTR(vendor, 0444, dasd_vendor_show, NULL);
+
+static ssize_t
+dasd_uid_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	char uid_string[DASD_UID_STRLEN];
+	struct dasd_device *device;
+	struct dasd_uid uid;
+	char ua_string[3];
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	uid_string[0] = 0;
+	if (IS_ERR(device))
+		return sysfs_emit(buf, "%s\n", uid_string);
+
+	if (device->discipline && device->discipline->get_uid &&
+	    !device->discipline->get_uid(device, &uid)) {
+		switch (uid.type) {
+		case UA_BASE_DEVICE:
+			snprintf(ua_string, sizeof(ua_string), "%02x",
+				 uid.real_unit_addr);
+			break;
+		case UA_BASE_PAV_ALIAS:
+			snprintf(ua_string, sizeof(ua_string), "%02x",
+				 uid.base_unit_addr);
+			break;
+		case UA_HYPER_PAV_ALIAS:
+			snprintf(ua_string, sizeof(ua_string), "xx");
+			break;
+		default:
+			/* should not happen, treat like base device */
+			snprintf(ua_string, sizeof(ua_string), "%02x",
+				 uid.real_unit_addr);
+			break;
+		}
+
+		snprintf(uid_string, sizeof(uid_string), "%s.%s.%04x.%s%s%s",
+			 uid.vendor, uid.serial, uid.ssid, ua_string,
+			 uid.vduit[0] ? "." : "", uid.vduit);
+	}
+	dasd_put_device(device);
+
+	return sysfs_emit(buf, "%s\n", uid_string);
+}
+static DEVICE_ATTR(uid, 0444, dasd_uid_show, NULL);
+
+/*
+ * extended error-reporting
+ */
+static ssize_t
+dasd_eer_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct dasd_devmap *devmap;
+	int eer_flag;
+
+	devmap = dasd_find_busid(dev_name(dev));
+	if (!IS_ERR(devmap) && devmap->device)
+		eer_flag = dasd_eer_enabled(devmap->device);
+	else
+		eer_flag = 0;
+	return sysfs_emit(buf, eer_flag ? "1\n" : "0\n");
+}
+
+static ssize_t
+dasd_eer_store(struct device *dev, struct device_attribute *attr,
+	       const char *buf, size_t count)
+{
+	struct dasd_device *device;
+	unsigned int val;
+	int rc = 0;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return PTR_ERR(device);
+
+	if (kstrtouint(buf, 0, &val) || val > 1)
+		return -EINVAL;
+
+	if (val)
+		rc = dasd_eer_enable(device);
+	else
+		dasd_eer_disable(device);
+
+	dasd_put_device(device);
+
+	return rc ? : count;
+}
+
+static DEVICE_ATTR(eer_enabled, 0644, dasd_eer_show, dasd_eer_store);
+
+/*
+ * aq_mask controls if the DASD should be quiesced on certain triggers
+ * The aq_mask attribute is interpreted as bitmap of the DASD_EER_* triggers.
+ */
+static ssize_t dasd_aq_mask_show(struct device *dev, struct device_attribute *attr,
+				 char *buf)
+{
+	struct dasd_devmap *devmap;
+	unsigned int aq_mask = 0;
+
+	devmap = dasd_find_busid(dev_name(dev));
+	if (!IS_ERR(devmap))
+		aq_mask = devmap->aq_mask;
+
+	return sysfs_emit(buf, "%d\n", aq_mask);
+}
+
+static ssize_t dasd_aq_mask_store(struct device *dev, struct device_attribute *attr,
+				  const char *buf, size_t count)
+{
+	struct dasd_devmap *devmap;
+	unsigned int val;
+
+	if (kstrtouint(buf, 0, &val) || val > DASD_EER_VALID)
+		return -EINVAL;
+
+	devmap = dasd_devmap_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(devmap))
+		return PTR_ERR(devmap);
+
+	spin_lock(&dasd_devmap_lock);
+	devmap->aq_mask = val;
+	if (devmap->device)
+		devmap->device->aq_mask = devmap->aq_mask;
+	spin_unlock(&dasd_devmap_lock);
+
+	return count;
+}
+
+static DEVICE_ATTR(aq_mask, 0644, dasd_aq_mask_show, dasd_aq_mask_store);
+
+/*
+ * aq_requeue controls if requests are returned to the blocklayer on quiesce
+ * or if requests are only not started
+ */
+static ssize_t dasd_aqr_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
+{
+	struct dasd_devmap *devmap;
+	int flag;
+
+	devmap = dasd_find_busid(dev_name(dev));
+	if (!IS_ERR(devmap))
+		flag = (devmap->features & DASD_FEATURE_REQUEUEQUIESCE) != 0;
+	else
+		flag = (DASD_FEATURE_DEFAULT &
+			DASD_FEATURE_REQUEUEQUIESCE) != 0;
+	return sysfs_emit(buf, "%d\n", flag);
+}
+
+static ssize_t dasd_aqr_store(struct device *dev, struct device_attribute *attr,
+			      const char *buf, size_t count)
+{
+	bool val;
+	int rc;
+
+	if (kstrtobool(buf, &val))
+		return -EINVAL;
+
+	rc = dasd_set_feature(to_ccwdev(dev), DASD_FEATURE_REQUEUEQUIESCE, val);
+
+	return rc ? : count;
+}
+
+static DEVICE_ATTR(aq_requeue, 0644, dasd_aqr_show, dasd_aqr_store);
+
+/*
+ * aq_timeouts controls how much retries have to time out until
+ * a device gets autoquiesced
+ */
+static ssize_t
+dasd_aq_timeouts_show(struct device *dev, struct device_attribute *attr,
+		      char *buf)
+{
+	struct dasd_device *device;
+	int len;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+	len = sysfs_emit(buf, "%u\n", device->aq_timeouts);
+	dasd_put_device(device);
+	return len;
+}
+
+static ssize_t
+dasd_aq_timeouts_store(struct device *dev, struct device_attribute *attr,
+		       const char *buf, size_t count)
+{
+	struct dasd_device *device;
+	unsigned int val;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+
+	if ((kstrtouint(buf, 10, &val) != 0) ||
+	    val > DASD_RETRIES_MAX || val == 0) {
+		dasd_put_device(device);
+		return -EINVAL;
+	}
+
+	if (val)
+		device->aq_timeouts = val;
+
+	dasd_put_device(device);
+	return count;
+}
+
+static DEVICE_ATTR(aq_timeouts, 0644, dasd_aq_timeouts_show,
+		   dasd_aq_timeouts_store);
+
+/*
+ * expiration time for default requests
+ */
+static ssize_t
+dasd_expires_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct dasd_device *device;
+	int len;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+	len = sysfs_emit(buf, "%lu\n", device->default_expires);
+	dasd_put_device(device);
+	return len;
+}
+
+static ssize_t
+dasd_expires_store(struct device *dev, struct device_attribute *attr,
+	       const char *buf, size_t count)
+{
+	struct dasd_device *device;
+	unsigned long val;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+
+	if ((kstrtoul(buf, 10, &val) != 0) ||
+	    (val > DASD_EXPIRES_MAX) || val == 0) {
+		dasd_put_device(device);
+		return -EINVAL;
+	}
+
+	if (val)
+		device->default_expires = val;
+
+	dasd_put_device(device);
+	return count;
+}
+
+static DEVICE_ATTR(expires, 0644, dasd_expires_show, dasd_expires_store);
+
+/* ESE fulltrack write aggressiveness knob (0..100, see DASD_FT_BIAS_*) */
+static ssize_t
+full_track_bias_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct dasd_device *device;
+	int len;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+	len = sysfs_emit(buf, "%u\n", device->ft_bias);
+	dasd_put_device(device);
+	return len;
+}
+
+static ssize_t full_track_bias_store(struct device *dev,
+				     struct device_attribute *attr,
+				     const char *buf, size_t count)
+{
+	struct dasd_device *device;
+	unsigned int val;
+
+	if (kstrtouint(buf, 0, &val) || val > DASD_FT_BIAS_MAX)
+		return -EINVAL;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+
+	/*
+	 * ft_bias is the tuning target; fulltrack is a best-effort mode hint
+	 * that the per-IO heuristic also updates locklessly. A racing writer can
+	 * at most leave a transient mismatch that self-corrects on the next IO,
+	 * never corruption, so the update is left unlocked.
+	 */
+	device->ft_bias = val;
+	dasd_ft_bias_apply(device);
+
+	dasd_put_device(device);
+	return count;
+}
+
+static DEVICE_ATTR_RW(full_track_bias);
+
+static const char * const dasd_ese_heu_state_names[] = {
+	[DASD_ESE_HEU_FT1_ACTIVE] = "fulltrack active",
+	[DASD_ESE_HEU_PROBING]    = "probing",
+	[DASD_ESE_HEU_FT0_STABLE] = "fulltrack inactive",
+};
+
+/* read-only: current full-track mode / adaptive FSM state, for observability */
+static ssize_t
+ese_heuristic_state_show(struct device *dev, struct device_attribute *attr,
+			 char *buf)
+{
+	struct dasd_device *device;
+	unsigned int state;
+	int len;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+	if (device->ft_bias == 0) {
+		len = sysfs_emit(buf, "fulltrack deactivated\n");
+	} else if (device->ft_bias >= DASD_FT_BIAS_MAX) {
+		len = sysfs_emit(buf, "fulltrack permanent active\n");
+	} else if (!dasd_ese_adaptive(device)) {
+		/* adaptive range but not ESE: the heuristic does not run */
+		len = sysfs_emit(buf, "fulltrack deactivated\n");
+	} else {
+		state = device->ese_probe_state;
+		if (state < ARRAY_SIZE(dasd_ese_heu_state_names))
+			len = sysfs_emit(buf, "%s\n", dasd_ese_heu_state_names[state]);
+		else
+			len = sysfs_emit(buf, "unknown\n");
+	}
+	dasd_put_device(device);
+	return len;
+}
+
+static DEVICE_ATTR_RO(ese_heuristic_state);
+
+static ssize_t
+dasd_retries_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct dasd_device *device;
+	int len;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+	len = sysfs_emit(buf, "%lu\n", device->default_retries);
+	dasd_put_device(device);
+	return len;
+}
+
+static ssize_t
+dasd_retries_store(struct device *dev, struct device_attribute *attr,
+		   const char *buf, size_t count)
+{
+	struct dasd_device *device;
+	unsigned long val;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+
+	if ((kstrtoul(buf, 10, &val) != 0) ||
+	    (val > DASD_RETRIES_MAX)) {
+		dasd_put_device(device);
+		return -EINVAL;
+	}
+
+	if (val)
+		device->default_retries = val;
+
+	dasd_put_device(device);
+	return count;
+}
+
+static DEVICE_ATTR(retries, 0644, dasd_retries_show, dasd_retries_store);
+
+static ssize_t
+dasd_timeout_show(struct device *dev, struct device_attribute *attr,
+		  char *buf)
+{
+	struct dasd_device *device;
+	int len;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+	len = sysfs_emit(buf, "%lu\n", device->blk_timeout);
+	dasd_put_device(device);
+	return len;
+}
+
+static ssize_t
+dasd_timeout_store(struct device *dev, struct device_attribute *attr,
+		   const char *buf, size_t count)
+{
+	struct dasd_device *device;
+	unsigned long val;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device) || !device->block)
+		return -ENODEV;
+
+	if ((kstrtoul(buf, 10, &val) != 0) ||
+	    val > UINT_MAX / HZ) {
+		dasd_put_device(device);
+		return -EINVAL;
+	}
+	if (!device->block->gdp) {
+		dasd_put_device(device);
+		return -ENODEV;
+	}
+
+	device->blk_timeout = val;
+	blk_queue_rq_timeout(device->block->gdp->queue, val * HZ);
+
+	dasd_put_device(device);
+	return count;
+}
+
+static DEVICE_ATTR(timeout, 0644,
+		   dasd_timeout_show, dasd_timeout_store);
+
+
+static ssize_t
+dasd_path_reset_store(struct device *dev, struct device_attribute *attr,
+		      const char *buf, size_t count)
+{
+	struct dasd_device *device;
+	unsigned int val;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+
+	if ((kstrtouint(buf, 16, &val) != 0) || val > 0xff)
+		val = 0;
+
+	if (device->discipline && device->discipline->reset_path)
+		device->discipline->reset_path(device, (__u8) val);
+
+	dasd_put_device(device);
+	return count;
+}
+
+static DEVICE_ATTR(path_reset, 0200, NULL, dasd_path_reset_store);
+
+static ssize_t dasd_hpf_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
+{
+	struct dasd_device *device;
+	int hpf;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+	if (!device->discipline || !device->discipline->hpf_enabled) {
+		dasd_put_device(device);
+		return sysfs_emit(buf, "%d\n", dasd_nofcx);
+	}
+	hpf = device->discipline->hpf_enabled(device);
+	dasd_put_device(device);
+	return sysfs_emit(buf, "%d\n", hpf);
+}
+
+static DEVICE_ATTR(hpf, 0444, dasd_hpf_show, NULL);
+
+static ssize_t dasd_reservation_policy_show(struct device *dev,
+					    struct device_attribute *attr,
+					    char *buf)
+{
+	struct dasd_devmap *devmap;
+	int rc = 0;
+
+	devmap = dasd_find_busid(dev_name(dev));
+	if (IS_ERR(devmap)) {
+		rc = sysfs_emit(buf, "ignore\n");
+	} else {
+		spin_lock(&dasd_devmap_lock);
+		if (devmap->features & DASD_FEATURE_FAILONSLCK)
+			rc = sysfs_emit(buf, "fail\n");
+		else
+			rc = sysfs_emit(buf, "ignore\n");
+		spin_unlock(&dasd_devmap_lock);
+	}
+	return rc;
+}
+
+static ssize_t dasd_reservation_policy_store(struct device *dev,
+					     struct device_attribute *attr,
+					     const char *buf, size_t count)
+{
+	struct ccw_device *cdev = to_ccwdev(dev);
+	int rc;
+
+	if (sysfs_streq("ignore", buf))
+		rc = dasd_set_feature(cdev, DASD_FEATURE_FAILONSLCK, 0);
+	else if (sysfs_streq("fail", buf))
+		rc = dasd_set_feature(cdev, DASD_FEATURE_FAILONSLCK, 1);
+	else
+		rc = -EINVAL;
+
+	return rc ? : count;
+}
+
+static DEVICE_ATTR(reservation_policy, 0644,
+		   dasd_reservation_policy_show, dasd_reservation_policy_store);
+
+static ssize_t dasd_reservation_state_show(struct device *dev,
+					   struct device_attribute *attr,
+					   char *buf)
+{
+	struct dasd_device *device;
+	int rc = 0;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return sysfs_emit(buf, "none\n");
+
+	if (test_bit(DASD_FLAG_IS_RESERVED, &device->flags))
+		rc = sysfs_emit(buf, "reserved\n");
+	else if (test_bit(DASD_FLAG_LOCK_STOLEN, &device->flags))
+		rc = sysfs_emit(buf, "lost\n");
+	else
+		rc = sysfs_emit(buf, "none\n");
+	dasd_put_device(device);
+	return rc;
+}
+
+static ssize_t dasd_reservation_state_store(struct device *dev,
+					    struct device_attribute *attr,
+					    const char *buf, size_t count)
+{
+	struct dasd_device *device;
+	int rc = 0;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+	if (sysfs_streq("reset", buf))
+		clear_bit(DASD_FLAG_LOCK_STOLEN, &device->flags);
+	else
+		rc = -EINVAL;
+	dasd_put_device(device);
+
+	if (rc)
+		return rc;
+	else
+		return count;
+}
+
+static DEVICE_ATTR(last_known_reservation_state, 0644,
+		   dasd_reservation_state_show, dasd_reservation_state_store);
+
+static ssize_t dasd_pm_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
+{
+	struct dasd_device *device;
+	u8 opm, nppm, cablepm, cuirpm, hpfpm, ifccpm;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return sysfs_emit(buf, "0\n");
+
+	opm = dasd_path_get_opm(device);
+	nppm = dasd_path_get_nppm(device);
+	cablepm = dasd_path_get_cablepm(device);
+	cuirpm = dasd_path_get_cuirpm(device);
+	hpfpm = dasd_path_get_hpfpm(device);
+	ifccpm = dasd_path_get_ifccpm(device);
+	dasd_put_device(device);
+
+	return sysfs_emit(buf, "%02x %02x %02x %02x %02x %02x\n", opm, nppm,
+			  cablepm, cuirpm, hpfpm, ifccpm);
+}
+
+static DEVICE_ATTR(path_masks, 0444, dasd_pm_show, NULL);
+
+/*
+ * threshold value for IFCC/CCC errors
+ */
+static ssize_t
+dasd_path_threshold_show(struct device *dev,
+			  struct device_attribute *attr, char *buf)
+{
+	struct dasd_device *device;
+	int len;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+	len = sysfs_emit(buf, "%lu\n", device->path_thrhld);
+	dasd_put_device(device);
+	return len;
+}
+
+static ssize_t
+dasd_path_threshold_store(struct device *dev, struct device_attribute *attr,
+			   const char *buf, size_t count)
+{
+	struct dasd_device *device;
+	unsigned long flags;
+	unsigned long val;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+
+	if (kstrtoul(buf, 10, &val) != 0 || val > DASD_THRHLD_MAX) {
+		dasd_put_device(device);
+		return -EINVAL;
+	}
+	spin_lock_irqsave(get_ccwdev_lock(to_ccwdev(dev)), flags);
+	device->path_thrhld = val;
+	spin_unlock_irqrestore(get_ccwdev_lock(to_ccwdev(dev)), flags);
+	dasd_put_device(device);
+	return count;
+}
+static DEVICE_ATTR(path_threshold, 0644, dasd_path_threshold_show,
+		   dasd_path_threshold_store);
+
+/*
+ * configure if path is disabled after IFCC/CCC error threshold is
+ * exceeded
+ */
+static ssize_t
+dasd_path_autodisable_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	struct dasd_devmap *devmap;
+	int flag;
+
+	devmap = dasd_find_busid(dev_name(dev));
+	if (!IS_ERR(devmap))
+		flag = (devmap->features & DASD_FEATURE_PATH_AUTODISABLE) != 0;
+	else
+		flag = (DASD_FEATURE_DEFAULT &
+			DASD_FEATURE_PATH_AUTODISABLE) != 0;
+	return sysfs_emit(buf, flag ? "1\n" : "0\n");
+}
+
+static ssize_t
+dasd_path_autodisable_store(struct device *dev,
+				    struct device_attribute *attr,
+				    const char *buf, size_t count)
+{
+	unsigned int val;
+	int rc;
+
+	if (kstrtouint(buf, 0, &val) || val > 1)
+		return -EINVAL;
+
+	rc = dasd_set_feature(to_ccwdev(dev),
+			      DASD_FEATURE_PATH_AUTODISABLE, val);
+
+	return rc ? : count;
+}
+
+static DEVICE_ATTR(path_autodisable, 0644,
+		   dasd_path_autodisable_show,
+		   dasd_path_autodisable_store);
+/*
+ * interval for IFCC/CCC checks
+ * meaning time with no IFCC/CCC error before the error counter
+ * gets reset
+ */
+static ssize_t
+dasd_path_interval_show(struct device *dev,
+			struct device_attribute *attr, char *buf)
+{
+	struct dasd_device *device;
+	int len;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+	len = sysfs_emit(buf, "%lu\n", device->path_interval);
+	dasd_put_device(device);
+	return len;
+}
+
+static ssize_t
+dasd_path_interval_store(struct device *dev, struct device_attribute *attr,
+	       const char *buf, size_t count)
+{
+	struct dasd_device *device;
+	unsigned long flags;
+	unsigned long val;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+
+	if ((kstrtoul(buf, 10, &val) != 0) ||
+	    (val > DASD_INTERVAL_MAX) || val == 0) {
+		dasd_put_device(device);
+		return -EINVAL;
+	}
+	spin_lock_irqsave(get_ccwdev_lock(to_ccwdev(dev)), flags);
+	if (val)
+		device->path_interval = val;
+	spin_unlock_irqrestore(get_ccwdev_lock(to_ccwdev(dev)), flags);
+	dasd_put_device(device);
+	return count;
+}
+
+static DEVICE_ATTR(path_interval, 0644, dasd_path_interval_show,
+		   dasd_path_interval_store);
+
+static ssize_t
+dasd_device_fcs_show(struct device *dev, struct device_attribute *attr,
+		     char *buf)
+{
+	struct dasd_device *device;
+	int fc_sec;
+	int rc;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+	fc_sec = dasd_path_get_fcs_device(device);
+	if (fc_sec == -EINVAL)
+		rc = sysfs_emit(buf, "Inconsistent\n");
+	else
+		rc = sysfs_emit(buf, "%s\n", dasd_path_get_fcs_str(fc_sec));
+	dasd_put_device(device);
+
+	return rc;
+}
+static DEVICE_ATTR(fc_security, 0444, dasd_device_fcs_show, NULL);
+
+static ssize_t
+dasd_path_fcs_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf)
+{
+	struct dasd_path *path = to_dasd_path(kobj);
+	unsigned int fc_sec = path->fc_security;
+
+	return sysfs_emit(buf, "%s\n", dasd_path_get_fcs_str(fc_sec));
+}
+
+static struct kobj_attribute path_fcs_attribute =
+	__ATTR(fc_security, 0444, dasd_path_fcs_show, NULL);
+
+/*
+ * print copy relation in the form
+ * primary,secondary[1] primary,secondary[2], ...
+ */
+static ssize_t
+dasd_copy_pair_show(struct device *dev,
+		    struct device_attribute *attr, char *buf)
+{
+	char prim_busid[DASD_BUS_ID_SIZE];
+	struct dasd_copy_relation *copy;
+	struct dasd_devmap *devmap;
+	int len = 0;
+	int i;
+
+	devmap = dasd_find_busid(dev_name(dev));
+	if (IS_ERR(devmap))
+		return -ENODEV;
+
+	if (!devmap->copy)
+		return -ENODEV;
+
+	copy = devmap->copy;
+	/* find primary */
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		if (copy->entry[i].configured && copy->entry[i].primary) {
+			strscpy(prim_busid, copy->entry[i].busid,
+				DASD_BUS_ID_SIZE);
+			break;
+		}
+	}
+	if (i == DASD_CP_ENTRIES)
+		goto out;
+
+	/* print all secondary */
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		if (copy->entry[i].configured && !copy->entry[i].primary)
+			len += sysfs_emit_at(buf, len, "%s,%s ", prim_busid,
+					     copy->entry[i].busid);
+	}
+
+	len += sysfs_emit_at(buf, len, "\n");
+out:
+	return len;
+}
+
+static int dasd_devmap_set_copy_relation(struct dasd_devmap *devmap,
+					 struct dasd_copy_relation *copy,
+					 char *busid, bool primary)
+{
+	int i;
+
+	/* find free entry */
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		/* current bus_id already included, nothing to do */
+		if (copy->entry[i].configured &&
+		    strncmp(copy->entry[i].busid, busid, DASD_BUS_ID_SIZE) == 0)
+			return 0;
+
+		if (!copy->entry[i].configured)
+			break;
+	}
+	if (i == DASD_CP_ENTRIES)
+		return -EINVAL;
+
+	copy->entry[i].configured = true;
+	strscpy(copy->entry[i].busid, busid, DASD_BUS_ID_SIZE);
+	if (primary) {
+		copy->active = &copy->entry[i];
+		copy->entry[i].primary = true;
+	}
+	if (!devmap->copy)
+		devmap->copy = copy;
+
+	return 0;
+}
+
+static void dasd_devmap_del_copy_relation(struct dasd_copy_relation *copy,
+					  char *busid)
+{
+	int i;
+
+	spin_lock(&dasd_devmap_lock);
+	/* find entry */
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		if (copy->entry[i].configured &&
+		    strncmp(copy->entry[i].busid, busid, DASD_BUS_ID_SIZE) == 0)
+			break;
+	}
+	if (i == DASD_CP_ENTRIES || !copy->entry[i].configured) {
+		spin_unlock(&dasd_devmap_lock);
+		return;
+	}
+
+	copy->entry[i].configured = false;
+	memset(copy->entry[i].busid, 0, DASD_BUS_ID_SIZE);
+	if (copy->active == &copy->entry[i]) {
+		copy->active = NULL;
+		copy->entry[i].primary = false;
+	}
+	spin_unlock(&dasd_devmap_lock);
+}
+
+static int dasd_devmap_clear_copy_relation(struct device *dev)
+{
+	struct dasd_copy_relation *copy;
+	struct dasd_devmap *devmap;
+	int i, rc = 1;
+
+	devmap = dasd_devmap_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(devmap))
+		return 1;
+
+	spin_lock(&dasd_devmap_lock);
+	if (!devmap->copy)
+		goto out;
+
+	copy = devmap->copy;
+	/* first check if all secondary devices are offline*/
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		if (!copy->entry[i].configured)
+			continue;
+
+		if (copy->entry[i].device == copy->active->device)
+			continue;
+
+		if (copy->entry[i].device)
+			goto out;
+	}
+	/* clear all devmap entries */
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		if (strlen(copy->entry[i].busid) == 0)
+			continue;
+		if (copy->entry[i].device) {
+			dasd_put_device(copy->entry[i].device);
+			copy->entry[i].device->copy = NULL;
+			copy->entry[i].device = NULL;
+		}
+		devmap = dasd_find_busid_locked(copy->entry[i].busid);
+		devmap->copy = NULL;
+		memset(copy->entry[i].busid, 0, DASD_BUS_ID_SIZE);
+	}
+	kfree(copy);
+	rc = 0;
+out:
+	spin_unlock(&dasd_devmap_lock);
+	return rc;
+}
+
+/*
+ * parse BUSIDs from a copy pair
+ */
+static int dasd_devmap_parse_busid(const char *buf, char *prim_busid,
+				   char *sec_busid)
+{
+	char *primary, *secondary, *tmp, *pt;
+	int id0, id1, id2;
+
+	pt =  kstrdup(buf, GFP_KERNEL);
+	tmp = pt;
+	if (!tmp)
+		return -ENOMEM;
+
+	primary = strsep(&tmp, ",");
+	if (!primary) {
+		kfree(pt);
+		return -EINVAL;
+	}
+	secondary = strsep(&tmp, ",");
+	if (!secondary) {
+		kfree(pt);
+		return -EINVAL;
+	}
+	if (dasd_busid(primary, &id0, &id1, &id2)) {
+		kfree(pt);
+		return -EINVAL;
+	}
+	sprintf(prim_busid, "%01x.%01x.%04x", id0, id1, id2);
+	if (dasd_busid(secondary, &id0, &id1, &id2)) {
+		kfree(pt);
+		return -EINVAL;
+	}
+	sprintf(sec_busid, "%01x.%01x.%04x", id0, id1, id2);
+	kfree(pt);
+
+	return 0;
+}
+
+static ssize_t dasd_copy_pair_store(struct device *dev,
+				    struct device_attribute *attr,
+				    const char *buf, size_t count)
+{
+	struct dasd_devmap *prim_devmap, *sec_devmap;
+	char prim_busid[DASD_BUS_ID_SIZE];
+	char sec_busid[DASD_BUS_ID_SIZE];
+	struct dasd_copy_relation *copy;
+	struct dasd_device *device;
+	bool pprc_enabled;
+	int rc;
+
+	if (strncmp(buf, "clear", strlen("clear")) == 0) {
+		if (dasd_devmap_clear_copy_relation(dev))
+			return -EINVAL;
+		return count;
+	}
+
+	rc = dasd_devmap_parse_busid(buf, prim_busid, sec_busid);
+	if (rc)
+		return rc;
+
+	if (strncmp(dev_name(dev), prim_busid, DASD_BUS_ID_SIZE) != 0 &&
+	    strncmp(dev_name(dev), sec_busid, DASD_BUS_ID_SIZE) != 0)
+		return -EINVAL;
+
+	/* allocate primary devmap if needed */
+	prim_devmap = dasd_find_busid(prim_busid);
+	if (IS_ERR(prim_devmap)) {
+		prim_devmap = dasd_add_busid(prim_busid, DASD_FEATURE_DEFAULT);
+		if (IS_ERR(prim_devmap))
+			return PTR_ERR(prim_devmap);
+	}
+
+	/* allocate secondary devmap if needed */
+	sec_devmap = dasd_find_busid(sec_busid);
+	if (IS_ERR(sec_devmap)) {
+		sec_devmap = dasd_add_busid(sec_busid, DASD_FEATURE_DEFAULT);
+		if (IS_ERR(sec_devmap))
+			return PTR_ERR(sec_devmap);
+	}
+
+	/* setting copy relation is only allowed for offline secondary */
+	if (sec_devmap->device)
+		return -EINVAL;
+
+	if (prim_devmap->copy) {
+		copy = prim_devmap->copy;
+	} else if (sec_devmap->copy) {
+		copy = sec_devmap->copy;
+	} else {
+		copy = kzalloc_obj(*copy);
+		if (!copy)
+			return -ENOMEM;
+	}
+	spin_lock(&dasd_devmap_lock);
+	rc = dasd_devmap_set_copy_relation(prim_devmap, copy, prim_busid, true);
+	if (rc) {
+		spin_unlock(&dasd_devmap_lock);
+		return rc;
+	}
+	rc = dasd_devmap_set_copy_relation(sec_devmap, copy, sec_busid, false);
+	if (rc) {
+		spin_unlock(&dasd_devmap_lock);
+		return rc;
+	}
+	spin_unlock(&dasd_devmap_lock);
+
+	/* if primary device is already online call device setup directly */
+	if (prim_devmap->device && !prim_devmap->device->copy) {
+		device = prim_devmap->device;
+		if (device->discipline->pprc_enabled) {
+			pprc_enabled = device->discipline->pprc_enabled(device);
+			rc = dasd_devmap_set_device_copy_relation(device->cdev,
+								  pprc_enabled);
+		} else {
+			rc = -EOPNOTSUPP;
+		}
+	}
+	if (rc) {
+		dasd_devmap_del_copy_relation(copy, prim_busid);
+		dasd_devmap_del_copy_relation(copy, sec_busid);
+		count = rc;
+	}
+
+	return count;
+}
+static DEVICE_ATTR(copy_pair, 0644, dasd_copy_pair_show,
+		   dasd_copy_pair_store);
+
+static ssize_t
+dasd_copy_role_show(struct device *dev,
+		    struct device_attribute *attr, char *buf)
+{
+	struct dasd_copy_relation *copy;
+	struct dasd_device *device;
+	int len, i;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+
+	if (!device->copy) {
+		len = sysfs_emit(buf, "none\n");
+		goto out;
+	}
+	copy = device->copy;
+	/* only the active device is primary */
+	if (copy->active->device == device) {
+		len = sysfs_emit(buf, "primary\n");
+		goto out;
+	}
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		if (copy->entry[i].device == device) {
+			len = sysfs_emit(buf, "secondary\n");
+			goto out;
+		}
+	}
+	/* not in the list, no COPY role */
+	len = sysfs_emit(buf, "none\n");
+out:
+	dasd_put_device(device);
+	return len;
+}
+static DEVICE_ATTR(copy_role, 0444, dasd_copy_role_show, NULL);
+
+static ssize_t dasd_device_ping(struct device *dev,
+				struct device_attribute *attr,
+				const char *buf, size_t count)
+{
+	struct dasd_device *device;
+	size_t rc;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return -ENODEV;
+
+	/*
+	 * do not try during offline processing
+	 * early check only
+	 * the sleep_on function itself checks for offline
+	 * processing again
+	 */
+	if (test_bit(DASD_FLAG_OFFLINE, &device->flags)) {
+		rc = -EBUSY;
+		goto out;
+	}
+	if (!device->discipline || !device->discipline->device_ping) {
+		rc = -EOPNOTSUPP;
+		goto out;
+	}
+	rc = device->discipline->device_ping(device);
+	if (!rc)
+		rc = count;
+out:
+	dasd_put_device(device);
+	return rc;
+}
+static DEVICE_ATTR(ping, 0200, NULL, dasd_device_ping);
+
+#define DASD_DEFINE_ATTR(_name, _func)					\
+static ssize_t dasd_##_name##_show(struct device *dev,			\
+				   struct device_attribute *attr,	\
+				   char *buf)				\
+{									\
+	struct ccw_device *cdev = to_ccwdev(dev);			\
+	struct dasd_device *device = dasd_device_from_cdev(cdev);	\
+	int val = 0;							\
+									\
+	if (IS_ERR(device))						\
+		return -ENODEV;						\
+	if (device->discipline && _func)				\
+		val = _func(device);					\
+	dasd_put_device(device);					\
+									\
+	return sysfs_emit(buf, "%d\n", val);			\
+}									\
+static DEVICE_ATTR(_name, 0444, dasd_##_name##_show, NULL);
+
+DASD_DEFINE_ATTR(ese, device->discipline->ese_capable);
+DASD_DEFINE_ATTR(on_demand_formatting, device->discipline->on_demand_format);
+DASD_DEFINE_ATTR(extent_size, device->discipline->ext_size);
+DASD_DEFINE_ATTR(pool_id, device->discipline->ext_pool_id);
+DASD_DEFINE_ATTR(space_configured, device->discipline->space_configured);
+DASD_DEFINE_ATTR(space_allocated, device->discipline->space_allocated);
+DASD_DEFINE_ATTR(logical_capacity, device->discipline->logical_capacity);
+DASD_DEFINE_ATTR(warn_threshold, device->discipline->ext_pool_warn_thrshld);
+DASD_DEFINE_ATTR(cap_at_warnlevel, device->discipline->ext_pool_cap_at_warnlevel);
+DASD_DEFINE_ATTR(pool_oos, device->discipline->ext_pool_oos);
+
+static struct attribute * dasd_attrs[] = {
+	&dev_attr_readonly.attr,
+	&dev_attr_discipline.attr,
+	&dev_attr_status.attr,
+	&dev_attr_alias.attr,
+	&dev_attr_vendor.attr,
+	&dev_attr_uid.attr,
+	&dev_attr_use_diag.attr,
+	&dev_attr_raw_track_access.attr,
+	&dev_attr_eer_enabled.attr,
+	&dev_attr_erplog.attr,
+	&dev_attr_failfast.attr,
+	&dev_attr_expires.attr,
+	&dev_attr_full_track_bias.attr,
+	&dev_attr_ese_heuristic_state.attr,
+	&dev_attr_retries.attr,
+	&dev_attr_timeout.attr,
+	&dev_attr_reservation_policy.attr,
+	&dev_attr_last_known_reservation_state.attr,
+	&dev_attr_safe_offline.attr,
+	&dev_attr_host_access_count.attr,
+	&dev_attr_path_masks.attr,
+	&dev_attr_path_threshold.attr,
+	&dev_attr_path_autodisable.attr,
+	&dev_attr_path_interval.attr,
+	&dev_attr_path_reset.attr,
+	&dev_attr_hpf.attr,
+	&dev_attr_ese.attr,
+	&dev_attr_on_demand_formatting.attr,
+	&dev_attr_fc_security.attr,
+	&dev_attr_copy_pair.attr,
+	&dev_attr_copy_role.attr,
+	&dev_attr_ping.attr,
+	&dev_attr_aq_mask.attr,
+	&dev_attr_aq_requeue.attr,
+	&dev_attr_aq_timeouts.attr,
+	NULL,
+};
+
+static const struct attribute_group dasd_attr_group = {
+	.attrs = dasd_attrs,
+};
+
+static struct attribute *capacity_attrs[] = {
+	&dev_attr_space_configured.attr,
+	&dev_attr_space_allocated.attr,
+	&dev_attr_logical_capacity.attr,
+	NULL,
+};
+
+static const struct attribute_group capacity_attr_group = {
+	.name = "capacity",
+	.attrs = capacity_attrs,
+};
+
+static struct attribute *ext_pool_attrs[] = {
+	&dev_attr_pool_id.attr,
+	&dev_attr_extent_size.attr,
+	&dev_attr_warn_threshold.attr,
+	&dev_attr_cap_at_warnlevel.attr,
+	&dev_attr_pool_oos.attr,
+	NULL,
+};
+
+static const struct attribute_group ext_pool_attr_group = {
+	.name = "extent_pool",
+	.attrs = ext_pool_attrs,
+};
+
+const struct attribute_group *dasd_dev_groups[] = {
+	&dasd_attr_group,
+	&capacity_attr_group,
+	&ext_pool_attr_group,
+	NULL,
+};
+EXPORT_SYMBOL_GPL(dasd_dev_groups);
+
+/*
+ * Return value of the specified feature.
+ */
+int
+dasd_get_feature(struct ccw_device *cdev, int feature)
+{
+	struct dasd_devmap *devmap;
+
+	devmap = dasd_find_busid(dev_name(&cdev->dev));
+	if (IS_ERR(devmap))
+		return PTR_ERR(devmap);
+
+	return ((devmap->features & feature) != 0);
+}
+
+/*
+ * Set / reset given feature.
+ * Flag indicates whether to set (!=0) or the reset (=0) the feature.
+ */
+int
+dasd_set_feature(struct ccw_device *cdev, int feature, int flag)
+{
+	struct dasd_devmap *devmap;
+
+	devmap = dasd_devmap_from_cdev(cdev);
+	if (IS_ERR(devmap))
+		return PTR_ERR(devmap);
+
+	spin_lock(&dasd_devmap_lock);
+	if (flag)
+		devmap->features |= feature;
+	else
+		devmap->features &= ~feature;
+	if (devmap->device)
+		devmap->device->features = devmap->features;
+	spin_unlock(&dasd_devmap_lock);
+	return 0;
+}
+EXPORT_SYMBOL(dasd_set_feature);
+
+static struct attribute *paths_info_attrs[] = {
+	&path_fcs_attribute.attr,
+	NULL,
+};
+ATTRIBUTE_GROUPS(paths_info);
+
+static struct kobj_type path_attr_type = {
+	.release	= dasd_path_release,
+	.default_groups	= paths_info_groups,
+	.sysfs_ops	= &kobj_sysfs_ops,
+};
+
+static void dasd_path_init_kobj(struct dasd_device *device, int chp)
+{
+	device->path[chp].kobj.kset = device->paths_info;
+	kobject_init(&device->path[chp].kobj, &path_attr_type);
+}
+
+void dasd_path_create_kobj(struct dasd_device *device, int chp)
+{
+	int rc;
+
+	if (test_bit(DASD_FLAG_OFFLINE, &device->flags))
+		return;
+	if (!device->paths_info) {
+		dev_warn(&device->cdev->dev, "Unable to create paths objects\n");
+		return;
+	}
+	if (device->path[chp].in_sysfs)
+		return;
+	if (!device->path[chp].conf_data)
+		return;
+
+	dasd_path_init_kobj(device, chp);
+
+	rc = kobject_add(&device->path[chp].kobj, NULL, "%x.%02x",
+			 device->path[chp].cssid, device->path[chp].chpid);
+	if (rc)
+		kobject_put(&device->path[chp].kobj);
+	device->path[chp].in_sysfs = true;
+}
+EXPORT_SYMBOL(dasd_path_create_kobj);
+
+void dasd_path_create_kobjects(struct dasd_device *device)
+{
+	u8 lpm, opm;
+
+	opm = dasd_path_get_opm(device);
+	for (lpm = 0x80; lpm; lpm >>= 1) {
+		if (!(lpm & opm))
+			continue;
+		dasd_path_create_kobj(device, pathmask_to_pos(lpm));
+	}
+}
+EXPORT_SYMBOL(dasd_path_create_kobjects);
+
+static void dasd_path_remove_kobj(struct dasd_device *device, int chp)
+{
+	if (device->path[chp].in_sysfs) {
+		kobject_put(&device->path[chp].kobj);
+		device->path[chp].in_sysfs = false;
+	}
+}
+
+/*
+ * As we keep kobjects for the lifetime of a device, this function must not be
+ * called anywhere but in the context of offlining a device.
+ */
+void dasd_path_remove_kobjects(struct dasd_device *device)
+{
+	int i;
+
+	for (i = 0; i < 8; i++)
+		dasd_path_remove_kobj(device, i);
+}
+EXPORT_SYMBOL(dasd_path_remove_kobjects);
+
+int
+dasd_devmap_init(void)
+{
+	int i;
+
+	/* Initialize devmap structures. */
+	dasd_max_devindex = 0;
+	for (i = 0; i < 256; i++)
+		INIT_LIST_HEAD(&dasd_hashlists[i]);
+	return 0;
+}
+
+void
+dasd_devmap_exit(void)
+{
+	dasd_forget_ranges();
+}
diff --git a/drivers/s390/block/dasd_diag.c b/drivers/s390/block/dasd_diag.c
new file mode 100644
index 000000000..af68c2ee6
--- /dev/null
+++ b/drivers/s390/block/dasd_diag.c
@@ -0,0 +1,680 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ * Based on.......: linux/drivers/s390/block/mdisk.c
+ * ...............: by Hartmunt Penner <hpenner@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2000
+ *
+ */
+
+#include <linux/kernel_stat.h>
+#include <linux/stddef.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/hdreg.h>
+#include <linux/bio.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/jiffies.h>
+#include <asm/asm-extable.h>
+#include <asm/machine.h>
+#include <asm/dasd.h>
+#include <asm/debug.h>
+#include <asm/diag.h>
+#include <asm/ebcdic.h>
+#include <linux/io.h>
+#include <asm/irq.h>
+#include <asm/vtoc.h>
+#include <asm/asm.h>
+
+#include "dasd_int.h"
+#include "dasd_diag.h"
+
+MODULE_DESCRIPTION("S/390 Support for DIAG access to DASD Disks");
+MODULE_LICENSE("GPL");
+
+/* The maximum number of blocks per request (max_blocks) is dependent on the
+ * amount of storage that is available in the static I/O buffer for each
+ * device. Currently each device gets 2 pages. We want to fit two requests
+ * into the available memory so that we can immediately start the next if one
+ * finishes. */
+#define DIAG_MAX_BLOCKS	(((2 * PAGE_SIZE - sizeof(struct dasd_ccw_req) - \
+			   sizeof(struct dasd_diag_req)) / \
+		           sizeof(struct dasd_diag_bio)) / 2)
+#define DIAG_MAX_RETRIES	32
+#define DIAG_TIMEOUT		50
+
+static struct dasd_discipline dasd_diag_discipline;
+
+struct dasd_diag_private {
+	struct dasd_diag_characteristics rdc_data;
+	struct dasd_diag_rw_io iob;
+	struct dasd_diag_init_io iib;
+	blocknum_t pt_block;
+	struct ccw_dev_id dev_id;
+};
+
+struct dasd_diag_req {
+	unsigned int block_count;
+	struct dasd_diag_bio bio[];
+};
+
+static const u8 DASD_DIAG_CMS1[] = { 0xc3, 0xd4, 0xe2, 0xf1 };/* EBCDIC CMS1 */
+
+/* Perform DIAG250 call with block I/O parameter list iob (input and output)
+ * and function code cmd.
+ * In case of an exception return 3. Otherwise return result of bitwise OR of
+ * resulting condition code and DIAG return code. */
+static inline int __dia250(void *iob, int cmd)
+{
+	union register_pair rx = { .even = (unsigned long)iob, };
+	int cc, exception;
+	typedef union {
+		struct dasd_diag_init_io init_io;
+		struct dasd_diag_rw_io rw_io;
+	} addr_type;
+
+	exception = 1;
+	asm_inline volatile(
+		"	diag	%[rx],%[cmd],0x250\n"
+		"0:	lhi	%[exc],0\n"
+		"1:\n"
+		CC_IPM(cc)
+		EX_TABLE(0b,1b)
+		: CC_OUT(cc, cc), [rx] "+d" (rx.pair),
+		  "+m" (*(addr_type *)iob), [exc] "+d" (exception)
+		: [cmd] "d" (cmd)
+		: CC_CLOBBER);
+	cc = exception ? 3 : CC_TRANSFORM(cc);
+	return cc | rx.odd;
+}
+
+static inline int dia250(void *iob, int cmd)
+{
+	diag_stat_inc(DIAG_STAT_X250);
+	return __dia250(iob, cmd);
+}
+
+/* Initialize block I/O to DIAG device using the specified blocksize and
+ * block offset. On success, return zero and set end_block to contain the
+ * number of blocks on the device minus the specified offset. Return non-zero
+ * otherwise. */
+static inline int
+mdsk_init_io(struct dasd_device *device, unsigned int blocksize,
+	     blocknum_t offset, blocknum_t *end_block)
+{
+	struct dasd_diag_private *private = device->private;
+	struct dasd_diag_init_io *iib = &private->iib;
+	int rc;
+
+	memset(iib, 0, sizeof (struct dasd_diag_init_io));
+
+	iib->dev_nr = private->dev_id.devno;
+	iib->block_size = blocksize;
+	iib->offset = offset;
+	iib->flaga = DASD_DIAG_FLAGA_DEFAULT;
+
+	rc = dia250(iib, INIT_BIO);
+
+	if ((rc & 3) == 0 && end_block)
+		*end_block = iib->end_block;
+
+	return rc;
+}
+
+/* Remove block I/O environment for device. Return zero on success, non-zero
+ * otherwise. */
+static inline int
+mdsk_term_io(struct dasd_device * device)
+{
+	struct dasd_diag_private *private = device->private;
+	struct dasd_diag_init_io *iib = &private->iib;
+	int rc;
+
+	memset(iib, 0, sizeof (struct dasd_diag_init_io));
+	iib->dev_nr = private->dev_id.devno;
+	rc = dia250(iib, TERM_BIO);
+	return rc;
+}
+
+/* Error recovery for failed DIAG requests - try to reestablish the DIAG
+ * environment. */
+static void
+dasd_diag_erp(struct dasd_device *device)
+{
+	int rc;
+
+	mdsk_term_io(device);
+	rc = mdsk_init_io(device, device->block->bp_block, 0, NULL);
+	if (rc == 4) {
+		if (!(test_and_set_bit(DASD_FLAG_DEVICE_RO, &device->flags)))
+			pr_warn("%s: The access mode of a DIAG device changed to read-only\n",
+				dev_name(&device->cdev->dev));
+		rc = 0;
+	}
+	if (rc)
+		pr_warn("%s: DIAG ERP failed with rc=%d\n",
+			dev_name(&device->cdev->dev), rc);
+}
+
+/* Start a given request at the device. Return zero on success, non-zero
+ * otherwise. */
+static int
+dasd_start_diag(struct dasd_ccw_req * cqr)
+{
+	struct dasd_device *device;
+	struct dasd_diag_private *private;
+	struct dasd_diag_req *dreq;
+	int rc;
+
+	device = cqr->startdev;
+	if (cqr->retries < 0) {
+		DBF_DEV_EVENT(DBF_ERR, device, "DIAG start_IO: request %p "
+			    "- no retry left)", cqr);
+		cqr->status = DASD_CQR_ERROR;
+		return -EIO;
+	}
+	private = device->private;
+	dreq = cqr->data;
+
+	private->iob.dev_nr = private->dev_id.devno;
+	private->iob.key = 0;
+	private->iob.flags = DASD_DIAG_RWFLAG_ASYNC;
+	private->iob.block_count = dreq->block_count;
+	private->iob.interrupt_params = (addr_t) cqr;
+	private->iob.bio_list = dreq->bio;
+	private->iob.flaga = DASD_DIAG_FLAGA_DEFAULT;
+
+	cqr->startclk = get_tod_clock();
+	cqr->starttime = jiffies;
+	cqr->retries--;
+
+	rc = dia250(&private->iob, RW_BIO);
+	switch (rc) {
+	case 0: /* Synchronous I/O finished successfully */
+		cqr->stopclk = get_tod_clock();
+		cqr->status = DASD_CQR_SUCCESS;
+		/* Indicate to calling function that only a dasd_schedule_bh()
+		   and no timer is needed */
+                rc = -EACCES;
+		break;
+	case 8: /* Asynchronous I/O was started */
+		cqr->status = DASD_CQR_IN_IO;
+		rc = 0;
+		break;
+	default: /* Error condition */
+		cqr->status = DASD_CQR_QUEUED;
+		DBF_DEV_EVENT(DBF_WARNING, device, "dia250 returned rc=%d", rc);
+		dasd_diag_erp(device);
+		rc = -EIO;
+		break;
+	}
+	cqr->intrc = rc;
+	return rc;
+}
+
+/* Terminate given request at the device. */
+static int
+dasd_diag_term_IO(struct dasd_ccw_req * cqr)
+{
+	struct dasd_device *device;
+
+	device = cqr->startdev;
+	mdsk_term_io(device);
+	mdsk_init_io(device, device->block->bp_block, 0, NULL);
+	cqr->status = DASD_CQR_CLEAR_PENDING;
+	cqr->stopclk = get_tod_clock();
+	dasd_schedule_device_bh(device);
+	return 0;
+}
+
+/* Handle external interruption. */
+static void dasd_ext_handler(struct ext_code ext_code,
+			     unsigned int param32, unsigned long param64)
+{
+	struct dasd_ccw_req *cqr, *next;
+	struct dasd_device *device;
+	unsigned long expires;
+	unsigned long flags;
+	addr_t ip;
+	int rc;
+
+	switch (ext_code.subcode >> 8) {
+	case DASD_DIAG_CODE_31BIT:
+		ip = (addr_t) param32;
+		break;
+	case DASD_DIAG_CODE_64BIT:
+		ip = (addr_t) param64;
+		break;
+	default:
+		return;
+	}
+	inc_irq_stat(IRQEXT_DSD);
+	if (!ip) {		/* no intparm: unsolicited interrupt */
+		DBF_EVENT(DBF_NOTICE, "%s", "caught unsolicited "
+			      "interrupt");
+		return;
+	}
+	cqr = (struct dasd_ccw_req *) ip;
+	device = (struct dasd_device *) cqr->startdev;
+	if (strncmp(device->discipline->ebcname, (char *) &cqr->magic, 4)) {
+		DBF_DEV_EVENT(DBF_WARNING, device,
+			    " magic number of dasd_ccw_req 0x%08X doesn't"
+			    " match discipline 0x%08X",
+			    cqr->magic, *(int *) (&device->discipline->name));
+		return;
+	}
+
+	/* get irq lock to modify request queue */
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+
+	/* Check for a pending clear operation */
+	if (cqr->status == DASD_CQR_CLEAR_PENDING) {
+		cqr->status = DASD_CQR_CLEARED;
+		dasd_device_clear_timer(device);
+		dasd_schedule_device_bh(device);
+		spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+		return;
+	}
+
+	cqr->stopclk = get_tod_clock();
+
+	expires = 0;
+	if ((ext_code.subcode & 0xff) == 0) {
+		cqr->status = DASD_CQR_SUCCESS;
+		/* Start first request on queue if possible -> fast_io. */
+		if (!list_empty(&device->ccw_queue)) {
+			next = list_entry(device->ccw_queue.next,
+					  struct dasd_ccw_req, devlist);
+			if (next->status == DASD_CQR_QUEUED) {
+				rc = dasd_start_diag(next);
+				if (rc == 0)
+					expires = next->expires;
+			}
+		}
+	} else {
+		cqr->status = DASD_CQR_QUEUED;
+		DBF_DEV_EVENT(DBF_DEBUG, device, "interrupt status for "
+			      "request %p was %d (%d retries left)", cqr,
+			      ext_code.subcode & 0xff, cqr->retries);
+		dasd_diag_erp(device);
+	}
+
+	if (expires != 0)
+		dasd_device_set_timer(device, expires);
+	else
+		dasd_device_clear_timer(device);
+	dasd_schedule_device_bh(device);
+
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+}
+
+/* Check whether device can be controlled by DIAG discipline. Return zero on
+ * success, non-zero otherwise. */
+static int
+dasd_diag_check_device(struct dasd_device *device)
+{
+	struct dasd_diag_private *private = device->private;
+	struct dasd_diag_characteristics *rdc_data;
+	struct vtoc_cms_label *label;
+	struct dasd_block *block;
+	struct dasd_diag_bio *bio;
+	unsigned int sb, bsize;
+	blocknum_t end_block;
+	int rc;
+
+	if (private == NULL) {
+		private = kzalloc_obj(*private);
+		if (private == NULL) {
+			DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+				"Allocating memory for private DASD data "
+				      "failed\n");
+			return -ENOMEM;
+		}
+		ccw_device_get_id(device->cdev, &private->dev_id);
+		device->private = private;
+	}
+	block = dasd_alloc_block();
+	if (IS_ERR(block)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "could not allocate dasd block structure");
+		device->private = NULL;
+		kfree(private);
+		return PTR_ERR(block);
+	}
+	device->block = block;
+	block->base = device;
+
+	/* Read Device Characteristics */
+	rdc_data = &private->rdc_data;
+	rdc_data->dev_nr = private->dev_id.devno;
+	rdc_data->rdc_len = sizeof (struct dasd_diag_characteristics);
+
+	rc = diag210((struct diag210 *) rdc_data);
+	if (rc) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "failed to retrieve device "
+			    "information (rc=%d)", rc);
+		rc = -EOPNOTSUPP;
+		goto out;
+	}
+
+	device->default_expires = DIAG_TIMEOUT;
+	device->default_retries = DIAG_MAX_RETRIES;
+
+	/* Figure out position of label block */
+	switch (private->rdc_data.vdev_class) {
+	case DEV_CLASS_FBA:
+		private->pt_block = 1;
+		break;
+	case DEV_CLASS_ECKD:
+		private->pt_block = 2;
+		break;
+	default:
+		pr_warn("%s: Device type %d is not supported in DIAG mode\n",
+			dev_name(&device->cdev->dev),
+			private->rdc_data.vdev_class);
+		rc = -EOPNOTSUPP;
+		goto out;
+	}
+
+	DBF_DEV_EVENT(DBF_INFO, device,
+		      "%04X: %04X on real %04X/%02X",
+		      rdc_data->dev_nr,
+		      rdc_data->vdev_type,
+		      rdc_data->rdev_type, rdc_data->rdev_model);
+
+	/* terminate all outstanding operations */
+	mdsk_term_io(device);
+
+	/* figure out blocksize of device */
+	label = (struct vtoc_cms_label *) get_zeroed_page(GFP_KERNEL);
+	if (label == NULL)  {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "No memory to allocate initialization request");
+		rc = -ENOMEM;
+		goto out;
+	}
+	bio = kzalloc_obj(*bio);
+	if (bio == NULL)  {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "No memory to allocate initialization bio");
+		rc = -ENOMEM;
+		goto out_label;
+	}
+	rc = 0;
+	end_block = 0;
+	/* try all sizes - needed for ECKD devices */
+	for (bsize = 512; bsize <= PAGE_SIZE; bsize <<= 1) {
+		mdsk_init_io(device, bsize, 0, &end_block);
+		memset(bio, 0, sizeof(*bio));
+		bio->type = MDSK_READ_REQ;
+		bio->block_number = private->pt_block + 1;
+		bio->buffer = label;
+		memset(&private->iob, 0, sizeof (struct dasd_diag_rw_io));
+		private->iob.dev_nr = rdc_data->dev_nr;
+		private->iob.key = 0;
+		private->iob.flags = 0;	/* do synchronous io */
+		private->iob.block_count = 1;
+		private->iob.interrupt_params = 0;
+		private->iob.bio_list = bio;
+		private->iob.flaga = DASD_DIAG_FLAGA_DEFAULT;
+		rc = dia250(&private->iob, RW_BIO);
+		if (rc == 3) {
+			pr_warn("%s: A 64-bit DIAG call failed\n",
+				dev_name(&device->cdev->dev));
+			rc = -EOPNOTSUPP;
+			goto out_bio;
+		}
+		mdsk_term_io(device);
+		if (rc == 0)
+			break;
+	}
+	if (bsize > PAGE_SIZE) {
+		pr_warn("%s: Accessing the DASD failed because of an incorrect format (rc=%d)\n",
+			dev_name(&device->cdev->dev), rc);
+		rc = -EIO;
+		goto out_bio;
+	}
+	/* check for label block */
+	if (memcmp(label->label_id, DASD_DIAG_CMS1,
+		  sizeof(DASD_DIAG_CMS1)) == 0) {
+		/* get formatted blocksize from label block */
+		bsize = (unsigned int) label->block_size;
+		block->blocks = (unsigned long) label->block_count;
+	} else
+		block->blocks = end_block;
+	block->bp_block = bsize;
+	block->s2b_shift = 0;	/* bits to shift 512 to get a block */
+	for (sb = 512; sb < bsize; sb = sb << 1)
+		block->s2b_shift++;
+	rc = mdsk_init_io(device, block->bp_block, 0, NULL);
+	if (rc && (rc != 4)) {
+		pr_warn("%s: DIAG initialization failed with rc=%d\n",
+			dev_name(&device->cdev->dev), rc);
+		rc = -EIO;
+	} else {
+		if (rc == 4)
+			set_bit(DASD_FLAG_DEVICE_RO, &device->flags);
+		pr_info("%s: New DASD with %ld byte/block, total size %ld "
+			"KB%s\n", dev_name(&device->cdev->dev),
+			(unsigned long) block->bp_block,
+			(unsigned long) (block->blocks <<
+					 block->s2b_shift) >> 1,
+			(rc == 4) ? ", read-only device" : "");
+		rc = 0;
+	}
+out_bio:
+	kfree(bio);
+out_label:
+	free_page((long) label);
+out:
+	if (rc) {
+		device->block = NULL;
+		dasd_free_block(block);
+		device->private = NULL;
+		kfree(private);
+	}
+	return rc;
+}
+
+/* Fill in virtual disk geometry for device. Return zero on success, non-zero
+ * otherwise. */
+static int
+dasd_diag_fill_geometry(struct dasd_block *block, struct hd_geometry *geo)
+{
+	if (dasd_check_blocksize(block->bp_block) != 0)
+		return -EINVAL;
+	geo->cylinders = (block->blocks << block->s2b_shift) >> 10;
+	geo->heads = 16;
+	geo->sectors = 128 >> block->s2b_shift;
+	return 0;
+}
+
+static dasd_erp_fn_t
+dasd_diag_erp_action(struct dasd_ccw_req * cqr)
+{
+	return dasd_default_erp_action;
+}
+
+static dasd_erp_fn_t
+dasd_diag_erp_postaction(struct dasd_ccw_req * cqr)
+{
+	return dasd_default_erp_postaction;
+}
+
+/* Create DASD request from block device request. Return pointer to new
+ * request on success, ERR_PTR otherwise. */
+static struct dasd_ccw_req *dasd_diag_build_cp(struct dasd_device *memdev,
+					       struct dasd_block *block,
+					       struct request *req)
+{
+	struct dasd_ccw_req *cqr;
+	struct dasd_diag_req *dreq;
+	struct dasd_diag_bio *dbio;
+	struct req_iterator iter;
+	struct bio_vec bv;
+	char *dst;
+	unsigned int count;
+	sector_t recid, first_rec, last_rec;
+	unsigned int blksize, off;
+	unsigned char rw_cmd;
+
+	if (rq_data_dir(req) == READ)
+		rw_cmd = MDSK_READ_REQ;
+	else if (rq_data_dir(req) == WRITE)
+		rw_cmd = MDSK_WRITE_REQ;
+	else
+		return ERR_PTR(-EINVAL);
+	blksize = block->bp_block;
+	/* Calculate record id of first and last block. */
+	first_rec = blk_rq_pos(req) >> block->s2b_shift;
+	last_rec =
+		(blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift;
+	/* Check struct bio and count the number of blocks for the request. */
+	count = 0;
+	rq_for_each_segment(bv, req, iter) {
+		if (bv.bv_len & (blksize - 1))
+			/* Fba can only do full blocks. */
+			return ERR_PTR(-EINVAL);
+		count += bv.bv_len >> (block->s2b_shift + 9);
+	}
+	/* Paranoia. */
+	if (count != last_rec - first_rec + 1)
+		return ERR_PTR(-EINVAL);
+	/* Build the request */
+	cqr = dasd_smalloc_request(DASD_DIAG_MAGIC, 0, struct_size(dreq, bio, count),
+				   memdev, blk_mq_rq_to_pdu(req));
+	if (IS_ERR(cqr))
+		return cqr;
+
+	dreq = (struct dasd_diag_req *) cqr->data;
+	dreq->block_count = count;
+	dbio = dreq->bio;
+	recid = first_rec;
+	rq_for_each_segment(bv, req, iter) {
+		dst = bvec_virt(&bv);
+		for (off = 0; off < bv.bv_len; off += blksize) {
+			memset(dbio, 0, sizeof (struct dasd_diag_bio));
+			dbio->type = rw_cmd;
+			dbio->block_number = recid + 1;
+			dbio->buffer = dst;
+			dbio++;
+			dst += blksize;
+			recid++;
+		}
+	}
+	cqr->retries = memdev->default_retries;
+	cqr->buildclk = get_tod_clock();
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->startdev = memdev;
+	cqr->memdev = memdev;
+	cqr->block = block;
+	cqr->expires = memdev->default_expires * HZ;
+	cqr->status = DASD_CQR_FILLED;
+	return cqr;
+}
+
+/* Release DASD request. Return non-zero if request was successful, zero
+ * otherwise. */
+static int
+dasd_diag_free_cp(struct dasd_ccw_req *cqr, struct request *req)
+{
+	int status;
+
+	status = cqr->status == DASD_CQR_DONE;
+	dasd_sfree_request(cqr, cqr->memdev);
+	return status;
+}
+
+static void dasd_diag_handle_terminated_request(struct dasd_ccw_req *cqr)
+{
+	if (cqr->retries < 0)
+		cqr->status = DASD_CQR_FAILED;
+	else
+		cqr->status = DASD_CQR_FILLED;
+};
+
+/* Fill in IOCTL data for device. */
+static int
+dasd_diag_fill_info(struct dasd_device * device,
+		    struct dasd_information2_t * info)
+{
+	struct dasd_diag_private *private = device->private;
+
+	info->label_block = (unsigned int) private->pt_block;
+	info->FBA_layout = 1;
+	info->format = DASD_FORMAT_LDL;
+	info->characteristics_size = sizeof(private->rdc_data);
+	memcpy(info->characteristics, &private->rdc_data,
+	       sizeof(private->rdc_data));
+	info->confdata_size = 0;
+	return 0;
+}
+
+static void
+dasd_diag_dump_sense(struct dasd_device *device, struct dasd_ccw_req * req,
+		     struct irb *stat)
+{
+	DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+		    "dump sense not available for DIAG data");
+}
+
+static unsigned int dasd_diag_max_sectors(struct dasd_block *block)
+{
+	return DIAG_MAX_BLOCKS << block->s2b_shift;
+}
+
+static int dasd_diag_pe_handler(struct dasd_device *device,
+				__u8 tbvpm, __u8 fcsecpm)
+{
+	return dasd_generic_verify_path(device, tbvpm);
+}
+
+static struct dasd_discipline dasd_diag_discipline = {
+	.owner = THIS_MODULE,
+	.name = "DIAG",
+	.ebcname = "DIAG",
+	.max_sectors = dasd_diag_max_sectors,
+	.check_device = dasd_diag_check_device,
+	.pe_handler = dasd_diag_pe_handler,
+	.fill_geometry = dasd_diag_fill_geometry,
+	.start_IO = dasd_start_diag,
+	.term_IO = dasd_diag_term_IO,
+	.handle_terminated_request = dasd_diag_handle_terminated_request,
+	.erp_action = dasd_diag_erp_action,
+	.erp_postaction = dasd_diag_erp_postaction,
+	.build_cp = dasd_diag_build_cp,
+	.free_cp = dasd_diag_free_cp,
+	.dump_sense = dasd_diag_dump_sense,
+	.fill_info = dasd_diag_fill_info,
+};
+
+static int __init
+dasd_diag_init(void)
+{
+	if (!machine_is_vm()) {
+		pr_info("Discipline %s cannot be used without z/VM\n",
+			dasd_diag_discipline.name);
+		return -ENODEV;
+	}
+	ASCEBC(dasd_diag_discipline.ebcname, 4);
+
+	irq_subclass_register(IRQ_SUBCLASS_SERVICE_SIGNAL);
+	register_external_irq(EXT_IRQ_CP_SERVICE, dasd_ext_handler);
+	dasd_diag_discipline_pointer = &dasd_diag_discipline;
+	return 0;
+}
+
+static void __exit
+dasd_diag_cleanup(void)
+{
+	unregister_external_irq(EXT_IRQ_CP_SERVICE, dasd_ext_handler);
+	irq_subclass_unregister(IRQ_SUBCLASS_SERVICE_SIGNAL);
+	dasd_diag_discipline_pointer = NULL;
+}
+
+module_init(dasd_diag_init);
+module_exit(dasd_diag_cleanup);
diff --git a/drivers/s390/block/dasd_diag.h b/drivers/s390/block/dasd_diag.h
new file mode 100644
index 000000000..405b6feed
--- /dev/null
+++ b/drivers/s390/block/dasd_diag.h
@@ -0,0 +1,81 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ * Based on.......: linux/drivers/s390/block/mdisk.h
+ * ...............: by Hartmunt Penner <hpenner@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2000
+ *
+ */
+
+#define MDSK_WRITE_REQ 0x01
+#define MDSK_READ_REQ  0x02
+
+#define INIT_BIO	0x00
+#define RW_BIO		0x01
+#define TERM_BIO	0x02
+
+#define DEV_CLASS_FBA	0x01
+#define DEV_CLASS_ECKD	0x04
+
+#define DASD_DIAG_CODE_31BIT		0x03
+#define DASD_DIAG_CODE_64BIT		0x07
+
+#define DASD_DIAG_RWFLAG_ASYNC		0x02
+#define DASD_DIAG_RWFLAG_NOCACHE	0x01
+
+#define DASD_DIAG_FLAGA_FORMAT_64BIT	0x80
+
+struct dasd_diag_characteristics {
+	u16 dev_nr;
+	u16 rdc_len;
+	u8 vdev_class;
+	u8 vdev_type;
+	u8 vdev_status;
+	u8 vdev_flags;
+	u8 rdev_class;
+	u8 rdev_type;
+	u8 rdev_model;
+	u8 rdev_features;
+} __attribute__ ((packed, aligned(4)));
+
+#define DASD_DIAG_FLAGA_DEFAULT		DASD_DIAG_FLAGA_FORMAT_64BIT
+
+typedef u64 blocknum_t;
+typedef s64 sblocknum_t;
+
+struct dasd_diag_bio {
+	u8 type;
+	u8 status;
+	u8 spare1[2];
+	u32 alet;
+	blocknum_t block_number;
+	void *buffer;
+} __attribute__ ((packed, aligned(8)));
+
+struct dasd_diag_init_io {
+	u16 dev_nr;
+	u8 flaga;
+	u8 spare1[21];
+	u32 block_size;
+	u8 spare2[4];
+	blocknum_t offset;
+	sblocknum_t start_block;
+	blocknum_t end_block;
+	u8  spare3[8];
+} __attribute__ ((packed, aligned(8)));
+
+struct dasd_diag_rw_io {
+	u16 dev_nr;
+	u8  flaga;
+	u8  spare1[21];
+	u8  key;
+	u8  flags;
+	u8  spare2[2];
+	u32 block_count;
+	u32 alet;
+	u8  spare3[4];
+	u64 interrupt_params;
+	struct dasd_diag_bio *bio_list;
+	u8  spare4[8];
+} __attribute__ ((packed, aligned(8)));
diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c
new file mode 100644
index 000000000..8d976dd58
--- /dev/null
+++ b/drivers/s390/block/dasd_eckd.c
@@ -0,0 +1,7975 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Carsten Otte <Cotte@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2009
+ * EMC Symmetrix ioctl Copyright EMC Corporation, 2008
+ * Author.........: Nigel Hislop <hislop_nigel@emc.com>
+ */
+
+#include <linux/stddef.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/hdreg.h>	/* HDIO_GETGEO			    */
+#include <linux/bio.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/seq_file.h>
+#include <linux/uaccess.h>
+#include <linux/utsname.h>
+#include <linux/io.h>
+#include <linux/overflow.h>
+
+#include <asm/css_chars.h>
+#include <asm/machine.h>
+#include <asm/debug.h>
+#include <asm/idals.h>
+#include <asm/ebcdic.h>
+#include <asm/cio.h>
+#include <asm/ccwdev.h>
+#include <asm/itcw.h>
+#include <asm/schid.h>
+#include <asm/chpid.h>
+
+#include "dasd_int.h"
+#include "dasd_eckd.h"
+
+/*
+ * raw track access always map to 64k in memory
+ * so it maps to 16 blocks of 4k per track
+ */
+#define DASD_RAW_BLOCK_PER_TRACK 16
+#define DASD_RAW_BLOCKSIZE 4096
+/* 64k are 128 x 512 byte sectors  */
+#define DASD_RAW_SECTORS_PER_TRACK 128
+
+MODULE_DESCRIPTION("S/390 DASD ECKD Disks device driver");
+MODULE_LICENSE("GPL");
+
+/*
+ * Default full-track write bias applied to every ESE volume at online time;
+ * individual volumes can be re-tuned afterwards through their per-device
+ * full_track_bias sysfs attribute. 0 disables full-track writes, 100 always
+ * uses them, 50 (the default) enables the adaptive heuristic. Values above
+ * DASD_FT_BIAS_MAX are capped when applied.
+ */
+static unsigned int full_track_bias = DASD_FT_BIAS_DEFAULT;
+module_param(full_track_bias, uint, 0644);
+MODULE_PARM_DESC(full_track_bias,
+		 "Default ESE full-track write bias 0..100 (0=off, 1..99=adaptive, 100=always)");
+
+static struct dasd_discipline dasd_eckd_discipline;
+
+/* The ccw bus type uses this table to find devices that it sends to
+ * dasd_eckd_probe */
+static struct ccw_device_id dasd_eckd_ids[] = {
+	{ CCW_DEVICE_DEVTYPE (0x3990, 0, 0x3390, 0), .driver_info = 0x1},
+	{ CCW_DEVICE_DEVTYPE (0x2105, 0, 0x3390, 0), .driver_info = 0x2},
+	{ CCW_DEVICE_DEVTYPE (0x3880, 0, 0x3380, 0), .driver_info = 0x3},
+	{ CCW_DEVICE_DEVTYPE (0x3990, 0, 0x3380, 0), .driver_info = 0x4},
+	{ CCW_DEVICE_DEVTYPE (0x2105, 0, 0x3380, 0), .driver_info = 0x5},
+	{ CCW_DEVICE_DEVTYPE (0x9343, 0, 0x9345, 0), .driver_info = 0x6},
+	{ CCW_DEVICE_DEVTYPE (0x2107, 0, 0x3390, 0), .driver_info = 0x7},
+	{ CCW_DEVICE_DEVTYPE (0x2107, 0, 0x3380, 0), .driver_info = 0x8},
+	{ CCW_DEVICE_DEVTYPE (0x1750, 0, 0x3390, 0), .driver_info = 0x9},
+	{ CCW_DEVICE_DEVTYPE (0x1750, 0, 0x3380, 0), .driver_info = 0xa},
+	{ /* end of list */ },
+};
+
+MODULE_DEVICE_TABLE(ccw, dasd_eckd_ids);
+
+static struct ccw_driver dasd_eckd_driver; /* see below */
+
+static void *rawpadpage;
+
+#define INIT_CQR_OK 0
+#define INIT_CQR_UNFORMATTED 1
+#define INIT_CQR_ERROR 2
+
+/* emergency request for reserve/release */
+static struct {
+	struct dasd_ccw_req cqr;
+	struct ccw1 ccw;
+	char data[32];
+} *dasd_reserve_req;
+static DEFINE_MUTEX(dasd_reserve_mutex);
+
+static struct {
+	struct dasd_ccw_req cqr;
+	struct ccw1 ccw[2];
+	char data[40];
+} *dasd_vol_info_req;
+static DEFINE_MUTEX(dasd_vol_info_mutex);
+
+struct ext_pool_exhaust_work_data {
+	struct work_struct worker;
+	struct dasd_device *device;
+	struct dasd_device *base;
+};
+
+/* definitions for the path verification worker */
+struct pe_handler_work_data {
+	struct work_struct worker;
+	struct dasd_device *device;
+	struct dasd_ccw_req cqr;
+	struct ccw1 ccw;
+	__u8 rcd_buffer[DASD_ECKD_RCD_DATA_SIZE];
+	int isglobal;
+	__u8 tbvpm;
+	__u8 fcsecpm;
+};
+static struct pe_handler_work_data *pe_handler_worker;
+static DEFINE_MUTEX(dasd_pe_handler_mutex);
+
+struct check_attention_work_data {
+	struct work_struct worker;
+	struct dasd_device *device;
+	__u8 lpum;
+};
+
+static int dasd_eckd_ext_pool_id(struct dasd_device *);
+static int prepare_itcw(struct itcw *, unsigned int, unsigned int, int,
+			struct dasd_device *, struct dasd_device *,
+			unsigned int, int, unsigned int, unsigned int,
+			unsigned int, unsigned int);
+static int dasd_eckd_query_pprc_status(struct dasd_device *,
+				       struct dasd_pprc_data_sc4 *);
+static struct dasd_ccw_req *dasd_eckd_build_cp_tpm_writefulltrack(struct dasd_device *,
+								  struct dasd_block *,
+								  struct request *,
+								  sector_t, sector_t,
+								  sector_t, sector_t,
+								  unsigned int, unsigned int,
+								  unsigned int, unsigned int,
+								  struct dasd_ccw_req *);
+
+/* initial attempt at a probe function. this can be simplified once
+ * the other detection code is gone */
+static int
+dasd_eckd_probe (struct ccw_device *cdev)
+{
+	int ret;
+
+	/* set ECKD specific ccw-device options */
+	ret = ccw_device_set_options(cdev, CCWDEV_ALLOW_FORCE |
+				     CCWDEV_DO_PATHGROUP | CCWDEV_DO_MULTIPATH);
+	if (ret) {
+		DBF_EVENT_DEVID(DBF_WARNING, cdev, "%s",
+				"dasd_eckd_probe: could not set "
+				"ccw-device options");
+		return ret;
+	}
+	ret = dasd_generic_probe(cdev);
+	return ret;
+}
+
+static int
+dasd_eckd_set_online(struct ccw_device *cdev)
+{
+	return dasd_generic_set_online(cdev, &dasd_eckd_discipline);
+}
+
+static const int sizes_trk0[] = { 28, 148, 84 };
+#define LABEL_SIZE 140
+
+/* head and record addresses of count_area read in analysis ccw */
+static const int count_area_head[] = { 0, 0, 0, 0, 1 };
+static const int count_area_rec[] = { 1, 2, 3, 4, 1 };
+
+static inline unsigned int
+ceil_quot(unsigned int d1, unsigned int d2)
+{
+	return (d1 + (d2 - 1)) / d2;
+}
+
+static unsigned int
+recs_per_track(struct dasd_eckd_characteristics * rdc,
+	       unsigned int kl, unsigned int dl)
+{
+	int dn, kn;
+
+	switch (rdc->dev_type) {
+	case 0x3380:
+		if (kl)
+			return 1499 / (15 + 7 + ceil_quot(kl + 12, 32) +
+				       ceil_quot(dl + 12, 32));
+		else
+			return 1499 / (15 + ceil_quot(dl + 12, 32));
+	case 0x3390:
+		dn = ceil_quot(dl + 6, 232) + 1;
+		if (kl) {
+			kn = ceil_quot(kl + 6, 232) + 1;
+			return 1729 / (10 + 9 + ceil_quot(kl + 6 * kn, 34) +
+				       9 + ceil_quot(dl + 6 * dn, 34));
+		} else
+			return 1729 / (10 + 9 + ceil_quot(dl + 6 * dn, 34));
+	case 0x9345:
+		dn = ceil_quot(dl + 6, 232) + 1;
+		if (kl) {
+			kn = ceil_quot(kl + 6, 232) + 1;
+			return 1420 / (18 + 7 + ceil_quot(kl + 6 * kn, 34) +
+				       ceil_quot(dl + 6 * dn, 34));
+		} else
+			return 1420 / (18 + 7 + ceil_quot(dl + 6 * dn, 34));
+	}
+	return 0;
+}
+
+static void set_ch_t(struct ch_t *geo, __u32 cyl, __u8 head)
+{
+	geo->cyl = (__u16) cyl;
+	geo->head = cyl >> DASD_EAV_CYL_HI_SHIFT;
+	geo->head <<= DASD_EAV_HEAD_HI_SHIFT;
+	geo->head |= head;
+}
+
+static void set_chr_t(void *addr, __u32 cyl, __u8 head, __u8 record)
+{
+	struct chr_t *geo = addr;
+
+	geo->cyl = (__u16)cyl;
+	geo->head = cyl >> DASD_EAV_CYL_HI_SHIFT;
+	geo->head <<= DASD_EAV_HEAD_HI_SHIFT;
+	geo->head |= head;
+	geo->record = record;
+}
+
+/*
+ * calculate failing track from sense data depending if
+ * it is an EAV device or not
+ */
+static int dasd_eckd_track_from_irb(struct irb *irb, struct dasd_device *device,
+				    sector_t *track)
+{
+	struct dasd_eckd_private *private = device->private;
+	u8 *sense = NULL;
+	u32 cyl;
+	u8 head;
+
+	sense = dasd_get_sense(irb);
+	if (!sense) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "ESE error no sense data\n");
+		return -EINVAL;
+	}
+	if (!(sense[27] & DASD_SENSE_BIT_2)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "ESE error no valid track data\n");
+		return -EINVAL;
+	}
+
+	if (sense[27] & DASD_SENSE_BIT_3) {
+		/* enhanced addressing */
+		cyl = sense[30] << 20;
+		cyl |= (sense[31] & 0xF0) << 12;
+		cyl |= sense[28] << 8;
+		cyl |= sense[29];
+	} else {
+		cyl = sense[29] << 8;
+		cyl |= sense[30];
+	}
+	head = sense[31] & 0x0F;
+	*track = cyl * private->rdc_data.trk_per_cyl + head;
+	return 0;
+}
+
+static int set_timestamp(struct ccw1 *ccw, struct DE_eckd_data *data,
+		     struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	int rc;
+
+	rc = get_phys_clock(&data->ep_sys_time);
+	/*
+	 * Ignore return code if XRC is not supported or
+	 * sync clock is switched off
+	 */
+	if ((rc && !private->rdc_data.facilities.XRC_supported) ||
+	    rc == -EOPNOTSUPP || rc == -EACCES)
+		return 0;
+
+	/* switch on System Time Stamp - needed for XRC Support */
+	data->ga_extended |= 0x08; /* switch on 'Time Stamp Valid'   */
+	data->ga_extended |= 0x02; /* switch on 'Extended Parameter' */
+
+	if (ccw) {
+		ccw->count = sizeof(struct DE_eckd_data);
+		ccw->flags |= CCW_FLAG_SLI;
+	}
+
+	return rc;
+}
+
+static int
+define_extent(struct ccw1 *ccw, struct DE_eckd_data *data, unsigned int trk,
+	      unsigned int totrk, int cmd, struct dasd_device *device,
+	      int blksize)
+{
+	struct dasd_eckd_private *private = device->private;
+	u16 heads, beghead, endhead;
+	u32 begcyl, endcyl;
+	int rc = 0;
+
+	if (ccw) {
+		ccw->cmd_code = DASD_ECKD_CCW_DEFINE_EXTENT;
+		ccw->flags = 0;
+		ccw->count = 16;
+		ccw->cda = virt_to_dma32(data);
+	}
+
+	memset(data, 0, sizeof(struct DE_eckd_data));
+	switch (cmd) {
+	case DASD_ECKD_CCW_READ_HOME_ADDRESS:
+	case DASD_ECKD_CCW_READ_RECORD_ZERO:
+	case DASD_ECKD_CCW_READ:
+	case DASD_ECKD_CCW_READ_MT:
+	case DASD_ECKD_CCW_READ_CKD:
+	case DASD_ECKD_CCW_READ_CKD_MT:
+	case DASD_ECKD_CCW_READ_KD:
+	case DASD_ECKD_CCW_READ_KD_MT:
+		data->mask.perm = 0x1;
+		data->attributes.operation = private->attrib.operation;
+		break;
+	case DASD_ECKD_CCW_READ_COUNT:
+		data->mask.perm = 0x1;
+		data->attributes.operation = DASD_BYPASS_CACHE;
+		break;
+	case DASD_ECKD_CCW_READ_TRACK:
+	case DASD_ECKD_CCW_READ_TRACK_DATA:
+		data->mask.perm = 0x1;
+		data->attributes.operation = private->attrib.operation;
+		data->blk_size = 0;
+		break;
+	case DASD_ECKD_CCW_WRITE:
+	case DASD_ECKD_CCW_WRITE_MT:
+	case DASD_ECKD_CCW_WRITE_KD:
+	case DASD_ECKD_CCW_WRITE_KD_MT:
+		data->mask.perm = 0x02;
+		data->attributes.operation = private->attrib.operation;
+		rc = set_timestamp(ccw, data, device);
+		break;
+	case DASD_ECKD_CCW_WRITE_CKD:
+	case DASD_ECKD_CCW_WRITE_CKD_MT:
+		data->attributes.operation = DASD_BYPASS_CACHE;
+		rc = set_timestamp(ccw, data, device);
+		break;
+	case DASD_ECKD_CCW_ERASE:
+	case DASD_ECKD_CCW_WRITE_HOME_ADDRESS:
+	case DASD_ECKD_CCW_WRITE_RECORD_ZERO:
+		data->mask.perm = 0x3;
+		data->mask.auth = 0x1;
+		data->attributes.operation = DASD_BYPASS_CACHE;
+		rc = set_timestamp(ccw, data, device);
+		break;
+	case DASD_ECKD_CCW_WRITE_FULL_TRACK:
+		data->mask.perm = 0x03;
+		data->attributes.operation = private->attrib.operation;
+		data->blk_size = 0;
+		break;
+	case DASD_ECKD_CCW_WRITE_TRACK_DATA:
+		data->mask.perm = 0x02;
+		data->attributes.operation = private->attrib.operation;
+		data->blk_size = blksize;
+		rc = set_timestamp(ccw, data, device);
+		break;
+	default:
+		dev_err(&device->cdev->dev,
+			"0x%x is not a known command\n", cmd);
+		break;
+	}
+
+	data->attributes.mode = 0x3;	/* ECKD */
+
+	if ((private->rdc_data.cu_type == 0x2105 ||
+	     private->rdc_data.cu_type == 0x2107 ||
+	     private->rdc_data.cu_type == 0x1750)
+	    && !(private->uses_cdl && trk < 2))
+		data->ga_extended |= 0x40; /* Regular Data Format Mode */
+
+	heads = private->rdc_data.trk_per_cyl;
+	begcyl = trk / heads;
+	beghead = trk % heads;
+	endcyl = totrk / heads;
+	endhead = totrk % heads;
+
+	/* check for sequential prestage - enhance cylinder range */
+	if (data->attributes.operation == DASD_SEQ_PRESTAGE ||
+	    data->attributes.operation == DASD_SEQ_ACCESS) {
+
+		if (endcyl + private->attrib.nr_cyl < private->real_cyl)
+			endcyl += private->attrib.nr_cyl;
+		else
+			endcyl = (private->real_cyl - 1);
+	}
+
+	set_ch_t(&data->beg_ext, begcyl, beghead);
+	set_ch_t(&data->end_ext, endcyl, endhead);
+	return rc;
+}
+
+
+static void locate_record_ext(struct ccw1 *ccw, struct LRE_eckd_data *data,
+			      unsigned int trk, unsigned int rec_on_trk,
+			      int count, int cmd, struct dasd_device *device,
+			      unsigned int reclen, unsigned int tlf)
+{
+	struct dasd_eckd_private *private = device->private;
+	int sector;
+	int dn, d;
+
+	if (ccw) {
+		ccw->cmd_code = DASD_ECKD_CCW_LOCATE_RECORD_EXT;
+		ccw->flags = 0;
+		if (cmd == DASD_ECKD_CCW_WRITE_FULL_TRACK)
+			ccw->count = 22;
+		else
+			ccw->count = 20;
+		ccw->cda = virt_to_dma32(data);
+	}
+
+	memset(data, 0, sizeof(*data));
+	sector = 0;
+	if (rec_on_trk) {
+		switch (private->rdc_data.dev_type) {
+		case 0x3390:
+			dn = ceil_quot(reclen + 6, 232);
+			d = 9 + ceil_quot(reclen + 6 * (dn + 1), 34);
+			sector = (49 + (rec_on_trk - 1) * (10 + d)) / 8;
+			break;
+		case 0x3380:
+			d = 7 + ceil_quot(reclen + 12, 32);
+			sector = (39 + (rec_on_trk - 1) * (8 + d)) / 7;
+			break;
+		}
+	}
+	data->sector = sector;
+	/* note: meaning of count depends on the operation
+	 *	 for record based I/O it's the number of records, but for
+	 *	 track based I/O it's the number of tracks
+	 */
+	data->count = count;
+	switch (cmd) {
+	case DASD_ECKD_CCW_WRITE_HOME_ADDRESS:
+		data->operation.orientation = 0x3;
+		data->operation.operation = 0x03;
+		break;
+	case DASD_ECKD_CCW_READ_HOME_ADDRESS:
+		data->operation.orientation = 0x3;
+		data->operation.operation = 0x16;
+		break;
+	case DASD_ECKD_CCW_WRITE_RECORD_ZERO:
+		data->operation.orientation = 0x1;
+		data->operation.operation = 0x03;
+		data->count++;
+		break;
+	case DASD_ECKD_CCW_READ_RECORD_ZERO:
+		data->operation.orientation = 0x3;
+		data->operation.operation = 0x16;
+		data->count++;
+		break;
+	case DASD_ECKD_CCW_WRITE:
+	case DASD_ECKD_CCW_WRITE_MT:
+	case DASD_ECKD_CCW_WRITE_KD:
+	case DASD_ECKD_CCW_WRITE_KD_MT:
+		data->auxiliary.length_valid = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x01;
+		break;
+	case DASD_ECKD_CCW_WRITE_CKD:
+	case DASD_ECKD_CCW_WRITE_CKD_MT:
+		data->auxiliary.length_valid = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x03;
+		break;
+	case DASD_ECKD_CCW_WRITE_FULL_TRACK:
+		data->operation.orientation = 0x0;
+		data->operation.operation = 0x3F;
+		data->extended_operation = 0x11;
+		data->length = 0;
+		data->extended_parameter_length = 0x02;
+		if (data->count > 8) {
+			data->extended_parameter[0] = 0xFF;
+			data->extended_parameter[1] = 0xFF;
+			data->extended_parameter[1] <<= (16 - count);
+		} else {
+			data->extended_parameter[0] = 0xFF;
+			data->extended_parameter[0] <<= (8 - count);
+			data->extended_parameter[1] = 0x00;
+		}
+		data->sector = 0xFF;
+		break;
+	case DASD_ECKD_CCW_WRITE_TRACK_DATA:
+		data->auxiliary.length_valid = 0x1;
+		data->length = reclen;	/* not tlf, as one might think */
+		data->operation.operation = 0x3F;
+		data->extended_operation = 0x23;
+		break;
+	case DASD_ECKD_CCW_READ:
+	case DASD_ECKD_CCW_READ_MT:
+	case DASD_ECKD_CCW_READ_KD:
+	case DASD_ECKD_CCW_READ_KD_MT:
+		data->auxiliary.length_valid = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x06;
+		break;
+	case DASD_ECKD_CCW_READ_CKD:
+	case DASD_ECKD_CCW_READ_CKD_MT:
+		data->auxiliary.length_valid = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x16;
+		break;
+	case DASD_ECKD_CCW_READ_COUNT:
+		data->operation.operation = 0x06;
+		break;
+	case DASD_ECKD_CCW_READ_TRACK:
+		data->operation.orientation = 0x1;
+		data->operation.operation = 0x0C;
+		data->extended_parameter_length = 0;
+		data->sector = 0xFF;
+		break;
+	case DASD_ECKD_CCW_READ_TRACK_DATA:
+		data->auxiliary.length_valid = 0x1;
+		data->length = tlf;
+		data->operation.operation = 0x0C;
+		break;
+	case DASD_ECKD_CCW_ERASE:
+		data->length = reclen;
+		data->auxiliary.length_valid = 0x1;
+		data->operation.operation = 0x0b;
+		break;
+	default:
+		DBF_DEV_EVENT(DBF_ERR, device,
+			    "fill LRE unknown opcode 0x%x", cmd);
+		BUG();
+	}
+	set_ch_t(&data->seek_addr,
+		 trk / private->rdc_data.trk_per_cyl,
+		 trk % private->rdc_data.trk_per_cyl);
+	data->search_arg.cyl = data->seek_addr.cyl;
+	data->search_arg.head = data->seek_addr.head;
+	data->search_arg.record = rec_on_trk;
+}
+
+static int prefix_LRE(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
+		      unsigned int trk, unsigned int totrk, int cmd,
+		      struct dasd_device *basedev, struct dasd_device *startdev,
+		      unsigned int format, unsigned int rec_on_trk, int count,
+		      unsigned int blksize, unsigned int tlf)
+{
+	struct dasd_eckd_private *basepriv, *startpriv;
+	struct LRE_eckd_data *lredata;
+	struct DE_eckd_data *dedata;
+	int rc = 0;
+
+	basepriv = basedev->private;
+	startpriv = startdev->private;
+	dedata = &pfxdata->define_extent;
+	lredata = &pfxdata->locate_record;
+
+	ccw->cmd_code = DASD_ECKD_CCW_PFX;
+	ccw->flags = 0;
+	if (cmd == DASD_ECKD_CCW_WRITE_FULL_TRACK) {
+		ccw->count = sizeof(*pfxdata) + 2;
+		ccw->cda = virt_to_dma32(pfxdata);
+		memset(pfxdata, 0, sizeof(*pfxdata) + 2);
+	} else {
+		ccw->count = sizeof(*pfxdata);
+		ccw->cda = virt_to_dma32(pfxdata);
+		memset(pfxdata, 0, sizeof(*pfxdata));
+	}
+
+	/* prefix data */
+	if (format > 1) {
+		DBF_DEV_EVENT(DBF_ERR, basedev,
+			      "PFX LRE unknown format 0x%x", format);
+		BUG();
+		return -EINVAL;
+	}
+	pfxdata->format = format;
+	/* cached copies - conf.ned may be freed under us by the reload worker */
+	pfxdata->base_address = READ_ONCE(basepriv->ned_ua);
+	pfxdata->base_lss = READ_ONCE(basepriv->ned_lss);
+	pfxdata->validity.define_extent = 1;
+
+	/* private uid is kept up to date, conf_data may be outdated */
+	if (startpriv->uid.type == UA_BASE_PAV_ALIAS)
+		pfxdata->validity.verify_base = 1;
+
+	if (startpriv->uid.type == UA_HYPER_PAV_ALIAS) {
+		pfxdata->validity.verify_base = 1;
+		pfxdata->validity.hyper_pav = 1;
+	}
+
+	rc = define_extent(NULL, dedata, trk, totrk, cmd, basedev, blksize);
+
+	/*
+	 * For some commands the System Time Stamp is set in the define extent
+	 * data when XRC is supported. The validity of the time stamp must be
+	 * reflected in the prefix data as well.
+	 */
+	if (dedata->ga_extended & 0x08 && dedata->ga_extended & 0x02)
+		pfxdata->validity.time_stamp = 1; /* 'Time Stamp Valid'   */
+
+	if (format == 1) {
+		locate_record_ext(NULL, lredata, trk, rec_on_trk, count, cmd,
+				  basedev, blksize, tlf);
+	}
+
+	return rc;
+}
+
+static int prefix(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
+		  unsigned int trk, unsigned int totrk, int cmd,
+		  struct dasd_device *basedev, struct dasd_device *startdev)
+{
+	return prefix_LRE(ccw, pfxdata, trk, totrk, cmd, basedev, startdev,
+			  0, 0, 0, 0, 0);
+}
+
+static void
+locate_record(struct ccw1 *ccw, struct LO_eckd_data *data, unsigned int trk,
+	      unsigned int rec_on_trk, int no_rec, int cmd,
+	      struct dasd_device * device, int reclen)
+{
+	struct dasd_eckd_private *private = device->private;
+	int sector;
+	int dn, d;
+
+	DBF_DEV_EVENT(DBF_INFO, device,
+		  "Locate: trk %d, rec %d, no_rec %d, cmd %d, reclen %d",
+		  trk, rec_on_trk, no_rec, cmd, reclen);
+
+	ccw->cmd_code = DASD_ECKD_CCW_LOCATE_RECORD;
+	ccw->flags = 0;
+	ccw->count = 16;
+	ccw->cda = virt_to_dma32(data);
+
+	memset(data, 0, sizeof(struct LO_eckd_data));
+	sector = 0;
+	if (rec_on_trk) {
+		switch (private->rdc_data.dev_type) {
+		case 0x3390:
+			dn = ceil_quot(reclen + 6, 232);
+			d = 9 + ceil_quot(reclen + 6 * (dn + 1), 34);
+			sector = (49 + (rec_on_trk - 1) * (10 + d)) / 8;
+			break;
+		case 0x3380:
+			d = 7 + ceil_quot(reclen + 12, 32);
+			sector = (39 + (rec_on_trk - 1) * (8 + d)) / 7;
+			break;
+		}
+	}
+	data->sector = sector;
+	data->count = no_rec;
+	switch (cmd) {
+	case DASD_ECKD_CCW_WRITE_HOME_ADDRESS:
+		data->operation.orientation = 0x3;
+		data->operation.operation = 0x03;
+		break;
+	case DASD_ECKD_CCW_READ_HOME_ADDRESS:
+		data->operation.orientation = 0x3;
+		data->operation.operation = 0x16;
+		break;
+	case DASD_ECKD_CCW_WRITE_RECORD_ZERO:
+		data->operation.orientation = 0x1;
+		data->operation.operation = 0x03;
+		data->count++;
+		break;
+	case DASD_ECKD_CCW_READ_RECORD_ZERO:
+		data->operation.orientation = 0x3;
+		data->operation.operation = 0x16;
+		data->count++;
+		break;
+	case DASD_ECKD_CCW_WRITE:
+	case DASD_ECKD_CCW_WRITE_MT:
+	case DASD_ECKD_CCW_WRITE_KD:
+	case DASD_ECKD_CCW_WRITE_KD_MT:
+		data->auxiliary.last_bytes_used = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x01;
+		break;
+	case DASD_ECKD_CCW_WRITE_CKD:
+	case DASD_ECKD_CCW_WRITE_CKD_MT:
+		data->auxiliary.last_bytes_used = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x03;
+		break;
+	case DASD_ECKD_CCW_READ:
+	case DASD_ECKD_CCW_READ_MT:
+	case DASD_ECKD_CCW_READ_KD:
+	case DASD_ECKD_CCW_READ_KD_MT:
+		data->auxiliary.last_bytes_used = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x06;
+		break;
+	case DASD_ECKD_CCW_READ_CKD:
+	case DASD_ECKD_CCW_READ_CKD_MT:
+		data->auxiliary.last_bytes_used = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x16;
+		break;
+	case DASD_ECKD_CCW_READ_COUNT:
+		data->operation.operation = 0x06;
+		break;
+	case DASD_ECKD_CCW_ERASE:
+		data->length = reclen;
+		data->auxiliary.last_bytes_used = 0x1;
+		data->operation.operation = 0x0b;
+		break;
+	default:
+		DBF_DEV_EVENT(DBF_ERR, device, "unknown locate record "
+			      "opcode 0x%x", cmd);
+	}
+	set_ch_t(&data->seek_addr,
+		 trk / private->rdc_data.trk_per_cyl,
+		 trk % private->rdc_data.trk_per_cyl);
+	data->search_arg.cyl = data->seek_addr.cyl;
+	data->search_arg.head = data->seek_addr.head;
+	data->search_arg.record = rec_on_trk;
+}
+
+/*
+ * Returns 1 if the block is one of the special blocks that needs
+ * to get read/written with the KD variant of the command.
+ * That is DASD_ECKD_READ_KD_MT instead of DASD_ECKD_READ_MT and
+ * DASD_ECKD_WRITE_KD_MT instead of DASD_ECKD_WRITE_MT.
+ * Luckily the KD variants differ only by one bit (0x08) from the
+ * normal variant. So don't wonder about code like:
+ * if (dasd_eckd_cdl_special(blk_per_trk, recid))
+ *         ccw->cmd_code |= 0x8;
+ */
+static inline int
+dasd_eckd_cdl_special(int blk_per_trk, int recid)
+{
+	if (recid < 3)
+		return 1;
+	if (recid < blk_per_trk)
+		return 0;
+	if (recid < 2 * blk_per_trk)
+		return 1;
+	return 0;
+}
+
+/*
+ * Returns the record size for the special blocks of the cdl format.
+ * Only returns something useful if dasd_eckd_cdl_special is true
+ * for the recid.
+ */
+static inline int
+dasd_eckd_cdl_reclen(int recid)
+{
+	if (recid < 3)
+		return sizes_trk0[recid];
+	return LABEL_SIZE;
+}
+/* create unique id from private structure. */
+static void create_uid(struct dasd_conf *conf, struct dasd_uid *uid)
+{
+	int count;
+
+	memset(uid, 0, sizeof(struct dasd_uid));
+	memcpy(uid->vendor, conf->ned->HDA_manufacturer,
+	       sizeof(uid->vendor) - 1);
+	EBCASC(uid->vendor, sizeof(uid->vendor) - 1);
+	memcpy(uid->serial, &conf->ned->serial,
+	       sizeof(uid->serial) - 1);
+	EBCASC(uid->serial, sizeof(uid->serial) - 1);
+	uid->ssid = conf->gneq->subsystemID;
+	uid->real_unit_addr = conf->ned->unit_addr;
+	if (conf->sneq) {
+		uid->type = conf->sneq->sua_flags;
+		if (uid->type == UA_BASE_PAV_ALIAS)
+			uid->base_unit_addr = conf->sneq->base_unit_addr;
+	} else {
+		uid->type = UA_BASE_DEVICE;
+	}
+	if (conf->vdsneq) {
+		for (count = 0; count < 16; count++) {
+			sprintf(uid->vduit+2*count, "%02x",
+				conf->vdsneq->uit[count]);
+		}
+	}
+}
+
+/*
+ * Generate device unique id that specifies the physical device.
+ */
+static int dasd_eckd_generate_uid(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	unsigned long flags;
+
+	if (!private)
+		return -ENODEV;
+	if (!private->conf.ned || !private->conf.gneq)
+		return -ENODEV;
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	create_uid(&private->conf, &private->uid);
+	/* cache LSS and unit address for the lockless CCW-build path */
+	WRITE_ONCE(private->ned_lss, private->conf.ned->ID);
+	WRITE_ONCE(private->ned_ua, private->conf.ned->unit_addr);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	return 0;
+}
+
+static int dasd_eckd_get_uid(struct dasd_device *device, struct dasd_uid *uid)
+{
+	struct dasd_eckd_private *private = device->private;
+	unsigned long flags;
+
+	if (private) {
+		spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+		*uid = private->uid;
+		spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+		return 0;
+	}
+	return -EINVAL;
+}
+
+/*
+ * compare device UID with data of a given dasd_eckd_private structure
+ * return 0 for match
+ */
+static int dasd_eckd_compare_path_uid(struct dasd_device *device,
+				      struct dasd_conf *path_conf)
+{
+	struct dasd_uid device_uid;
+	struct dasd_uid path_uid;
+
+	create_uid(path_conf, &path_uid);
+	dasd_eckd_get_uid(device, &device_uid);
+
+	return memcmp(&device_uid, &path_uid, sizeof(struct dasd_uid));
+}
+
+static void dasd_eckd_fill_rcd_cqr(struct dasd_device *device,
+				   struct dasd_ccw_req *cqr,
+				   __u8 *rcd_buffer,
+				   __u8 lpm)
+{
+	struct ccw1 *ccw;
+	/*
+	 * buffer has to start with EBCDIC "V1.0" to show
+	 * support for virtual device SNEQ
+	 */
+	rcd_buffer[0] = 0xE5;
+	rcd_buffer[1] = 0xF1;
+	rcd_buffer[2] = 0x4B;
+	rcd_buffer[3] = 0xF0;
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_RCD;
+	ccw->flags = 0;
+	ccw->cda = virt_to_dma32(rcd_buffer);
+	ccw->count = DASD_ECKD_RCD_DATA_SIZE;
+	cqr->magic = DASD_ECKD_MAGIC;
+
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->expires = 10*HZ;
+	cqr->lpm = lpm;
+	cqr->retries = 256;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	set_bit(DASD_CQR_VERIFY_PATH, &cqr->flags);
+}
+
+/*
+ * Wakeup helper for read_conf
+ * if the cqr is not done and needs some error recovery
+ * the buffer has to be re-initialized with the EBCDIC "V1.0"
+ * to show support for virtual device SNEQ
+ */
+static void read_conf_cb(struct dasd_ccw_req *cqr, void *data)
+{
+	struct ccw1 *ccw;
+	__u8 *rcd_buffer;
+
+	if (cqr->status !=  DASD_CQR_DONE) {
+		ccw = cqr->cpaddr;
+		rcd_buffer = dma32_to_virt(ccw->cda);
+		memset(rcd_buffer, 0, sizeof(*rcd_buffer));
+
+		rcd_buffer[0] = 0xE5;
+		rcd_buffer[1] = 0xF1;
+		rcd_buffer[2] = 0x4B;
+		rcd_buffer[3] = 0xF0;
+	}
+	dasd_wakeup_cb(cqr, data);
+}
+
+static int dasd_eckd_read_conf_immediately(struct dasd_device *device,
+					   struct dasd_ccw_req *cqr,
+					   __u8 *rcd_buffer,
+					   __u8 lpm)
+{
+	struct ciw *ciw;
+	int rc;
+	/*
+	 * sanity check: scan for RCD command in extended SenseID data
+	 * some devices do not support RCD
+	 */
+	ciw = ccw_device_get_ciw(device->cdev, CIW_TYPE_RCD);
+	if (!ciw || ciw->cmd != DASD_ECKD_CCW_RCD)
+		return -EOPNOTSUPP;
+
+	dasd_eckd_fill_rcd_cqr(device, cqr, rcd_buffer, lpm);
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	set_bit(DASD_CQR_ALLOW_SLOCK, &cqr->flags);
+	cqr->retries = 5;
+	cqr->callback = read_conf_cb;
+	rc = dasd_sleep_on_immediatly(cqr);
+	return rc;
+}
+
+static int dasd_eckd_read_conf_lpm(struct dasd_device *device,
+				   void **rcd_buffer,
+				   int *rcd_buffer_size, __u8 lpm)
+{
+	struct ciw *ciw;
+	char *rcd_buf = NULL;
+	int ret;
+	struct dasd_ccw_req *cqr;
+
+	/*
+	 * sanity check: scan for RCD command in extended SenseID data
+	 * some devices do not support RCD
+	 */
+	ciw = ccw_device_get_ciw(device->cdev, CIW_TYPE_RCD);
+	if (!ciw || ciw->cmd != DASD_ECKD_CCW_RCD) {
+		ret = -EOPNOTSUPP;
+		goto out_error;
+	}
+	rcd_buf = kzalloc(DASD_ECKD_RCD_DATA_SIZE, GFP_KERNEL | GFP_DMA);
+	if (!rcd_buf) {
+		ret = -ENOMEM;
+		goto out_error;
+	}
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* RCD */,
+				   0, /* use rcd_buf as data ara */
+				   device, NULL);
+	if (IS_ERR(cqr)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "Could not allocate RCD request");
+		ret = -ENOMEM;
+		goto out_error;
+	}
+	dasd_eckd_fill_rcd_cqr(device, cqr, rcd_buf, lpm);
+	cqr->callback = read_conf_cb;
+	ret = dasd_sleep_on(cqr);
+	/*
+	 * on success we update the user input parms
+	 */
+	dasd_sfree_request(cqr, cqr->memdev);
+	if (ret)
+		goto out_error;
+
+	*rcd_buffer_size = DASD_ECKD_RCD_DATA_SIZE;
+	*rcd_buffer = rcd_buf;
+	return 0;
+out_error:
+	kfree(rcd_buf);
+	*rcd_buffer = NULL;
+	*rcd_buffer_size = 0;
+	return ret;
+}
+
+static int dasd_eckd_identify_conf_parts(struct dasd_conf *conf)
+{
+
+	struct dasd_sneq *sneq;
+	int i, count;
+
+	conf->ned = NULL;
+	conf->sneq = NULL;
+	conf->vdsneq = NULL;
+	conf->gneq = NULL;
+	count = conf->len / sizeof(struct dasd_sneq);
+	sneq = (struct dasd_sneq *)conf->data;
+	for (i = 0; i < count; ++i) {
+		if (sneq->flags.identifier == 1 && sneq->format == 1)
+			conf->sneq = sneq;
+		else if (sneq->flags.identifier == 1 && sneq->format == 4)
+			conf->vdsneq = (struct vd_sneq *)sneq;
+		else if (sneq->flags.identifier == 2)
+			conf->gneq = (struct dasd_gneq *)sneq;
+		else if (sneq->flags.identifier == 3 && sneq->res1 == 1)
+			conf->ned = (struct dasd_ned *)sneq;
+		sneq++;
+	}
+	if (!conf->ned || !conf->gneq) {
+		conf->ned = NULL;
+		conf->sneq = NULL;
+		conf->vdsneq = NULL;
+		conf->gneq = NULL;
+		return -EINVAL;
+	}
+	return 0;
+
+};
+
+static unsigned char dasd_eckd_path_access(void *conf_data, int conf_len)
+{
+	struct dasd_gneq *gneq;
+	int i, count, found;
+
+	count = conf_len / sizeof(*gneq);
+	gneq = (struct dasd_gneq *)conf_data;
+	found = 0;
+	for (i = 0; i < count; ++i) {
+		if (gneq->flags.identifier == 2) {
+			found = 1;
+			break;
+		}
+		gneq++;
+	}
+	if (found)
+		return ((char *)gneq)[18] & 0x07;
+	else
+		return 0;
+}
+
+static void dasd_eckd_store_conf_data(struct dasd_device *device,
+				      struct dasd_conf_data *conf_data, int chp)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct channel_path_desc_fmt0 *chp_desc;
+	struct subchannel_id sch_id;
+	void *cdp;
+
+	/*
+	 * path handling and read_conf allocate data
+	 * free it before replacing the pointer
+	 * also replace the old private->conf_data pointer
+	 * with the new one if this points to the same data
+	 */
+	cdp = device->path[chp].conf_data;
+	if (private->conf.data == cdp) {
+		private->conf.data = (void *)conf_data;
+		dasd_eckd_identify_conf_parts(&private->conf);
+	}
+	ccw_device_get_schid(device->cdev, &sch_id);
+	device->path[chp].conf_data = conf_data;
+	device->path[chp].cssid = sch_id.cssid;
+	device->path[chp].ssid = sch_id.ssid;
+	chp_desc = ccw_device_get_chp_desc(device->cdev, chp);
+	if (chp_desc)
+		device->path[chp].chpid = chp_desc->chpid;
+	kfree(chp_desc);
+	kfree(cdp);
+}
+
+static void dasd_eckd_clear_conf_data(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	int i;
+
+	private->conf.data = NULL;
+	private->conf.len = 0;
+	for (i = 0; i < 8; i++) {
+		kfree(device->path[i].conf_data);
+		device->path[i].conf_data = NULL;
+		device->path[i].cssid = 0;
+		device->path[i].ssid = 0;
+		device->path[i].chpid = 0;
+		dasd_path_notoper(device, i);
+	}
+}
+
+static void dasd_eckd_read_fc_security(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	u8 esm_valid;
+	u8 esm[8];
+	int chp;
+	int rc;
+
+	rc = chsc_scud(private->uid.ssid, (u64 *)esm, &esm_valid);
+	if (rc) {
+		for (chp = 0; chp < 8; chp++)
+			device->path[chp].fc_security = 0;
+		return;
+	}
+
+	for (chp = 0; chp < 8; chp++) {
+		if (esm_valid & (0x80 >> chp))
+			device->path[chp].fc_security = esm[chp];
+		else
+			device->path[chp].fc_security = 0;
+	}
+}
+
+static void dasd_eckd_get_uid_string(struct dasd_conf *conf, char *print_uid)
+{
+	struct dasd_uid uid;
+
+	create_uid(conf, &uid);
+	snprintf(print_uid, DASD_UID_STRLEN, "%s.%s.%04x.%02x%s%s",
+		 uid.vendor, uid.serial, uid.ssid, uid.real_unit_addr,
+		 uid.vduit[0] ? "." : "", uid.vduit);
+}
+
+static int dasd_eckd_check_cabling(struct dasd_device *device,
+				   void *conf_data, __u8 lpm)
+{
+	char print_path_uid[DASD_UID_STRLEN], print_device_uid[DASD_UID_STRLEN];
+	struct dasd_eckd_private *private = device->private;
+	struct dasd_conf path_conf;
+
+	path_conf.data = conf_data;
+	path_conf.len = DASD_ECKD_RCD_DATA_SIZE;
+	if (dasd_eckd_identify_conf_parts(&path_conf))
+		return 1;
+
+	if (dasd_eckd_compare_path_uid(device, &path_conf)) {
+		dasd_eckd_get_uid_string(&path_conf, print_path_uid);
+		dasd_eckd_get_uid_string(&private->conf, print_device_uid);
+		dev_err(&device->cdev->dev,
+			"Not all channel paths lead to the same device, path %02X leads to device %s instead of %s\n",
+			lpm, print_path_uid, print_device_uid);
+		return 1;
+	}
+
+	return 0;
+}
+
+static int dasd_eckd_read_conf(struct dasd_device *device)
+{
+	void *conf_data;
+	int conf_len, conf_data_saved;
+	int rc, path_err, pos;
+	__u8 lpm, opm;
+	struct dasd_eckd_private *private;
+
+	private = device->private;
+	opm = ccw_device_get_path_mask(device->cdev);
+	conf_data_saved = 0;
+	path_err = 0;
+	/* get configuration data per operational path */
+	for (lpm = 0x80; lpm; lpm>>= 1) {
+		if (!(lpm & opm))
+			continue;
+		rc = dasd_eckd_read_conf_lpm(device, &conf_data,
+					     &conf_len, lpm);
+		if (rc && rc != -EOPNOTSUPP) {	/* -EOPNOTSUPP is ok */
+			DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
+					"Read configuration data returned "
+					"error %d", rc);
+			return rc;
+		}
+		if (conf_data == NULL) {
+			DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+					"No configuration data "
+					"retrieved");
+			/* no further analysis possible */
+			dasd_path_add_opm(device, opm);
+			continue;	/* no error */
+		}
+		/* save first valid configuration data */
+		if (!conf_data_saved) {
+			/* initially clear previously stored conf_data */
+			dasd_eckd_clear_conf_data(device);
+			private->conf.data = conf_data;
+			private->conf.len = conf_len;
+			if (dasd_eckd_identify_conf_parts(&private->conf)) {
+				private->conf.data = NULL;
+				private->conf.len = 0;
+				kfree(conf_data);
+				continue;
+			}
+			/*
+			 * build device UID that other path data
+			 * can be compared to it
+			 */
+			dasd_eckd_generate_uid(device);
+			conf_data_saved++;
+		} else if (dasd_eckd_check_cabling(device, conf_data, lpm)) {
+			dasd_path_add_cablepm(device, lpm);
+			path_err = -EINVAL;
+			kfree(conf_data);
+			continue;
+		}
+
+		pos = pathmask_to_pos(lpm);
+		dasd_eckd_store_conf_data(device, conf_data, pos);
+
+		switch (dasd_eckd_path_access(conf_data, conf_len)) {
+		case 0x02:
+			dasd_path_add_nppm(device, lpm);
+			break;
+		case 0x03:
+			dasd_path_add_ppm(device, lpm);
+			break;
+		}
+		if (!dasd_path_get_opm(device)) {
+			dasd_path_set_opm(device, lpm);
+			dasd_generic_path_operational(device);
+		} else {
+			dasd_path_add_opm(device, lpm);
+		}
+	}
+
+	return path_err;
+}
+
+static u32 get_fcx_max_data(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	int fcx_in_css, fcx_in_gneq, fcx_in_features;
+	unsigned int mdc;
+	int tpm;
+
+	if (dasd_nofcx)
+		return 0;
+	/* is transport mode supported? */
+	fcx_in_css = css_general_characteristics.fcx;
+	fcx_in_gneq = private->conf.gneq->reserved2[7] & 0x04;
+	fcx_in_features = private->features.feature[40] & 0x80;
+	tpm = fcx_in_css && fcx_in_gneq && fcx_in_features;
+
+	if (!tpm)
+		return 0;
+
+	mdc = ccw_device_get_mdc(device->cdev, 0);
+	if (mdc == 0) {
+		dev_warn(&device->cdev->dev, "Detecting the maximum supported data size for zHPF requests failed\n");
+		return 0;
+	} else {
+		return (u32)mdc * FCX_MAX_DATA_FACTOR;
+	}
+}
+
+static int verify_fcx_max_data(struct dasd_device *device, __u8 lpm)
+{
+	struct dasd_eckd_private *private = device->private;
+	unsigned int mdc;
+	u32 fcx_max_data;
+
+	if (private->fcx_max_data) {
+		mdc = ccw_device_get_mdc(device->cdev, lpm);
+		if (mdc == 0) {
+			dev_warn(&device->cdev->dev,
+				 "Detecting the maximum data size for zHPF "
+				 "requests failed (rc=%d) for a new path %x\n",
+				 mdc, lpm);
+			return mdc;
+		}
+		fcx_max_data = (u32)mdc * FCX_MAX_DATA_FACTOR;
+		if (fcx_max_data < private->fcx_max_data) {
+			dev_warn(&device->cdev->dev,
+				 "The maximum data size for zHPF requests %u "
+				 "on a new path %x is below the active maximum "
+				 "%u\n", fcx_max_data, lpm,
+				 private->fcx_max_data);
+			return -EACCES;
+		}
+	}
+	return 0;
+}
+
+static int rebuild_device_uid(struct dasd_device *device,
+			      struct pe_handler_work_data *data)
+{
+	struct dasd_eckd_private *private = device->private;
+	__u8 lpm, opm = dasd_path_get_opm(device);
+	int rc = -ENODEV;
+
+	for (lpm = 0x80; lpm; lpm >>= 1) {
+		if (!(lpm & opm))
+			continue;
+		memset(&data->rcd_buffer, 0, sizeof(data->rcd_buffer));
+		memset(&data->cqr, 0, sizeof(data->cqr));
+		data->cqr.cpaddr = &data->ccw;
+		rc = dasd_eckd_read_conf_immediately(device, &data->cqr,
+						     data->rcd_buffer,
+						     lpm);
+
+		if (rc) {
+			if (rc == -EOPNOTSUPP) /* -EOPNOTSUPP is ok */
+				continue;
+			DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
+					"Read configuration data "
+					"returned error %d", rc);
+			break;
+		}
+		memcpy(private->conf.data, data->rcd_buffer,
+		       DASD_ECKD_RCD_DATA_SIZE);
+		if (dasd_eckd_identify_conf_parts(&private->conf)) {
+			rc = -ENODEV;
+		} else /* first valid path is enough */
+			break;
+	}
+
+	if (!rc)
+		rc = dasd_eckd_generate_uid(device);
+
+	return rc;
+}
+
+static void dasd_eckd_path_available_action(struct dasd_device *device,
+					    struct pe_handler_work_data *data)
+{
+	__u8 path_rcd_buf[DASD_ECKD_RCD_DATA_SIZE];
+	__u8 lpm, opm, npm, ppm, epm, hpfpm, cablepm;
+	struct dasd_conf_data *conf_data;
+	char print_uid[DASD_UID_STRLEN];
+	struct dasd_conf path_conf;
+	unsigned long flags;
+	int rc, pos;
+
+	opm = 0;
+	npm = 0;
+	ppm = 0;
+	epm = 0;
+	hpfpm = 0;
+	cablepm = 0;
+
+	for (lpm = 0x80; lpm; lpm >>= 1) {
+		if (!(lpm & data->tbvpm))
+			continue;
+		memset(&data->rcd_buffer, 0, sizeof(data->rcd_buffer));
+		memset(&data->cqr, 0, sizeof(data->cqr));
+		data->cqr.cpaddr = &data->ccw;
+		rc = dasd_eckd_read_conf_immediately(device, &data->cqr,
+						     data->rcd_buffer,
+						     lpm);
+		if (!rc) {
+			switch (dasd_eckd_path_access(data->rcd_buffer,
+						      DASD_ECKD_RCD_DATA_SIZE)
+				) {
+			case 0x02:
+				npm |= lpm;
+				break;
+			case 0x03:
+				ppm |= lpm;
+				break;
+			}
+			opm |= lpm;
+		} else if (rc == -EOPNOTSUPP) {
+			DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+					"path verification: No configuration "
+					"data retrieved");
+			opm |= lpm;
+		} else if (rc == -EAGAIN) {
+			DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+					"path verification: device is stopped,"
+					" try again later");
+			epm |= lpm;
+		} else {
+			dev_warn(&device->cdev->dev,
+				 "Reading device feature codes failed "
+				 "(rc=%d) for new path %x\n", rc, lpm);
+			continue;
+		}
+		if (verify_fcx_max_data(device, lpm)) {
+			opm &= ~lpm;
+			npm &= ~lpm;
+			ppm &= ~lpm;
+			hpfpm |= lpm;
+			continue;
+		}
+
+		/*
+		 * save conf_data for comparison after
+		 * rebuild_device_uid may have changed
+		 * the original data
+		 */
+		memcpy(&path_rcd_buf, data->rcd_buffer,
+		       DASD_ECKD_RCD_DATA_SIZE);
+		path_conf.data = (void *)&path_rcd_buf;
+		path_conf.len = DASD_ECKD_RCD_DATA_SIZE;
+		if (dasd_eckd_identify_conf_parts(&path_conf)) {
+			path_conf.data = NULL;
+			path_conf.len = 0;
+			continue;
+		}
+
+		/*
+		 * compare path UID with device UID only if at least
+		 * one valid path is left
+		 * in other case the device UID may have changed and
+		 * the first working path UID will be used as device UID
+		 */
+		if (dasd_path_get_opm(device) &&
+		    dasd_eckd_compare_path_uid(device, &path_conf)) {
+			/*
+			 * the comparison was not successful
+			 * rebuild the device UID with at least one
+			 * known path in case a z/VM hyperswap command
+			 * has changed the device
+			 *
+			 * after this compare again
+			 *
+			 * if either the rebuild or the recompare fails
+			 * the path can not be used
+			 */
+			if (rebuild_device_uid(device, data) ||
+			    dasd_eckd_compare_path_uid(
+				    device, &path_conf)) {
+				dasd_eckd_get_uid_string(&path_conf, print_uid);
+				dev_err(&device->cdev->dev,
+					"The newly added channel path %02X "
+					"will not be used because it leads "
+					"to a different device %s\n",
+					lpm, print_uid);
+				opm &= ~lpm;
+				npm &= ~lpm;
+				ppm &= ~lpm;
+				cablepm |= lpm;
+				continue;
+			}
+		}
+
+		conf_data = kzalloc(DASD_ECKD_RCD_DATA_SIZE, GFP_KERNEL);
+		if (conf_data) {
+			memcpy(conf_data, data->rcd_buffer,
+			       DASD_ECKD_RCD_DATA_SIZE);
+		} else {
+			/*
+			 * path is operational but path config data could not
+			 * be stored due to low mem condition
+			 * add it to the error path mask and schedule a path
+			 * verification later that this could be added again
+			 */
+			epm |= lpm;
+		}
+		pos = pathmask_to_pos(lpm);
+		dasd_eckd_store_conf_data(device, conf_data, pos);
+
+		/*
+		 * There is a small chance that a path is lost again between
+		 * above path verification and the following modification of
+		 * the device opm mask. We could avoid that race here by using
+		 * yet another path mask, but we rather deal with this unlikely
+		 * situation in dasd_start_IO.
+		 */
+		spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+		if (!dasd_path_get_opm(device) && opm) {
+			dasd_path_set_opm(device, opm);
+			dasd_generic_path_operational(device);
+		} else {
+			dasd_path_add_opm(device, opm);
+		}
+		dasd_path_add_nppm(device, npm);
+		dasd_path_add_ppm(device, ppm);
+		if (epm) {
+			dasd_path_add_tbvpm(device, epm);
+			dasd_device_set_timer(device, 50);
+		}
+		dasd_path_add_cablepm(device, cablepm);
+		dasd_path_add_nohpfpm(device, hpfpm);
+		spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+
+		dasd_path_create_kobj(device, pos);
+	}
+}
+
+static void do_pe_handler_work(struct work_struct *work)
+__context_unsafe(/* Conditional locking */)
+{
+	struct pe_handler_work_data *data;
+	struct dasd_device *device;
+
+	data = container_of(work, struct pe_handler_work_data, worker);
+	device = data->device;
+
+	/* delay path verification until device was resumed */
+	if (test_bit(DASD_FLAG_SUSPENDED, &device->flags)) {
+		schedule_work(work);
+		return;
+	}
+	/* check if path verification already running and delay if so */
+	if (test_and_set_bit(DASD_FLAG_PATH_VERIFY, &device->flags)) {
+		schedule_work(work);
+		return;
+	}
+
+	if (data->tbvpm)
+		dasd_eckd_path_available_action(device, data);
+	if (data->fcsecpm)
+		dasd_eckd_read_fc_security(device);
+
+	clear_bit(DASD_FLAG_PATH_VERIFY, &device->flags);
+	dasd_put_device(device);
+	if (data->isglobal)
+		mutex_unlock(&dasd_pe_handler_mutex);
+	else
+		kfree(data);
+}
+
+static int dasd_eckd_pe_handler(struct dasd_device *device,
+				__u8 tbvpm, __u8 fcsecpm)
+__context_unsafe(/* Conditional locking */)
+{
+	struct pe_handler_work_data *data;
+
+	data = kzalloc_obj(*data, GFP_ATOMIC | GFP_DMA);
+	if (!data) {
+		if (mutex_trylock(&dasd_pe_handler_mutex)) {
+			data = pe_handler_worker;
+			data->isglobal = 1;
+		} else {
+			return -ENOMEM;
+		}
+	}
+	INIT_WORK(&data->worker, do_pe_handler_work);
+	dasd_get_device(device);
+	data->device = device;
+	data->tbvpm = tbvpm;
+	data->fcsecpm = fcsecpm;
+	schedule_work(&data->worker);
+	return 0;
+}
+
+static void dasd_eckd_reset_path(struct dasd_device *device, __u8 pm)
+{
+	struct dasd_eckd_private *private = device->private;
+	unsigned long flags;
+
+	if (!private)
+		return;
+	if (!private->fcx_max_data)
+		private->fcx_max_data = get_fcx_max_data(device);
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	dasd_path_set_tbvpm(device, pm ? : dasd_path_get_notoperpm(device));
+	dasd_schedule_device_bh(device);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+}
+
+static int dasd_eckd_read_features(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct dasd_psf_prssd_data *prssdp;
+	struct dasd_rssd_features *features;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int rc;
+
+	memset(&private->features, 0, sizeof(struct dasd_rssd_features));
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */	+ 1 /* RSSD */,
+				   (sizeof(struct dasd_psf_prssd_data) +
+				    sizeof(struct dasd_rssd_features)),
+				   device, NULL);
+	if (IS_ERR(cqr)) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s", "Could not "
+				"allocate initialization request");
+		return PTR_ERR(cqr);
+	}
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->retries = 256;
+	cqr->expires = 10 * HZ;
+
+	/* Prepare for Read Subsystem Data */
+	prssdp = (struct dasd_psf_prssd_data *) cqr->data;
+	memset(prssdp, 0, sizeof(struct dasd_psf_prssd_data));
+	prssdp->order = PSF_ORDER_PRSSD;
+	prssdp->suborder = 0x41;	/* Read Feature Codes */
+	/* all other bytes of prssdp must be zero */
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->count = sizeof(struct dasd_psf_prssd_data);
+	ccw->flags |= CCW_FLAG_CC;
+	ccw->cda = virt_to_dma32(prssdp);
+
+	/* Read Subsystem Data - feature codes */
+	features = (struct dasd_rssd_features *) (prssdp + 1);
+	memset(features, 0, sizeof(struct dasd_rssd_features));
+
+	ccw++;
+	ccw->cmd_code = DASD_ECKD_CCW_RSSD;
+	ccw->count = sizeof(struct dasd_rssd_features);
+	ccw->cda = virt_to_dma32(features);
+
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	rc = dasd_sleep_on(cqr);
+	if (rc == 0) {
+		prssdp = (struct dasd_psf_prssd_data *) cqr->data;
+		features = (struct dasd_rssd_features *) (prssdp + 1);
+		memcpy(&private->features, features,
+		       sizeof(struct dasd_rssd_features));
+	} else
+		dev_warn(&device->cdev->dev, "Reading device feature codes"
+			 " failed with rc=%d\n", rc);
+	dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+/* Read Volume Information - Volume Storage Query */
+static int dasd_eckd_read_vol_info(struct dasd_device *device)
+__context_unsafe(/* Conditional locking */)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct dasd_psf_prssd_data *prssdp;
+	struct dasd_rssd_vsq *vsq;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int useglobal;
+	int rc;
+
+	/* This command cannot be executed on an alias device */
+	if (private->uid.type == UA_BASE_PAV_ALIAS ||
+	    private->uid.type == UA_HYPER_PAV_ALIAS)
+		return 0;
+
+	useglobal = 0;
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 2 /* PSF + RSSD */,
+				   sizeof(*prssdp) + sizeof(*vsq), device, NULL);
+	if (IS_ERR(cqr)) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+				"Could not allocate initialization request");
+		mutex_lock(&dasd_vol_info_mutex);
+		useglobal = 1;
+		cqr = &dasd_vol_info_req->cqr;
+		memset(cqr, 0, sizeof(*cqr));
+		memset(dasd_vol_info_req, 0, sizeof(*dasd_vol_info_req));
+		cqr->cpaddr = &dasd_vol_info_req->ccw;
+		cqr->data = &dasd_vol_info_req->data;
+		cqr->magic = DASD_ECKD_MAGIC;
+	}
+
+	/* Prepare for Read Subsystem Data */
+	prssdp = cqr->data;
+	prssdp->order = PSF_ORDER_PRSSD;
+	prssdp->suborder = PSF_SUBORDER_VSQ;	/* Volume Storage Query */
+	prssdp->lss = READ_ONCE(private->ned_lss);
+	prssdp->volume = READ_ONCE(private->ned_ua);
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->count = sizeof(*prssdp);
+	ccw->flags |= CCW_FLAG_CC;
+	ccw->cda = virt_to_dma32(prssdp);
+
+	/* Read Subsystem Data - Volume Storage Query */
+	vsq = (struct dasd_rssd_vsq *)(prssdp + 1);
+	memset(vsq, 0, sizeof(*vsq));
+
+	ccw++;
+	ccw->cmd_code = DASD_ECKD_CCW_RSSD;
+	ccw->count = sizeof(*vsq);
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->cda = virt_to_dma32(vsq);
+
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->retries = 256;
+	cqr->expires = device->default_expires * HZ;
+	/* The command might not be supported. Suppress the error output */
+	__set_bit(DASD_CQR_SUPPRESS_CR, &cqr->flags);
+
+	rc = dasd_sleep_on_interruptible(cqr);
+	if (rc == 0) {
+		memcpy(&private->vsq, vsq, sizeof(*vsq));
+	} else {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
+				"Reading the volume storage information failed with rc=%d", rc);
+	}
+
+	if (useglobal)
+		mutex_unlock(&dasd_vol_info_mutex);
+	else
+		dasd_sfree_request(cqr, cqr->memdev);
+
+	return rc;
+}
+
+/* Hardware/volume ESE capability, from the Volume Storage Query. */
+static int dasd_eckd_ese_capable(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	if (!private)
+		return 0;
+
+	return private->vsq.vol_info.ese;
+}
+
+/*
+ * Whether the volume is to be handled as ESE (thin). This reflects the state
+ * of the data, not the hardware: a volume copied off ESE storage onto other
+ * hardware still needs ESE handling. The on-disk format label is authoritative
+ * when present; without it (e.g. a volume formatted by an older driver) fall
+ * back to the hardware ESE field.
+ *
+ * Only the F_ESE flag gates this. An ESE volume is thin regardless of whether
+ * it was quick- or full-formatted (tracks are allocated on write, and discard
+ * re-thins a full one).
+ */
+static int dasd_eckd_is_ese(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	/* sysfs may read this during set_online before private is allocated */
+	if (!private)
+		return 0;
+
+	if (private->ese_label_valid)
+		return !!(private->ese_label.flags & DASD_ESE_LABEL_F_ESE);
+
+	return dasd_eckd_ese_capable(device);
+}
+
+/*
+ * Whether the volume is formatted on demand (thin), as opposed to fully
+ * formatted. This is the format mode, not the hardware ESE capability. When a
+ * label is present it is authoritative (F_QUICK). Without a label the mode is
+ * unknown, but an ESE volume is still handled on demand (NRF triggers the
+ * format), so fall back to the ESE state to stay consistent with the driver's
+ * behavior on older, label-less volumes.
+ */
+static int dasd_eckd_on_demand_format(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	/* sysfs may read this during set_online before private is allocated */
+	if (!private)
+		return 0;
+
+	if (private->ese_label_valid)
+		return !!(private->ese_label.flags & DASD_ESE_LABEL_F_QUICK);
+
+	return dasd_eckd_is_ese(device);
+}
+
+static int dasd_eckd_ext_pool_id(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	if (!private)
+		return 0;
+
+	return private->vsq.extent_pool_id;
+}
+
+/*
+ * This value represents the total amount of available space. As more space is
+ * allocated by ESE volumes, this value will decrease.
+ * The data for this value is therefore updated on any call.
+ */
+static int dasd_eckd_space_configured(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	int rc;
+
+	if (!private)
+		return 0;
+
+	rc = dasd_eckd_read_vol_info(device);
+
+	return rc ? : private->vsq.space_configured;
+}
+
+/*
+ * The value of space allocated by an ESE volume may have changed and is
+ * therefore updated on any call.
+ */
+static int dasd_eckd_space_allocated(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	int rc;
+
+	if (!private)
+		return 0;
+
+	rc = dasd_eckd_read_vol_info(device);
+
+	return rc ? : private->vsq.space_allocated;
+}
+
+static int dasd_eckd_logical_capacity(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	if (!private)
+		return 0;
+
+	return private->vsq.logical_capacity;
+}
+
+static void dasd_eckd_ext_pool_exhaust_work(struct work_struct *work)
+{
+	struct ext_pool_exhaust_work_data *data;
+	struct dasd_device *device;
+	struct dasd_device *base;
+
+	data = container_of(work, struct ext_pool_exhaust_work_data, worker);
+	device = data->device;
+	base = data->base;
+
+	if (!base)
+		base = device;
+	if (dasd_eckd_space_configured(base) != 0) {
+		dasd_generic_space_avail(device);
+	} else {
+		dev_warn(&device->cdev->dev, "No space left in the extent pool\n");
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s", "out of space");
+	}
+
+	dasd_put_device(device);
+	kfree(data);
+}
+
+static int dasd_eckd_ext_pool_exhaust(struct dasd_device *device,
+				      struct dasd_ccw_req *cqr)
+{
+	struct ext_pool_exhaust_work_data *data;
+
+	data = kzalloc_obj(*data, GFP_ATOMIC);
+	if (!data)
+		return -ENOMEM;
+	INIT_WORK(&data->worker, dasd_eckd_ext_pool_exhaust_work);
+	dasd_get_device(device);
+	data->device = device;
+
+	if (cqr->block)
+		data->base = cqr->block->base;
+	else if (cqr->basedev)
+		data->base = cqr->basedev;
+	else
+		data->base = NULL;
+
+	schedule_work(&data->worker);
+
+	return 0;
+}
+
+static void dasd_eckd_cpy_ext_pool_data(struct dasd_device *device,
+					struct dasd_rssd_lcq *lcq)
+{
+	struct dasd_eckd_private *private = device->private;
+	int pool_id = dasd_eckd_ext_pool_id(device);
+	struct dasd_ext_pool_sum eps;
+	int i;
+
+	for (i = 0; i < lcq->pool_count; i++) {
+		eps = lcq->ext_pool_sum[i];
+		if (eps.pool_id == pool_id) {
+			memcpy(&private->eps, &eps,
+			       sizeof(struct dasd_ext_pool_sum));
+		}
+	}
+}
+
+/* Read Extent Pool Information - Logical Configuration Query */
+static int dasd_eckd_read_ext_pool_info(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct dasd_psf_prssd_data *prssdp;
+	struct dasd_rssd_lcq *lcq;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int rc;
+
+	/* This command cannot be executed on an alias device */
+	if (private->uid.type == UA_BASE_PAV_ALIAS ||
+	    private->uid.type == UA_HYPER_PAV_ALIAS)
+		return 0;
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 2 /* PSF + RSSD */,
+				   sizeof(*prssdp) + sizeof(*lcq), device, NULL);
+	if (IS_ERR(cqr)) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+				"Could not allocate initialization request");
+		return PTR_ERR(cqr);
+	}
+
+	/* Prepare for Read Subsystem Data */
+	prssdp = cqr->data;
+	memset(prssdp, 0, sizeof(*prssdp));
+	prssdp->order = PSF_ORDER_PRSSD;
+	prssdp->suborder = PSF_SUBORDER_LCQ;	/* Logical Configuration Query */
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->count = sizeof(*prssdp);
+	ccw->flags |= CCW_FLAG_CC;
+	ccw->cda = virt_to_dma32(prssdp);
+
+	lcq = (struct dasd_rssd_lcq *)(prssdp + 1);
+	memset(lcq, 0, sizeof(*lcq));
+
+	ccw++;
+	ccw->cmd_code = DASD_ECKD_CCW_RSSD;
+	ccw->count = sizeof(*lcq);
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->cda = virt_to_dma32(lcq);
+
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->retries = 256;
+	cqr->expires = device->default_expires * HZ;
+	/* The command might not be supported. Suppress the error output */
+	__set_bit(DASD_CQR_SUPPRESS_CR, &cqr->flags);
+
+	rc = dasd_sleep_on_interruptible(cqr);
+	if (rc == 0) {
+		dasd_eckd_cpy_ext_pool_data(device, lcq);
+	} else {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
+				"Reading the logical configuration failed with rc=%d", rc);
+	}
+
+	dasd_sfree_request(cqr, cqr->memdev);
+
+	return rc;
+}
+
+/*
+ * Depending on the device type, the extent size is specified either as
+ * cylinders per extent (CKD) or size per extent (FBA)
+ * A 1GB size corresponds to 1113cyl, and 16MB to 21cyl.
+ */
+static int dasd_eckd_ext_size(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct dasd_ext_pool_sum eps;
+
+	if (!private)
+		return 0;
+	eps = private->eps;
+
+	if (!eps.flags.extent_size_valid)
+		return 0;
+	if (eps.extent_size.size_1G)
+		return 1113;
+	if (eps.extent_size.size_16M)
+		return 21;
+
+	return 0;
+}
+
+static int dasd_eckd_ext_pool_warn_thrshld(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	if (!private)
+		return 0;
+
+	return private->eps.warn_thrshld;
+}
+
+static int dasd_eckd_ext_pool_cap_at_warnlevel(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	if (!private)
+		return 0;
+
+	return private->eps.flags.capacity_at_warnlevel;
+}
+
+/*
+ * Extent Pool out of space
+ */
+static int dasd_eckd_ext_pool_oos(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	if (!private)
+		return 0;
+
+	return private->eps.flags.pool_oos;
+}
+
+/*
+ * Build CP for Perform Subsystem Function - SSC.
+ */
+static struct dasd_ccw_req *dasd_eckd_build_psf_ssc(struct dasd_device *device,
+						    int enable_pav)
+{
+	struct dasd_ccw_req *cqr;
+	struct dasd_psf_ssc_data *psf_ssc_data;
+	struct ccw1 *ccw;
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */ ,
+				  sizeof(struct dasd_psf_ssc_data),
+				   device, NULL);
+
+	if (IS_ERR(cqr)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			   "Could not allocate PSF-SSC request");
+		return cqr;
+	}
+	psf_ssc_data = (struct dasd_psf_ssc_data *)cqr->data;
+	psf_ssc_data->order = PSF_ORDER_SSC;
+	psf_ssc_data->suborder = 0xc0;
+	if (enable_pav) {
+		psf_ssc_data->suborder |= 0x08;
+		psf_ssc_data->reserved[0] = 0x88;
+	}
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->cda = virt_to_dma32(psf_ssc_data);
+	ccw->count = 66;
+
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->retries = 256;
+	cqr->expires = 10*HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	return cqr;
+}
+
+/*
+ * Perform Subsystem Function.
+ * It is necessary to trigger CIO for channel revalidation since this
+ * call might change behaviour of DASD devices.
+ */
+static int
+dasd_eckd_psf_ssc(struct dasd_device *device, int enable_pav,
+		  unsigned long flags)
+{
+	struct dasd_ccw_req *cqr;
+	int rc;
+
+	cqr = dasd_eckd_build_psf_ssc(device, enable_pav);
+	if (IS_ERR(cqr))
+		return PTR_ERR(cqr);
+
+	/*
+	 * set flags e.g. turn on failfast, to prevent blocking
+	 * the calling function should handle failed requests
+	 */
+	cqr->flags |= flags;
+
+	rc = dasd_sleep_on(cqr);
+	if (!rc)
+		/* trigger CIO to reprobe devices */
+		css_schedule_reprobe();
+	else if (cqr->intrc == -EAGAIN)
+		rc = -EAGAIN;
+
+	dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+/*
+ * Valide storage server of current device.
+ */
+static int dasd_eckd_validate_server(struct dasd_device *device,
+				     unsigned long flags)
+{
+	struct dasd_eckd_private *private = device->private;
+	int enable_pav, rc;
+
+	if (private->uid.type == UA_BASE_PAV_ALIAS ||
+	    private->uid.type == UA_HYPER_PAV_ALIAS)
+		return 0;
+	if (dasd_nopav || machine_is_vm())
+		enable_pav = 0;
+	else
+		enable_pav = 1;
+	rc = dasd_eckd_psf_ssc(device, enable_pav, flags);
+
+	/* may be requested feature is not available on server,
+	 * therefore just report error and go ahead */
+	DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "PSF-SSC for SSID %04x "
+			"returned rc=%d", private->uid.ssid, rc);
+	return rc;
+}
+
+/*
+ * worker to do a validate server in case of a lost pathgroup
+ */
+static void dasd_eckd_do_validate_server(struct work_struct *work)
+{
+	struct dasd_device *device = container_of(work, struct dasd_device,
+						  kick_validate);
+	unsigned long flags = 0;
+
+	set_bit(DASD_CQR_FLAGS_FAILFAST, &flags);
+	if (dasd_eckd_validate_server(device, flags)
+	    == -EAGAIN) {
+		/* schedule worker again if failed */
+		schedule_work(&device->kick_validate);
+		return;
+	}
+
+	dasd_put_device(device);
+}
+
+static void dasd_eckd_kick_validate_server(struct dasd_device *device)
+{
+	dasd_get_device(device);
+	/* exit if device not online or in offline processing */
+	if (test_bit(DASD_FLAG_OFFLINE, &device->flags) ||
+	   device->state < DASD_STATE_ONLINE) {
+		dasd_put_device(device);
+		return;
+	}
+	/* queue call to do_validate_server to the kernel event daemon. */
+	if (!schedule_work(&device->kick_validate))
+		dasd_put_device(device);
+}
+
+/*
+ * return if the device is the copy relation primary if a copy relation is active
+ */
+static int dasd_device_is_primary(struct dasd_device *device)
+{
+	if (!device->copy)
+		return 1;
+
+	if (device->copy->active->device == device)
+		return 1;
+
+	return 0;
+}
+
+static int dasd_eckd_alloc_block(struct dasd_device *device)
+{
+	struct dasd_block *block;
+	struct dasd_uid temp_uid;
+
+	if (!dasd_device_is_primary(device))
+		return 0;
+
+	dasd_eckd_get_uid(device, &temp_uid);
+	if (temp_uid.type == UA_BASE_DEVICE) {
+		block = dasd_alloc_block();
+		if (IS_ERR(block)) {
+			DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+					"could not allocate dasd block structure");
+			return PTR_ERR(block);
+		}
+		device->block = block;
+		block->base = device;
+	}
+	return 0;
+}
+
+static bool dasd_eckd_pprc_enabled(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	return private->rdc_data.facilities.PPRC_enabled;
+}
+
+/*
+ * Read the on-disk format label from track 0, record 4. On a formatted volume
+ * R4 holds the label as its record data; on an unformatted (fresh ESE) or
+ * label-less volume the read returns No Record Found, which is expected and
+ * leaves the cache invalid so is_ese() falls back to the hardware field.
+ */
+static void dasd_eckd_read_format_label(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct dasd_format_label *label;
+	struct DE_eckd_data *dedata;
+	struct LO_eckd_data *lodata;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+
+	private->ese_label_valid = false;
+
+	/* The label lives on the base volume; aliases have none of their own. */
+	if (private->uid.type == UA_BASE_PAV_ALIAS ||
+	    private->uid.type == UA_HYPER_PAV_ALIAS)
+		return;
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 3 /* DE + LO + READ */,
+				   sizeof(*dedata) + sizeof(*lodata) +
+				   sizeof(*label), device, NULL);
+	if (IS_ERR(cqr))
+		return;
+
+	dedata = cqr->data;
+	lodata = (struct LO_eckd_data *)(dedata + 1);
+	label = (struct dasd_format_label *)(lodata + 1);
+
+	ccw = cqr->cpaddr;
+	define_extent(ccw++, dedata, 0, 0, DASD_ECKD_CCW_READ, device, 0);
+	ccw[-1].flags |= CCW_FLAG_CC;
+	locate_record(ccw++, lodata, 0, 4, 1, DASD_ECKD_CCW_READ, device,
+		      sizeof(*label));
+	ccw[-1].flags |= CCW_FLAG_CC;
+	ccw->cmd_code = DASD_ECKD_CCW_READ;
+	ccw->count = sizeof(*label);
+	ccw->flags = CCW_FLAG_SLI;
+	ccw->cda = virt_to_dma32(label);
+
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->retries = 256;
+	cqr->expires = 10 * HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	/* R4 may be absent (unformatted) or larger than the label. */
+	set_bit(DASD_CQR_SUPPRESS_NRF, &cqr->flags);
+	set_bit(DASD_CQR_SUPPRESS_IL, &cqr->flags);
+
+	if (!dasd_sleep_on(cqr) &&
+	    label->magic == DASD_ESE_LABEL_MAGIC &&
+	    label->version == DASD_ESE_LABEL_VERSION) {
+		private->ese_label = *label;
+		private->ese_label_valid = true;
+	}
+	dasd_sfree_request(cqr, device);
+}
+
+/*
+ * Check device characteristics.
+ * If the device is accessible using ECKD discipline, the device is enabled.
+ */
+static int
+dasd_eckd_check_characteristics(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	int rc, i;
+	int readonly;
+	unsigned long value;
+
+	/* setup work queue for validate server*/
+	INIT_WORK(&device->kick_validate, dasd_eckd_do_validate_server);
+	/* setup work queue for summary unit check */
+	INIT_WORK(&device->suc_work, dasd_alias_handle_summary_unit_check);
+
+	if (!ccw_device_is_pathgroup(device->cdev)) {
+		dev_warn(&device->cdev->dev,
+			 "A channel path group could not be established\n");
+		return -EIO;
+	}
+	if (!ccw_device_is_multipath(device->cdev)) {
+		dev_info(&device->cdev->dev,
+			 "The DASD is not operating in multipath mode\n");
+	}
+	if (!private) {
+		private = kzalloc_obj(*private, GFP_KERNEL | GFP_DMA);
+		if (!private) {
+			dev_warn(&device->cdev->dev,
+				 "Allocating memory for private DASD data "
+				 "failed\n");
+			return -ENOMEM;
+		}
+		device->private = private;
+	} else {
+		memset(private, 0, sizeof(*private));
+	}
+	/* Invalidate status of initial analysis. */
+	private->init_cqr_status = -1;
+	/* Set default cache operations. */
+	private->attrib.operation = DASD_NORMAL_CACHE;
+	private->attrib.nr_cyl = 0;
+
+	/* Read Configuration Data */
+	rc = dasd_eckd_read_conf(device);
+	if (rc)
+		goto out_err1;
+
+	/* set some default values */
+	device->default_expires = DASD_EXPIRES;
+	device->default_retries = DASD_RETRIES;
+	device->path_thrhld = DASD_ECKD_PATH_THRHLD;
+	device->path_interval = DASD_ECKD_PATH_INTERVAL;
+	device->aq_timeouts = DASD_RETRIES_MAX;
+
+	if (private->conf.gneq) {
+		value = 1;
+		for (i = 0; i < private->conf.gneq->timeout.value; i++)
+			value = 10 * value;
+		value = value * private->conf.gneq->timeout.number;
+		/* do not accept useless values */
+		if (value != 0 && value <= DASD_EXPIRES_MAX)
+			device->default_expires = value;
+	}
+
+	/* Read Device Characteristics */
+	rc = dasd_generic_read_dev_chars(device, DASD_ECKD_MAGIC,
+					 &private->rdc_data, 64);
+	if (rc) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
+				"Read device characteristic failed, rc=%d", rc);
+		goto out_err1;
+	}
+
+	/* setup PPRC for device from devmap */
+	rc = dasd_devmap_set_device_copy_relation(device->cdev,
+						  dasd_eckd_pprc_enabled(device));
+	if (rc) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
+				"copy relation setup failed, rc=%d", rc);
+		goto out_err1;
+	}
+
+	/* check if block device is needed and allocate in case */
+	rc = dasd_eckd_alloc_block(device);
+	if (rc)
+		goto out_err1;
+
+	/* register lcu with alias handling, enable PAV */
+	rc = dasd_alias_make_device_known_to_lcu(device);
+	if (rc)
+		goto out_err2;
+
+	dasd_eckd_validate_server(device, 0);
+
+	/* device may report different configuration data after LCU setup */
+	rc = dasd_eckd_read_conf(device);
+	if (rc)
+		goto out_err3;
+
+	dasd_eckd_read_fc_security(device);
+	dasd_path_create_kobjects(device);
+
+	/* Read Feature Codes */
+	dasd_eckd_read_features(device);
+
+	/* Read Volume Information */
+	dasd_eckd_read_vol_info(device);
+
+	/*
+	 * Advertise discard through the device feature so the block layer sets
+	 * up discard limits. Discard releases allocated space, so require a thin
+	 * (ESE) volume whose storage reports support for the space-release
+	 * function. Raw-track access bypasses the normal block CCW path (discard
+	 * would reach the raw builder, which has no record data), so exclude it.
+	 */
+	if (dasd_eckd_ese_capable(device) &&
+	    (private->features.feature[56] & 0x01) &&
+	    !(device->features & DASD_FEATURE_USERAW))
+		device->features |= DASD_FEATURE_DISCARD;
+
+	/* Read the on-disk format label for ESE detection */
+	dasd_eckd_read_format_label(device);
+
+	/*
+	 * is_ese() now reflects the real ESE state (vsq + on-disk label), so
+	 * the adaptive heuristic can be derived correctly for this device.
+	 */
+	device->ft_bias = min_t(unsigned int, full_track_bias, DASD_FT_BIAS_MAX);
+	dasd_ft_bias_apply(device);
+
+	/* Read Extent Pool Information */
+	dasd_eckd_read_ext_pool_info(device);
+
+	if ((device->features & DASD_FEATURE_USERAW) &&
+	    !(private->rdc_data.facilities.RT_in_LR)) {
+		dev_err(&device->cdev->dev, "The storage server does not "
+			"support raw-track access\n");
+		rc = -EINVAL;
+		goto out_err3;
+	}
+
+	/* find the valid cylinder size */
+	if (private->rdc_data.no_cyl == LV_COMPAT_CYL &&
+	    private->rdc_data.long_no_cyl)
+		private->real_cyl = private->rdc_data.long_no_cyl;
+	else
+		private->real_cyl = private->rdc_data.no_cyl;
+
+	private->fcx_max_data = get_fcx_max_data(device);
+
+	readonly = dasd_device_is_ro(device);
+	if (readonly)
+		set_bit(DASD_FLAG_DEVICE_RO, &device->flags);
+
+	dev_info(&device->cdev->dev, "New DASD %04X/%02X (CU %04X/%02X) "
+		 "with %d cylinders, %d heads, %d sectors%s\n",
+		 private->rdc_data.dev_type,
+		 private->rdc_data.dev_model,
+		 private->rdc_data.cu_type,
+		 private->rdc_data.cu_model.model,
+		 private->real_cyl,
+		 private->rdc_data.trk_per_cyl,
+		 private->rdc_data.sec_per_trk,
+		 readonly ? ", read-only device" : "");
+	return 0;
+
+out_err3:
+	dasd_alias_disconnect_device_from_lcu(device);
+out_err2:
+	dasd_free_block(device->block);
+	device->block = NULL;
+out_err1:
+	dasd_eckd_clear_conf_data(device);
+	dasd_path_remove_kobjects(device);
+	kfree(device->private);
+	device->private = NULL;
+	return rc;
+}
+
+static void dasd_eckd_uncheck_device(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	if (!private)
+		return;
+
+	dasd_alias_disconnect_device_from_lcu(device);
+	private->conf.ned = NULL;
+	private->conf.sneq = NULL;
+	private->conf.vdsneq = NULL;
+	private->conf.gneq = NULL;
+	dasd_eckd_clear_conf_data(device);
+	dasd_path_remove_kobjects(device);
+}
+
+static struct dasd_ccw_req *
+dasd_eckd_analysis_ccw(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct eckd_count *count_data;
+	struct LO_eckd_data *LO_data;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int cplength, datasize;
+	int i;
+
+	cplength = 8;
+	datasize = sizeof(struct DE_eckd_data) + 2*sizeof(struct LO_eckd_data);
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, cplength, datasize, device,
+				   NULL);
+	if (IS_ERR(cqr))
+		return cqr;
+	ccw = cqr->cpaddr;
+	/* Define extent for the first 2 tracks. */
+	define_extent(ccw++, cqr->data, 0, 1,
+		      DASD_ECKD_CCW_READ_COUNT, device, 0);
+	LO_data = cqr->data + sizeof(struct DE_eckd_data);
+	/* Locate record for the first 4 records on track 0. */
+	ccw[-1].flags |= CCW_FLAG_CC;
+	locate_record(ccw++, LO_data++, 0, 0, 4,
+		      DASD_ECKD_CCW_READ_COUNT, device, 0);
+
+	count_data = private->count_area;
+	for (i = 0; i < 4; i++) {
+		ccw[-1].flags |= CCW_FLAG_CC;
+		ccw->cmd_code = DASD_ECKD_CCW_READ_COUNT;
+		ccw->flags = 0;
+		ccw->count = 8;
+		ccw->cda = virt_to_dma32(count_data);
+		ccw++;
+		count_data++;
+	}
+
+	/* Locate record for the first record on track 1. */
+	ccw[-1].flags |= CCW_FLAG_CC;
+	locate_record(ccw++, LO_data++, 1, 0, 1,
+		      DASD_ECKD_CCW_READ_COUNT, device, 0);
+	/* Read count ccw. */
+	ccw[-1].flags |= CCW_FLAG_CC;
+	ccw->cmd_code = DASD_ECKD_CCW_READ_COUNT;
+	ccw->flags = 0;
+	ccw->count = 8;
+	ccw->cda = virt_to_dma32(count_data);
+
+	cqr->block = NULL;
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->retries = 255;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	/* Set flags to suppress output for expected errors */
+	set_bit(DASD_CQR_SUPPRESS_NRF, &cqr->flags);
+	set_bit(DASD_CQR_SUPPRESS_IT, &cqr->flags);
+
+	return cqr;
+}
+
+/* differentiate between 'no record found' and any other error */
+static int dasd_eckd_analysis_evaluation(struct dasd_ccw_req *init_cqr)
+{
+	char *sense;
+	if (init_cqr->status == DASD_CQR_DONE)
+		return INIT_CQR_OK;
+	else if (init_cqr->status == DASD_CQR_NEED_ERP ||
+		 init_cqr->status == DASD_CQR_FAILED) {
+		sense = dasd_get_sense(&init_cqr->irb);
+		if (sense && (sense[1] & SNS1_NO_REC_FOUND))
+			return INIT_CQR_UNFORMATTED;
+		else
+			return INIT_CQR_ERROR;
+	} else
+		return INIT_CQR_ERROR;
+}
+
+/*
+ * This is the callback function for the init_analysis cqr. It saves
+ * the status of the initial analysis ccw before it frees it and kicks
+ * the device to continue the startup sequence. This will call
+ * dasd_eckd_do_analysis again (if the devices has not been marked
+ * for deletion in the meantime).
+ */
+static void dasd_eckd_analysis_callback(struct dasd_ccw_req *init_cqr,
+					void *data)
+{
+	struct dasd_device *device = init_cqr->startdev;
+	struct dasd_eckd_private *private = device->private;
+
+	private->init_cqr_status = dasd_eckd_analysis_evaluation(init_cqr);
+	dasd_sfree_request(init_cqr, device);
+	dasd_kick_device(device);
+}
+
+static int dasd_eckd_start_analysis(struct dasd_block *block)
+{
+	struct dasd_ccw_req *init_cqr;
+
+	init_cqr = dasd_eckd_analysis_ccw(block->base);
+	if (IS_ERR(init_cqr))
+		return PTR_ERR(init_cqr);
+	init_cqr->callback = dasd_eckd_analysis_callback;
+	init_cqr->callback_data = NULL;
+	init_cqr->expires = 5*HZ;
+	/* first try without ERP, so we can later handle unformatted
+	 * devices as special case
+	 */
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &init_cqr->flags);
+	init_cqr->retries = 0;
+	dasd_add_request_head(init_cqr);
+	return -EAGAIN;
+}
+
+static int dasd_eckd_end_analysis(struct dasd_block *block)
+{
+	struct dasd_device *device = block->base;
+	struct dasd_eckd_private *private = device->private;
+	struct eckd_count *count_area;
+	const char *ese_str, *fmt_str;
+	unsigned int sb, blk_per_trk;
+	int status, i;
+	struct dasd_ccw_req *init_cqr;
+
+	status = private->init_cqr_status;
+	private->init_cqr_status = -1;
+	if (status == INIT_CQR_ERROR) {
+		/* try again, this time with full ERP */
+		init_cqr = dasd_eckd_analysis_ccw(device);
+		dasd_sleep_on(init_cqr);
+		status = dasd_eckd_analysis_evaluation(init_cqr);
+		dasd_sfree_request(init_cqr, device);
+	}
+
+	if (device->features & DASD_FEATURE_USERAW) {
+		block->bp_block = DASD_RAW_BLOCKSIZE;
+		blk_per_trk = DASD_RAW_BLOCK_PER_TRACK;
+		block->s2b_shift = 3;
+		goto raw;
+	}
+
+	if (status == INIT_CQR_UNFORMATTED) {
+		dev_warn(&device->cdev->dev, "The DASD is not formatted\n");
+		return -EMEDIUMTYPE;
+	} else if (status == INIT_CQR_ERROR) {
+		dev_err(&device->cdev->dev,
+			"Detecting the DASD disk layout failed because "
+			"of an I/O error\n");
+		return -EIO;
+	}
+
+	private->uses_cdl = 1;
+	/* Check Track 0 for Compatible Disk Layout */
+	count_area = NULL;
+	for (i = 0; i < 3; i++) {
+		if (private->count_area[i].kl != 4 ||
+		    private->count_area[i].dl != dasd_eckd_cdl_reclen(i) - 4 ||
+		    private->count_area[i].cyl != 0 ||
+		    private->count_area[i].head != count_area_head[i] ||
+		    private->count_area[i].record != count_area_rec[i]) {
+			private->uses_cdl = 0;
+			break;
+		}
+	}
+	if (i == 3)
+		count_area = &private->count_area[3];
+
+	if (private->uses_cdl == 0) {
+		for (i = 0; i < 5; i++) {
+			if ((private->count_area[i].kl != 0) ||
+			    (private->count_area[i].dl !=
+			     private->count_area[0].dl) ||
+			    private->count_area[i].cyl !=  0 ||
+			    private->count_area[i].head != count_area_head[i] ||
+			    private->count_area[i].record != count_area_rec[i])
+				break;
+		}
+		if (i == 5)
+			count_area = &private->count_area[0];
+	} else {
+		if (private->count_area[3].record == 1)
+			dev_warn(&device->cdev->dev,
+				 "Track 0 has no records following the VTOC\n");
+	}
+
+	if (count_area != NULL && count_area->kl == 0) {
+		/* we found nothing violating our disk layout */
+		if (dasd_check_blocksize(count_area->dl) == 0)
+			block->bp_block = count_area->dl;
+	}
+	if (block->bp_block == 0) {
+		dev_warn(&device->cdev->dev,
+			 "The disk layout of the DASD is not supported\n");
+		return -EMEDIUMTYPE;
+	}
+	block->s2b_shift = 0;	/* bits to shift 512 to get a block */
+	for (sb = 512; sb < block->bp_block; sb = sb << 1)
+		block->s2b_shift++;
+
+	blk_per_trk = recs_per_track(&private->rdc_data, 0, block->bp_block);
+
+raw:
+	block->blocks = ((unsigned long) private->real_cyl *
+			  private->rdc_data.trk_per_cyl *
+			  blk_per_trk);
+
+	/*
+	 * Report the ESE hardware capability and the format mode. The mode
+	 * comes from dasd_eckd_on_demand_format() (the on-disk label, or the
+	 * ESE state when no label is present), matching the on_demand_formatting
+	 * sysfs attribute.
+	 */
+	ese_str = dasd_eckd_ese_capable(device) ? ", ESE" : "";
+	fmt_str = "";
+	if (dasd_eckd_on_demand_format(device))
+		fmt_str = ", on-demand format";
+	else if (dasd_eckd_ese_capable(device))
+		fmt_str = ", full format";
+
+	dev_info(&device->cdev->dev,
+		 "DASD with %u KB/block, %lu KB total size, %u KB/track, %s%s%s\n",
+		 (block->bp_block >> 10),
+		 (((unsigned long) private->real_cyl *
+		   private->rdc_data.trk_per_cyl *
+		   blk_per_trk * (block->bp_block >> 9)) >> 1),
+		 ((blk_per_trk * block->bp_block) >> 10),
+		 private->uses_cdl ?
+		 "compatible disk layout" : "linux disk layout",
+		 ese_str, fmt_str);
+
+	return 0;
+}
+
+static int dasd_eckd_do_analysis(struct dasd_block *block)
+{
+	struct dasd_eckd_private *private = block->base->private;
+
+	if (private->init_cqr_status < 0)
+		return dasd_eckd_start_analysis(block);
+	else
+		return dasd_eckd_end_analysis(block);
+}
+
+static int dasd_eckd_basic_to_ready(struct dasd_device *device)
+{
+	return dasd_alias_add_device(device);
+};
+
+static int dasd_eckd_online_to_ready(struct dasd_device *device)
+{
+	if (cancel_work_sync(&device->reload_device))
+		dasd_put_device(device);
+	if (cancel_work_sync(&device->kick_validate))
+		dasd_put_device(device);
+
+	return 0;
+};
+
+static int dasd_eckd_basic_to_known(struct dasd_device *device)
+{
+	return dasd_alias_remove_device(device);
+};
+
+static int
+dasd_eckd_fill_geometry(struct dasd_block *block, struct hd_geometry *geo)
+{
+	struct dasd_eckd_private *private = block->base->private;
+
+	if (dasd_check_blocksize(block->bp_block) == 0) {
+		geo->sectors = recs_per_track(&private->rdc_data,
+					      0, block->bp_block);
+	}
+	geo->cylinders = private->rdc_data.no_cyl;
+	geo->heads = private->rdc_data.trk_per_cyl;
+	return 0;
+}
+
+/*
+ * Build the TCW request for the format check
+ */
+static struct dasd_ccw_req *
+dasd_eckd_build_check_tcw(struct dasd_device *base, struct format_data_t *fdata,
+			  int enable_pav, struct eckd_count *fmt_buffer,
+			  int rpt)
+{
+	struct dasd_eckd_private *start_priv;
+	struct dasd_device *startdev = NULL;
+	struct tidaw *last_tidaw = NULL;
+	struct dasd_ccw_req *cqr;
+	struct itcw *itcw;
+	int itcw_size;
+	int count;
+	int rc;
+	int i;
+
+	if (enable_pav)
+		startdev = dasd_alias_get_start_dev(base);
+
+	if (!startdev)
+		startdev = base;
+
+	start_priv = startdev->private;
+
+	count = rpt * (fdata->stop_unit - fdata->start_unit + 1);
+
+	/*
+	 * we're adding 'count' amount of tidaw to the itcw.
+	 * calculate the corresponding itcw_size
+	 */
+	itcw_size = itcw_calc_size(0, count, 0);
+
+	cqr = dasd_fmalloc_request(DASD_ECKD_MAGIC, 0, itcw_size, startdev);
+	if (IS_ERR(cqr))
+		return cqr;
+
+	start_priv->count++;
+
+	itcw = itcw_init(cqr->data, itcw_size, ITCW_OP_READ, 0, count, 0);
+	if (IS_ERR(itcw)) {
+		rc = -EINVAL;
+		goto out_err;
+	}
+
+	cqr->cpaddr = itcw_get_tcw(itcw);
+	rc = prepare_itcw(itcw, fdata->start_unit, fdata->stop_unit,
+			  DASD_ECKD_CCW_READ_COUNT_MT, base, startdev, 0, count,
+			  sizeof(struct eckd_count),
+			  count * sizeof(struct eckd_count), 0, rpt);
+	if (rc)
+		goto out_err;
+
+	for (i = 0; i < count; i++) {
+		last_tidaw = itcw_add_tidaw(itcw, 0, fmt_buffer++,
+					    sizeof(struct eckd_count));
+		if (IS_ERR(last_tidaw)) {
+			rc = -EINVAL;
+			goto out_err;
+		}
+	}
+
+	last_tidaw->flags |= TIDAW_FLAGS_LAST;
+	itcw_finalize(itcw);
+
+	cqr->cpmode = 1;
+	cqr->startdev = startdev;
+	cqr->memdev = startdev;
+	cqr->basedev = base;
+	cqr->retries = startdev->default_retries;
+	cqr->expires = startdev->default_expires * HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	/* Set flags to suppress output for expected errors */
+	set_bit(DASD_CQR_SUPPRESS_IL, &cqr->flags);
+
+	return cqr;
+
+out_err:
+	dasd_sfree_request(cqr, startdev);
+
+	return ERR_PTR(rc);
+}
+
+/*
+ * Build the CCW request for the format check
+ */
+static struct dasd_ccw_req *
+dasd_eckd_build_check(struct dasd_device *base, struct format_data_t *fdata,
+		      int enable_pav, struct eckd_count *fmt_buffer, int rpt)
+{
+	struct dasd_eckd_private *start_priv;
+	struct dasd_eckd_private *base_priv;
+	struct dasd_device *startdev = NULL;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	void *data;
+	int cplength, datasize;
+	int use_prefix;
+	int count;
+	int i;
+
+	if (enable_pav)
+		startdev = dasd_alias_get_start_dev(base);
+
+	if (!startdev)
+		startdev = base;
+
+	start_priv = startdev->private;
+	base_priv = base->private;
+
+	count = rpt * (fdata->stop_unit - fdata->start_unit + 1);
+
+	use_prefix = base_priv->features.feature[8] & 0x01;
+
+	if (use_prefix) {
+		cplength = 1;
+		datasize = sizeof(struct PFX_eckd_data);
+	} else {
+		cplength = 2;
+		datasize = sizeof(struct DE_eckd_data) +
+			sizeof(struct LO_eckd_data);
+	}
+	cplength += count;
+
+	cqr = dasd_fmalloc_request(DASD_ECKD_MAGIC, cplength, datasize, startdev);
+	if (IS_ERR(cqr))
+		return cqr;
+
+	start_priv->count++;
+	data = cqr->data;
+	ccw = cqr->cpaddr;
+
+	if (use_prefix) {
+		prefix_LRE(ccw++, data, fdata->start_unit, fdata->stop_unit,
+			   DASD_ECKD_CCW_READ_COUNT, base, startdev, 1, 0,
+			   count, 0, 0);
+	} else {
+		define_extent(ccw++, data, fdata->start_unit, fdata->stop_unit,
+			      DASD_ECKD_CCW_READ_COUNT, startdev, 0);
+
+		data += sizeof(struct DE_eckd_data);
+		ccw[-1].flags |= CCW_FLAG_CC;
+
+		locate_record(ccw++, data, fdata->start_unit, 0, count,
+			      DASD_ECKD_CCW_READ_COUNT, base, 0);
+	}
+
+	for (i = 0; i < count; i++) {
+		ccw[-1].flags |= CCW_FLAG_CC;
+		ccw->cmd_code = DASD_ECKD_CCW_READ_COUNT;
+		ccw->flags = CCW_FLAG_SLI;
+		ccw->count = 8;
+		ccw->cda = virt_to_dma32(fmt_buffer);
+		ccw++;
+		fmt_buffer++;
+	}
+
+	cqr->startdev = startdev;
+	cqr->memdev = startdev;
+	cqr->basedev = base;
+	cqr->retries = DASD_RETRIES;
+	cqr->expires = startdev->default_expires * HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	/* Set flags to suppress output for expected errors */
+	set_bit(DASD_CQR_SUPPRESS_NRF, &cqr->flags);
+
+	return cqr;
+}
+
+/* Fill the format label into a R4 record buffer, zero-padded to blksize. */
+static void dasd_eckd_fill_format_label(struct dasd_device *device, void *data,
+					unsigned int blksize)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct dasd_format_label *label = data;
+
+	memset(label, 0, blksize);
+	label->magic = DASD_ESE_LABEL_MAGIC;
+	label->version = DASD_ESE_LABEL_VERSION;
+	/*
+	 * F_ESE records the hardware capability at format time, not is_ese():
+	 * is_ese() is derived from the label, so using it here would let the
+	 * flag flip on repeated quick/full reformats.
+	 */
+	if (dasd_eckd_ese_capable(device))
+		label->flags |= DASD_ESE_LABEL_F_ESE;
+	if (private->ese_format_quick)
+		label->flags |= DASD_ESE_LABEL_F_QUICK;
+	else
+		label->flags |= DASD_ESE_LABEL_F_FULL;
+	label->blksize = blksize;
+	label->format_tod = get_tod_clock();
+	strscpy(label->kernel_version, init_utsname()->release,
+		sizeof(label->kernel_version));
+
+	/*
+	 * Populate the cache directly from the bytes just computed instead of
+	 * synchronously reading them back from disk after the write lands.
+	 */
+	private->ese_label = *label;
+	private->ese_label_valid = true;
+}
+
+static struct dasd_ccw_req *
+dasd_eckd_build_format(struct dasd_device *base, struct dasd_device *startdev,
+		       struct format_data_t *fdata, int enable_pav)
+{
+	struct dasd_eckd_private *base_priv;
+	struct dasd_eckd_private *start_priv;
+	struct dasd_ccw_req *fcp;
+	struct eckd_count *ect;
+	struct ch_t address;
+	struct ccw1 *ccw;
+	void *data;
+	int rpt;
+	int cplength, datasize;
+	int i, j;
+	int intensity = 0;
+	int r0_perm;
+	int nr_tracks;
+	int use_prefix;
+	int write_label;
+
+	if (enable_pav)
+		startdev = dasd_alias_get_start_dev(base);
+
+	if (!startdev)
+		startdev = base;
+
+	start_priv = startdev->private;
+	base_priv = base->private;
+
+	rpt = recs_per_track(&base_priv->rdc_data, 0, fdata->blksize);
+
+	nr_tracks = fdata->stop_unit - fdata->start_unit + 1;
+
+	/*
+	 * fdata->intensity is a bit string that tells us what to do:
+	 *   Bit 0: write record zero
+	 *   Bit 1: write home address, currently not supported
+	 *   Bit 2: invalidate tracks
+	 *   Bit 3: use OS/390 compatible disk layout (cdl)
+	 *   Bit 4: do not allow storage subsystem to modify record zero
+	 * Only some bit combinations do make sense.
+	 */
+	if (fdata->intensity & 0x10) {
+		r0_perm = 0;
+		intensity = fdata->intensity & ~0x10;
+	} else {
+		r0_perm = 1;
+		intensity = fdata->intensity;
+	}
+
+	use_prefix = base_priv->features.feature[8] & 0x01;
+
+	/*
+	 * Stamp the format label into R4 of the very first track. Only for CDL
+	 * (R4 is the first non-special record there), only when this request
+	 * covers track 0, only for the record-writing format intensities (not
+	 * track invalidation), and only if the track actually has an R4.
+	 */
+	write_label = (intensity & 0x08) && !((intensity & ~0x08) & 0x04) &&
+		      fdata->start_unit == 0 && rpt > 3;
+
+	switch (intensity) {
+	case 0x00:	/* Normal format */
+	case 0x08:	/* Normal format, use cdl. */
+		cplength = 2 + (rpt*nr_tracks);
+		if (use_prefix)
+			datasize = sizeof(struct PFX_eckd_data) +
+				sizeof(struct LO_eckd_data) +
+				rpt * nr_tracks * sizeof(struct eckd_count);
+		else
+			datasize = sizeof(struct DE_eckd_data) +
+				sizeof(struct LO_eckd_data) +
+				rpt * nr_tracks * sizeof(struct eckd_count);
+		break;
+	case 0x01:	/* Write record zero and format track. */
+	case 0x09:	/* Write record zero and format track, use cdl. */
+		cplength = 2 + rpt * nr_tracks;
+		if (use_prefix)
+			datasize = sizeof(struct PFX_eckd_data) +
+				sizeof(struct LO_eckd_data) +
+				sizeof(struct eckd_count) +
+				rpt * nr_tracks * sizeof(struct eckd_count);
+		else
+			datasize = sizeof(struct DE_eckd_data) +
+				sizeof(struct LO_eckd_data) +
+				sizeof(struct eckd_count) +
+				rpt * nr_tracks * sizeof(struct eckd_count);
+		break;
+	case 0x04:	/* Invalidate track. */
+	case 0x0c:	/* Invalidate track, use cdl. */
+		cplength = 3;
+		if (use_prefix)
+			datasize = sizeof(struct PFX_eckd_data) +
+				sizeof(struct LO_eckd_data) +
+				sizeof(struct eckd_count);
+		else
+			datasize = sizeof(struct DE_eckd_data) +
+				sizeof(struct LO_eckd_data) +
+				sizeof(struct eckd_count);
+		break;
+	default:
+		dev_warn(&startdev->cdev->dev,
+			 "An I/O control call used incorrect flags 0x%x\n",
+			 fdata->intensity);
+		return ERR_PTR(-EINVAL);
+	}
+
+	/* room for the label data that R4 carries in addition to its count */
+	if (write_label)
+		datasize += fdata->blksize;
+
+	fcp = dasd_fmalloc_request(DASD_ECKD_MAGIC, cplength, datasize, startdev);
+	if (IS_ERR(fcp))
+		return fcp;
+
+	start_priv->count++;
+	data = fcp->data;
+	ccw = fcp->cpaddr;
+
+	switch (intensity & ~0x08) {
+	case 0x00: /* Normal format. */
+		if (use_prefix) {
+			prefix(ccw++, (struct PFX_eckd_data *) data,
+			       fdata->start_unit, fdata->stop_unit,
+			       DASD_ECKD_CCW_WRITE_CKD, base, startdev);
+			/* grant subsystem permission to format R0 */
+			if (r0_perm)
+				((struct PFX_eckd_data *)data)
+					->define_extent.ga_extended |= 0x04;
+			data += sizeof(struct PFX_eckd_data);
+		} else {
+			define_extent(ccw++, (struct DE_eckd_data *) data,
+				      fdata->start_unit, fdata->stop_unit,
+				      DASD_ECKD_CCW_WRITE_CKD, startdev, 0);
+			/* grant subsystem permission to format R0 */
+			if (r0_perm)
+				((struct DE_eckd_data *) data)
+					->ga_extended |= 0x04;
+			data += sizeof(struct DE_eckd_data);
+		}
+		ccw[-1].flags |= CCW_FLAG_CC;
+		locate_record(ccw++, (struct LO_eckd_data *) data,
+			      fdata->start_unit, 0, rpt*nr_tracks,
+			      DASD_ECKD_CCW_WRITE_CKD, base,
+			      fdata->blksize);
+		data += sizeof(struct LO_eckd_data);
+		break;
+	case 0x01: /* Write record zero + format track. */
+		if (use_prefix) {
+			prefix(ccw++, (struct PFX_eckd_data *) data,
+			       fdata->start_unit, fdata->stop_unit,
+			       DASD_ECKD_CCW_WRITE_RECORD_ZERO,
+			       base, startdev);
+			data += sizeof(struct PFX_eckd_data);
+		} else {
+			define_extent(ccw++, (struct DE_eckd_data *) data,
+			       fdata->start_unit, fdata->stop_unit,
+			       DASD_ECKD_CCW_WRITE_RECORD_ZERO, startdev, 0);
+			data += sizeof(struct DE_eckd_data);
+		}
+		ccw[-1].flags |= CCW_FLAG_CC;
+		locate_record(ccw++, (struct LO_eckd_data *) data,
+			      fdata->start_unit, 0, rpt * nr_tracks + 1,
+			      DASD_ECKD_CCW_WRITE_RECORD_ZERO, base,
+			      base->block->bp_block);
+		data += sizeof(struct LO_eckd_data);
+		break;
+	case 0x04: /* Invalidate track. */
+		if (use_prefix) {
+			prefix(ccw++, (struct PFX_eckd_data *) data,
+			       fdata->start_unit, fdata->stop_unit,
+			       DASD_ECKD_CCW_WRITE_CKD, base, startdev);
+			data += sizeof(struct PFX_eckd_data);
+		} else {
+			define_extent(ccw++, (struct DE_eckd_data *) data,
+			       fdata->start_unit, fdata->stop_unit,
+			       DASD_ECKD_CCW_WRITE_CKD, startdev, 0);
+			data += sizeof(struct DE_eckd_data);
+		}
+		ccw[-1].flags |= CCW_FLAG_CC;
+		locate_record(ccw++, (struct LO_eckd_data *) data,
+			      fdata->start_unit, 0, 1,
+			      DASD_ECKD_CCW_WRITE_CKD, base, 8);
+		data += sizeof(struct LO_eckd_data);
+		break;
+	}
+
+	for (j = 0; j < nr_tracks; j++) {
+		/* calculate cylinder and head for the current track */
+		set_ch_t(&address,
+			 (fdata->start_unit + j) /
+			 base_priv->rdc_data.trk_per_cyl,
+			 (fdata->start_unit + j) %
+			 base_priv->rdc_data.trk_per_cyl);
+		if (intensity & 0x01) {	/* write record zero */
+			ect = (struct eckd_count *) data;
+			data += sizeof(struct eckd_count);
+			ect->cyl = address.cyl;
+			ect->head = address.head;
+			ect->record = 0;
+			ect->kl = 0;
+			ect->dl = 8;
+			ccw[-1].flags |= CCW_FLAG_CC;
+			ccw->cmd_code = DASD_ECKD_CCW_WRITE_RECORD_ZERO;
+			ccw->flags = CCW_FLAG_SLI;
+			ccw->count = 8;
+			ccw->cda = virt_to_dma32(ect);
+			ccw++;
+		}
+		if ((intensity & ~0x08) & 0x04) {	/* erase track */
+			ect = (struct eckd_count *) data;
+			data += sizeof(struct eckd_count);
+			ect->cyl = address.cyl;
+			ect->head = address.head;
+			ect->record = 1;
+			ect->kl = 0;
+			ect->dl = 0;
+			ccw[-1].flags |= CCW_FLAG_CC;
+			ccw->cmd_code = DASD_ECKD_CCW_WRITE_CKD;
+			ccw->flags = CCW_FLAG_SLI;
+			ccw->count = 8;
+			ccw->cda = virt_to_dma32(ect);
+		} else {		/* write remaining records */
+			for (i = 0; i < rpt; i++) {
+				ect = (struct eckd_count *) data;
+				data += sizeof(struct eckd_count);
+				ect->cyl = address.cyl;
+				ect->head = address.head;
+				ect->record = i + 1;
+				ect->kl = 0;
+				ect->dl = fdata->blksize;
+				/*
+				 * Check for special tracks 0-1
+				 * when formatting CDL
+				 */
+				if ((intensity & 0x08) &&
+				    address.cyl == 0 && address.head == 0) {
+					if (i < 3) {
+						ect->kl = 4;
+						ect->dl = sizes_trk0[i] - 4;
+					}
+				}
+				if ((intensity & 0x08) &&
+				    address.cyl == 0 && address.head == 1) {
+					ect->kl = 44;
+					ect->dl = LABEL_SIZE - 44;
+				}
+				ccw[-1].flags |= CCW_FLAG_CC;
+				if (i != 0 || j == 0)
+					ccw->cmd_code =
+						DASD_ECKD_CCW_WRITE_CKD;
+				else
+					ccw->cmd_code =
+						DASD_ECKD_CCW_WRITE_CKD_MT;
+				ccw->flags = CCW_FLAG_SLI;
+				if (write_label && address.cyl == 0 &&
+				    address.head == 0 && i == 3) {
+					/*
+					 * R4 carries the label as its record
+					 * data; it follows ect contiguously so
+					 * the CCW transfers count + data.
+					 */
+					dasd_eckd_fill_format_label(base,
+								    data,
+								    fdata->blksize);
+					data += fdata->blksize;
+					ccw->count = 8 + fdata->blksize;
+				} else {
+					ccw->count = 8;
+				}
+				ccw->cda = virt_to_dma32(ect);
+				ccw++;
+			}
+		}
+	}
+
+	fcp->startdev = startdev;
+	fcp->memdev = startdev;
+	fcp->basedev = base;
+	fcp->retries = 256;
+	fcp->expires = startdev->default_expires * HZ;
+	fcp->buildclk = get_tod_clock();
+	fcp->status = DASD_CQR_FILLED;
+
+	return fcp;
+}
+
+/*
+ * Wrapper function to build a CCW request depending on input data
+ */
+static struct dasd_ccw_req *
+dasd_eckd_format_build_ccw_req(struct dasd_device *base,
+			       struct format_data_t *fdata, int enable_pav,
+			       int tpm, struct eckd_count *fmt_buffer, int rpt)
+{
+	struct dasd_ccw_req *ccw_req;
+
+	if (!fmt_buffer) {
+		ccw_req = dasd_eckd_build_format(base, NULL, fdata, enable_pav);
+	} else {
+		if (tpm)
+			ccw_req = dasd_eckd_build_check_tcw(base, fdata,
+							    enable_pav,
+							    fmt_buffer, rpt);
+		else
+			ccw_req = dasd_eckd_build_check(base, fdata, enable_pav,
+							fmt_buffer, rpt);
+	}
+
+	return ccw_req;
+}
+
+/*
+ * Sanity checks on format_data
+ */
+static int dasd_eckd_format_sanity_checks(struct dasd_device *base,
+					  struct format_data_t *fdata)
+{
+	struct dasd_eckd_private *private = base->private;
+
+	if (fdata->start_unit >=
+	    (private->real_cyl * private->rdc_data.trk_per_cyl)) {
+		dev_warn(&base->cdev->dev,
+			 "Start track number %u used in formatting is too big\n",
+			 fdata->start_unit);
+		return -EINVAL;
+	}
+	if (fdata->stop_unit >=
+	    (private->real_cyl * private->rdc_data.trk_per_cyl)) {
+		dev_warn(&base->cdev->dev,
+			 "Stop track number %u used in formatting is too big\n",
+			 fdata->stop_unit);
+		return -EINVAL;
+	}
+	if (fdata->start_unit > fdata->stop_unit) {
+		dev_warn(&base->cdev->dev,
+			 "Start track %u used in formatting exceeds end track\n",
+			 fdata->start_unit);
+		return -EINVAL;
+	}
+	if (dasd_check_blocksize(fdata->blksize) != 0) {
+		dev_warn(&base->cdev->dev,
+			 "The DASD cannot be formatted with block size %u\n",
+			 fdata->blksize);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/*
+ * This function will process format_data originally coming from an IOCTL
+ */
+static int dasd_eckd_format_process_data(struct dasd_device *base,
+					 struct format_data_t *fdata,
+					 int enable_pav, int tpm,
+					 struct eckd_count *fmt_buffer, int rpt,
+					 struct irb *irb)
+{
+	struct dasd_eckd_private *private = base->private;
+	struct dasd_ccw_req *cqr, *n;
+	struct list_head format_queue;
+	struct dasd_device *device;
+	char *sense = NULL;
+	int old_start, old_stop, format_step;
+	int step, retry;
+	int rc;
+
+	rc = dasd_eckd_format_sanity_checks(base, fdata);
+	if (rc)
+		return rc;
+
+	INIT_LIST_HEAD(&format_queue);
+
+	old_start = fdata->start_unit;
+	old_stop = fdata->stop_unit;
+
+	if (!tpm && fmt_buffer != NULL) {
+		/* Command Mode / Format Check */
+		format_step = 1;
+	} else if (tpm && fmt_buffer != NULL) {
+		/* Transport Mode / Format Check */
+		format_step = DASD_CQR_MAX_CCW / rpt;
+	} else {
+		/* Normal Formatting */
+		format_step = DASD_CQR_MAX_CCW /
+			recs_per_track(&private->rdc_data, 0, fdata->blksize);
+	}
+
+	do {
+		retry = 0;
+		while (fdata->start_unit <= old_stop) {
+			step = fdata->stop_unit - fdata->start_unit + 1;
+			if (step > format_step) {
+				fdata->stop_unit =
+					fdata->start_unit + format_step - 1;
+			}
+
+			cqr = dasd_eckd_format_build_ccw_req(base, fdata,
+							     enable_pav, tpm,
+							     fmt_buffer, rpt);
+			if (IS_ERR(cqr)) {
+				rc = PTR_ERR(cqr);
+				if (rc == -ENOMEM) {
+					if (list_empty(&format_queue))
+						goto out;
+					/*
+					 * not enough memory available, start
+					 * requests retry after first requests
+					 * were finished
+					 */
+					retry = 1;
+					break;
+				}
+				goto out_err;
+			}
+			list_add_tail(&cqr->blocklist, &format_queue);
+
+			if (fmt_buffer) {
+				step = fdata->stop_unit - fdata->start_unit + 1;
+				fmt_buffer += rpt * step;
+			}
+			fdata->start_unit = fdata->stop_unit + 1;
+			fdata->stop_unit = old_stop;
+		}
+
+		rc = dasd_sleep_on_queue(&format_queue);
+
+out_err:
+		list_for_each_entry_safe(cqr, n, &format_queue, blocklist) {
+			device = cqr->startdev;
+			private = device->private;
+
+			if (cqr->status == DASD_CQR_FAILED) {
+				/*
+				 * Only get sense data if called by format
+				 * check
+				 */
+				if (fmt_buffer && irb) {
+					sense = dasd_get_sense(&cqr->irb);
+					memcpy(irb, &cqr->irb, sizeof(*irb));
+				}
+				rc = -EIO;
+			}
+			list_del_init(&cqr->blocklist);
+			dasd_ffree_request(cqr, device);
+			private->count--;
+		}
+
+		if (rc && rc != -EIO)
+			goto out;
+		if (rc == -EIO) {
+			/*
+			 * In case fewer than the expected records are on the
+			 * track, we will most likely get a 'No Record Found'
+			 * error (in command mode) or a 'File Protected' error
+			 * (in transport mode). Those particular cases shouldn't
+			 * pass the -EIO to the IOCTL, therefore reset the rc
+			 * and continue.
+			 */
+			if (sense &&
+			    (sense[1] & SNS1_NO_REC_FOUND ||
+			     sense[1] & SNS1_FILE_PROTECTED))
+				retry = 1;
+			else
+				goto out;
+		}
+
+	} while (retry);
+
+out:
+	fdata->start_unit = old_start;
+	fdata->stop_unit = old_stop;
+
+	return rc;
+}
+
+static int dasd_eckd_format_device(struct dasd_device *base,
+				   struct format_data_t *fdata, int enable_pav)
+{
+	struct dasd_eckd_private *private = base->private;
+	int rc;
+
+	rc = dasd_eckd_format_process_data(base, fdata, enable_pav, 0, NULL,
+					   0, NULL);
+	if (fdata->start_unit != 0)
+		return rc;
+
+	if (rc) {
+		/*
+		 * The format failed, so the label cached speculatively during
+		 * CCW build may not match the disk; drop it so is_ese() falls
+		 * back to the hardware field until the next successful format
+		 * or bring-up.
+		 */
+		private->ese_label_valid = false;
+		return rc;
+	}
+
+	/*
+	 * The quick-format indicator was consumed by the label stamped into
+	 * track 0; clear it so a later format that is not preceded by a full
+	 * space release is recorded as a full format.
+	 */
+	private->ese_format_quick = 0;
+
+	/*
+	 * A full format returns the device to a fully sparse state and has just
+	 * committed a fresh label; restart the heuristic from ft1 on the now
+	 * current is_ese state, without an offline cycle.
+	 */
+	dasd_ft_bias_apply(base);
+
+	return rc;
+}
+
+static bool test_and_set_format_track(sector_t start, sector_t end,
+				      struct dasd_ccw_req *cqr,
+				      struct dasd_block *block,
+				      struct dasd_device *device,
+				      struct dasd_format_entry **entry)
+{
+	struct dasd_format_entry *to_format, *format;
+	unsigned long flags;
+	bool rc = false;
+	int i = 0;
+
+	/* marked as a collision by dasd_return_cqr_cb last round: retry */
+	if (cqr && READ_ONCE(cqr->collision)) {
+		WRITE_ONCE(cqr->collision, false);
+		return true;
+	}
+	spin_lock_irqsave(&block->format_lock, flags);
+	while (i < DASD_NR_FORMAT_ENTRIES &&
+	       READ_ONCE(device->format_entry[i].cqr))
+		i++;
+
+	if (i >= DASD_NR_FORMAT_ENTRIES) {
+		rc = true;
+		goto out;
+	}
+
+	list_for_each_entry(format, &block->format_list, list) {
+		if (!(end < format->start_trk || format->end_trk < start)) {
+			rc = true;
+			goto out;
+		}
+	}
+	to_format = &device->format_entry[i];
+	to_format->start_trk = start;
+	to_format->end_trk = end;
+	to_format->cqr = cqr;
+	list_add_tail(&to_format->list, &block->format_list);
+	*entry = to_format;
+
+out:
+	spin_unlock_irqrestore(&block->format_lock, flags);
+	return rc;
+}
+
+static void clear_format_track(struct dasd_format_entry *format,
+			       struct dasd_block *block)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&block->format_lock, flags);
+	list_del_init(&format->list);
+	format->cqr = NULL;
+	spin_unlock_irqrestore(&block->format_lock, flags);
+}
+
+/*
+ * Adaptive ft_bias heuristic, called once per IO from dasd_eckd_build_cp().
+ * Probes the device formatting state by briefly switching to ft0 and measuring
+ * the NRF rate; parameters are derived from ft_bias.
+ */
+static void dasd_ese_heuristic_tick(struct dasd_device *basedev)
+{
+	int ios, nrf, rate;
+
+	if (atomic_inc_return(&basedev->ese_io_cnt) < (int)basedev->ese_probe_interval)
+		return;
+
+	/*
+	 * One wins the race to evaluate, the rest see ios == 0 after the
+	 * xchg and return early, preventing redundant state transitions.
+	 */
+	ios = atomic_xchg(&basedev->ese_io_cnt, 0);
+	if (ios <= 0)
+		return;
+
+	switch (basedev->ese_probe_state) {
+	case DASD_ESE_HEU_FT1_ACTIVE:
+		/* Start ft0 probe window, reset NRF counter for clean measurement */
+		basedev->fulltrack          = 0;
+		basedev->ese_probe_state    = DASD_ESE_HEU_PROBING;
+		basedev->ese_probe_interval = basedev->ese_heu_probe_window;
+		atomic_set(&basedev->ese_nrf_window, 0);
+		break;
+
+	case DASD_ESE_HEU_PROBING:
+	case DASD_ESE_HEU_FT0_STABLE:
+		nrf  = atomic_xchg(&basedev->ese_nrf_window, 0);
+		rate = (int)((u64)nrf * 1000 / ios);
+		if (rate > (int)basedev->ese_heu_nrf_high) {
+			/* NRF rate high: device still sparse, ft1 is better */
+			basedev->fulltrack          = 1;
+			basedev->ese_probe_state    = DASD_ESE_HEU_FT1_ACTIVE;
+			basedev->ese_probe_interval = basedev->ese_heu_start_interval;
+		} else if (basedev->ese_probe_state == DASD_ESE_HEU_PROBING) {
+			/*
+			 * NRF rate low: device mostly formatted, ft0 is faster.
+			 * Re-probe frequently at first, then back off below.
+			 */
+			basedev->fulltrack          = 0;
+			basedev->ese_probe_state    = DASD_ESE_HEU_FT0_STABLE;
+			basedev->ese_probe_interval = basedev->ese_heu_probe_window;
+		} else {
+			/*
+			 * Still stable in ft0: re-assert plain-write mode so a
+			 * fulltrack value left behind by a racing sysfs write
+			 * self-corrects, and back off the re-probe interval
+			 * (double it, capped at max_interval) so a long-lived
+			 * formatted device is not probed more often than needed.
+			 */
+			basedev->fulltrack          = 0;
+			basedev->ese_probe_interval =
+				min(basedev->ese_probe_interval * 2,
+				    basedev->ese_heu_max_interval);
+		}
+		break;
+	}
+}
+
+static void dasd_eckd_ese_format(struct dasd_device *startdev, struct dasd_ccw_req *cqr,
+				 struct irb *irb)
+{
+	struct dasd_format_entry *format = NULL;
+	unsigned int first_offs, last_offs;
+	struct dasd_eckd_private *private;
+	struct dasd_ccw_req *base_cqr;
+	sector_t first_rec, last_rec;
+	sector_t first_trk, last_trk;
+	unsigned int proc_bytes = 0;
+	struct dasd_ccw_req *fcqr;
+	unsigned int recs_per_trk;
+	struct dasd_device *base;
+	struct dasd_block *block;
+	unsigned int blksize;
+	struct request *req;
+	sector_t curr_trk;
+	unsigned int diff;
+	int rc;
+
+	req = dasd_get_callback_data(cqr);
+	block = cqr->block;
+	base = block->base;
+	private = base->private;
+	if (dasd_ese_adaptive(base))
+		atomic_inc(&base->ese_nrf_window);
+	blksize = block->bp_block;
+	recs_per_trk = recs_per_track(&private->rdc_data, 0, blksize);
+
+	/* Calculate record id of first and last block. */
+	first_rec = blk_rq_pos(req) >> block->s2b_shift;
+	first_trk = first_rec;
+	first_offs = sector_div(first_trk, recs_per_trk);
+	last_rec = (blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift;
+	last_trk = last_rec;
+	last_offs = sector_div(last_trk, recs_per_trk);
+
+	/*
+	 * detect if some data has already been processed and the unformatted track is
+	 * within the request.
+	 * If so, finish the request first with the already processed bytes and let the
+	 * blocklayer only redrive unformatted part.
+	 * With this we ensure that there is no overlap of existing data with unformatted
+	 * zero blocks
+	 */
+	rc = dasd_eckd_track_from_irb(irb, base, &curr_trk);
+	if (rc) {
+		/* sense data could not be parsed - this will not resolve by retrying */
+		cqr->status = DASD_CQR_ERROR;
+		goto out;
+	}
+	if (curr_trk >= (sector_t)private->real_cyl * private->rdc_data.trk_per_cyl) {
+		DBF_DEV_EVENT(DBF_WARNING, startdev,
+			      "ESE error track %llu exceeds device geometry\n",
+			      curr_trk);
+		cqr->status = DASD_CQR_ERROR;
+		goto out;
+	}
+	if (curr_trk < first_trk || curr_trk > last_trk) {
+		DBF_DEV_EVENT(DBF_WARNING, startdev,
+			      "ESE error track %llu not within range %llu - %llu\n",
+			      curr_trk, first_trk, last_trk);
+		cqr->status = DASD_CQR_ERROR;
+		goto out;
+	}
+	if (curr_trk != first_trk) {
+		proc_bytes = ((curr_trk - first_trk) * recs_per_trk - first_offs) * blksize;
+		cqr->proc_bytes = proc_bytes;
+		cqr->status = DASD_CQR_SUCCESS;
+		cqr->stopclk = get_tod_clock();
+		goto out;
+	}
+
+	/*
+	 * If there are multiple tracks to be format-written, we can not write
+	 * the partial last track since we do not know if it is already formatted
+	 * or not so skip the partial last track for now. Return the partial
+	 * completion to blocklayer and let it redo the remainder
+	 */
+	if (first_trk != last_trk && last_offs + 1 < recs_per_trk) {
+		diff = last_offs + 1;
+		last_rec = last_rec - diff;
+		last_trk = last_rec;
+		last_offs = sector_div(last_trk, recs_per_trk);
+		proc_bytes = (last_rec - first_rec + 1) * blksize;
+	}
+	if (first_offs > 0 || last_offs + 1 < recs_per_trk) {
+		/* test if tracks are already in formatting by another thread */
+		if (test_and_set_format_track(first_trk, last_trk, cqr,
+					      cqr->block, cqr->startdev, &format)) {
+			/* this is no real error so do not count down retries */
+			cqr->retries++;
+			goto out_retry;
+		}
+	}
+
+	fcqr = dasd_eckd_build_cp_tpm_writefulltrack(startdev, block, req,
+						     first_rec, last_rec,
+						     first_trk, last_trk,
+						     first_offs, last_offs,
+						     recs_per_trk, blksize, cqr);
+	if (IS_ERR(fcqr)) {
+		if (format)
+			clear_format_track(format, cqr->block);
+		if (PTR_ERR(fcqr) == -EINVAL) {
+			/* permanent build failure - fail instead of retrying */
+			cqr->status = DASD_CQR_ERROR;
+			goto out;
+		}
+		/*
+		 * Transient conditions - the XRC clock is not in sync (-EAGAIN)
+		 * or the format request pool is momentarily exhausted under load
+		 * (-ENOMEM). Retry the origin without counting down its retries.
+		 */
+		if (PTR_ERR(fcqr) == -EAGAIN || PTR_ERR(fcqr) == -ENOMEM)
+			cqr->retries++;
+		goto out_retry;
+	}
+
+	if (format) {
+		/* occupancy marker; the free-slot scan reads it with READ_ONCE */
+		WRITE_ONCE(format->cqr, fcqr);
+		fcqr->format = format;
+	}
+
+	/*
+	 * cqr may be an ERP request; dq and the owning request are only set on
+	 * the base request at the end of the ERP chain, so copy from there.
+	 */
+	base_cqr = cqr;
+	while (base_cqr->refers)
+		base_cqr = base_cqr->refers;
+	fcqr->dq = base_cqr->dq;
+	fcqr->callback_data = base_cqr->callback_data;
+	if (proc_bytes)
+		fcqr->proc_bytes = proc_bytes;
+	fcqr->status = DASD_CQR_FILLED;
+	((struct dasd_eckd_private *)fcqr->memdev->private)->count++;
+	/*
+	 * stage under ese_lock; dasd_block_tasklet splices it into ccw_queue.
+	 * Direct enqueue here would invert queue_lock / ccwdev_lock.
+	 */
+	spin_lock(&block->ese_lock);
+	list_add(&fcqr->blocklist, &block->ese_staging);
+	spin_unlock(&block->ese_lock);
+	/* mark origin CQR as aborted; ccwdev_lock is held by the IRQ handler */
+	cqr->status = DASD_CQR_ABORT;
+	goto out;
+
+out_retry:
+	/*
+	 * If we can't format now, let the request go
+	 * one extra round. Maybe we can format later.
+	 * re-queue at the end to let potential format collision finish first
+	 */
+	list_move_tail(&cqr->devlist, &cqr->startdev->ccw_queue);
+	cqr->status = DASD_CQR_QUEUED;
+out:
+	dasd_device_clear_timer(startdev);
+	dasd_schedule_block_bh(block);
+	dasd_schedule_device_bh(startdev);
+	return;
+}
+
+/*
+ * When data is read from an unformatted area of an ESE volume, this function
+ * returns zeroed data and thereby mimics a read of zero data.
+ *
+ * The first unformatted track is the one that got the NRF error, the address is
+ * encoded in the sense data.
+ *
+ * All tracks before have returned valid data and should not be touched.
+ * All tracks after the unformatted track might be formatted or not. This is
+ * currently not known, remember the processed data and return the remainder of
+ * the request to the blocklayer in __dasd_cleanup_cqr().
+ */
+static int dasd_eckd_ese_read(struct dasd_ccw_req *cqr, struct irb *irb)
+{
+	struct dasd_eckd_private *private;
+	sector_t first_trk, last_trk;
+	sector_t first_blk, last_blk;
+	unsigned int blksize, off;
+	unsigned int recs_per_trk;
+	struct dasd_device *base;
+	struct req_iterator iter;
+	struct dasd_block *block;
+	unsigned int skip_block;
+	unsigned int blk_count;
+	struct request *req;
+	struct bio_vec bv;
+	sector_t curr_trk;
+	sector_t end_blk;
+	char *dst;
+	int rc;
+
+	req = (struct request *) cqr->callback_data;
+	base = cqr->block->base;
+	blksize = base->block->bp_block;
+	block =  cqr->block;
+	private = base->private;
+	skip_block = 0;
+	blk_count = 0;
+
+	recs_per_trk = recs_per_track(&private->rdc_data, 0, blksize);
+	first_trk = first_blk = blk_rq_pos(req) >> block->s2b_shift;
+	sector_div(first_trk, recs_per_trk);
+	last_trk = last_blk =
+		(blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift;
+	sector_div(last_trk, recs_per_trk);
+	rc = dasd_eckd_track_from_irb(irb, base, &curr_trk);
+	if (rc)
+		return rc;
+
+	/* sanity check if the current track from sense data is valid */
+	if (curr_trk < first_trk || curr_trk > last_trk) {
+		DBF_DEV_EVENT(DBF_WARNING, base,
+			      "ESE error track %llu not within range %llu - %llu\n",
+			      curr_trk, first_trk, last_trk);
+		return -EINVAL;
+	}
+
+	/*
+	 * if not the first track got the NRF error we have to skip over valid
+	 * blocks
+	 */
+	if (curr_trk != first_trk)
+		skip_block = curr_trk * recs_per_trk - first_blk;
+
+	/* we have no information beyond the current track */
+	end_blk = (curr_trk + 1) * recs_per_trk;
+
+	rq_for_each_segment(bv, req, iter) {
+		dst = bvec_virt(&bv);
+		for (off = 0; off < bv.bv_len; off += blksize) {
+			if (first_blk + blk_count >= end_blk) {
+				cqr->proc_bytes = blk_count * blksize;
+				return 0;
+			}
+			if (dst && !skip_block)
+				memset(dst, 0, blksize);
+			else
+				skip_block--;
+			dst += blksize;
+			blk_count++;
+		}
+	}
+	return 0;
+}
+
+/*
+ * Helper function to count consecutive records of a single track.
+ */
+static int dasd_eckd_count_records(struct eckd_count *fmt_buffer, int start,
+				   int max)
+{
+	int head;
+	int i;
+
+	head = fmt_buffer[start].head;
+
+	/*
+	 * There are 3 conditions where we stop counting:
+	 * - if data reoccurs (same head and record may reoccur), which may
+	 *   happen due to the way DASD_ECKD_CCW_READ_COUNT works
+	 * - when the head changes, because we're iterating over several tracks
+	 *   then (DASD_ECKD_CCW_READ_COUNT_MT)
+	 * - when we've reached the end of sensible data in the buffer (the
+	 *   record will be 0 then)
+	 */
+	for (i = start; i < max; i++) {
+		if (i > start) {
+			if ((fmt_buffer[i].head == head &&
+			    fmt_buffer[i].record == 1) ||
+			    fmt_buffer[i].head != head ||
+			    fmt_buffer[i].record == 0)
+				break;
+		}
+	}
+
+	return i - start;
+}
+
+/*
+ * Evaluate a given range of tracks. Data like number of records, blocksize,
+ * record ids, and key length are compared with expected data.
+ *
+ * If a mismatch occurs, the corresponding error bit is set, as well as
+ * additional information, depending on the error.
+ */
+static void dasd_eckd_format_evaluate_tracks(struct eckd_count *fmt_buffer,
+					     struct format_check_t *cdata,
+					     int rpt_max, int rpt_exp,
+					     int trk_per_cyl, int tpm)
+{
+	struct ch_t geo;
+	int max_entries;
+	int count = 0;
+	int trkcount;
+	int blksize;
+	int pos = 0;
+	int i, j;
+	int kl;
+
+	trkcount = cdata->expect.stop_unit - cdata->expect.start_unit + 1;
+	max_entries = trkcount * rpt_max;
+
+	for (i = cdata->expect.start_unit; i <= cdata->expect.stop_unit; i++) {
+		/* Calculate the correct next starting position in the buffer */
+		if (tpm) {
+			while (fmt_buffer[pos].record == 0 &&
+			       fmt_buffer[pos].dl == 0) {
+				if (pos++ > max_entries)
+					break;
+			}
+		} else {
+			if (i != cdata->expect.start_unit)
+				pos += rpt_max - count;
+		}
+
+		/* Calculate the expected geo values for the current track */
+		set_ch_t(&geo, i / trk_per_cyl, i % trk_per_cyl);
+
+		/* Count and check number of records */
+		count = dasd_eckd_count_records(fmt_buffer, pos, pos + rpt_max);
+
+		if (count < rpt_exp) {
+			cdata->result = DASD_FMT_ERR_TOO_FEW_RECORDS;
+			break;
+		}
+		if (count > rpt_exp) {
+			cdata->result = DASD_FMT_ERR_TOO_MANY_RECORDS;
+			break;
+		}
+
+		for (j = 0; j < count; j++, pos++) {
+			blksize = cdata->expect.blksize;
+			kl = 0;
+
+			/*
+			 * Set special values when checking CDL formatted
+			 * devices.
+			 */
+			if ((cdata->expect.intensity & 0x08) &&
+			    geo.cyl == 0 && geo.head == 0) {
+				if (j < 3) {
+					blksize = sizes_trk0[j] - 4;
+					kl = 4;
+				}
+			}
+			if ((cdata->expect.intensity & 0x08) &&
+			    geo.cyl == 0 && geo.head == 1) {
+				blksize = LABEL_SIZE - 44;
+				kl = 44;
+			}
+
+			/* Check blocksize */
+			if (fmt_buffer[pos].dl != blksize) {
+				cdata->result = DASD_FMT_ERR_BLKSIZE;
+				goto out;
+			}
+			/* Check if key length is 0 */
+			if (fmt_buffer[pos].kl != kl) {
+				cdata->result = DASD_FMT_ERR_KEY_LENGTH;
+				goto out;
+			}
+			/* Check if record_id is correct */
+			if (fmt_buffer[pos].cyl != geo.cyl ||
+			    fmt_buffer[pos].head != geo.head ||
+			    fmt_buffer[pos].record != (j + 1)) {
+				cdata->result = DASD_FMT_ERR_RECORD_ID;
+				goto out;
+			}
+		}
+	}
+
+out:
+	/*
+	 * In case of no errors, we need to decrease by one
+	 * to get the correct positions.
+	 */
+	if (!cdata->result) {
+		i--;
+		pos--;
+	}
+
+	cdata->unit = i;
+	cdata->num_records = count;
+	cdata->rec = fmt_buffer[pos].record;
+	cdata->blksize = fmt_buffer[pos].dl;
+	cdata->key_length = fmt_buffer[pos].kl;
+}
+
+/*
+ * Check the format of a range of tracks of a DASD.
+ */
+static int dasd_eckd_check_device_format(struct dasd_device *base,
+					 struct format_check_t *cdata,
+					 int enable_pav)
+{
+	struct dasd_eckd_private *private = base->private;
+	struct eckd_count *fmt_buffer;
+	size_t fmt_buffer_size;
+	unsigned int trkcount;
+	int rpt_max, rpt_exp;
+	struct irb irb;
+	int trk_per_cyl;
+	int tpm = 0;
+	int rc;
+
+	trk_per_cyl = private->rdc_data.trk_per_cyl;
+
+	/* Get maximum and expected amount of records per track */
+	rpt_max = recs_per_track(&private->rdc_data, 0, 512) + 1;
+	rpt_exp = recs_per_track(&private->rdc_data, 0, cdata->expect.blksize);
+
+	trkcount = cdata->expect.stop_unit - cdata->expect.start_unit + 1;
+	if (check_mul_overflow(trkcount, rpt_max, &fmt_buffer_size) ||
+	    check_mul_overflow(fmt_buffer_size, sizeof(struct eckd_count), &fmt_buffer_size))
+		return -EINVAL;
+
+	fmt_buffer = kzalloc(fmt_buffer_size, GFP_KERNEL | GFP_DMA);
+	if (!fmt_buffer)
+		return -ENOMEM;
+
+	/*
+	 * A certain FICON feature subset is needed to operate in transport
+	 * mode. Additionally, the support for transport mode is implicitly
+	 * checked by comparing the buffer size with fcx_max_data. As long as
+	 * the buffer size is smaller we can operate in transport mode and
+	 * process multiple tracks. If not, only one track at once is being
+	 * processed using command mode.
+	 */
+	if ((private->features.feature[40] & 0x04) &&
+	    fmt_buffer_size <= private->fcx_max_data)
+		tpm = 1;
+
+	rc = dasd_eckd_format_process_data(base, &cdata->expect, enable_pav,
+					   tpm, fmt_buffer, rpt_max, &irb);
+	if (rc && rc != -EIO)
+		goto out;
+	if (rc == -EIO) {
+		/*
+		 * If our first attempt with transport mode enabled comes back
+		 * with an incorrect length error, we're going to retry the
+		 * check with command mode.
+		 */
+		if (tpm && scsw_cstat(&irb.scsw) == 0x40) {
+			tpm = 0;
+			rc = dasd_eckd_format_process_data(base, &cdata->expect,
+							   enable_pav, tpm,
+							   fmt_buffer, rpt_max,
+							   &irb);
+			if (rc)
+				goto out;
+		} else {
+			goto out;
+		}
+	}
+
+	dasd_eckd_format_evaluate_tracks(fmt_buffer, cdata, rpt_max, rpt_exp,
+					 trk_per_cyl, tpm);
+
+out:
+	kfree(fmt_buffer);
+
+	return rc;
+}
+
+static void dasd_eckd_handle_terminated_request(struct dasd_ccw_req *cqr)
+{
+	if (cqr->retries < 0) {
+		cqr->status = DASD_CQR_FAILED;
+		return;
+	}
+	cqr->status = DASD_CQR_FILLED;
+	if (cqr->block && (cqr->startdev != cqr->block->base)) {
+		dasd_eckd_reset_ccw_to_base_io(cqr);
+		cqr->startdev = cqr->block->base;
+		cqr->lpm = dasd_path_get_opm(cqr->block->base);
+	}
+};
+
+static dasd_erp_fn_t
+dasd_eckd_erp_action(struct dasd_ccw_req * cqr)
+{
+	struct dasd_device *device = (struct dasd_device *) cqr->startdev;
+	struct ccw_device *cdev = device->cdev;
+
+	switch (cdev->id.cu_type) {
+	case 0x3990:
+	case 0x2105:
+	case 0x2107:
+	case 0x1750:
+		return dasd_3990_erp_action;
+	case 0x9343:
+	case 0x3880:
+	default:
+		return dasd_default_erp_action;
+	}
+}
+
+static dasd_erp_fn_t
+dasd_eckd_erp_postaction(struct dasd_ccw_req * cqr)
+{
+	return dasd_default_erp_postaction;
+}
+
+static void dasd_eckd_check_for_device_change(struct dasd_device *device,
+					      struct dasd_ccw_req *cqr,
+					      struct irb *irb)
+{
+	char mask;
+	char *sense = NULL;
+	struct dasd_eckd_private *private = device->private;
+
+	/* first of all check for state change pending interrupt */
+	mask = DEV_STAT_ATTENTION | DEV_STAT_DEV_END | DEV_STAT_UNIT_EXCEP;
+	if ((scsw_dstat(&irb->scsw) & mask) == mask) {
+		/*
+		 * for alias only, not in offline processing
+		 * and only if not suspended
+		 */
+		if (!device->block && private->lcu &&
+		    device->state == DASD_STATE_ONLINE &&
+		    !test_bit(DASD_FLAG_OFFLINE, &device->flags) &&
+		    !test_bit(DASD_FLAG_SUSPENDED, &device->flags)) {
+			/* schedule worker to reload device */
+			dasd_reload_device(device);
+		}
+		dasd_generic_handle_state_change(device);
+		return;
+	}
+
+	sense = dasd_get_sense(irb);
+	if (!sense)
+		return;
+
+	/* summary unit check */
+	if ((sense[27] & DASD_SENSE_BIT_0) && (sense[7] == 0x0D) &&
+	    (scsw_dstat(&irb->scsw) & DEV_STAT_UNIT_CHECK)) {
+		if (test_and_set_bit(DASD_FLAG_SUC, &device->flags)) {
+			DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+				      "eckd suc: device already notified");
+			return;
+		}
+		sense = dasd_get_sense(irb);
+		if (!sense) {
+			DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+				      "eckd suc: no reason code available");
+			clear_bit(DASD_FLAG_SUC, &device->flags);
+			return;
+
+		}
+		private->suc_reason = sense[8];
+		DBF_DEV_EVENT(DBF_NOTICE, device, "%s %x",
+			      "eckd handle summary unit check: reason",
+			      private->suc_reason);
+		dasd_get_device(device);
+		if (!schedule_work(&device->suc_work))
+			dasd_put_device(device);
+
+		return;
+	}
+
+	/* service information message SIM */
+	if (!cqr && !(sense[27] & DASD_SENSE_BIT_0) &&
+	    ((sense[6] & DASD_SIM_SENSE) == DASD_SIM_SENSE)) {
+		dasd_3990_erp_handle_sim(device, sense);
+		return;
+	}
+
+	/* loss of device reservation is handled via base devices only
+	 * as alias devices may be used with several bases
+	 */
+	if (device->block && (sense[27] & DASD_SENSE_BIT_0) &&
+	    (sense[7] == 0x3F) &&
+	    (scsw_dstat(&irb->scsw) & DEV_STAT_UNIT_CHECK) &&
+	    test_bit(DASD_FLAG_IS_RESERVED, &device->flags)) {
+		if (device->features & DASD_FEATURE_FAILONSLCK)
+			set_bit(DASD_FLAG_LOCK_STOLEN, &device->flags);
+		clear_bit(DASD_FLAG_IS_RESERVED, &device->flags);
+		dev_err(&device->cdev->dev,
+			"The device reservation was lost\n");
+	}
+}
+
+static int dasd_eckd_ras_sanity_checks(struct dasd_device *device,
+				       unsigned int first_trk,
+				       unsigned int last_trk)
+{
+	struct dasd_eckd_private *private = device->private;
+	unsigned int trks_per_vol;
+	int rc = 0;
+
+	trks_per_vol = private->real_cyl * private->rdc_data.trk_per_cyl;
+
+	if (first_trk >= trks_per_vol) {
+		dev_warn(&device->cdev->dev,
+			 "Start track number %u used in the space release command is too big\n",
+			 first_trk);
+		rc = -EINVAL;
+	} else if (last_trk >= trks_per_vol) {
+		dev_warn(&device->cdev->dev,
+			 "Stop track number %u used in the space release command is too big\n",
+			 last_trk);
+		rc = -EINVAL;
+	} else if (first_trk > last_trk) {
+		dev_warn(&device->cdev->dev,
+			 "Start track %u used in the space release command exceeds the end track\n",
+			 first_trk);
+		rc = -EINVAL;
+	}
+	return rc;
+}
+
+/*
+ * Number of extents the track range [from, to] spans. Extent n covers tracks
+ * [n * trks_per_ext, (n + 1) * trks_per_ext - 1], so the range touches the
+ * extents from (from / trks_per_ext) to (to / trks_per_ext) inclusive.
+ */
+static int count_exts(unsigned int from, unsigned int to, int trks_per_ext)
+{
+	return to / trks_per_ext - from / trks_per_ext + 1;
+}
+
+static int dasd_in_copy_relation(struct dasd_device *device)
+{
+	struct dasd_pprc_data_sc4 *temp;
+	int rc;
+
+	if (!dasd_eckd_pprc_enabled(device))
+		return 0;
+
+	temp = kzalloc_obj(*temp);
+	if (!temp)
+		return -ENOMEM;
+
+	rc = dasd_eckd_query_pprc_status(device, temp);
+	if (!rc)
+		rc = temp->dev_info[0].state;
+
+	kfree(temp);
+	return rc;
+}
+
+/*
+ * Release allocated space for a given range or an entire volume.
+ */
+static struct dasd_ccw_req *
+dasd_eckd_dso_ras(struct dasd_device *device, struct dasd_block *block,
+		  struct request *req, unsigned int first_trk,
+		  unsigned int last_trk, int by_extent)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct dasd_dso_ras_ext_range *ras_range;
+	struct dasd_rssd_features *features;
+	struct dasd_dso_ras_data *ras_data;
+	u16 heads, beg_head, end_head;
+	int cur_to_trk, cur_from_trk;
+	struct dasd_ccw_req *cqr;
+	u32 beg_cyl, end_cyl;
+	int copy_relation;
+	struct ccw1 *ccw;
+	int trks_per_ext;
+	size_t ras_size;
+	size_t size;
+	int nr_exts;
+	void *rq;
+	int i;
+
+	if (dasd_eckd_ras_sanity_checks(device, first_trk, last_trk))
+		return ERR_PTR(-EINVAL);
+
+	/*
+	 * The block-layer discard path (req != NULL) runs in atomic context, so
+	 * it must not issue the sleeping copy-relation (PPRC) query. It also
+	 * leaves guarantee_init off - discard does not promise zeroing anyway.
+	 */
+	copy_relation = 0;
+	if (!req) {
+		copy_relation = dasd_in_copy_relation(device);
+		if (copy_relation < 0)
+			return ERR_PTR(copy_relation);
+	}
+
+	rq = req ? blk_mq_rq_to_pdu(req) : NULL;
+
+	features = &private->features;
+
+	trks_per_ext = dasd_eckd_ext_size(device) * private->rdc_data.trk_per_cyl;
+	nr_exts = 0;
+	if (by_extent)
+		nr_exts = count_exts(first_trk, last_trk, trks_per_ext);
+	ras_size = sizeof(*ras_data);
+	size = ras_size + (nr_exts * sizeof(*ras_range));
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1, size, device, rq);
+	if (IS_ERR(cqr)) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+				"Could not allocate RAS request");
+		return cqr;
+	}
+
+	ras_data = cqr->data;
+	memset(ras_data, 0, size);
+
+	ras_data->order = DSO_ORDER_RAS;
+	ras_data->flags.vol_type = 0; /* CKD volume */
+	/* Release specified extents or entire volume */
+	ras_data->op_flags.by_extent = by_extent;
+	/*
+	 * This bit guarantees initialisation of tracks within an extent that is
+	 * not fully specified, but is only supported with a certain feature
+	 * subset and for devices not in a copy relation.
+	 */
+	if (!req && features->feature[56] & 0x01 && !copy_relation)
+		ras_data->op_flags.guarantee_init = 1;
+
+	/* cached copies - conf.ned may be freed under us by the reload worker */
+	ras_data->lss = READ_ONCE(private->ned_lss);
+	ras_data->dev_addr = READ_ONCE(private->ned_ua);
+	ras_data->nr_exts = nr_exts;
+
+	if (by_extent) {
+		heads = private->rdc_data.trk_per_cyl;
+		cur_from_trk = first_trk;
+		cur_to_trk = first_trk + trks_per_ext -
+			(first_trk % trks_per_ext) - 1;
+		if (cur_to_trk > last_trk)
+			cur_to_trk = last_trk;
+		ras_range = (struct dasd_dso_ras_ext_range *)(cqr->data + ras_size);
+
+		for (i = 0; i < nr_exts; i++) {
+			beg_cyl = cur_from_trk / heads;
+			beg_head = cur_from_trk % heads;
+			end_cyl = cur_to_trk / heads;
+			end_head = cur_to_trk % heads;
+
+			set_ch_t(&ras_range->beg_ext, beg_cyl, beg_head);
+			set_ch_t(&ras_range->end_ext, end_cyl, end_head);
+
+			cur_from_trk = cur_to_trk + 1;
+			cur_to_trk = cur_from_trk + trks_per_ext - 1;
+			if (cur_to_trk > last_trk)
+				cur_to_trk = last_trk;
+			ras_range++;
+		}
+	}
+
+	ccw = cqr->cpaddr;
+	ccw->cda = virt_to_dma32(cqr->data);
+	ccw->cmd_code = DASD_ECKD_CCW_DSO;
+	ccw->count = size;
+
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = block;
+	cqr->retries = 256;
+	cqr->expires = device->default_expires * HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	return cqr;
+}
+
+static int dasd_eckd_release_space_full(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private;
+	struct dasd_ccw_req *cqr;
+	int rc;
+
+	cqr = dasd_eckd_dso_ras(device, NULL, NULL, 0, 0, 0);
+	if (IS_ERR(cqr))
+		return PTR_ERR(cqr);
+
+	rc = dasd_sleep_on_interruptible(cqr);
+
+	if (!rc) {
+		/*
+		 * Releasing all space (RAS) wipes every track and the device is
+		 * fully sparse again, so restart the heuristic from ft1.
+		 */
+		dasd_ft_bias_apply(device);
+		/*
+		 * A full release is what makes a subsequent format a quick
+		 * (thin) one; remember it so the format label records that.
+		 */
+		private = device->private;
+		private->ese_format_quick = 1;
+	}
+
+	dasd_sfree_request(cqr, cqr->memdev);
+
+	return rc;
+}
+
+static int dasd_eckd_release_space_trks(struct dasd_device *device,
+					unsigned int from, unsigned int to)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct dasd_block *block = device->block;
+	struct dasd_ccw_req *cqr, *n;
+	struct list_head ras_queue;
+	unsigned int device_exts;
+	int trks_per_ext;
+	int stop, step;
+	int cur_pos;
+	int rc = 0;
+	int retry;
+
+	INIT_LIST_HEAD(&ras_queue);
+
+	if (dasd_eckd_ext_size(device) == 0)
+		return -EINVAL;
+
+	device_exts = private->real_cyl / dasd_eckd_ext_size(device);
+	trks_per_ext = dasd_eckd_ext_size(device) * private->rdc_data.trk_per_cyl;
+
+	/* Make sure device limits are not exceeded */
+	step = trks_per_ext * min(device_exts, DASD_ECKD_RAS_EXTS_MAX);
+	cur_pos = from;
+
+	do {
+		retry = 0;
+		while (cur_pos < to) {
+			stop = cur_pos + step -
+				((cur_pos + step) % trks_per_ext) - 1;
+			if (stop > to)
+				stop = to;
+
+			cqr = dasd_eckd_dso_ras(device, NULL, NULL, cur_pos, stop, 1);
+			if (IS_ERR(cqr)) {
+				rc = PTR_ERR(cqr);
+				if (rc == -ENOMEM) {
+					if (list_empty(&ras_queue))
+						goto out;
+					retry = 1;
+					break;
+				}
+				goto err_out;
+			}
+
+			spin_lock_irq(&block->queue_lock);
+			list_add_tail(&cqr->blocklist, &ras_queue);
+			spin_unlock_irq(&block->queue_lock);
+			cur_pos = stop + 1;
+		}
+
+		rc = dasd_sleep_on_queue_interruptible(&ras_queue);
+
+err_out:
+		list_for_each_entry_safe(cqr, n, &ras_queue, blocklist) {
+			device = cqr->startdev;
+			private = device->private;
+
+			spin_lock_irq(&block->queue_lock);
+			list_del_init(&cqr->blocklist);
+			spin_unlock_irq(&block->queue_lock);
+			dasd_sfree_request(cqr, device);
+			private->count--;
+		}
+	} while (retry);
+
+out:
+	return rc;
+}
+
+static int dasd_eckd_release_space(struct dasd_device *device,
+				   struct format_data_t *rdata)
+{
+	/*
+	 * Space release (and thus a quick format) requires real ESE hardware.
+	 * is_ese() may be true from a copied label on non-ESE hardware, so gate
+	 * on the hardware capability, not on is_ese().
+	 */
+	if (!dasd_eckd_ese_capable(device))
+		return -EOPNOTSUPP;
+
+	if (rdata->intensity & DASD_FMT_INT_ESE_FULL)
+		return dasd_eckd_release_space_full(device);
+	else if (rdata->intensity == 0)
+		return dasd_eckd_release_space_trks(device, rdata->start_unit,
+						    rdata->stop_unit);
+	else
+		return -EINVAL;
+}
+
+static struct dasd_ccw_req *dasd_eckd_build_cp_cmd_single(
+					       struct dasd_device *startdev,
+					       struct dasd_block *block,
+					       struct request *req,
+					       sector_t first_rec,
+					       sector_t last_rec,
+					       sector_t first_trk,
+					       sector_t last_trk,
+					       unsigned int first_offs,
+					       unsigned int last_offs,
+					       unsigned int blk_per_trk,
+					       unsigned int blksize)
+{
+	struct dasd_eckd_private *private;
+	dma64_t *idaws;
+	struct LO_eckd_data *LO_data;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	struct req_iterator iter;
+	struct bio_vec bv;
+	char *dst;
+	unsigned int off;
+	int count, cidaw, cplength, datasize;
+	sector_t recid;
+	unsigned char cmd, rcmd;
+	int use_prefix;
+	struct dasd_device *basedev;
+
+	basedev = block->base;
+	private = basedev->private;
+	if (rq_data_dir(req) == READ)
+		cmd = DASD_ECKD_CCW_READ_MT;
+	else if (rq_data_dir(req) == WRITE)
+		cmd = DASD_ECKD_CCW_WRITE_MT;
+	else
+		return ERR_PTR(-EINVAL);
+
+	/* Check struct bio and count the number of blocks for the request. */
+	count = 0;
+	cidaw = 0;
+	rq_for_each_segment(bv, req, iter) {
+		if (bv.bv_len & (blksize - 1))
+			/* Eckd can only do full blocks. */
+			return ERR_PTR(-EINVAL);
+		count += bv.bv_len >> (block->s2b_shift + 9);
+		if (idal_is_needed (page_address(bv.bv_page), bv.bv_len))
+			cidaw += bv.bv_len >> (block->s2b_shift + 9);
+	}
+	/* Paranoia. */
+	if (count != last_rec - first_rec + 1)
+		return ERR_PTR(-EINVAL);
+
+	/* use the prefix command if available */
+	use_prefix = private->features.feature[8] & 0x01;
+	if (use_prefix) {
+		/* 1x prefix + number of blocks */
+		cplength = 2 + count;
+		/* 1x prefix + cidaws*sizeof(long) */
+		datasize = sizeof(struct PFX_eckd_data) +
+			sizeof(struct LO_eckd_data) +
+			cidaw * sizeof(unsigned long);
+	} else {
+		/* 1x define extent + 1x locate record + number of blocks */
+		cplength = 2 + count;
+		/* 1x define extent + 1x locate record + cidaws*sizeof(long) */
+		datasize = sizeof(struct DE_eckd_data) +
+			sizeof(struct LO_eckd_data) +
+			cidaw * sizeof(unsigned long);
+	}
+	/* Find out the number of additional locate record ccws for cdl. */
+	if (private->uses_cdl && first_rec < 2*blk_per_trk) {
+		if (last_rec >= 2*blk_per_trk)
+			count = 2*blk_per_trk - first_rec;
+		cplength += count;
+		datasize += count*sizeof(struct LO_eckd_data);
+	}
+	/* Allocate the ccw request. */
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, cplength, datasize,
+				   startdev, blk_mq_rq_to_pdu(req));
+	if (IS_ERR(cqr))
+		return cqr;
+	ccw = cqr->cpaddr;
+	/* First ccw is define extent or prefix. */
+	if (use_prefix) {
+		if (prefix(ccw++, cqr->data, first_trk,
+			   last_trk, cmd, basedev, startdev) == -EAGAIN) {
+			/* Clock not in sync and XRC is enabled.
+			 * Try again later.
+			 */
+			dasd_sfree_request(cqr, startdev);
+			return ERR_PTR(-EAGAIN);
+		}
+		idaws = (dma64_t *)(cqr->data + sizeof(struct PFX_eckd_data));
+	} else {
+		if (define_extent(ccw++, cqr->data, first_trk,
+				  last_trk, cmd, basedev, 0) == -EAGAIN) {
+			/* Clock not in sync and XRC is enabled.
+			 * Try again later.
+			 */
+			dasd_sfree_request(cqr, startdev);
+			return ERR_PTR(-EAGAIN);
+		}
+		idaws = (dma64_t *)(cqr->data + sizeof(struct DE_eckd_data));
+	}
+	/* Build locate_record+read/write/ccws. */
+	LO_data = (struct LO_eckd_data *) (idaws + cidaw);
+	recid = first_rec;
+	if (private->uses_cdl == 0 || recid > 2*blk_per_trk) {
+		/* Only standard blocks so there is just one locate record. */
+		ccw[-1].flags |= CCW_FLAG_CC;
+		locate_record(ccw++, LO_data++, first_trk, first_offs + 1,
+			      last_rec - recid + 1, cmd, basedev, blksize);
+	}
+	rq_for_each_segment(bv, req, iter) {
+		dst = bvec_virt(&bv);
+		if (dasd_page_cache) {
+			char *copy = kmem_cache_alloc(dasd_page_cache,
+						      GFP_DMA | __GFP_NOWARN);
+			if (copy && rq_data_dir(req) == WRITE)
+				memcpy(copy + bv.bv_offset, dst, bv.bv_len);
+			if (copy)
+				dst = copy + bv.bv_offset;
+		}
+		for (off = 0; off < bv.bv_len; off += blksize) {
+			sector_t trkid = recid;
+			unsigned int recoffs = sector_div(trkid, blk_per_trk);
+			rcmd = cmd;
+			count = blksize;
+			/* Locate record for cdl special block ? */
+			if (private->uses_cdl && recid < 2*blk_per_trk) {
+				if (dasd_eckd_cdl_special(blk_per_trk, recid)){
+					rcmd |= 0x8;
+					count = dasd_eckd_cdl_reclen(recid);
+					if (count < blksize &&
+					    rq_data_dir(req) == READ)
+						memset(dst + count, 0xe5,
+						       blksize - count);
+				}
+				ccw[-1].flags |= CCW_FLAG_CC;
+				locate_record(ccw++, LO_data++,
+					      trkid, recoffs + 1,
+					      1, rcmd, basedev, count);
+			}
+			/* Locate record for standard blocks ? */
+			if (private->uses_cdl && recid == 2*blk_per_trk) {
+				ccw[-1].flags |= CCW_FLAG_CC;
+				locate_record(ccw++, LO_data++,
+					      trkid, recoffs + 1,
+					      last_rec - recid + 1,
+					      cmd, basedev, count);
+			}
+			/* Read/write ccw. */
+			ccw[-1].flags |= CCW_FLAG_CC;
+			ccw->cmd_code = rcmd;
+			ccw->count = count;
+			if (idal_is_needed(dst, blksize)) {
+				ccw->cda = virt_to_dma32(idaws);
+				ccw->flags = CCW_FLAG_IDA;
+				idaws = idal_create_words(idaws, dst, blksize);
+			} else {
+				ccw->cda = virt_to_dma32(dst);
+				ccw->flags = 0;
+			}
+			ccw++;
+			dst += blksize;
+			recid++;
+		}
+	}
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->startdev = startdev;
+	cqr->memdev = startdev;
+	cqr->block = block;
+	cqr->expires = startdev->default_expires * HZ;	/* default 5 minutes */
+	cqr->lpm = dasd_path_get_ppm(startdev);
+	cqr->retries = startdev->default_retries;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	/* Set flags to suppress output for expected errors */
+	if (dasd_eckd_is_ese(basedev)) {
+		set_bit(DASD_CQR_SUPPRESS_NRF, &cqr->flags);
+	}
+
+	return cqr;
+}
+
+static struct dasd_ccw_req *dasd_eckd_build_cp_cmd_track(
+					       struct dasd_device *startdev,
+					       struct dasd_block *block,
+					       struct request *req,
+					       sector_t first_rec,
+					       sector_t last_rec,
+					       sector_t first_trk,
+					       sector_t last_trk,
+					       unsigned int first_offs,
+					       unsigned int last_offs,
+					       unsigned int blk_per_trk,
+					       unsigned int blksize)
+{
+	dma64_t *idaws;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	struct req_iterator iter;
+	struct bio_vec bv;
+	char *dst, *idaw_dst;
+	unsigned int cidaw, cplength, datasize;
+	unsigned int tlf;
+	sector_t recid;
+	unsigned char cmd;
+	struct dasd_device *basedev;
+	unsigned int trkcount, count, count_to_trk_end;
+	unsigned int idaw_len, seg_len, part_len, len_to_track_end;
+	unsigned char new_track, end_idaw;
+	sector_t trkid;
+	unsigned int recoffs;
+
+	basedev = block->base;
+	if (rq_data_dir(req) == READ)
+		cmd = DASD_ECKD_CCW_READ_TRACK_DATA;
+	else if (rq_data_dir(req) == WRITE)
+		cmd = DASD_ECKD_CCW_WRITE_TRACK_DATA;
+	else
+		return ERR_PTR(-EINVAL);
+
+	/* Track based I/O needs IDAWs for each page, and not just for
+	 * 64 bit addresses. We need additional idals for pages
+	 * that get filled from two tracks, so we use the number
+	 * of records as upper limit.
+	 */
+	cidaw = last_rec - first_rec + 1;
+	trkcount = last_trk - first_trk + 1;
+
+	/* 1x prefix + one read/write ccw per track */
+	cplength = 1 + trkcount;
+
+	datasize = sizeof(struct PFX_eckd_data) + cidaw * sizeof(unsigned long);
+
+	/* Allocate the ccw request. */
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, cplength, datasize,
+				   startdev, blk_mq_rq_to_pdu(req));
+	if (IS_ERR(cqr))
+		return cqr;
+	ccw = cqr->cpaddr;
+	/* transfer length factor: how many bytes to read from the last track */
+	if (first_trk == last_trk)
+		tlf = last_offs - first_offs + 1;
+	else
+		tlf = last_offs + 1;
+	tlf *= blksize;
+
+	if (prefix_LRE(ccw++, cqr->data, first_trk,
+		       last_trk, cmd, basedev, startdev,
+		       1 /* format */, first_offs + 1,
+		       trkcount, blksize,
+		       tlf) == -EAGAIN) {
+		/* Clock not in sync and XRC is enabled.
+		 * Try again later.
+		 */
+		dasd_sfree_request(cqr, startdev);
+		return ERR_PTR(-EAGAIN);
+	}
+
+	/*
+	 * The translation of request into ccw programs must meet the
+	 * following conditions:
+	 * - all idaws but the first and the last must address full pages
+	 *   (or 2K blocks on 31-bit)
+	 * - the scope of a ccw and it's idal ends with the track boundaries
+	 */
+	idaws = (dma64_t *)(cqr->data + sizeof(struct PFX_eckd_data));
+	recid = first_rec;
+	new_track = 1;
+	end_idaw = 0;
+	len_to_track_end = 0;
+	idaw_dst = NULL;
+	idaw_len = 0;
+	rq_for_each_segment(bv, req, iter) {
+		dst = bvec_virt(&bv);
+		seg_len = bv.bv_len;
+		while (seg_len) {
+			if (new_track) {
+				trkid = recid;
+				recoffs = sector_div(trkid, blk_per_trk);
+				count_to_trk_end = blk_per_trk - recoffs;
+				count = min((last_rec - recid + 1),
+					    (sector_t)count_to_trk_end);
+				len_to_track_end = count * blksize;
+				ccw[-1].flags |= CCW_FLAG_CC;
+				ccw->cmd_code = cmd;
+				ccw->count = len_to_track_end;
+				ccw->cda = virt_to_dma32(idaws);
+				ccw->flags = CCW_FLAG_IDA;
+				ccw++;
+				recid += count;
+				new_track = 0;
+				/* first idaw for a ccw may start anywhere */
+				if (!idaw_dst)
+					idaw_dst = dst;
+			}
+			/* If we start a new idaw, we must make sure that it
+			 * starts on an IDA_BLOCK_SIZE boundary.
+			 * If we continue an idaw, we must make sure that the
+			 * current segment begins where the so far accumulated
+			 * idaw ends
+			 */
+			if (!idaw_dst) {
+				if ((unsigned long)(dst) & (IDA_BLOCK_SIZE - 1)) {
+					dasd_sfree_request(cqr, startdev);
+					return ERR_PTR(-ERANGE);
+				} else
+					idaw_dst = dst;
+			}
+			if ((idaw_dst + idaw_len) != dst) {
+				dasd_sfree_request(cqr, startdev);
+				return ERR_PTR(-ERANGE);
+			}
+			part_len = min(seg_len, len_to_track_end);
+			seg_len -= part_len;
+			dst += part_len;
+			idaw_len += part_len;
+			len_to_track_end -= part_len;
+			/* collected memory area ends on an IDA_BLOCK border,
+			 * -> create an idaw
+			 * idal_create_words will handle cases where idaw_len
+			 * is larger then IDA_BLOCK_SIZE
+			 */
+			if (!((unsigned long)(idaw_dst + idaw_len) & (IDA_BLOCK_SIZE - 1)))
+				end_idaw = 1;
+			/* We also need to end the idaw at track end */
+			if (!len_to_track_end) {
+				new_track = 1;
+				end_idaw = 1;
+			}
+			if (end_idaw) {
+				idaws = idal_create_words(idaws, idaw_dst,
+							  idaw_len);
+				idaw_dst = NULL;
+				idaw_len = 0;
+				end_idaw = 0;
+			}
+		}
+	}
+
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->startdev = startdev;
+	cqr->memdev = startdev;
+	cqr->block = block;
+	cqr->expires = startdev->default_expires * HZ;	/* default 5 minutes */
+	cqr->lpm = dasd_path_get_ppm(startdev);
+	cqr->retries = startdev->default_retries;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	/* Set flags to suppress output for expected errors */
+	if (dasd_eckd_is_ese(basedev))
+		set_bit(DASD_CQR_SUPPRESS_NRF, &cqr->flags);
+
+	return cqr;
+}
+
+static int prepare_itcw(struct itcw *itcw,
+			unsigned int trk, unsigned int totrk, int cmd,
+			struct dasd_device *basedev,
+			struct dasd_device *startdev,
+			unsigned int rec_on_trk, int count,
+			unsigned int blksize,
+			unsigned int total_data_size,
+			unsigned int tlf,
+			unsigned int blk_per_trk)
+{
+	u8 pfxbuf[sizeof(struct PFX_eckd_data) + 2] __aligned(8);
+	struct PFX_eckd_data *pfxdata = (struct PFX_eckd_data *)pfxbuf;
+	struct dasd_eckd_private *basepriv, *startpriv;
+	struct DE_eckd_data *dedata;
+	struct LRE_eckd_data *lredata;
+	struct dcw *dcw;
+	int pfxsize;
+
+	u32 begcyl, endcyl;
+	u16 heads, beghead, endhead;
+	u8 pfx_cmd;
+
+	int rc = 0;
+	int sector = 0;
+	int dn, d;
+
+	pfxsize = sizeof(struct PFX_eckd_data);
+	/* prefix + LRE extended data */
+	if (cmd == DASD_ECKD_CCW_WRITE_FULL_TRACK)
+		pfxsize += 2;
+
+	memset(pfxbuf, 0, pfxsize);
+
+	/* setup prefix data */
+	basepriv = basedev->private;
+	startpriv = startdev->private;
+	dedata = &pfxdata->define_extent;
+	lredata = &pfxdata->locate_record;
+
+	pfxdata->format = 1; /* PFX with LRE */
+	/* cached copies - conf.ned may be freed under us by the reload worker */
+	pfxdata->base_address = READ_ONCE(basepriv->ned_ua);
+	pfxdata->base_lss = READ_ONCE(basepriv->ned_lss);
+	pfxdata->validity.define_extent = 1;
+
+	/* private uid is kept up to date, conf_data may be outdated */
+	if (startpriv->uid.type == UA_BASE_PAV_ALIAS)
+		pfxdata->validity.verify_base = 1;
+
+	if (startpriv->uid.type == UA_HYPER_PAV_ALIAS) {
+		pfxdata->validity.verify_base = 1;
+		pfxdata->validity.hyper_pav = 1;
+	}
+
+	switch (cmd) {
+	case DASD_ECKD_CCW_READ_TRACK_DATA:
+		dedata->mask.perm = 0x1;
+		dedata->attributes.operation = basepriv->attrib.operation;
+		dedata->blk_size = blksize;
+		dedata->ga_extended |= 0x42;
+		lredata->operation.orientation = 0x0;
+		lredata->operation.operation = 0x0C;
+		lredata->auxiliary.check_bytes = 0x01;
+		pfx_cmd = DASD_ECKD_CCW_PFX_READ;
+		break;
+	case DASD_ECKD_CCW_WRITE_TRACK_DATA:
+		dedata->mask.perm = 0x02;
+		dedata->attributes.operation = basepriv->attrib.operation;
+		dedata->blk_size = blksize;
+		rc = set_timestamp(NULL, dedata, basedev);
+		dedata->ga_extended |= 0x42;
+		lredata->operation.orientation = 0x0;
+		lredata->operation.operation = 0x3F;
+		lredata->extended_operation = 0x23;
+		lredata->auxiliary.check_bytes = 0x2;
+		/*
+		 * If XRC is supported the System Time Stamp is set. The
+		 * validity of the time stamp must be reflected in the prefix
+		 * data as well.
+		 */
+		if (dedata->ga_extended & 0x08 && dedata->ga_extended & 0x02)
+			pfxdata->validity.time_stamp = 1; /* 'Time Stamp Valid' */
+		pfx_cmd = DASD_ECKD_CCW_PFX;
+		break;
+	case DASD_ECKD_CCW_WRITE_FULL_TRACK:
+		dedata->mask.perm = 0x3;
+		dedata->mask.auth = 0x00;
+		dedata->attributes.operation = basepriv->attrib.operation;
+		dedata->blk_size = blksize;
+		dedata->ga_extended |= 0x42;
+		rc = set_timestamp(NULL, dedata, basedev);
+		lredata->operation.orientation = 0x0;
+		lredata->operation.operation = 0x3F;
+		lredata->extended_operation = 0x11;
+		lredata->auxiliary.check_bytes = 0x2;
+		lredata->extended_parameter_length = 0x02;
+		if (count > 8) {
+			lredata->extended_parameter[0] = 0xFF;
+			lredata->extended_parameter[1] = 0xFF;
+			lredata->extended_parameter[1] <<= (16 - count);
+		} else {
+			lredata->extended_parameter[0] = 0xFF;
+			lredata->extended_parameter[0] <<= (8 - count);
+			lredata->extended_parameter[1] = 0x00;
+		}
+		sector = 0xFF;
+		/*
+		 * If XRC is supported the System Time Stamp is set. The
+		 * validity of the time stamp must be reflected in the prefix
+		 * data as well.
+		 */
+		if (dedata->ga_extended & 0x08 && dedata->ga_extended & 0x02)
+			pfxdata->validity.time_stamp = 1; /* 'Time Stamp Valid' */
+		pfx_cmd = DASD_ECKD_CCW_PFX;
+		break;
+	case DASD_ECKD_CCW_READ_COUNT_MT:
+		dedata->mask.perm = 0x1;
+		dedata->attributes.operation = DASD_BYPASS_CACHE;
+		dedata->ga_extended |= 0x42;
+		dedata->blk_size = blksize;
+		lredata->operation.orientation = 0x2;
+		lredata->operation.operation = 0x16;
+		lredata->auxiliary.check_bytes = 0x01;
+		pfx_cmd = DASD_ECKD_CCW_PFX_READ;
+		break;
+	default:
+		DBF_DEV_EVENT(DBF_ERR, basedev,
+			      "prepare itcw, unknown opcode 0x%x", cmd);
+		BUG();
+		break;
+	}
+	if (rc)
+		return rc;
+
+	dedata->attributes.mode = 0x3;	/* ECKD */
+
+	heads = basepriv->rdc_data.trk_per_cyl;
+	begcyl = trk / heads;
+	beghead = trk % heads;
+	endcyl = totrk / heads;
+	endhead = totrk % heads;
+
+	/* check for sequential prestage - enhance cylinder range */
+	if (dedata->attributes.operation == DASD_SEQ_PRESTAGE ||
+	    dedata->attributes.operation == DASD_SEQ_ACCESS) {
+
+		if (endcyl + basepriv->attrib.nr_cyl < basepriv->real_cyl)
+			endcyl += basepriv->attrib.nr_cyl;
+		else
+			endcyl = (basepriv->real_cyl - 1);
+	}
+
+	set_ch_t(&dedata->beg_ext, begcyl, beghead);
+	set_ch_t(&dedata->end_ext, endcyl, endhead);
+
+	dedata->ep_format = 0x20; /* records per track is valid */
+	dedata->ep_rec_per_track = blk_per_trk;
+
+	if (rec_on_trk) {
+		switch (basepriv->rdc_data.dev_type) {
+		case 0x3390:
+			dn = ceil_quot(blksize + 6, 232);
+			d = 9 + ceil_quot(blksize + 6 * (dn + 1), 34);
+			sector = (49 + (rec_on_trk - 1) * (10 + d)) / 8;
+			break;
+		case 0x3380:
+			d = 7 + ceil_quot(blksize + 12, 32);
+			sector = (39 + (rec_on_trk - 1) * (8 + d)) / 7;
+			break;
+		}
+	}
+
+	if (cmd == DASD_ECKD_CCW_READ_COUNT_MT) {
+		lredata->auxiliary.length_valid = 0;
+		lredata->auxiliary.length_scope = 0;
+		lredata->sector = 0xff;
+	} else {
+		lredata->auxiliary.length_valid = 1;
+		lredata->auxiliary.length_scope = 1;
+		lredata->sector = sector;
+	}
+	lredata->auxiliary.imbedded_ccw_valid = 1;
+	lredata->length = tlf;
+	lredata->imbedded_ccw = cmd;
+	lredata->count = count;
+	set_ch_t(&lredata->seek_addr, begcyl, beghead);
+	lredata->search_arg.cyl = lredata->seek_addr.cyl;
+	lredata->search_arg.head = lredata->seek_addr.head;
+	lredata->search_arg.record = rec_on_trk;
+
+	dcw = itcw_add_dcw(itcw, pfx_cmd, 0,
+			   pfxdata, pfxsize, total_data_size);
+	return PTR_ERR_OR_ZERO(dcw);
+}
+
+static struct dasd_ccw_req *dasd_eckd_build_cp_tpm_track(
+					       struct dasd_device *startdev,
+					       struct dasd_block *block,
+					       struct request *req,
+					       sector_t first_rec,
+					       sector_t last_rec,
+					       sector_t first_trk,
+					       sector_t last_trk,
+					       unsigned int first_offs,
+					       unsigned int last_offs,
+					       unsigned int blk_per_trk,
+					       unsigned int blksize)
+{
+	struct dasd_ccw_req *cqr;
+	struct req_iterator iter;
+	struct bio_vec bv;
+	char *dst;
+	unsigned int trkcount, ctidaw;
+	unsigned char cmd;
+	struct dasd_device *basedev;
+	unsigned int tlf;
+	struct itcw *itcw;
+	struct tidaw *last_tidaw = NULL;
+	int itcw_op;
+	size_t itcw_size;
+	u8 tidaw_flags;
+	unsigned int seg_len, part_len, len_to_track_end;
+	unsigned char new_track;
+	sector_t recid, trkid;
+	unsigned int offs;
+	unsigned int count, count_to_trk_end;
+	int ret;
+
+	basedev = block->base;
+	if (rq_data_dir(req) == READ) {
+		cmd = DASD_ECKD_CCW_READ_TRACK_DATA;
+		itcw_op = ITCW_OP_READ;
+	} else if (rq_data_dir(req) == WRITE) {
+		cmd = DASD_ECKD_CCW_WRITE_TRACK_DATA;
+		itcw_op = ITCW_OP_WRITE;
+	} else
+		return ERR_PTR(-EINVAL);
+
+	/* trackbased I/O needs address all memory via TIDAWs,
+	 * not just for 64 bit addresses. This allows us to map
+	 * each segment directly to one tidaw.
+	 * In the case of write requests, additional tidaws may
+	 * be needed when a segment crosses a track boundary.
+	 */
+	trkcount = last_trk - first_trk + 1;
+	ctidaw = 0;
+	rq_for_each_segment(bv, req, iter) {
+		++ctidaw;
+	}
+	if (rq_data_dir(req) == WRITE)
+		ctidaw += (last_trk - first_trk);
+
+	/* Allocate the ccw request. */
+	itcw_size = itcw_calc_size(0, ctidaw, 0);
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 0, itcw_size, startdev,
+				   blk_mq_rq_to_pdu(req));
+	if (IS_ERR(cqr))
+		return cqr;
+
+	/* transfer length factor: how many bytes to read from the last track */
+	if (first_trk == last_trk)
+		tlf = last_offs - first_offs + 1;
+	else
+		tlf = last_offs + 1;
+	tlf *= blksize;
+
+	itcw = itcw_init(cqr->data, itcw_size, itcw_op, 0, ctidaw, 0);
+	if (IS_ERR(itcw)) {
+		ret = -EINVAL;
+		goto out_error;
+	}
+	cqr->cpaddr = itcw_get_tcw(itcw);
+	if (prepare_itcw(itcw, first_trk, last_trk,
+			 cmd, basedev, startdev,
+			 first_offs + 1,
+			 trkcount, blksize,
+			 (last_rec - first_rec + 1) * blksize,
+			 tlf, blk_per_trk) == -EAGAIN) {
+		/* Clock not in sync and XRC is enabled.
+		 * Try again later.
+		 */
+		ret = -EAGAIN;
+		goto out_error;
+	}
+	len_to_track_end = 0;
+	/*
+	 * A tidaw can address 4k of memory, but must not cross page boundaries
+	 * We can let the block layer handle this by setting seg_boundary_mask
+	 * to page boundaries and max_segment_size to page size when setting up
+	 * the request queue.
+	 * For write requests, a TIDAW must not cross track boundaries, because
+	 * we have to set the CBC flag on the last tidaw for each track.
+	 */
+	if (rq_data_dir(req) == WRITE) {
+		new_track = 1;
+		recid = first_rec;
+		rq_for_each_segment(bv, req, iter) {
+			dst = bvec_virt(&bv);
+			seg_len = bv.bv_len;
+			while (seg_len) {
+				if (new_track) {
+					trkid = recid;
+					offs = sector_div(trkid, blk_per_trk);
+					count_to_trk_end = blk_per_trk - offs;
+					count = min((last_rec - recid + 1),
+						    (sector_t)count_to_trk_end);
+					len_to_track_end = count * blksize;
+					recid += count;
+					new_track = 0;
+				}
+				part_len = min(seg_len, len_to_track_end);
+				seg_len -= part_len;
+				len_to_track_end -= part_len;
+				/* We need to end the tidaw at track end */
+				if (!len_to_track_end) {
+					new_track = 1;
+					tidaw_flags = TIDAW_FLAGS_INSERT_CBC;
+				} else
+					tidaw_flags = 0;
+				last_tidaw = itcw_add_tidaw(itcw, tidaw_flags,
+							    dst, part_len);
+				if (IS_ERR(last_tidaw)) {
+					ret = -EINVAL;
+					goto out_error;
+				}
+				dst += part_len;
+			}
+		}
+	} else {
+		rq_for_each_segment(bv, req, iter) {
+			dst = bvec_virt(&bv);
+			last_tidaw = itcw_add_tidaw(itcw, 0x00,
+						    dst, bv.bv_len);
+			if (IS_ERR(last_tidaw)) {
+				ret = -EINVAL;
+				goto out_error;
+			}
+		}
+	}
+	last_tidaw->flags |= TIDAW_FLAGS_LAST;
+	last_tidaw->flags &= ~TIDAW_FLAGS_INSERT_CBC;
+	itcw_finalize(itcw);
+
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->cpmode = 1;
+	cqr->startdev = startdev;
+	cqr->memdev = startdev;
+	cqr->block = block;
+	cqr->expires = startdev->default_expires * HZ;	/* default 5 minutes */
+	cqr->lpm = dasd_path_get_ppm(startdev);
+	cqr->retries = startdev->default_retries;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	/* Set flags to suppress output for expected errors */
+	if (dasd_eckd_is_ese(basedev)) {
+		set_bit(DASD_CQR_SUPPRESS_NRF, &cqr->flags);
+		set_bit(DASD_CQR_SUPPRESS_IT, &cqr->flags);
+	}
+
+	return cqr;
+out_error:
+	dasd_sfree_request(cqr, startdev);
+	return ERR_PTR(ret);
+}
+
+static __always_inline bool crosses_page(const void *addr, size_t len)
+{
+	return len && (offset_in_page(addr) + len > PAGE_SIZE);
+}
+
+static __always_inline void *reserve_nocross(char **p, size_t *space, size_t len)
+{
+	size_t pad = crosses_page(*p, len) ? PAGE_SIZE - offset_in_page(*p) : 0;
+	void *ret;
+
+	if (*space < pad + len)
+		return NULL;	/* out of space */
+
+	*p += pad;
+	*space -= pad;
+	ret = *p;
+	*p += len;
+	*space -= len;
+	return ret;
+}
+
+/*
+ * Helpers for dasd_eckd_build_cp_tpm_writefulltrack(): append the TIDAWs for
+ * one track-image element (R0 header, a count + data record, or the trailing
+ * pseudo track end count) to the itcw. Return the last TIDAW, or NULL on failure.
+ */
+static struct tidaw *add_track_r0(struct itcw *itcw, char **fill,
+				  size_t *fillsize, u32 cyl, u16 head)
+{
+	struct tidaw *tidaw;
+	struct eckd_r0 *r0;
+
+	r0 = reserve_nocross(fill, fillsize, sizeof(*r0));
+	if (WARN_ON_ONCE(!r0))
+		return NULL;
+	set_chr_t(r0, cyl, head, 0);
+	r0->count.dl = 8;
+	tidaw = itcw_add_tidaw(itcw, 0, r0, sizeof(*r0));
+	return IS_ERR_OR_NULL(tidaw) ? NULL : tidaw;
+}
+
+static struct tidaw *add_track_record(struct itcw *itcw, char **fill,
+				      size_t *fillsize, u32 cyl, u16 head,
+				      u8 rec, void *data, u32 dl)
+{
+	struct eckd_count *count;
+	struct tidaw *tidaw;
+
+	count = reserve_nocross(fill, fillsize, sizeof(*count));
+	if (WARN_ON_ONCE(!count))
+		return NULL;
+	set_chr_t(count, cyl, head, rec);
+	count->dl = dl;
+	tidaw = itcw_add_tidaw(itcw, 0, count, sizeof(*count));
+	if (IS_ERR_OR_NULL(tidaw))
+		return NULL;
+	tidaw = itcw_add_tidaw(itcw, 0, data, dl);
+	return IS_ERR_OR_NULL(tidaw) ? NULL : tidaw;
+}
+
+static struct tidaw *add_track_end(struct itcw *itcw, char **fill,
+				   size_t *fillsize)
+{
+	struct eckd_count *count;
+	struct tidaw *tidaw;
+
+	count = reserve_nocross(fill, fillsize, sizeof(*count));
+	if (WARN_ON_ONCE(!count))
+		return NULL;
+	count->cyl = 0xffff;
+	count->head = 0xffff;
+	count->dl = 0xffff;
+	count->record = 0xff;
+	count->kl = 0xff;
+	tidaw = itcw_add_tidaw(itcw, TIDAW_FLAGS_INSERT_CBC, count, sizeof(*count));
+	return IS_ERR_OR_NULL(tidaw) ? NULL : tidaw;
+}
+
+static struct dasd_ccw_req *
+dasd_eckd_build_cp_tpm_writefulltrack(struct dasd_device *startdev,
+				      struct dasd_block *block,
+				      struct request *req,
+				      sector_t first_rec,
+				      sector_t last_rec,
+				      sector_t first_trk,
+				      sector_t last_trk,
+				      unsigned int first_offs,
+				      unsigned int last_offs,
+				      unsigned int blk_per_trk,
+				      unsigned int blksize,
+				      struct dasd_ccw_req *ocqr)
+{
+	struct dasd_eckd_private *private = block->base->private;
+	unsigned int seg_len, part_len, len_to_track_end;
+	unsigned int count, count_to_trk_end, offs;
+	unsigned int trkcount, ctidaw, tlf;
+	int itcw_op, rec_count, datasize;
+	struct tidaw *last_tidaw = NULL;
+	sector_t recid, trkid, curr_trk;
+	unsigned char cmd, new_track;
+	struct dasd_device *basedev;
+	size_t itcw_size, fillsize;
+	struct dasd_ccw_req *cqr;
+	struct req_iterator iter;
+	char *dst, *filldata;
+	unsigned long flags;
+	struct itcw *itcw;
+	struct bio_vec bv;
+	int ret = -EINVAL;
+	void *nullrecord;
+	u16 heads, head;
+	u32 cyl;
+	u8 rec;
+
+	basedev = block->base;
+	cmd = DASD_ECKD_CCW_WRITE_FULL_TRACK;
+	itcw_op = ITCW_OP_WRITE;
+
+	/*
+	 * trackbased I/O needs address all memory via TIDAWs,
+	 * not just for 64 bit addresses. This allows us to map
+	 * each segment directly to one tidaw.
+	 * In the case of write requests, additional tidaws may
+	 * be needed when a segment crosses a track boundary.
+	 * Per track we emit one R0 tidaw, two tidaws per record (count field
+	 * plus data - a record never crosses a track or page boundary, as
+	 * part_len is clamped to both blksize and the track end), and one track
+	 * end tidaw: 2 * blk_per_trk + 2.
+	 * Round the +2 up to blk_per_trk-independent headroom via 2 * (blk_per_trk + 2).
+	 */
+	trkcount = last_trk - first_trk + 1;
+	ctidaw = trkcount * 2 * (blk_per_trk + 2);
+
+	/*
+	 * build_cp (ocqr == NULL): the request owns its CCW program - block in
+	 * the pdu, ITCW in ccw_chunks. ese_format (ocqr != NULL): the failing
+	 * origin still owns its pdu, so take the replacement from ese_chunks.
+	 */
+	itcw_size = itcw_calc_size(0, ctidaw, 0);
+	if (ocqr)
+		cqr = dasd_fmalloc_request(DASD_ECKD_MAGIC, 0, itcw_size, startdev);
+	else
+		cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 0, itcw_size, startdev,
+					   blk_mq_rq_to_pdu(req));
+	if (IS_ERR(cqr))
+		return cqr;
+	fillsize = trkcount * (sizeof(struct eckd_r0) +
+			       (sizeof(struct eckd_count) * (blk_per_trk + 2)));
+	/*
+	 * reserve_nocross() pads elements away from page boundaries and draws
+	 * that padding from fillsize; budget one element per page the buffer
+	 * may span so it never runs short.
+	 */
+	fillsize += (fillsize / PAGE_SIZE + 1) * sizeof(struct eckd_r0);
+	spin_lock_irqsave(&startdev->mem_lock, flags);
+	filldata  = dasd_alloc_chunk(&startdev->fill_chunks, fillsize);
+	spin_unlock_irqrestore(&startdev->mem_lock, flags);
+	if (!filldata) {
+		ret = -ENOMEM;
+		goto out_error;
+	}
+	memset(filldata, 0, fillsize);
+	cqr->filldata = filldata;
+
+	nullrecord = startdev->nulldata;
+
+	/* count + data for each record, plus r0 and the pseudo count */
+	tlf = blk_per_trk * (blksize + sizeof(struct eckd_count));
+	tlf += sizeof(struct eckd_r0) + sizeof(struct eckd_count);
+
+	itcw = itcw_init(cqr->data, itcw_size, itcw_op, 0, ctidaw, 0);
+	if (IS_ERR(itcw)) {
+		ret = -EINVAL;
+		goto out_error;
+	}
+	cqr->cpaddr = itcw_get_tcw(itcw);
+	datasize = trkcount * tlf;
+	if (prepare_itcw(itcw, first_trk, last_trk,
+			 cmd, basedev, startdev,
+			 0,
+			 trkcount, blksize,
+			 datasize,
+			 tlf,
+			 blk_per_trk) == -EAGAIN) {
+		/* Clock not in sync and XRC is enabled.
+		 * Try again later.
+		 */
+		ret = -EAGAIN;
+		goto out_error;
+	}
+	heads = private->rdc_data.trk_per_cyl;
+	/*
+	 * A tidaw can address 4k of memory, but must not cross page boundaries
+	 * We can let the block layer handle this by setting seg_boundary_mask
+	 * to page boundaries and max_segment_size to page size when setting up
+	 * the request queue.
+	 */
+	curr_trk = first_trk;
+	recid = first_rec;
+	trkid = recid;
+	offs = sector_div(trkid, blk_per_trk);
+	count = blk_per_trk;
+	len_to_track_end = count * blksize;
+	recid += count - first_offs;
+	new_track = 0;
+
+	/* the R0 header of the first track */
+	cyl = curr_trk / heads;
+	head = curr_trk % heads;
+	last_tidaw = add_track_r0(itcw, &filldata, &fillsize, cyl, head);
+	if (!last_tidaw)
+		goto out_error;
+
+	/* empty records before the first data record */
+	for (int i = 1; i <= first_offs; i++) {
+		len_to_track_end -= blksize;
+		last_tidaw = add_track_record(itcw, &filldata, &fillsize,
+					      cyl, head, i, nullrecord, blksize);
+		if (!last_tidaw)
+			goto out_error;
+	}
+
+	/* process data records */
+	rec = first_offs + 1;
+	rec_count = 0;
+	rq_for_each_segment(bv, req, iter) {
+		dst = bvec_virt(&bv);
+		seg_len = bv.bv_len;
+		while (seg_len) {
+			if (new_track) {
+				trkid = recid;
+				offs = sector_div(trkid, blk_per_trk);
+				count_to_trk_end = blk_per_trk - offs;
+				count = min((last_rec - recid + 1),
+					    (sector_t)count_to_trk_end);
+				/*
+				 * Size to the physical track end: a short last
+				 * track is padded in out_skip, so the track-end
+				 * marker must not be emitted early here.
+				 */
+				len_to_track_end = count_to_trk_end * blksize;
+				recid += count;
+				new_track = 0;
+				/* the R0 header of the next track */
+				cyl = curr_trk / heads;
+				head = curr_trk % heads;
+				last_tidaw = add_track_r0(itcw, &filldata,
+							  &fillsize, cyl, head);
+				if (!last_tidaw)
+					goto out_error;
+				rec = 1;
+			}
+			/*
+			 * One count + data record per block: a bvec segment can
+			 * be up to a page, so clamp to blksize - otherwise the
+			 * count field would describe one oversized record instead
+			 * of several blksize ones for sub-page block sizes.
+			 */
+			part_len = min(seg_len, len_to_track_end);
+			part_len = min(part_len, blksize);
+			seg_len -= part_len;
+			len_to_track_end -= part_len;
+			/*
+			 * This block ends the track; the next one starts a new
+			 * track. The track-end marker emitted below carries the
+			 * CBC flag.
+			 */
+			if (!len_to_track_end)
+				new_track = 1;
+
+			last_tidaw = add_track_record(itcw, &filldata, &fillsize,
+						      cyl, head, rec, dst, part_len);
+			if (!last_tidaw)
+				goto out_error;
+
+			if (new_track) {
+				/* add track end marker */
+				last_tidaw = add_track_end(itcw, &filldata,
+							   &fillsize);
+				if (!last_tidaw)
+					goto out_error;
+				curr_trk++;
+			}
+			rec++;
+			dst += part_len;
+			rec_count++;
+			if (rec_count >= (last_rec - first_rec + 1))
+				goto out_skip;
+		}
+	}
+
+out_skip:
+	new_track = 0;
+	/* empty records after the last data record */
+	for (int i = last_offs + 2; i <= blk_per_trk; i++) {
+		len_to_track_end -= blksize;
+		last_tidaw = add_track_record(itcw, &filldata, &fillsize,
+					      cyl, head, i, nullrecord, blksize);
+		if (!last_tidaw)
+			goto out_error;
+		new_track = 1;
+	}
+
+	/* add track end marker */
+	if (new_track) {
+		last_tidaw = add_track_end(itcw, &filldata, &fillsize);
+		if (!last_tidaw)
+			goto out_error;
+	}
+
+	last_tidaw->flags |= TIDAW_FLAGS_LAST;
+	last_tidaw->flags &= ~TIDAW_FLAGS_INSERT_CBC;
+	itcw_finalize(itcw);
+
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->cpmode = 1;
+	cqr->startdev = startdev;
+	cqr->memdev = startdev;
+	cqr->block = block;
+	cqr->expires = startdev->default_expires * HZ;	/* default 5 minutes */
+	cqr->lpm = dasd_path_get_ppm(startdev);
+	cqr->retries = startdev->default_retries;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	return cqr;
+out_error:
+	/* dasd_sfree_request frees from the right pool via cqr->mem_chunk */
+	dasd_sfree_request(cqr, startdev);
+	return ERR_PTR(ret);
+}
+
+static struct dasd_ccw_req *
+dasd_eckd_build_cp_discard(struct dasd_device *device, struct dasd_block *block,
+			   struct request *req, sector_t first_trk,
+			   sector_t last_trk, unsigned int first_offs,
+			   unsigned int last_offs, unsigned int blk_per_trk)
+{
+	struct dasd_eckd_private *private = device->private;
+	sector_t first_ext_trk, last_ext_end, last_ext_trk;
+	unsigned int trks_per_ext;
+
+	trks_per_ext = dasd_eckd_ext_size(device) * private->rdc_data.trk_per_cyl;
+	if (!trks_per_ext)
+		return ERR_PTR(-EOPNOTSUPP);
+
+	/*
+	 * A discard range is rarely track-aligned: fstrim is FS-block granular
+	 * and discard_granularity is only a hint. If it starts or ends mid-track,
+	 * that boundary track still holds live records outside the range, so drop
+	 * it from the whole-track span first. Otherwise a partial boundary track
+	 * that happens to sit on an extent boundary would be released together
+	 * with its live records resulting in silent data loss
+	 */
+	if (first_offs)				/* partial first track */
+		first_trk++;
+	if (last_offs != blk_per_trk - 1) {	/* partial last track */
+		if (!last_trk)
+			return ERR_PTR(-EOPNOTSUPP);
+		last_trk--;
+	}
+	if (first_trk > last_trk)
+		return ERR_PTR(-EOPNOTSUPP);	/* no whole track fully covered */
+
+	/*
+	 * RAS releases whole extents. Only release extents that lie entirely
+	 * within the (now whole-track) discard range by rounding inward to extent
+	 * boundaries - an extent shared with a live allocation must never be
+	 * released. If no whole extent is covered there is nothing to release
+	 * safely (e.g. a sub-extent discard, unavoidable with large extents), so
+	 * reject the request rather than release too much.
+	 */
+	first_ext_trk = roundup(first_trk, trks_per_ext);
+	/* one past the last whole extent inside the range (exclusive) */
+	last_ext_end = rounddown(last_trk + 1, trks_per_ext);
+	if (first_ext_trk >= last_ext_end)
+		return ERR_PTR(-EOPNOTSUPP);
+	/* inclusive last track; the guard above keeps this from underflowing */
+	last_ext_trk = last_ext_end - 1;
+
+	return dasd_eckd_dso_ras(device, block, req, first_ext_trk,
+				 last_ext_trk, 1);
+}
+
+static struct dasd_ccw_req *dasd_eckd_build_cp(struct dasd_device *startdev,
+					       struct dasd_block *block,
+					       struct request *req)
+{
+	int cmdrtd, cmdwtd;
+	int use_prefix;
+	int fcx_multitrack;
+	struct dasd_eckd_private *private;
+	struct dasd_device *basedev;
+	sector_t first_rec, last_rec;
+	sector_t first_trk, last_trk;
+	unsigned int first_offs, last_offs;
+	unsigned int blk_per_trk, blksize;
+	int cdlspecial;
+	unsigned int data_size;
+	struct dasd_ccw_req *cqr;
+
+	basedev = block->base;
+	if (dasd_ese_adaptive(basedev))
+		dasd_ese_heuristic_tick(basedev);
+	else
+		/* re-assert the endpoint mode: a stale heuristic write cannot stick */
+		basedev->fulltrack = (basedev->ft_bias >= DASD_FT_BIAS_MAX) ? 1 : 0;
+	private = basedev->private;
+
+	/* Calculate number of blocks/records per track. */
+	blksize = block->bp_block;
+	blk_per_trk = recs_per_track(&private->rdc_data, 0, blksize);
+	if (blk_per_trk == 0)
+		return ERR_PTR(-EINVAL);
+	/* Calculate record id of first and last block. */
+	first_rec = first_trk = blk_rq_pos(req) >> block->s2b_shift;
+	first_offs = sector_div(first_trk, blk_per_trk);
+	last_rec = last_trk =
+		(blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift;
+	last_offs = sector_div(last_trk, blk_per_trk);
+	cdlspecial = (private->uses_cdl && first_rec < 2*blk_per_trk);
+
+	if (req_op(req) == REQ_OP_DISCARD)
+		return dasd_eckd_build_cp_discard(startdev, block, req,
+						  first_trk, last_trk,
+						  first_offs, last_offs,
+						  blk_per_trk);
+
+	fcx_multitrack = private->features.feature[40] & 0x20;
+	data_size = blk_rq_bytes(req);
+	if (data_size % blksize || data_size == 0)
+		return ERR_PTR(-EINVAL);
+	/* tpm write request add CBC data on each track boundary */
+	if (rq_data_dir(req) == WRITE)
+		data_size += (last_trk - first_trk) * 4;
+
+	/* is read track data and write track data in command mode supported? */
+	cmdrtd = private->features.feature[9] & 0x20;
+	cmdwtd = private->features.feature[12] & 0x40;
+	use_prefix = private->features.feature[8] & 0x01;
+
+	cqr = NULL;
+	if (cdlspecial || dasd_page_cache) {
+		/* do nothing, just fall through to the cmd mode single case */
+	} else if ((data_size <= private->fcx_max_data)
+		   && (fcx_multitrack || (first_trk == last_trk))) {
+		if (!first_offs && (last_offs + 1 == blk_per_trk) &&
+		    rq_data_dir(req) == WRITE && basedev->fulltrack) {
+			cqr = dasd_eckd_build_cp_tpm_writefulltrack(startdev, block, req,
+								    first_rec, last_rec,
+								    first_trk, last_trk,
+								    first_offs, last_offs,
+								    blk_per_trk, blksize, NULL);
+		} else {
+			cqr = dasd_eckd_build_cp_tpm_track(startdev, block, req,
+							   first_rec, last_rec,
+							   first_trk, last_trk,
+							   first_offs, last_offs,
+							   blk_per_trk, blksize);
+		}
+		if (IS_ERR(cqr) && (PTR_ERR(cqr) != -EAGAIN) &&
+		    (PTR_ERR(cqr) != -ENOMEM))
+			cqr = NULL;
+	} else if (use_prefix &&
+		   (((rq_data_dir(req) == READ) && cmdrtd) ||
+		    ((rq_data_dir(req) == WRITE) && cmdwtd))) {
+		cqr = dasd_eckd_build_cp_cmd_track(startdev, block, req,
+						   first_rec, last_rec,
+						   first_trk, last_trk,
+						   first_offs, last_offs,
+						   blk_per_trk, blksize);
+		if (IS_ERR(cqr) && (PTR_ERR(cqr) != -EAGAIN) &&
+		    (PTR_ERR(cqr) != -ENOMEM))
+			cqr = NULL;
+	}
+	if (!cqr)
+		cqr = dasd_eckd_build_cp_cmd_single(startdev, block, req,
+						    first_rec, last_rec,
+						    first_trk, last_trk,
+						    first_offs, last_offs,
+						    blk_per_trk, blksize);
+
+	if (!IS_ERR(cqr)) {
+		cqr->start_trk = first_trk;
+		cqr->end_trk = last_trk;
+	}
+	return cqr;
+}
+
+static struct dasd_ccw_req *dasd_eckd_build_cp_raw(struct dasd_device *startdev,
+						   struct dasd_block *block,
+						   struct request *req)
+{
+	sector_t start_padding_sectors, end_sector_offset, end_padding_sectors;
+	unsigned int seg_len, len_to_track_end;
+	unsigned int cidaw, cplength, datasize;
+	sector_t first_trk, last_trk, sectors;
+	struct dasd_eckd_private *base_priv;
+	struct dasd_device *basedev;
+	struct req_iterator iter;
+	struct dasd_ccw_req *cqr;
+	unsigned int trkcount;
+	unsigned int size;
+	unsigned char cmd;
+	struct bio_vec bv;
+	struct ccw1 *ccw;
+	dma64_t *idaws;
+	int use_prefix;
+	void *data;
+	char *dst;
+
+	/*
+	 * raw track access needs to be mutiple of 64k and on 64k boundary
+	 * For read requests we can fix an incorrect alignment by padding
+	 * the request with dummy pages.
+	 */
+	start_padding_sectors = blk_rq_pos(req) % DASD_RAW_SECTORS_PER_TRACK;
+	end_sector_offset = (blk_rq_pos(req) + blk_rq_sectors(req)) %
+		DASD_RAW_SECTORS_PER_TRACK;
+	end_padding_sectors = (DASD_RAW_SECTORS_PER_TRACK - end_sector_offset) %
+		DASD_RAW_SECTORS_PER_TRACK;
+	basedev = block->base;
+	if ((start_padding_sectors || end_padding_sectors) &&
+	    (rq_data_dir(req) == WRITE)) {
+		DBF_DEV_EVENT(DBF_ERR, basedev,
+			      "raw write not track aligned (%llu,%llu) req %p",
+			      start_padding_sectors, end_padding_sectors, req);
+		return ERR_PTR(-EINVAL);
+	}
+
+	first_trk = blk_rq_pos(req) / DASD_RAW_SECTORS_PER_TRACK;
+	last_trk = (blk_rq_pos(req) + blk_rq_sectors(req) - 1) /
+		DASD_RAW_SECTORS_PER_TRACK;
+	trkcount = last_trk - first_trk + 1;
+
+	if (rq_data_dir(req) == READ)
+		cmd = DASD_ECKD_CCW_READ_TRACK;
+	else if (rq_data_dir(req) == WRITE)
+		cmd = DASD_ECKD_CCW_WRITE_FULL_TRACK;
+	else
+		return ERR_PTR(-EINVAL);
+
+	/*
+	 * Raw track based I/O needs IDAWs for each page,
+	 * and not just for 64 bit addresses.
+	 */
+	cidaw = trkcount * DASD_RAW_BLOCK_PER_TRACK;
+
+	/*
+	 * struct PFX_eckd_data and struct LRE_eckd_data can have up to 2 bytes
+	 * of extended parameter. This is needed for write full track.
+	 */
+	base_priv = basedev->private;
+	use_prefix = base_priv->features.feature[8] & 0x01;
+	if (use_prefix) {
+		cplength = 1 + trkcount;
+		size = sizeof(struct PFX_eckd_data) + 2;
+	} else {
+		cplength = 2 + trkcount;
+		size = sizeof(struct DE_eckd_data) +
+			sizeof(struct LRE_eckd_data) + 2;
+	}
+	size = ALIGN(size, 8);
+
+	datasize = size + cidaw * sizeof(unsigned long);
+
+	/* Allocate the ccw request. */
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, cplength,
+				   datasize, startdev, blk_mq_rq_to_pdu(req));
+	if (IS_ERR(cqr))
+		return cqr;
+
+	ccw = cqr->cpaddr;
+	data = cqr->data;
+
+	if (use_prefix) {
+		prefix_LRE(ccw++, data, first_trk, last_trk, cmd, basedev,
+			   startdev, 1, 0, trkcount, 0, 0);
+	} else {
+		define_extent(ccw++, data, first_trk, last_trk, cmd, basedev, 0);
+		ccw[-1].flags |= CCW_FLAG_CC;
+
+		data += sizeof(struct DE_eckd_data);
+		locate_record_ext(ccw++, data, first_trk, 0,
+				  trkcount, cmd, basedev, 0, 0);
+	}
+
+	idaws = (dma64_t *)(cqr->data + size);
+	len_to_track_end = 0;
+	if (start_padding_sectors) {
+		ccw[-1].flags |= CCW_FLAG_CC;
+		ccw->cmd_code = cmd;
+		/* maximum 3390 track size */
+		ccw->count = 57326;
+		/* 64k map to one track */
+		len_to_track_end = 65536 - start_padding_sectors * 512;
+		ccw->cda = virt_to_dma32(idaws);
+		ccw->flags |= CCW_FLAG_IDA;
+		ccw->flags |= CCW_FLAG_SLI;
+		ccw++;
+		for (sectors = 0; sectors < start_padding_sectors; sectors += 8)
+			idaws = idal_create_words(idaws, rawpadpage, PAGE_SIZE);
+	}
+	rq_for_each_segment(bv, req, iter) {
+		dst = bvec_virt(&bv);
+		seg_len = bv.bv_len;
+		if (cmd == DASD_ECKD_CCW_READ_TRACK)
+			memset(dst, 0, seg_len);
+		if (!len_to_track_end) {
+			ccw[-1].flags |= CCW_FLAG_CC;
+			ccw->cmd_code = cmd;
+			/* maximum 3390 track size */
+			ccw->count = 57326;
+			/* 64k map to one track */
+			len_to_track_end = 65536;
+			ccw->cda = virt_to_dma32(idaws);
+			ccw->flags |= CCW_FLAG_IDA;
+			ccw->flags |= CCW_FLAG_SLI;
+			ccw++;
+		}
+		len_to_track_end -= seg_len;
+		idaws = idal_create_words(idaws, dst, seg_len);
+	}
+	for (sectors = 0; sectors < end_padding_sectors; sectors += 8)
+		idaws = idal_create_words(idaws, rawpadpage, PAGE_SIZE);
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->startdev = startdev;
+	cqr->memdev = startdev;
+	cqr->block = block;
+	cqr->expires = startdev->default_expires * HZ;
+	cqr->lpm = dasd_path_get_ppm(startdev);
+	cqr->retries = startdev->default_retries;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	return cqr;
+}
+
+
+static int
+dasd_eckd_free_cp(struct dasd_ccw_req *cqr, struct request *req)
+{
+	struct dasd_eckd_private *private;
+	struct ccw1 *ccw;
+	struct req_iterator iter;
+	struct bio_vec bv;
+	char *dst, *cda;
+	unsigned int blksize, blk_per_trk, off;
+	sector_t recid;
+	int status;
+
+	/*
+	 * A format-aborted request finished nothing - its replacement
+	 * completes the block request - so report ABORTED instead of DONE,
+	 * but still release its bounce buffers like any other request.
+	 */
+	if (cqr->status == DASD_CQR_ABORTED)
+		status = DASD_CQR_ABORTED;
+	else
+		status = cqr->status == DASD_CQR_DONE;
+	/* transport mode has no dasd_page_cache bounce buffers to release */
+	if (!dasd_page_cache || cqr->cpmode)
+		goto out;
+	private = cqr->block->base->private;
+	blksize = cqr->block->bp_block;
+	blk_per_trk = recs_per_track(&private->rdc_data, 0, blksize);
+	recid = blk_rq_pos(req) >> cqr->block->s2b_shift;
+	ccw = cqr->cpaddr;
+	/* Skip over define extent & locate record. */
+	ccw++;
+	if (private->uses_cdl == 0 || recid > 2*blk_per_trk)
+		ccw++;
+	rq_for_each_segment(bv, req, iter) {
+		dst = bvec_virt(&bv);
+		for (off = 0; off < bv.bv_len; off += blksize) {
+			/* Skip locate record. */
+			if (private->uses_cdl && recid <= 2*blk_per_trk)
+				ccw++;
+			if (dst) {
+				if (ccw->flags & CCW_FLAG_IDA)
+					cda = dma64_to_virt(*((dma64_t *)dma32_to_virt(ccw->cda)));
+				else
+					cda = dma32_to_virt(ccw->cda);
+				if (dst != cda) {
+					if (rq_data_dir(req) == READ)
+						memcpy(dst, cda, bv.bv_len);
+					kmem_cache_free(dasd_page_cache,
+					    (void *)((addr_t)cda & PAGE_MASK));
+				}
+				dst = NULL;
+			}
+			ccw++;
+			recid++;
+		}
+	}
+out:
+	dasd_sfree_request(cqr, cqr->memdev);
+	return status;
+}
+
+/*
+ * Modify ccw/tcw in cqr so it can be started on a base device.
+ *
+ * Note that this is not enough to restart the cqr!
+ * Either reset cqr->startdev as well (summary unit check handling)
+ * or restart via separate cqr (as in ERP handling).
+ */
+void dasd_eckd_reset_ccw_to_base_io(struct dasd_ccw_req *cqr)
+{
+	struct ccw1 *ccw;
+	struct PFX_eckd_data *pfxdata;
+	struct tcw *tcw;
+	struct tccb *tccb;
+	struct dcw *dcw;
+
+	if (cqr->cpmode == 1) {
+		tcw = cqr->cpaddr;
+		tccb = tcw_get_tccb(tcw);
+		dcw = (struct dcw *)&tccb->tca[0];
+		pfxdata = (struct PFX_eckd_data *)&dcw->cd[0];
+		pfxdata->validity.verify_base = 0;
+		pfxdata->validity.hyper_pav = 0;
+	} else {
+		ccw = cqr->cpaddr;
+		pfxdata = cqr->data;
+		if (ccw->cmd_code == DASD_ECKD_CCW_PFX) {
+			pfxdata->validity.verify_base = 0;
+			pfxdata->validity.hyper_pav = 0;
+		}
+	}
+}
+
+#define DASD_ECKD_CHANQ_MAX_SIZE 4
+
+static struct dasd_ccw_req *dasd_eckd_build_alias_cp(struct dasd_device *base,
+						     struct dasd_block *block,
+						     struct request *req)
+{
+	struct dasd_eckd_private *private;
+	struct dasd_device *startdev = NULL;
+	unsigned long flags;
+	struct dasd_ccw_req *cqr;
+
+	/* Discard requests (space release) can only run on the base device. */
+	if (req_op(req) != REQ_OP_DISCARD)
+		startdev = dasd_alias_get_start_dev(base);
+	if (!startdev)
+		startdev = base;
+	private = startdev->private;
+	if (private->count >= DASD_ECKD_CHANQ_MAX_SIZE)
+		return ERR_PTR(-EBUSY);
+
+	spin_lock_irqsave(get_ccwdev_lock(startdev->cdev), flags);
+	private->count++;
+	if ((base->features & DASD_FEATURE_USERAW))
+		cqr = dasd_eckd_build_cp_raw(startdev, block, req);
+	else
+		cqr = dasd_eckd_build_cp(startdev, block, req);
+	if (IS_ERR(cqr))
+		private->count--;
+	spin_unlock_irqrestore(get_ccwdev_lock(startdev->cdev), flags);
+	return cqr;
+}
+
+static int dasd_eckd_free_alias_cp(struct dasd_ccw_req *cqr,
+				   struct request *req)
+{
+	struct dasd_eckd_private *private;
+	unsigned long flags;
+
+	spin_lock_irqsave(get_ccwdev_lock(cqr->memdev->cdev), flags);
+	private = cqr->memdev->private;
+	private->count--;
+	spin_unlock_irqrestore(get_ccwdev_lock(cqr->memdev->cdev), flags);
+	if (cqr->format)
+		clear_format_track(cqr->format, cqr->block);
+	return dasd_eckd_free_cp(cqr, req);
+}
+
+static int
+dasd_eckd_fill_info(struct dasd_device * device,
+		    struct dasd_information2_t * info)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	info->label_block = 2;
+	info->FBA_layout = private->uses_cdl ? 0 : 1;
+	info->format = private->uses_cdl ? DASD_FORMAT_CDL : DASD_FORMAT_LDL;
+	info->characteristics_size = sizeof(private->rdc_data);
+	memcpy(info->characteristics, &private->rdc_data,
+	       sizeof(private->rdc_data));
+	info->confdata_size = min_t(unsigned long, private->conf.len,
+				    sizeof(info->configuration_data));
+	memcpy(info->configuration_data, private->conf.data,
+	       info->confdata_size);
+	return 0;
+}
+
+/*
+ * SECTION: ioctl functions for eckd devices.
+ */
+
+/*
+ * Release device ioctl.
+ * Buils a channel programm to releases a prior reserved
+ * (see dasd_eckd_reserve) device.
+ */
+static int
+dasd_eckd_release(struct dasd_device *device)
+__context_unsafe(/* Conditional locking */)
+{
+	struct dasd_ccw_req *cqr;
+	int rc;
+	struct ccw1 *ccw;
+	int useglobal;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	useglobal = 0;
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1, 32, device, NULL);
+	if (IS_ERR(cqr)) {
+		mutex_lock(&dasd_reserve_mutex);
+		useglobal = 1;
+		cqr = &dasd_reserve_req->cqr;
+		memset(cqr, 0, sizeof(*cqr));
+		memset(&dasd_reserve_req->ccw, 0,
+		       sizeof(dasd_reserve_req->ccw));
+		cqr->cpaddr = &dasd_reserve_req->ccw;
+		cqr->data = &dasd_reserve_req->data;
+		cqr->magic = DASD_ECKD_MAGIC;
+	}
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_RELEASE;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->count = 32;
+	ccw->cda = virt_to_dma32(cqr->data);
+	cqr->startdev = device;
+	cqr->memdev = device;
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->retries = 2;	/* set retry counter to enable basic ERP */
+	cqr->expires = 2 * HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	rc = dasd_sleep_on_immediatly(cqr);
+	if (!rc)
+		clear_bit(DASD_FLAG_IS_RESERVED, &device->flags);
+
+	if (useglobal)
+		mutex_unlock(&dasd_reserve_mutex);
+	else
+		dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+/*
+ * Reserve device ioctl.
+ * Options are set to 'synchronous wait for interrupt' and
+ * 'timeout the request'. This leads to a terminate IO if
+ * the interrupt is outstanding for a certain time.
+ */
+static int
+dasd_eckd_reserve(struct dasd_device *device)
+__context_unsafe(/* Conditional locking */)
+{
+	struct dasd_ccw_req *cqr;
+	int rc;
+	struct ccw1 *ccw;
+	int useglobal;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	useglobal = 0;
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1, 32, device, NULL);
+	if (IS_ERR(cqr)) {
+		mutex_lock(&dasd_reserve_mutex);
+		useglobal = 1;
+		cqr = &dasd_reserve_req->cqr;
+		memset(cqr, 0, sizeof(*cqr));
+		memset(&dasd_reserve_req->ccw, 0,
+		       sizeof(dasd_reserve_req->ccw));
+		cqr->cpaddr = &dasd_reserve_req->ccw;
+		cqr->data = &dasd_reserve_req->data;
+		cqr->magic = DASD_ECKD_MAGIC;
+	}
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_RESERVE;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->count = 32;
+	ccw->cda = virt_to_dma32(cqr->data);
+	cqr->startdev = device;
+	cqr->memdev = device;
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->retries = 2;	/* set retry counter to enable basic ERP */
+	cqr->expires = 2 * HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	rc = dasd_sleep_on_immediatly(cqr);
+	if (!rc)
+		set_bit(DASD_FLAG_IS_RESERVED, &device->flags);
+
+	if (useglobal)
+		mutex_unlock(&dasd_reserve_mutex);
+	else
+		dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+/*
+ * Steal lock ioctl - unconditional reserve device.
+ * Buils a channel programm to break a device's reservation.
+ * (unconditional reserve)
+ */
+static int
+dasd_eckd_steal_lock(struct dasd_device *device)
+__context_unsafe(/* Conditional locking */)
+{
+	struct dasd_ccw_req *cqr;
+	int rc;
+	struct ccw1 *ccw;
+	int useglobal;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	useglobal = 0;
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1, 32, device, NULL);
+	if (IS_ERR(cqr)) {
+		mutex_lock(&dasd_reserve_mutex);
+		useglobal = 1;
+		cqr = &dasd_reserve_req->cqr;
+		memset(cqr, 0, sizeof(*cqr));
+		memset(&dasd_reserve_req->ccw, 0,
+		       sizeof(dasd_reserve_req->ccw));
+		cqr->cpaddr = &dasd_reserve_req->ccw;
+		cqr->data = &dasd_reserve_req->data;
+		cqr->magic = DASD_ECKD_MAGIC;
+	}
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_SLCK;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->count = 32;
+	ccw->cda = virt_to_dma32(cqr->data);
+	cqr->startdev = device;
+	cqr->memdev = device;
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->retries = 2;	/* set retry counter to enable basic ERP */
+	cqr->expires = 2 * HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	rc = dasd_sleep_on_immediatly(cqr);
+	if (!rc)
+		set_bit(DASD_FLAG_IS_RESERVED, &device->flags);
+
+	if (useglobal)
+		mutex_unlock(&dasd_reserve_mutex);
+	else
+		dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+/*
+ * SNID - Sense Path Group ID
+ * This ioctl may be used in situations where I/O is stalled due to
+ * a reserve, so if the normal dasd_smalloc_request fails, we use the
+ * preallocated dasd_reserve_req.
+ */
+static int dasd_eckd_snid(struct dasd_device *device,
+			  void __user *argp)
+__context_unsafe(/* Conditional locking */)
+{
+	struct dasd_ccw_req *cqr;
+	int rc;
+	struct ccw1 *ccw;
+	int useglobal;
+	struct dasd_snid_ioctl_data usrparm;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	if (copy_from_user(&usrparm, argp, sizeof(usrparm)))
+		return -EFAULT;
+
+	useglobal = 0;
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1,
+				   sizeof(struct dasd_snid_data), device,
+				   NULL);
+	if (IS_ERR(cqr)) {
+		mutex_lock(&dasd_reserve_mutex);
+		useglobal = 1;
+		cqr = &dasd_reserve_req->cqr;
+		memset(cqr, 0, sizeof(*cqr));
+		memset(&dasd_reserve_req->ccw, 0,
+		       sizeof(dasd_reserve_req->ccw));
+		cqr->cpaddr = &dasd_reserve_req->ccw;
+		cqr->data = &dasd_reserve_req->data;
+		cqr->magic = DASD_ECKD_MAGIC;
+	}
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_SNID;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->count = 12;
+	ccw->cda = virt_to_dma32(cqr->data);
+	cqr->startdev = device;
+	cqr->memdev = device;
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	set_bit(DASD_CQR_ALLOW_SLOCK, &cqr->flags);
+	cqr->retries = 5;
+	cqr->expires = 10 * HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	cqr->lpm = usrparm.path_mask;
+
+	rc = dasd_sleep_on_immediatly(cqr);
+	/* verify that I/O processing didn't modify the path mask */
+	if (!rc && usrparm.path_mask && (cqr->lpm != usrparm.path_mask))
+		rc = -EIO;
+	if (!rc) {
+		usrparm.data = *((struct dasd_snid_data *)cqr->data);
+		if (copy_to_user(argp, &usrparm, sizeof(usrparm)))
+			rc = -EFAULT;
+	}
+
+	if (useglobal)
+		mutex_unlock(&dasd_reserve_mutex);
+	else
+		dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+/*
+ * Read performance statistics
+ */
+static int
+dasd_eckd_performance(struct dasd_device *device, void __user *argp)
+{
+	struct dasd_psf_prssd_data *prssdp;
+	struct dasd_rssd_perf_stats_t *stats;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int rc;
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */  + 1 /* RSSD */,
+				   (sizeof(struct dasd_psf_prssd_data) +
+				    sizeof(struct dasd_rssd_perf_stats_t)),
+				   device, NULL);
+	if (IS_ERR(cqr)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "Could not allocate initialization request");
+		return PTR_ERR(cqr);
+	}
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->retries = 0;
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	cqr->expires = 10 * HZ;
+
+	/* Prepare for Read Subsystem Data */
+	prssdp = (struct dasd_psf_prssd_data *) cqr->data;
+	memset(prssdp, 0, sizeof(struct dasd_psf_prssd_data));
+	prssdp->order = PSF_ORDER_PRSSD;
+	prssdp->suborder = 0x01;	/* Performance Statistics */
+	prssdp->varies[1] = 0x01;	/* Perf Statistics for the Subsystem */
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->count = sizeof(struct dasd_psf_prssd_data);
+	ccw->flags |= CCW_FLAG_CC;
+	ccw->cda = virt_to_dma32(prssdp);
+
+	/* Read Subsystem Data - Performance Statistics */
+	stats = (struct dasd_rssd_perf_stats_t *) (prssdp + 1);
+	memset(stats, 0, sizeof(struct dasd_rssd_perf_stats_t));
+
+	ccw++;
+	ccw->cmd_code = DASD_ECKD_CCW_RSSD;
+	ccw->count = sizeof(struct dasd_rssd_perf_stats_t);
+	ccw->cda = virt_to_dma32(stats);
+
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	rc = dasd_sleep_on(cqr);
+	if (rc == 0) {
+		prssdp = (struct dasd_psf_prssd_data *) cqr->data;
+		stats = (struct dasd_rssd_perf_stats_t *) (prssdp + 1);
+		if (copy_to_user(argp, stats,
+				 sizeof(struct dasd_rssd_perf_stats_t)))
+			rc = -EFAULT;
+	}
+	dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+/*
+ * Get attributes (cache operations)
+ * Returnes the cache attributes used in Define Extend (DE).
+ */
+static int
+dasd_eckd_get_attrib(struct dasd_device *device, void __user *argp)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct attrib_data_t attrib = private->attrib;
+	int rc;
+
+        if (!capable(CAP_SYS_ADMIN))
+                return -EACCES;
+	if (!argp)
+                return -EINVAL;
+
+	rc = 0;
+	if (copy_to_user(argp, (long *) &attrib,
+			 sizeof(struct attrib_data_t)))
+		rc = -EFAULT;
+
+	return rc;
+}
+
+/*
+ * Set attributes (cache operations)
+ * Stores the attributes for cache operation to be used in Define Extend (DE).
+ */
+static int
+dasd_eckd_set_attrib(struct dasd_device *device, void __user *argp)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct attrib_data_t attrib;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+	if (!argp)
+		return -EINVAL;
+
+	if (copy_from_user(&attrib, argp, sizeof(struct attrib_data_t)))
+		return -EFAULT;
+	private->attrib = attrib;
+
+	dev_info(&device->cdev->dev,
+		 "The DASD cache mode was set to %x (%i cylinder prestage)\n",
+		 private->attrib.operation, private->attrib.nr_cyl);
+	return 0;
+}
+
+/*
+ * Issue syscall I/O to EMC Symmetrix array.
+ * CCWs are PSF and RSSD
+ */
+static int dasd_symm_io(struct dasd_device *device, void __user *argp)
+{
+	struct dasd_symmio_parms usrparm;
+	char *psf_data, *rssd_result;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	char psf0, psf1;
+	int rc;
+
+	if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RAWIO))
+		return -EACCES;
+	psf0 = psf1 = 0;
+
+	/* Copy parms from caller */
+	rc = -EFAULT;
+	if (copy_from_user(&usrparm, argp, sizeof(usrparm)))
+		goto out;
+	/* at least 2 bytes are accessed and should be allocated */
+	if (usrparm.psf_data_len < 2) {
+		DBF_DEV_EVENT(DBF_WARNING, device,
+			      "Symmetrix ioctl invalid data length %d",
+			      usrparm.psf_data_len);
+		rc = -EINVAL;
+		goto out;
+	}
+	/* alloc I/O data area */
+	psf_data = kzalloc(usrparm.psf_data_len, GFP_KERNEL | GFP_DMA);
+	rssd_result = kzalloc(usrparm.rssd_result_len, GFP_KERNEL | GFP_DMA);
+	if (!psf_data || !rssd_result) {
+		rc = -ENOMEM;
+		goto out_free;
+	}
+
+	/* get syscall header from user space */
+	rc = -EFAULT;
+	if (copy_from_user(psf_data,
+			   (void __user *)(unsigned long) usrparm.psf_data,
+			   usrparm.psf_data_len))
+		goto out_free;
+	psf0 = psf_data[0];
+	psf1 = psf_data[1];
+
+	/* setup CCWs for PSF + RSSD */
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 2, 0, device, NULL);
+	if (IS_ERR(cqr)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			"Could not allocate initialization request");
+		rc = PTR_ERR(cqr);
+		goto out_free;
+	}
+
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->retries = 3;
+	cqr->expires = 10 * HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	/* Build the ccws */
+	ccw = cqr->cpaddr;
+
+	/* PSF ccw */
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->count = usrparm.psf_data_len;
+	ccw->flags |= CCW_FLAG_CC;
+	ccw->cda = virt_to_dma32(psf_data);
+
+	ccw++;
+
+	/* RSSD ccw  */
+	ccw->cmd_code = DASD_ECKD_CCW_RSSD;
+	ccw->count = usrparm.rssd_result_len;
+	ccw->flags = CCW_FLAG_SLI ;
+	ccw->cda = virt_to_dma32(rssd_result);
+
+	rc = dasd_sleep_on(cqr);
+	if (rc)
+		goto out_sfree;
+
+	rc = -EFAULT;
+	if (copy_to_user((void __user *)(unsigned long) usrparm.rssd_result,
+			   rssd_result, usrparm.rssd_result_len))
+		goto out_sfree;
+	rc = 0;
+
+out_sfree:
+	dasd_sfree_request(cqr, cqr->memdev);
+out_free:
+	kfree(rssd_result);
+	kfree(psf_data);
+out:
+	DBF_DEV_EVENT(DBF_WARNING, device,
+		      "Symmetrix ioctl (0x%02x 0x%02x): rc=%d",
+		      (int) psf0, (int) psf1, rc);
+	return rc;
+}
+
+static int
+dasd_eckd_ioctl(struct dasd_block *block, unsigned int cmd, void __user *argp)
+{
+	struct dasd_device *device = block->base;
+
+	switch (cmd) {
+	case BIODASDGATTR:
+		return dasd_eckd_get_attrib(device, argp);
+	case BIODASDSATTR:
+		return dasd_eckd_set_attrib(device, argp);
+	case BIODASDPSRD:
+		return dasd_eckd_performance(device, argp);
+	case BIODASDRLSE:
+		return dasd_eckd_release(device);
+	case BIODASDRSRV:
+		return dasd_eckd_reserve(device);
+	case BIODASDSLCK:
+		return dasd_eckd_steal_lock(device);
+	case BIODASDSNID:
+		return dasd_eckd_snid(device, argp);
+	case BIODASDSYMMIO:
+		return dasd_symm_io(device, argp);
+	default:
+		return -ENOTTY;
+	}
+}
+
+/*
+ * Dump the range of CCWs into 'page' buffer
+ * and return number of printed chars.
+ */
+static void
+dasd_eckd_dump_ccw_range(struct dasd_device *device, struct ccw1 *from,
+			 struct ccw1 *to, char *page)
+{
+	int len, count;
+	char *datap;
+
+	len = 0;
+	while (from <= to) {
+		len += sprintf(page + len, "CCW %px: %08X %08X DAT:",
+			       from, ((int *) from)[0], ((int *) from)[1]);
+
+		/* get pointer to data (consider IDALs) */
+		if (from->flags & CCW_FLAG_IDA)
+			datap = dma64_to_virt(*((dma64_t *)dma32_to_virt(from->cda)));
+		else
+			datap = dma32_to_virt(from->cda);
+
+		/* dump data (max 128 bytes) */
+		for (count = 0; count < from->count && count < 128; count++) {
+			if (count % 32 == 0)
+				len += sprintf(page + len, "\n");
+			if (count % 8 == 0)
+				len += sprintf(page + len, " ");
+			if (count % 4 == 0)
+				len += sprintf(page + len, " ");
+			len += sprintf(page + len, "%02x", datap[count]);
+		}
+		len += sprintf(page + len, "\n");
+		from++;
+	}
+	if (len > 0)
+		dev_err(&device->cdev->dev, "%s", page);
+}
+
+static void
+dasd_eckd_dump_sense_dbf(struct dasd_device *device, struct irb *irb,
+			 char *reason)
+{
+	u64 *sense;
+	u64 *stat;
+
+	sense = (u64 *) dasd_get_sense(irb);
+	stat = (u64 *) &irb->scsw;
+	if (sense) {
+		DBF_DEV_EVENT(DBF_EMERG, device, "%s: %016llx %08x : "
+			      "%016llx %016llx %016llx %016llx",
+			      reason, *stat, *((u32 *) (stat + 1)),
+			      sense[0], sense[1], sense[2], sense[3]);
+	} else {
+		DBF_DEV_EVENT(DBF_EMERG, device, "%s: %016llx %08x : %s",
+			      reason, *stat, *((u32 *) (stat + 1)),
+			      "NO VALID SENSE");
+	}
+}
+
+/*
+ * Print sense data and related channel program.
+ * Parts are printed because printk buffer is only 1024 bytes.
+ */
+static void dasd_eckd_dump_sense_ccw(struct dasd_device *device,
+				 struct dasd_ccw_req *req, struct irb *irb)
+{
+	struct ccw1 *first, *last, *fail, *from, *to;
+	struct device *dev;
+	int len, sl, sct;
+	char *page;
+
+	dev = &device->cdev->dev;
+
+	page = kzalloc(PAGE_SIZE, GFP_ATOMIC);
+	if (page == NULL) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "No memory to dump sense data\n");
+		return;
+	}
+	/* dump the sense data */
+	len = sprintf(page, "I/O status report:\n");
+	len += sprintf(page + len,
+		       "in req: %px CC:%02X FC:%02X AC:%02X SC:%02X DS:%02X CS:%02X RC:%d\n",
+		       req, scsw_cc(&irb->scsw), scsw_fctl(&irb->scsw),
+		       scsw_actl(&irb->scsw), scsw_stctl(&irb->scsw),
+		       scsw_dstat(&irb->scsw), scsw_cstat(&irb->scsw),
+		       req ? req->intrc : 0);
+	len += sprintf(page + len, "Failing CCW: %px\n",
+		       dma32_to_virt(irb->scsw.cmd.cpa));
+	if (irb->esw.esw0.erw.cons) {
+		for (sl = 0; sl < 4; sl++) {
+			len += sprintf(page + len, "Sense(hex) %2d-%2d:",
+				       (8 * sl), ((8 * sl) + 7));
+
+			for (sct = 0; sct < 8; sct++) {
+				len += sprintf(page + len, " %02x",
+					       irb->ecw[8 * sl + sct]);
+			}
+			len += sprintf(page + len, "\n");
+		}
+
+		if (irb->ecw[27] & DASD_SENSE_BIT_0) {
+			/* 24 Byte Sense Data */
+			sprintf(page + len,
+				"24 Byte: %x MSG %x, %s MSGb to SYSOP\n",
+				irb->ecw[7] >> 4, irb->ecw[7] & 0x0f,
+				irb->ecw[1] & 0x10 ? "" : "no");
+		} else {
+			/* 32 Byte Sense Data */
+			sprintf(page + len,
+				"32 Byte: Format: %x Exception class %x\n",
+				irb->ecw[6] & 0x0f, irb->ecw[22] >> 4);
+		}
+	} else {
+		sprintf(page + len, "SORRY - NO VALID SENSE AVAILABLE\n");
+	}
+	dev_err(dev, "%s", page);
+
+	if (req) {
+		/* req == NULL for unsolicited interrupts */
+		/* dump the Channel Program (max 140 Bytes per line) */
+		/* Count CCW and print first CCWs (maximum 7) */
+		first = req->cpaddr;
+		for (last = first; last->flags & (CCW_FLAG_CC | CCW_FLAG_DC); last++);
+		to = min(first + 6, last);
+		dev_err(dev, "Related CP in req: %px\n", req);
+		dasd_eckd_dump_ccw_range(device, first, to, page);
+
+		/* print failing CCW area (maximum 4) */
+		/* scsw->cda is either valid or zero  */
+		from = ++to;
+		fail = dma32_to_virt(irb->scsw.cmd.cpa); /* failing CCW */
+		if (from <  fail - 2) {
+			from = fail - 2;     /* there is a gap - print header */
+			dev_err(dev, "......\n");
+		}
+		to = min(fail + 1, last);
+		dasd_eckd_dump_ccw_range(device, from, to, page + len);
+
+		/* print last CCWs (maximum 2) */
+		len = 0;
+		from = max(from, ++to);
+		if (from < last - 1) {
+			from = last - 1;     /* there is a gap - print header */
+			dev_err(dev, "......\n");
+		}
+		dasd_eckd_dump_ccw_range(device, from, last, page + len);
+	}
+	kfree(page);
+}
+
+
+/*
+ * Print sense data from a tcw.
+ */
+static void dasd_eckd_dump_sense_tcw(struct dasd_device *device,
+				 struct dasd_ccw_req *req, struct irb *irb)
+{
+	char *page;
+	int len, sl, sct, residual;
+	struct tsb *tsb;
+	u8 *sense, *rcq;
+
+	page = kzalloc(PAGE_SIZE, GFP_ATOMIC);
+	if (page == NULL) {
+		DBF_DEV_EVENT(DBF_WARNING, device, " %s",
+			    "No memory to dump sense data");
+		return;
+	}
+	/* dump the sense data */
+	len = sprintf(page, "I/O status report:\n");
+	len += sprintf(page + len,
+		       "in req: %px CC:%02X FC:%02X AC:%02X SC:%02X DS:%02X "
+		       "CS:%02X fcxs:%02X schxs:%02X RC:%d\n",
+		       req, scsw_cc(&irb->scsw), scsw_fctl(&irb->scsw),
+		       scsw_actl(&irb->scsw), scsw_stctl(&irb->scsw),
+		       scsw_dstat(&irb->scsw), scsw_cstat(&irb->scsw),
+		       irb->scsw.tm.fcxs,
+		       (irb->scsw.tm.ifob << 7) | irb->scsw.tm.sesq,
+		       req ? req->intrc : 0);
+	len += sprintf(page + len, "Failing TCW: %px\n",
+		       dma32_to_virt(irb->scsw.tm.tcw));
+
+	tsb = NULL;
+	sense = NULL;
+	if (irb->scsw.tm.tcw && (irb->scsw.tm.fcxs & 0x01))
+		tsb = tcw_get_tsb(dma32_to_virt(irb->scsw.tm.tcw));
+
+	if (tsb) {
+		len += sprintf(page + len, "tsb->length %d\n", tsb->length);
+		len += sprintf(page + len, "tsb->flags %x\n", tsb->flags);
+		len += sprintf(page + len, "tsb->dcw_offset %d\n", tsb->dcw_offset);
+		len += sprintf(page + len, "tsb->count %d\n", tsb->count);
+		residual = tsb->count - 28;
+		len += sprintf(page + len, "residual %d\n", residual);
+
+		switch (tsb->flags & 0x07) {
+		case 1:	/* tsa_iostat */
+			len += sprintf(page + len, "tsb->tsa.iostat.dev_time %d\n",
+				       tsb->tsa.iostat.dev_time);
+			len += sprintf(page + len, "tsb->tsa.iostat.def_time %d\n",
+				       tsb->tsa.iostat.def_time);
+			len += sprintf(page + len, "tsb->tsa.iostat.queue_time %d\n",
+				       tsb->tsa.iostat.queue_time);
+			len += sprintf(page + len, "tsb->tsa.iostat.dev_busy_time %d\n",
+				       tsb->tsa.iostat.dev_busy_time);
+			len += sprintf(page + len, "tsb->tsa.iostat.dev_act_time %d\n",
+				       tsb->tsa.iostat.dev_act_time);
+			sense = tsb->tsa.iostat.sense;
+			break;
+		case 2: /* ts_ddpc */
+			len += sprintf(page + len, "tsb->tsa.ddpc.rc %d\n",
+				       tsb->tsa.ddpc.rc);
+			for (sl = 0; sl < 2; sl++) {
+				len += sprintf(page + len,
+					       "tsb->tsa.ddpc.rcq %2d-%2d: ",
+					       (8 * sl), ((8 * sl) + 7));
+				rcq = tsb->tsa.ddpc.rcq;
+				for (sct = 0; sct < 8; sct++) {
+					len += sprintf(page + len, "%02x",
+						       rcq[8 * sl + sct]);
+				}
+				len += sprintf(page + len, "\n");
+			}
+			sense = tsb->tsa.ddpc.sense;
+			break;
+		case 3: /* tsa_intrg */
+			len += sprintf(page + len,
+				      "tsb->tsa.intrg.: not supported yet\n");
+			break;
+		}
+
+		if (sense) {
+			for (sl = 0; sl < 4; sl++) {
+				len += sprintf(page + len,
+					       "Sense(hex) %2d-%2d:",
+					       (8 * sl), ((8 * sl) + 7));
+				for (sct = 0; sct < 8; sct++) {
+					len += sprintf(page + len, " %02x",
+						       sense[8 * sl + sct]);
+				}
+				len += sprintf(page + len, "\n");
+			}
+
+			if (sense[27] & DASD_SENSE_BIT_0) {
+				/* 24 Byte Sense Data */
+				sprintf(page + len,
+					"24 Byte: %x MSG %x, %s MSGb to SYSOP\n",
+					sense[7] >> 4, sense[7] & 0x0f,
+					sense[1] & 0x10 ? "" : "no");
+			} else {
+				/* 32 Byte Sense Data */
+				sprintf(page + len,
+					"32 Byte: Format: %x Exception class %x\n",
+					sense[6] & 0x0f, sense[22] >> 4);
+			}
+		} else {
+			sprintf(page + len, "SORRY - NO VALID SENSE AVAILABLE\n");
+		}
+	} else {
+		sprintf(page + len, "SORRY - NO TSB DATA AVAILABLE\n");
+	}
+	dev_err(&device->cdev->dev, "%s", page);
+	kfree(page);
+}
+
+static void dasd_eckd_dump_sense(struct dasd_device *device,
+				 struct dasd_ccw_req *req, struct irb *irb)
+{
+	u8 *sense = dasd_get_sense(irb);
+
+	/*
+	 * In some cases certain errors might be expected and
+	 * log messages shouldn't be written then.
+	 * Check if the according suppress bit is set.
+	 */
+	if (sense && (sense[1] & SNS1_INV_TRACK_FORMAT) &&
+	    !(sense[2] & SNS2_ENV_DATA_PRESENT) &&
+	    test_bit(DASD_CQR_SUPPRESS_IT, &req->flags))
+		return;
+
+	if (sense && sense[0] & SNS0_CMD_REJECT &&
+	    test_bit(DASD_CQR_SUPPRESS_CR, &req->flags))
+		return;
+
+	if (sense && sense[1] & SNS1_NO_REC_FOUND &&
+	    test_bit(DASD_CQR_SUPPRESS_NRF, &req->flags))
+		return;
+
+	if (scsw_cstat(&irb->scsw) == 0x40 &&
+	    test_bit(DASD_CQR_SUPPRESS_IL, &req->flags))
+		return;
+
+	if (scsw_is_tm(&irb->scsw))
+		dasd_eckd_dump_sense_tcw(device, req, irb);
+	else
+		dasd_eckd_dump_sense_ccw(device, req, irb);
+}
+
+static int dasd_eckd_reload_device(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	char print_uid[DASD_UID_STRLEN];
+	int rc, old_base;
+	struct dasd_uid uid;
+	unsigned long flags;
+
+	/*
+	 * remove device from alias handling to prevent new requests
+	 * from being scheduled on the wrong alias device
+	 */
+	dasd_alias_remove_device(device);
+
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	old_base = private->uid.base_unit_addr;
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+
+	/* Read Configuration Data */
+	rc = dasd_eckd_read_conf(device);
+	if (rc)
+		goto out_err;
+
+	dasd_eckd_read_fc_security(device);
+
+	rc = dasd_eckd_generate_uid(device);
+	if (rc)
+		goto out_err;
+	/*
+	 * update unit address configuration and
+	 * add device to alias management
+	 */
+	dasd_alias_update_add_device(device);
+
+	dasd_eckd_get_uid(device, &uid);
+
+	if (old_base != uid.base_unit_addr) {
+		dasd_eckd_get_uid_string(&private->conf, print_uid);
+		dev_info(&device->cdev->dev,
+			 "An Alias device was reassigned to a new base device "
+			 "with UID: %s\n", print_uid);
+	}
+	return 0;
+
+out_err:
+	return -1;
+}
+
+static int dasd_eckd_read_message_buffer(struct dasd_device *device,
+					 struct dasd_rssd_messages *messages,
+					 __u8 lpum)
+{
+	struct dasd_rssd_messages *message_buf;
+	struct dasd_psf_prssd_data *prssdp;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int rc;
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */	+ 1 /* RSSD */,
+				   (sizeof(struct dasd_psf_prssd_data) +
+				    sizeof(struct dasd_rssd_messages)),
+				   device, NULL);
+	if (IS_ERR(cqr)) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+				"Could not allocate read message buffer request");
+		return PTR_ERR(cqr);
+	}
+
+	cqr->lpm = lpum;
+retry:
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->expires = 10 * HZ;
+	set_bit(DASD_CQR_VERIFY_PATH, &cqr->flags);
+	/* dasd_sleep_on_immediatly does not do complex error
+	 * recovery so clear erp flag and set retry counter to
+	 * do basic erp */
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	cqr->retries = 256;
+
+	/* Prepare for Read Subsystem Data */
+	prssdp = (struct dasd_psf_prssd_data *) cqr->data;
+	memset(prssdp, 0, sizeof(struct dasd_psf_prssd_data));
+	prssdp->order = PSF_ORDER_PRSSD;
+	prssdp->suborder = 0x03;	/* Message Buffer */
+	/* all other bytes of prssdp must be zero */
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->count = sizeof(struct dasd_psf_prssd_data);
+	ccw->flags |= CCW_FLAG_CC;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->cda = virt_to_dma32(prssdp);
+
+	/* Read Subsystem Data - message buffer */
+	message_buf = (struct dasd_rssd_messages *) (prssdp + 1);
+	memset(message_buf, 0, sizeof(struct dasd_rssd_messages));
+
+	ccw++;
+	ccw->cmd_code = DASD_ECKD_CCW_RSSD;
+	ccw->count = sizeof(struct dasd_rssd_messages);
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->cda = virt_to_dma32(message_buf);
+
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	rc = dasd_sleep_on_immediatly(cqr);
+	if (rc == 0) {
+		prssdp = (struct dasd_psf_prssd_data *) cqr->data;
+		message_buf = (struct dasd_rssd_messages *)
+			(prssdp + 1);
+		memcpy(messages, message_buf,
+		       sizeof(struct dasd_rssd_messages));
+	} else if (cqr->lpm) {
+		/*
+		 * on z/VM we might not be able to do I/O on the requested path
+		 * but instead we get the required information on any path
+		 * so retry with open path mask
+		 */
+		cqr->lpm = 0;
+		goto retry;
+	} else
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
+				"Reading messages failed with rc=%d\n"
+				, rc);
+	dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+static int dasd_eckd_query_host_access(struct dasd_device *device,
+				       struct dasd_psf_query_host_access *data)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct dasd_psf_query_host_access *host_access;
+	struct dasd_psf_prssd_data *prssdp;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int rc;
+
+	if (!private)
+		return -ENODEV;
+
+	/* not available for HYPER PAV alias devices */
+	if (!device->block && private->lcu && private->lcu->pav == HYPER_PAV)
+		return -EOPNOTSUPP;
+
+	/* may not be supported by the storage server */
+	if (!(private->features.feature[14] & 0x80))
+		return -EOPNOTSUPP;
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */	+ 1 /* RSSD */,
+				   sizeof(struct dasd_psf_prssd_data) + 1,
+				   device, NULL);
+	if (IS_ERR(cqr)) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+				"Could not allocate read message buffer request");
+		return PTR_ERR(cqr);
+	}
+	host_access = kzalloc_obj(*host_access, GFP_KERNEL | GFP_DMA);
+	if (!host_access) {
+		dasd_sfree_request(cqr, device);
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+				"Could not allocate host_access buffer");
+		return -ENOMEM;
+	}
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->retries = 256;
+	cqr->expires = 10 * HZ;
+
+	/* Prepare for Read Subsystem Data */
+	prssdp = (struct dasd_psf_prssd_data *) cqr->data;
+	memset(prssdp, 0, sizeof(struct dasd_psf_prssd_data));
+	prssdp->order = PSF_ORDER_PRSSD;
+	prssdp->suborder = PSF_SUBORDER_QHA;	/* query host access */
+	/* LSS and Volume that will be queried */
+	prssdp->lss = READ_ONCE(private->ned_lss);
+	prssdp->volume = READ_ONCE(private->ned_ua);
+	/* all other bytes of prssdp must be zero */
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->count = sizeof(struct dasd_psf_prssd_data);
+	ccw->flags |= CCW_FLAG_CC;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->cda = virt_to_dma32(prssdp);
+
+	/* Read Subsystem Data - query host access */
+	ccw++;
+	ccw->cmd_code = DASD_ECKD_CCW_RSSD;
+	ccw->count = sizeof(struct dasd_psf_query_host_access);
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->cda = virt_to_dma32(host_access);
+
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	/* the command might not be supported, suppress error message */
+	__set_bit(DASD_CQR_SUPPRESS_CR, &cqr->flags);
+	rc = dasd_sleep_on_interruptible(cqr);
+	if (rc == 0) {
+		*data = *host_access;
+	} else {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
+				"Reading host access data failed with rc=%d\n",
+				rc);
+		rc = -EOPNOTSUPP;
+	}
+
+	dasd_sfree_request(cqr, cqr->memdev);
+	kfree(host_access);
+	return rc;
+}
+/*
+ * return number of grouped devices
+ */
+static int dasd_eckd_host_access_count(struct dasd_device *device)
+{
+	struct dasd_psf_query_host_access *access;
+	struct dasd_ckd_path_group_entry *entry;
+	struct dasd_ckd_host_information *info;
+	int count = 0;
+	int rc, i;
+
+	access = kzalloc_obj(*access, GFP_NOIO);
+	if (!access) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+				"Could not allocate access buffer");
+		return -ENOMEM;
+	}
+	rc = dasd_eckd_query_host_access(device, access);
+	if (rc) {
+		kfree(access);
+		return rc;
+	}
+
+	info = (struct dasd_ckd_host_information *)
+		access->host_access_information;
+	for (i = 0; i < info->entry_count; i++) {
+		entry = (struct dasd_ckd_path_group_entry *)
+			(info->entry + i * info->entry_size);
+		if (entry->status_flags & DASD_ECKD_PG_GROUPED)
+			count++;
+	}
+
+	kfree(access);
+	return count;
+}
+
+/*
+ * write host access information to a sequential file
+ */
+static int dasd_hosts_print(struct dasd_device *device, struct seq_file *m)
+{
+	struct dasd_psf_query_host_access *access;
+	struct dasd_ckd_path_group_entry *entry;
+	struct dasd_ckd_host_information *info;
+	char sysplex[9] = "";
+	int rc, i;
+
+	access = kzalloc_obj(*access, GFP_NOIO);
+	if (!access) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+				"Could not allocate access buffer");
+		return -ENOMEM;
+	}
+	rc = dasd_eckd_query_host_access(device, access);
+	if (rc) {
+		kfree(access);
+		return rc;
+	}
+
+	info = (struct dasd_ckd_host_information *)
+		access->host_access_information;
+	for (i = 0; i < info->entry_count; i++) {
+		entry = (struct dasd_ckd_path_group_entry *)
+			(info->entry + i * info->entry_size);
+		/* PGID */
+		seq_printf(m, "pgid %*phN\n", 11, entry->pgid);
+		/* FLAGS */
+		seq_printf(m, "status_flags %02x\n", entry->status_flags);
+		/* SYSPLEX NAME */
+		memcpy(&sysplex, &entry->sysplex_name, sizeof(sysplex) - 1);
+		EBCASC(sysplex, sizeof(sysplex));
+		seq_printf(m, "sysplex_name %8s\n", sysplex);
+		/* SUPPORTED CYLINDER */
+		seq_printf(m, "supported_cylinder %d\n", entry->cylinder);
+		/* TIMESTAMP */
+		seq_printf(m, "timestamp %lu\n", (unsigned long)
+			   entry->timestamp);
+	}
+	kfree(access);
+
+	return 0;
+}
+
+static struct dasd_device
+*copy_relation_find_device(struct dasd_copy_relation *copy,
+			   char *busid)
+{
+	int i;
+
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		if (copy->entry[i].configured &&
+		    strncmp(copy->entry[i].busid, busid, DASD_BUS_ID_SIZE) == 0)
+			return copy->entry[i].device;
+	}
+	return NULL;
+}
+
+/*
+ * set the new active/primary device
+ */
+static void copy_pair_set_active(struct dasd_copy_relation *copy, char *new_busid,
+				 char *old_busid)
+{
+	int i;
+
+	for (i = 0; i < DASD_CP_ENTRIES; i++) {
+		if (copy->entry[i].configured &&
+		    strncmp(copy->entry[i].busid, new_busid,
+			    DASD_BUS_ID_SIZE) == 0) {
+			copy->active = &copy->entry[i];
+			copy->entry[i].primary = true;
+		} else if (copy->entry[i].configured &&
+			   strncmp(copy->entry[i].busid, old_busid,
+				   DASD_BUS_ID_SIZE) == 0) {
+			copy->entry[i].primary = false;
+		}
+	}
+}
+
+/*
+ * The function will swap the role of a given copy pair.
+ * During the swap operation the relation of the blockdevice is disconnected
+ * from the old primary and connected to the new.
+ *
+ * IO is paused on the block queue before swap and may be resumed afterwards.
+ */
+static int dasd_eckd_copy_pair_swap(struct dasd_device *device, char *prim_busid,
+				    char *sec_busid)
+{
+	struct dasd_eckd_private *prim_priv, *sec_priv;
+	struct dasd_device *primary, *secondary;
+	struct dasd_copy_relation *copy;
+	struct dasd_block *block;
+	struct gendisk *gdp;
+	int rc;
+
+	copy = device->copy;
+	if (!copy)
+		return DASD_COPYPAIRSWAP_INVALID;
+	primary = copy->active->device;
+	if (!primary)
+		return DASD_COPYPAIRSWAP_INVALID;
+	/* double check if swap has correct primary */
+	if (strncmp(dev_name(&primary->cdev->dev), prim_busid, DASD_BUS_ID_SIZE) != 0)
+		return DASD_COPYPAIRSWAP_PRIMARY;
+
+	secondary = copy_relation_find_device(copy, sec_busid);
+	if (!secondary)
+		return DASD_COPYPAIRSWAP_SECONDARY;
+
+	prim_priv = primary->private;
+	sec_priv = secondary->private;
+
+	/*
+	 * usually the device should be quiesced for swap
+	 * for paranoia stop device and requeue requests again
+	 */
+	dasd_device_set_stop_bits(primary, DASD_STOPPED_PPRC);
+	dasd_device_set_stop_bits(secondary, DASD_STOPPED_PPRC);
+	dasd_generic_requeue_all_requests(primary);
+
+	/* swap DASD internal device <> block assignment */
+	block = primary->block;
+	primary->block = NULL;
+	secondary->block = block;
+	block->base = secondary;
+	/* set new primary device in COPY relation */
+	copy_pair_set_active(copy, sec_busid, prim_busid);
+
+	/* swap blocklayer device link */
+	gdp = block->gdp;
+	dasd_add_link_to_gendisk(gdp, secondary);
+	rc = device_move(disk_to_dev(gdp), &secondary->cdev->dev, DPM_ORDER_NONE);
+	if (rc) {
+		dev_err(&primary->cdev->dev,
+			"copy_pair_swap: moving blockdevice parent %s->%s failed (%d)\n",
+			dev_name(&primary->cdev->dev),
+			dev_name(&secondary->cdev->dev), rc);
+	}
+
+	if (primary->stopped & DASD_STOPPED_QUIESCE) {
+		dasd_device_set_stop_bits(secondary, DASD_STOPPED_QUIESCE);
+		dasd_device_remove_stop_bits(primary, DASD_STOPPED_QUIESCE);
+	}
+
+	/*
+	 * The secondary device never got through format detection, but since it
+	 * is a copy of the primary device, the format is exactly the same;
+	 * therefore, the detected layout can simply be copied.
+	 */
+	sec_priv->uses_cdl = prim_priv->uses_cdl;
+
+	/* re-enable device */
+	dasd_device_remove_stop_bits(primary, DASD_STOPPED_PPRC);
+	dasd_device_remove_stop_bits(secondary, DASD_STOPPED_PPRC);
+	dasd_schedule_device_bh(secondary);
+
+	return DASD_COPYPAIRSWAP_SUCCESS;
+}
+
+/*
+ * Perform Subsystem Function - Peer-to-Peer Remote Copy Extended Query
+ */
+static int dasd_eckd_query_pprc_status(struct dasd_device *device,
+				       struct dasd_pprc_data_sc4 *data)
+{
+	struct dasd_pprc_data_sc4 *pprc_data;
+	struct dasd_psf_prssd_data *prssdp;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int rc;
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */	+ 1 /* RSSD */,
+				   sizeof(*prssdp) + sizeof(*pprc_data) + 1,
+				   device, NULL);
+	if (IS_ERR(cqr)) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+				"Could not allocate query PPRC status request");
+		return PTR_ERR(cqr);
+	}
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->retries = 256;
+	cqr->expires = 10 * HZ;
+
+	/* Prepare for Read Subsystem Data */
+	prssdp = (struct dasd_psf_prssd_data *)cqr->data;
+	memset(prssdp, 0, sizeof(struct dasd_psf_prssd_data));
+	prssdp->order = PSF_ORDER_PRSSD;
+	prssdp->suborder = PSF_SUBORDER_PPRCEQ;
+	prssdp->varies[0] = PPRCEQ_SCOPE_4;
+	pprc_data = (struct dasd_pprc_data_sc4 *)(prssdp + 1);
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->count = sizeof(struct dasd_psf_prssd_data);
+	ccw->flags |= CCW_FLAG_CC;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->cda = virt_to_dma32(prssdp);
+
+	/* Read Subsystem Data - query host access */
+	ccw++;
+	ccw->cmd_code = DASD_ECKD_CCW_RSSD;
+	ccw->count = sizeof(*pprc_data);
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->cda = virt_to_dma32(pprc_data);
+
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	rc = dasd_sleep_on_interruptible(cqr);
+	if (rc == 0) {
+		*data = *pprc_data;
+	} else {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
+				"PPRC Extended Query failed with rc=%d\n",
+				rc);
+		rc = -EOPNOTSUPP;
+	}
+
+	dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+/*
+ * ECKD NOP - no operation
+ */
+static int dasd_eckd_nop(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int rc;
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1, 1, device, NULL);
+	if (IS_ERR(cqr)) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
+				"Could not allocate NOP request");
+		return PTR_ERR(cqr);
+	}
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->retries = 1;
+	cqr->expires = 10 * HZ;
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_NOP;
+	ccw->flags |= CCW_FLAG_SLI;
+
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	rc = dasd_sleep_on_interruptible(cqr);
+	if (rc != 0) {
+		DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
+				"NOP failed with rc=%d\n", rc);
+		rc = -EOPNOTSUPP;
+	}
+	dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+static int dasd_eckd_device_ping(struct dasd_device *device)
+{
+	return dasd_eckd_nop(device);
+}
+
+/*
+ * Perform Subsystem Function - CUIR response
+ */
+static int
+dasd_eckd_psf_cuir_response(struct dasd_device *device, int response,
+			    __u32 message_id, __u8 lpum)
+{
+	struct dasd_psf_cuir_response *psf_cuir;
+	int pos = pathmask_to_pos(lpum);
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int rc;
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */ ,
+				   sizeof(struct dasd_psf_cuir_response),
+				   device, NULL);
+
+	if (IS_ERR(cqr)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			   "Could not allocate PSF-CUIR request");
+		return PTR_ERR(cqr);
+	}
+
+	psf_cuir = (struct dasd_psf_cuir_response *)cqr->data;
+	psf_cuir->order = PSF_ORDER_CUIR_RESPONSE;
+	psf_cuir->cc = response;
+	psf_cuir->chpid = device->path[pos].chpid;
+	psf_cuir->message_id = message_id;
+	psf_cuir->cssid = device->path[pos].cssid;
+	psf_cuir->ssid = device->path[pos].ssid;
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->cda = virt_to_dma32(psf_cuir);
+	ccw->flags = CCW_FLAG_SLI;
+	ccw->count = sizeof(struct dasd_psf_cuir_response);
+
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->retries = 256;
+	cqr->expires = 10*HZ;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	set_bit(DASD_CQR_VERIFY_PATH, &cqr->flags);
+
+	rc = dasd_sleep_on(cqr);
+
+	dasd_sfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+/*
+ * return configuration data that is referenced by record selector
+ * if a record selector is specified or per default return the
+ * conf_data pointer for the path specified by lpum
+ */
+static struct dasd_conf_data *dasd_eckd_get_ref_conf(struct dasd_device *device,
+						     __u8 lpum,
+						     struct dasd_cuir_message *cuir)
+{
+	struct dasd_conf_data *conf_data;
+	int path, pos;
+
+	if (cuir->record_selector == 0)
+		goto out;
+	for (path = 0x80, pos = 0; path; path >>= 1, pos++) {
+		conf_data = device->path[pos].conf_data;
+		if (conf_data->gneq.record_selector ==
+		    cuir->record_selector)
+			return conf_data;
+	}
+out:
+	return device->path[pathmask_to_pos(lpum)].conf_data;
+}
+
+/*
+ * This function determines the scope of a reconfiguration request by
+ * analysing the path and device selection data provided in the CUIR request.
+ * Returns a path mask containing CUIR affected paths for the give device.
+ *
+ * If the CUIR request does not contain the required information return the
+ * path mask of the path the attention message for the CUIR request was reveived
+ * on.
+ */
+static int dasd_eckd_cuir_scope(struct dasd_device *device, __u8 lpum,
+				struct dasd_cuir_message *cuir)
+{
+	struct dasd_conf_data *ref_conf_data;
+	unsigned long bitmask = 0, mask = 0;
+	struct dasd_conf_data *conf_data;
+	unsigned int pos, path;
+	char *ref_gneq, *gneq;
+	char *ref_ned, *ned;
+	int tbcpm = 0;
+
+	/* if CUIR request does not specify the scope use the path
+	   the attention message was presented on */
+	if (!cuir->ned_map ||
+	    !(cuir->neq_map[0] | cuir->neq_map[1] | cuir->neq_map[2]))
+		return lpum;
+
+	/* get reference conf data */
+	ref_conf_data = dasd_eckd_get_ref_conf(device, lpum, cuir);
+	/* reference ned is determined by ned_map field */
+	pos = 8 - ffs(cuir->ned_map);
+	ref_ned = (char *)&ref_conf_data->neds[pos];
+	ref_gneq = (char *)&ref_conf_data->gneq;
+	/* transfer 24 bit neq_map to mask */
+	mask = cuir->neq_map[2];
+	mask |= cuir->neq_map[1] << 8;
+	mask |= cuir->neq_map[0] << 16;
+
+	for (path = 0; path < 8; path++) {
+		/* initialise data per path */
+		bitmask = mask;
+		conf_data = device->path[path].conf_data;
+		pos = 8 - ffs(cuir->ned_map);
+		ned = (char *) &conf_data->neds[pos];
+		/* compare reference ned and per path ned */
+		if (memcmp(ref_ned, ned, sizeof(*ned)) != 0)
+			continue;
+		gneq = (char *)&conf_data->gneq;
+		/* compare reference gneq and per_path gneq under
+		   24 bit mask where mask bit 0 equals byte 7 of
+		   the gneq and mask bit 24 equals byte 31 */
+		while (bitmask) {
+			pos = ffs(bitmask) - 1;
+			if (memcmp(&ref_gneq[31 - pos], &gneq[31 - pos], 1)
+			    != 0)
+				break;
+			clear_bit(pos, &bitmask);
+		}
+		if (bitmask)
+			continue;
+		/* device and path match the reference values
+		   add path to CUIR scope */
+		tbcpm |= 0x80 >> path;
+	}
+	return tbcpm;
+}
+
+static void dasd_eckd_cuir_notify_user(struct dasd_device *device,
+				       unsigned long paths, int action)
+{
+	int pos;
+
+	while (paths) {
+		/* get position of bit in mask */
+		pos = 8 - ffs(paths);
+		/* get channel path descriptor from this position */
+		if (action == CUIR_QUIESCE)
+			pr_warn("Service on the storage server caused path %x.%02x to go offline",
+				device->path[pos].cssid,
+				device->path[pos].chpid);
+		else if (action == CUIR_RESUME)
+			pr_info("Path %x.%02x is back online after service on the storage server",
+				device->path[pos].cssid,
+				device->path[pos].chpid);
+		clear_bit(7 - pos, &paths);
+	}
+}
+
+static int dasd_eckd_cuir_remove_path(struct dasd_device *device, __u8 lpum,
+				      struct dasd_cuir_message *cuir)
+{
+	unsigned long tbcpm;
+
+	tbcpm = dasd_eckd_cuir_scope(device, lpum, cuir);
+	/* nothing to do if path is not in use */
+	if (!(dasd_path_get_opm(device) & tbcpm))
+		return 0;
+	if (!(dasd_path_get_opm(device) & ~tbcpm)) {
+		/* no path would be left if the CUIR action is taken
+		   return error */
+		return -EINVAL;
+	}
+	/* remove device from operational path mask */
+	dasd_path_remove_opm(device, tbcpm);
+	dasd_path_add_cuirpm(device, tbcpm);
+	return tbcpm;
+}
+
+/*
+ * walk through all devices and build a path mask to quiesce them
+ * return an error if the last path to a device would be removed
+ *
+ * if only part of the devices are quiesced and an error
+ * occurs no onlining necessary, the storage server will
+ * notify the already set offline devices again
+ */
+static int dasd_eckd_cuir_quiesce(struct dasd_device *device, __u8 lpum,
+				  struct dasd_cuir_message *cuir)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct alias_pav_group *pavgroup, *tempgroup;
+	struct dasd_device *dev, *n;
+	unsigned long paths = 0;
+	unsigned long flags;
+	int tbcpm;
+
+	/* active devices */
+	list_for_each_entry_safe(dev, n, &private->lcu->active_devices,
+				 alias_list) {
+		spin_lock_irqsave(get_ccwdev_lock(dev->cdev), flags);
+		tbcpm = dasd_eckd_cuir_remove_path(dev, lpum, cuir);
+		spin_unlock_irqrestore(get_ccwdev_lock(dev->cdev), flags);
+		if (tbcpm < 0)
+			goto out_err;
+		paths |= tbcpm;
+	}
+	/* inactive devices */
+	list_for_each_entry_safe(dev, n, &private->lcu->inactive_devices,
+				 alias_list) {
+		spin_lock_irqsave(get_ccwdev_lock(dev->cdev), flags);
+		tbcpm = dasd_eckd_cuir_remove_path(dev, lpum, cuir);
+		spin_unlock_irqrestore(get_ccwdev_lock(dev->cdev), flags);
+		if (tbcpm < 0)
+			goto out_err;
+		paths |= tbcpm;
+	}
+	/* devices in PAV groups */
+	list_for_each_entry_safe(pavgroup, tempgroup,
+				 &private->lcu->grouplist, group) {
+		list_for_each_entry_safe(dev, n, &pavgroup->baselist,
+					 alias_list) {
+			spin_lock_irqsave(get_ccwdev_lock(dev->cdev), flags);
+			tbcpm = dasd_eckd_cuir_remove_path(dev, lpum, cuir);
+			spin_unlock_irqrestore(
+				get_ccwdev_lock(dev->cdev), flags);
+			if (tbcpm < 0)
+				goto out_err;
+			paths |= tbcpm;
+		}
+		list_for_each_entry_safe(dev, n, &pavgroup->aliaslist,
+					 alias_list) {
+			spin_lock_irqsave(get_ccwdev_lock(dev->cdev), flags);
+			tbcpm = dasd_eckd_cuir_remove_path(dev, lpum, cuir);
+			spin_unlock_irqrestore(
+				get_ccwdev_lock(dev->cdev), flags);
+			if (tbcpm < 0)
+				goto out_err;
+			paths |= tbcpm;
+		}
+	}
+	/* notify user about all paths affected by CUIR action */
+	dasd_eckd_cuir_notify_user(device, paths, CUIR_QUIESCE);
+	return 0;
+out_err:
+	return tbcpm;
+}
+
+static int dasd_eckd_cuir_resume(struct dasd_device *device, __u8 lpum,
+				 struct dasd_cuir_message *cuir)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct alias_pav_group *pavgroup, *tempgroup;
+	struct dasd_device *dev, *n;
+	unsigned long paths = 0;
+	int tbcpm;
+
+	/*
+	 * the path may have been added through a generic path event before
+	 * only trigger path verification if the path is not already in use
+	 */
+	list_for_each_entry_safe(dev, n,
+				 &private->lcu->active_devices,
+				 alias_list) {
+		tbcpm = dasd_eckd_cuir_scope(dev, lpum, cuir);
+		paths |= tbcpm;
+		if (!(dasd_path_get_opm(dev) & tbcpm)) {
+			dasd_path_add_tbvpm(dev, tbcpm);
+			dasd_schedule_device_bh(dev);
+		}
+	}
+	list_for_each_entry_safe(dev, n,
+				 &private->lcu->inactive_devices,
+				 alias_list) {
+		tbcpm = dasd_eckd_cuir_scope(dev, lpum, cuir);
+		paths |= tbcpm;
+		if (!(dasd_path_get_opm(dev) & tbcpm)) {
+			dasd_path_add_tbvpm(dev, tbcpm);
+			dasd_schedule_device_bh(dev);
+		}
+	}
+	/* devices in PAV groups */
+	list_for_each_entry_safe(pavgroup, tempgroup,
+				 &private->lcu->grouplist,
+				 group) {
+		list_for_each_entry_safe(dev, n,
+					 &pavgroup->baselist,
+					 alias_list) {
+			tbcpm = dasd_eckd_cuir_scope(dev, lpum, cuir);
+			paths |= tbcpm;
+			if (!(dasd_path_get_opm(dev) & tbcpm)) {
+				dasd_path_add_tbvpm(dev, tbcpm);
+				dasd_schedule_device_bh(dev);
+			}
+		}
+		list_for_each_entry_safe(dev, n,
+					 &pavgroup->aliaslist,
+					 alias_list) {
+			tbcpm = dasd_eckd_cuir_scope(dev, lpum, cuir);
+			paths |= tbcpm;
+			if (!(dasd_path_get_opm(dev) & tbcpm)) {
+				dasd_path_add_tbvpm(dev, tbcpm);
+				dasd_schedule_device_bh(dev);
+			}
+		}
+	}
+	/* notify user about all paths affected by CUIR action */
+	dasd_eckd_cuir_notify_user(device, paths, CUIR_RESUME);
+	return 0;
+}
+
+static void dasd_eckd_handle_cuir(struct dasd_device *device, void *messages,
+				 __u8 lpum)
+{
+	struct dasd_cuir_message *cuir = messages;
+	int response;
+
+	DBF_DEV_EVENT(DBF_WARNING, device,
+		      "CUIR request: %016llx %016llx %016llx %08x",
+		      ((u64 *)cuir)[0], ((u64 *)cuir)[1], ((u64 *)cuir)[2],
+		      ((u32 *)cuir)[3]);
+
+	if (cuir->code == CUIR_QUIESCE) {
+		/* quiesce */
+		if (dasd_eckd_cuir_quiesce(device, lpum, cuir))
+			response = PSF_CUIR_LAST_PATH;
+		else
+			response = PSF_CUIR_COMPLETED;
+	} else if (cuir->code == CUIR_RESUME) {
+		/* resume */
+		dasd_eckd_cuir_resume(device, lpum, cuir);
+		response = PSF_CUIR_COMPLETED;
+	} else
+		response = PSF_CUIR_NOT_SUPPORTED;
+
+	dasd_eckd_psf_cuir_response(device, response,
+				    cuir->message_id, lpum);
+	DBF_DEV_EVENT(DBF_WARNING, device,
+		      "CUIR response: %d on message ID %08x", response,
+		      cuir->message_id);
+	/* to make sure there is no attention left schedule work again */
+	device->discipline->check_attention(device, lpum);
+}
+
+static void dasd_eckd_oos_resume(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+	struct alias_pav_group *pavgroup, *tempgroup;
+	struct dasd_device *dev, *n;
+	unsigned long flags;
+
+	spin_lock_irqsave(&private->lcu->lock, flags);
+	list_for_each_entry_safe(dev, n, &private->lcu->active_devices,
+				 alias_list) {
+		if (dev->stopped & DASD_STOPPED_NOSPC)
+			dasd_generic_space_avail(dev);
+	}
+	list_for_each_entry_safe(dev, n, &private->lcu->inactive_devices,
+				 alias_list) {
+		if (dev->stopped & DASD_STOPPED_NOSPC)
+			dasd_generic_space_avail(dev);
+	}
+	/* devices in PAV groups */
+	list_for_each_entry_safe(pavgroup, tempgroup,
+				 &private->lcu->grouplist,
+				 group) {
+		list_for_each_entry_safe(dev, n, &pavgroup->baselist,
+					 alias_list) {
+			if (dev->stopped & DASD_STOPPED_NOSPC)
+				dasd_generic_space_avail(dev);
+		}
+		list_for_each_entry_safe(dev, n, &pavgroup->aliaslist,
+					 alias_list) {
+			if (dev->stopped & DASD_STOPPED_NOSPC)
+				dasd_generic_space_avail(dev);
+		}
+	}
+	spin_unlock_irqrestore(&private->lcu->lock, flags);
+}
+
+static void dasd_eckd_handle_oos(struct dasd_device *device, void *messages,
+				 __u8 lpum)
+{
+	struct dasd_oos_message *oos = messages;
+
+	switch (oos->code) {
+	case REPO_WARN:
+	case POOL_WARN:
+		dev_warn(&device->cdev->dev,
+			 "Extent pool usage has reached a critical value\n");
+		dasd_eckd_oos_resume(device);
+		break;
+	case REPO_EXHAUST:
+	case POOL_EXHAUST:
+		dev_warn(&device->cdev->dev,
+			 "Extent pool is exhausted\n");
+		break;
+	case REPO_RELIEVE:
+	case POOL_RELIEVE:
+		dev_info(&device->cdev->dev,
+			 "Extent pool physical space constraint has been relieved\n");
+		break;
+	}
+
+	/* In any case, update related data */
+	dasd_eckd_read_ext_pool_info(device);
+
+	/* to make sure there is no attention left schedule work again */
+	device->discipline->check_attention(device, lpum);
+}
+
+static void dasd_eckd_check_attention_work(struct work_struct *work)
+{
+	struct check_attention_work_data *data;
+	struct dasd_rssd_messages *messages;
+	struct dasd_device *device;
+	int rc;
+
+	data = container_of(work, struct check_attention_work_data, worker);
+	device = data->device;
+	messages = kzalloc_obj(*messages);
+	if (!messages) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "Could not allocate attention message buffer");
+		goto out;
+	}
+	rc = dasd_eckd_read_message_buffer(device, messages, data->lpum);
+	if (rc)
+		goto out;
+
+	if (messages->length == ATTENTION_LENGTH_CUIR &&
+	    messages->format == ATTENTION_FORMAT_CUIR)
+		dasd_eckd_handle_cuir(device, messages, data->lpum);
+	if (messages->length == ATTENTION_LENGTH_OOS &&
+	    messages->format == ATTENTION_FORMAT_OOS)
+		dasd_eckd_handle_oos(device, messages, data->lpum);
+
+out:
+	dasd_put_device(device);
+	kfree(messages);
+	kfree(data);
+}
+
+static int dasd_eckd_check_attention(struct dasd_device *device, __u8 lpum)
+{
+	struct check_attention_work_data *data;
+
+	data = kzalloc_obj(*data, GFP_ATOMIC);
+	if (!data)
+		return -ENOMEM;
+	INIT_WORK(&data->worker, dasd_eckd_check_attention_work);
+	dasd_get_device(device);
+	data->device = device;
+	data->lpum = lpum;
+	schedule_work(&data->worker);
+	return 0;
+}
+
+static int dasd_eckd_disable_hpf_path(struct dasd_device *device, __u8 lpum)
+{
+	if (~lpum & dasd_path_get_opm(device)) {
+		dasd_path_add_nohpfpm(device, lpum);
+		dasd_path_remove_opm(device, lpum);
+		dev_err(&device->cdev->dev,
+			"Channel path %02X lost HPF functionality and is disabled\n",
+			lpum);
+		return 1;
+	}
+	return 0;
+}
+
+static void dasd_eckd_disable_hpf_device(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	dev_err(&device->cdev->dev,
+		"High Performance FICON disabled\n");
+	private->fcx_max_data = 0;
+}
+
+static int dasd_eckd_hpf_enabled(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	if (!private)
+		return 0;
+
+	return private->fcx_max_data ? 1 : 0;
+}
+
+static void dasd_eckd_handle_hpf_error(struct dasd_device *device,
+				       struct irb *irb)
+{
+	struct dasd_eckd_private *private = device->private;
+
+	if (!private->fcx_max_data) {
+		/* sanity check for no HPF, the error makes no sense */
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "Trying to disable HPF for a non HPF device");
+		return;
+	}
+	if (irb->scsw.tm.sesq == SCSW_SESQ_DEV_NOFCX) {
+		dasd_eckd_disable_hpf_device(device);
+	} else if (irb->scsw.tm.sesq == SCSW_SESQ_PATH_NOFCX) {
+		if (dasd_eckd_disable_hpf_path(device, irb->esw.esw1.lpum))
+			return;
+		dasd_eckd_disable_hpf_device(device);
+		dasd_path_set_tbvpm(device,
+				  dasd_path_get_hpfpm(device));
+	}
+	/*
+	 * prevent that any new I/O ist started on the device and schedule a
+	 * requeue of existing requests
+	 */
+	dasd_device_set_stop_bits(device, DASD_STOPPED_NOT_ACC);
+	dasd_schedule_requeue(device);
+}
+
+static unsigned int dasd_eckd_max_sectors(struct dasd_block *block)
+{
+	if (block->base->features & DASD_FEATURE_USERAW) {
+		/*
+		 * the max_blocks value for raw_track access is 256
+		 * it is higher than the native ECKD value because we
+		 * only need one ccw per track
+		 * so the max_hw_sectors are
+		 * 2048 x 512B = 1024kB = 16 tracks
+		 */
+		return DASD_ECKD_MAX_BLOCKS_RAW << block->s2b_shift;
+	}
+
+	return DASD_ECKD_MAX_BLOCKS << block->s2b_shift;
+}
+
+/*
+ * Discard on ECKD releases space through RAS, which works on whole extents.
+ * Advertise extent granularity so the block layer only sends extent-aligned
+ * discards (avoiding partially specified extents), and only for volumes on ESE
+ * hardware. Non-ESE devices are left without discard limits.
+ */
+static void dasd_eckd_disc_limits(struct dasd_block *block,
+				  struct queue_limits *lim)
+{
+	struct dasd_device *device = block->base;
+	struct dasd_eckd_private *private = device->private;
+	unsigned int logical_block_size = block->bp_block;
+	unsigned int max_discard_sectors, max_bytes, ext_bytes;
+	int recs_per_trk, trks_per_cyl, ext_limit, ext_size;
+
+	if (!dasd_eckd_ese_capable(device) || dasd_eckd_ext_size(device) == 0)
+		return;
+
+	trks_per_cyl = private->rdc_data.trk_per_cyl;
+	recs_per_trk = recs_per_track(&private->rdc_data, 0, logical_block_size);
+
+	ext_size = dasd_eckd_ext_size(device);
+	ext_limit = min(private->real_cyl / ext_size, DASD_ECKD_RAS_EXTS_MAX);
+	ext_bytes = ext_size * trks_per_cyl * recs_per_trk * logical_block_size;
+	if (!ext_bytes)		/* malformed RDC data - leave discard unset */
+		return;
+	max_bytes = UINT_MAX - (UINT_MAX % ext_bytes);
+	if (max_bytes / ext_bytes > ext_limit)
+		max_bytes = ext_bytes * ext_limit;
+
+	max_discard_sectors = max_bytes / 512;
+
+	lim->max_hw_discard_sectors = max_discard_sectors;
+	/*
+	 * ext_bytes is the hardware extent size and is not a power of two, so
+	 * the block layer's power-of-two round_up()/round_down() alignment
+	 * helpers compute it only approximately. That is a hint, not a
+	 * correctness requirement: RAS safety is enforced in the CCW builder,
+	 * which rounds the range inward to whole extents and rejects a request
+	 * that covers no whole extent, so a misaligned range is never
+	 * over-released. At worst a few sub-extent discards are declined.
+	 */
+	lim->discard_granularity = ext_bytes;
+}
+
+static struct ccw_driver dasd_eckd_driver = {
+	.driver = {
+		.name	= "dasd-eckd",
+		.owner	= THIS_MODULE,
+		.dev_groups = dasd_dev_groups,
+	},
+	.ids	     = dasd_eckd_ids,
+	.probe	     = dasd_eckd_probe,
+	.remove      = dasd_generic_remove,
+	.set_offline = dasd_generic_set_offline,
+	.set_online  = dasd_eckd_set_online,
+	.notify      = dasd_generic_notify,
+	.path_event  = dasd_generic_path_event,
+	.shutdown    = dasd_generic_shutdown,
+	.uc_handler  = dasd_generic_uc_handler,
+	.int_class   = IRQIO_DAS,
+};
+
+static struct dasd_discipline dasd_eckd_discipline = {
+	.owner = THIS_MODULE,
+	.name = "ECKD",
+	.ebcname = "ECKD",
+	.disc_limits = dasd_eckd_disc_limits,
+	.check_device = dasd_eckd_check_characteristics,
+	.uncheck_device = dasd_eckd_uncheck_device,
+	.do_analysis = dasd_eckd_do_analysis,
+	.pe_handler = dasd_eckd_pe_handler,
+	.basic_to_ready = dasd_eckd_basic_to_ready,
+	.online_to_ready = dasd_eckd_online_to_ready,
+	.basic_to_known = dasd_eckd_basic_to_known,
+	.max_sectors = dasd_eckd_max_sectors,
+	.fill_geometry = dasd_eckd_fill_geometry,
+	.start_IO = dasd_start_IO,
+	.term_IO = dasd_term_IO,
+	.handle_terminated_request = dasd_eckd_handle_terminated_request,
+	.format_device = dasd_eckd_format_device,
+	.check_device_format = dasd_eckd_check_device_format,
+	.erp_action = dasd_eckd_erp_action,
+	.erp_postaction = dasd_eckd_erp_postaction,
+	.check_for_device_change = dasd_eckd_check_for_device_change,
+	.build_cp = dasd_eckd_build_alias_cp,
+	.free_cp = dasd_eckd_free_alias_cp,
+	.dump_sense = dasd_eckd_dump_sense,
+	.dump_sense_dbf = dasd_eckd_dump_sense_dbf,
+	.fill_info = dasd_eckd_fill_info,
+	.ioctl = dasd_eckd_ioctl,
+	.reload = dasd_eckd_reload_device,
+	.get_uid = dasd_eckd_get_uid,
+	.kick_validate = dasd_eckd_kick_validate_server,
+	.check_attention = dasd_eckd_check_attention,
+	.host_access_count = dasd_eckd_host_access_count,
+	.hosts_print = dasd_hosts_print,
+	.handle_hpf_error = dasd_eckd_handle_hpf_error,
+	.disable_hpf = dasd_eckd_disable_hpf_device,
+	.hpf_enabled = dasd_eckd_hpf_enabled,
+	.reset_path = dasd_eckd_reset_path,
+	.is_ese = dasd_eckd_is_ese,
+	.ese_capable = dasd_eckd_ese_capable,
+	.on_demand_format = dasd_eckd_on_demand_format,
+	.space_allocated = dasd_eckd_space_allocated,
+	.space_configured = dasd_eckd_space_configured,
+	.logical_capacity = dasd_eckd_logical_capacity,
+	.release_space = dasd_eckd_release_space,
+	.ext_pool_id = dasd_eckd_ext_pool_id,
+	.ext_size = dasd_eckd_ext_size,
+	.ext_pool_cap_at_warnlevel = dasd_eckd_ext_pool_cap_at_warnlevel,
+	.ext_pool_warn_thrshld = dasd_eckd_ext_pool_warn_thrshld,
+	.ext_pool_oos = dasd_eckd_ext_pool_oos,
+	.ext_pool_exhaust = dasd_eckd_ext_pool_exhaust,
+	.ese_format = dasd_eckd_ese_format,
+	.ese_read = dasd_eckd_ese_read,
+	.pprc_status = dasd_eckd_query_pprc_status,
+	.pprc_enabled = dasd_eckd_pprc_enabled,
+	.copy_pair_swap = dasd_eckd_copy_pair_swap,
+	.device_ping = dasd_eckd_device_ping,
+};
+
+static int __init
+dasd_eckd_init(void)
+{
+	int ret;
+
+	ASCEBC(dasd_eckd_discipline.ebcname, 4);
+	dasd_reserve_req = kmalloc_obj(*dasd_reserve_req, GFP_KERNEL | GFP_DMA);
+	if (!dasd_reserve_req)
+		return -ENOMEM;
+	dasd_vol_info_req = kmalloc_obj(*dasd_vol_info_req,
+					GFP_KERNEL | GFP_DMA);
+	if (!dasd_vol_info_req) {
+		kfree(dasd_reserve_req);
+		return -ENOMEM;
+	}
+	pe_handler_worker = kmalloc_obj(*pe_handler_worker,
+					GFP_KERNEL | GFP_DMA);
+	if (!pe_handler_worker) {
+		kfree(dasd_reserve_req);
+		kfree(dasd_vol_info_req);
+		return -ENOMEM;
+	}
+	rawpadpage = (void *)__get_free_page(GFP_KERNEL);
+	if (!rawpadpage) {
+		kfree(pe_handler_worker);
+		kfree(dasd_reserve_req);
+		kfree(dasd_vol_info_req);
+		return -ENOMEM;
+	}
+	ret = ccw_driver_register(&dasd_eckd_driver);
+	if (!ret)
+		wait_for_device_probe();
+	else {
+		kfree(pe_handler_worker);
+		kfree(dasd_reserve_req);
+		kfree(dasd_vol_info_req);
+		free_page((unsigned long)rawpadpage);
+	}
+	return ret;
+}
+
+static void __exit
+dasd_eckd_cleanup(void)
+{
+	ccw_driver_unregister(&dasd_eckd_driver);
+	kfree(pe_handler_worker);
+	kfree(dasd_reserve_req);
+	free_page((unsigned long)rawpadpage);
+}
+
+module_init(dasd_eckd_init);
+module_exit(dasd_eckd_cleanup);
diff --git a/drivers/s390/block/dasd_eckd.h b/drivers/s390/block/dasd_eckd.h
new file mode 100644
index 000000000..8e6f09e9c
--- /dev/null
+++ b/drivers/s390/block/dasd_eckd.h
@@ -0,0 +1,777 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2000
+ *
+ */
+
+#ifndef DASD_ECKD_H
+#define DASD_ECKD_H
+
+/*****************************************************************************
+ * SECTION: CCW Definitions
+ ****************************************************************************/
+#define DASD_ECKD_CCW_NOP		 0x03
+#define DASD_ECKD_CCW_WRITE		 0x05
+#define DASD_ECKD_CCW_READ		 0x06
+#define DASD_ECKD_CCW_WRITE_HOME_ADDRESS 0x09
+#define DASD_ECKD_CCW_READ_HOME_ADDRESS	 0x0a
+#define DASD_ECKD_CCW_WRITE_KD		 0x0d
+#define DASD_ECKD_CCW_READ_KD		 0x0e
+#define DASD_ECKD_CCW_ERASE		 0x11
+#define DASD_ECKD_CCW_READ_COUNT	 0x12
+#define DASD_ECKD_CCW_SLCK		 0x14
+#define DASD_ECKD_CCW_WRITE_RECORD_ZERO	 0x15
+#define DASD_ECKD_CCW_READ_RECORD_ZERO	 0x16
+#define DASD_ECKD_CCW_WRITE_CKD		 0x1d
+#define DASD_ECKD_CCW_READ_CKD		 0x1e
+#define DASD_ECKD_CCW_PSF		 0x27
+#define DASD_ECKD_CCW_SNID		 0x34
+#define DASD_ECKD_CCW_RSSD		 0x3e
+#define DASD_ECKD_CCW_LOCATE_RECORD	 0x47
+#define DASD_ECKD_CCW_LOCATE_RECORD_EXT	 0x4b
+#define DASD_ECKD_CCW_SNSS		 0x54
+#define DASD_ECKD_CCW_DEFINE_EXTENT	 0x63
+#define DASD_ECKD_CCW_WRITE_MT		 0x85
+#define DASD_ECKD_CCW_READ_MT		 0x86
+#define DASD_ECKD_CCW_WRITE_KD_MT	 0x8d
+#define DASD_ECKD_CCW_READ_KD_MT	 0x8e
+#define DASD_ECKD_CCW_READ_COUNT_MT	 0x92
+#define DASD_ECKD_CCW_RELEASE		 0x94
+#define DASD_ECKD_CCW_WRITE_FULL_TRACK	 0x95
+#define DASD_ECKD_CCW_READ_CKD_MT	 0x9e
+#define DASD_ECKD_CCW_WRITE_CKD_MT	 0x9d
+#define DASD_ECKD_CCW_WRITE_TRACK_DATA	 0xA5
+#define DASD_ECKD_CCW_READ_TRACK_DATA	 0xA6
+#define DASD_ECKD_CCW_RESERVE		 0xB4
+#define DASD_ECKD_CCW_READ_TRACK	 0xDE
+#define DASD_ECKD_CCW_PFX		 0xE7
+#define DASD_ECKD_CCW_PFX_READ		 0xEA
+#define DASD_ECKD_CCW_RSCK		 0xF9
+#define DASD_ECKD_CCW_RCD		 0xFA
+#define DASD_ECKD_CCW_DSO		 0xF7
+
+/* Define Subsystem Function / Orders */
+#define DSO_ORDER_RAS			 0x81
+
+/*
+ * Perform Subsystem Function / Orders
+ */
+#define PSF_ORDER_PRSSD			 0x18
+#define PSF_ORDER_CUIR_RESPONSE		 0x1A
+#define PSF_ORDER_SSC			 0x1D
+
+/*
+ * Perform Subsystem Function / Sub-Orders
+ */
+#define PSF_SUBORDER_QHA		 0x1C /* Query Host Access */
+#define PSF_SUBORDER_PPRCEQ		 0x50 /* PPRC Extended Query */
+#define PSF_SUBORDER_VSQ		 0x52 /* Volume Storage Query */
+#define PSF_SUBORDER_LCQ		 0x53 /* Logical Configuration Query */
+
+/*
+ * PPRC Extended Query Scopes
+ */
+#define PPRCEQ_SCOPE_4			 0x04 /* Scope 4 for PPRC Extended Query */
+
+/*
+ * CUIR response condition codes
+ */
+#define PSF_CUIR_INVALID		 0x00
+#define PSF_CUIR_COMPLETED		 0x01
+#define PSF_CUIR_NOT_SUPPORTED		 0x02
+#define PSF_CUIR_ERROR_IN_REQ		 0x03
+#define PSF_CUIR_DENIED			 0x04
+#define PSF_CUIR_LAST_PATH		 0x05
+#define PSF_CUIR_DEVICE_ONLINE		 0x06
+#define PSF_CUIR_VARY_FAILURE		 0x07
+#define PSF_CUIR_SOFTWARE_FAILURE	 0x08
+#define PSF_CUIR_NOT_RECOGNIZED		 0x09
+
+/*
+ * CUIR codes
+ */
+#define CUIR_QUIESCE			 0x01
+#define CUIR_RESUME			 0x02
+
+/*
+ * Out-of-space (OOS) Codes
+ */
+#define REPO_WARN			 0x01
+#define REPO_EXHAUST			 0x02
+#define POOL_WARN			 0x03
+#define POOL_EXHAUST			 0x04
+#define REPO_RELIEVE			 0x05
+#define POOL_RELIEVE			 0x06
+
+/*
+ * attention message definitions
+ */
+#define ATTENTION_LENGTH_CUIR		 0x0e
+#define ATTENTION_FORMAT_CUIR		 0x01
+#define ATTENTION_LENGTH_OOS		 0x10
+#define ATTENTION_FORMAT_OOS		 0x06
+
+#define DASD_ECKD_PG_GROUPED		 0x10
+
+/*
+ * Size that is reported for large volumes in the old 16-bit no_cyl field
+ */
+#define LV_COMPAT_CYL 0xFFFE
+
+
+#define FCX_MAX_DATA_FACTOR 65536
+#define DASD_ECKD_RCD_DATA_SIZE 256
+
+#define DASD_ECKD_PATH_THRHLD		 256
+#define DASD_ECKD_PATH_INTERVAL		 300
+
+/*
+ * Maximum number of blocks to be chained
+ */
+#define DASD_ECKD_MAX_BLOCKS		 180
+#define DASD_ECKD_MAX_BLOCKS_RAW	 256
+
+/*****************************************************************************
+ * SECTION: Type Definitions
+ ****************************************************************************/
+
+struct eckd_count {
+	__u16 cyl;
+	__u16 head;
+	__u8 record;
+	__u8 kl;
+	__u16 dl;
+} __attribute__ ((packed));
+
+struct eckd_r0 {
+	struct eckd_count count;
+	__u8 data[8];
+} __packed;
+
+/*
+ * Extended Address Volume track address: the head field carries the actual
+ * head in its low-order 4 bits; the cylinder bits that do not fit the 16-bit
+ * cyl field are shifted in just above them.
+ */
+#define DASD_EAV_CYL_HI_SHIFT	16	/* cylinder bits beyond the 16-bit cyl field */
+#define DASD_EAV_HEAD_HI_SHIFT	4	/* head occupies the low-order 4 bits of head */
+
+/*
+ * On-disk DASD format label.
+ *
+ * Written into track 0, head 0, record 4 (R4 - the first non-special CDL
+ * record) as part of the same channel program that formats track 0, so it is
+ * stored atomically with the track: either both the track format and the label
+ * make it to disk or neither does. Its presence with a valid magic therefore
+ * marks a completed format and can be used for format detection.
+ *
+ * The structure is exactly the smallest supported block size (512 bytes) so it
+ * always fits into a single record.
+ * For larger block sizes the rest of the record is zero padded.
+ * The magic together with the version is used to recognise a valid label.
+ */
+#define DASD_ESE_LABEL_MAGIC	0xC4C1E2C4C6D4E3F1ULL	/* EBCDIC "DASDFMT1" */
+#define DASD_ESE_LABEL_VERSION	1
+
+/* dasd_format_label.flags */
+#define DASD_ESE_LABEL_F_ESE	0x00000001	/* volume is extent space efficient */
+#define DASD_ESE_LABEL_F_QUICK	0x00000002	/* quick (space released) format */
+#define DASD_ESE_LABEL_F_FULL	0x00000004	/* full format */
+
+struct dasd_format_label {
+	__u64 magic;			/* DASD_ESE_LABEL_MAGIC */
+	__u32 version;			/* DASD_ESE_LABEL_VERSION */
+	__u32 flags;			/* DASD_ESE_LABEL_F_* */
+	__u32 blksize;			/* block size the volume was formatted with */
+	__u32 reserved0;
+	__u64 format_tod;		/* TOD clock at format time */
+	__u8  kernel_version[64];	/* NUL terminated kernel release (uname -r) */
+	__u8  reserved[416];		/* pad the struct to 512 bytes */
+} __packed;
+
+struct ch_t {
+	__u16 cyl;
+	__u16 head;
+} __attribute__ ((packed));
+
+struct chr_t {
+	__u16 cyl;
+	__u16 head;
+	__u8 record;
+} __attribute__ ((packed));
+
+struct DE_eckd_data {
+	struct {
+		unsigned char perm:2;	/* Permissions on this extent */
+		unsigned char reserved:1;
+		unsigned char seek:2;	/* Seek control */
+		unsigned char auth:2;	/* Access authorization */
+		unsigned char pci:1;	/* PCI Fetch mode */
+	} __attribute__ ((packed)) mask;
+	struct {
+		unsigned char mode:2;	/* Architecture mode */
+		unsigned char ckd:1;	/* CKD Conversion */
+		unsigned char operation:3;	/* Operation mode */
+		unsigned char cfw:1;	/* Cache fast write */
+		unsigned char dfw:1;	/* DASD fast write */
+	} __attribute__ ((packed)) attributes;
+	__u16 blk_size;		/* Blocksize */
+	__u16 fast_write_id;
+	__u8 ga_additional;	/* Global Attributes Additional */
+	__u8 ga_extended;	/* Global Attributes Extended	*/
+	struct ch_t beg_ext;
+	struct ch_t end_ext;
+	unsigned long ep_sys_time; /* Ext Parameter - System Time Stamp */
+	__u8 ep_format;        /* Extended Parameter format byte       */
+	__u8 ep_prio;          /* Extended Parameter priority I/O byte */
+	__u8 ep_reserved1;     /* Extended Parameter Reserved	       */
+	__u8 ep_rec_per_track; /* Number of records on a track	       */
+	__u8 ep_reserved[4];   /* Extended Parameter Reserved	       */
+} __attribute__ ((packed));
+
+struct LO_eckd_data {
+	struct {
+		unsigned char orientation:2;
+		unsigned char operation:6;
+	} __attribute__ ((packed)) operation;
+	struct {
+		unsigned char last_bytes_used:1;
+		unsigned char reserved:6;
+		unsigned char read_count_suffix:1;
+	} __attribute__ ((packed)) auxiliary;
+	__u8 unused;
+	__u8 count;
+	struct ch_t seek_addr;
+	struct chr_t search_arg;
+	__u8 sector;
+	__u16 length;
+} __attribute__ ((packed));
+
+struct LRE_eckd_data {
+	struct {
+		unsigned char orientation:2;
+		unsigned char operation:6;
+	} __attribute__ ((packed)) operation;
+	struct {
+		unsigned char length_valid:1;
+		unsigned char length_scope:1;
+		unsigned char imbedded_ccw_valid:1;
+		unsigned char check_bytes:2;
+		unsigned char imbedded_count_valid:1;
+		unsigned char reserved:1;
+		unsigned char read_count_suffix:1;
+	} __attribute__ ((packed)) auxiliary;
+	__u8 imbedded_ccw;
+	__u8 count;
+	struct ch_t seek_addr;
+	struct chr_t search_arg;
+	__u8 sector;
+	__u16 length;
+	__u8 imbedded_count;
+	__u8 extended_operation;
+	__u16 extended_parameter_length;
+	__u8 extended_parameter[];
+} __attribute__ ((packed));
+
+/* Prefix data for format 0x00 and 0x01 */
+struct PFX_eckd_data {
+	unsigned char format;
+	struct {
+		unsigned char define_extent:1;
+		unsigned char time_stamp:1;
+		unsigned char verify_base:1;
+		unsigned char hyper_pav:1;
+		unsigned char reserved:4;
+	} __attribute__ ((packed)) validity;
+	__u8 base_address;
+	__u8 aux;
+	__u8 base_lss;
+	__u8 reserved[7];
+	struct DE_eckd_data define_extent;
+	struct LRE_eckd_data locate_record;
+} __attribute__ ((packed));
+
+struct dasd_eckd_characteristics {
+	__u16 cu_type;
+	struct {
+		unsigned char support:2;
+		unsigned char async:1;
+		unsigned char reserved:1;
+		unsigned char cache_info:1;
+		unsigned char model:3;
+	} __attribute__ ((packed)) cu_model;
+	__u16 dev_type;
+	__u8 dev_model;
+	struct {
+		unsigned char mult_burst:1;
+		unsigned char RT_in_LR:1;
+		unsigned char reserved1:1;
+		unsigned char RD_IN_LR:1;
+		unsigned char reserved2:4;
+		unsigned char reserved3:8;
+		unsigned char defect_wr:1;
+		unsigned char XRC_supported:1;
+		unsigned char PPRC_enabled:1;
+		unsigned char striping:1;
+		unsigned char reserved5:4;
+		unsigned char cfw:1;
+		unsigned char reserved6:2;
+		unsigned char cache:1;
+		unsigned char dual_copy:1;
+		unsigned char dfw:1;
+		unsigned char reset_alleg:1;
+		unsigned char sense_down:1;
+	} __attribute__ ((packed)) facilities;
+	__u8 dev_class;
+	__u8 unit_type;
+	__u16 no_cyl;
+	__u16 trk_per_cyl;
+	__u8 sec_per_trk;
+	__u8 byte_per_track[3];
+	__u16 home_bytes;
+	__u8 formula;
+	union {
+		struct {
+			__u8 f1;
+			__u16 f2;
+			__u16 f3;
+		} __attribute__ ((packed)) f_0x01;
+		struct {
+			__u8 f1;
+			__u8 f2;
+			__u8 f3;
+			__u8 f4;
+			__u8 f5;
+		} __attribute__ ((packed)) f_0x02;
+	} __attribute__ ((packed)) factors;
+	__u16 first_alt_trk;
+	__u16 no_alt_trk;
+	__u16 first_dia_trk;
+	__u16 no_dia_trk;
+	__u16 first_sup_trk;
+	__u16 no_sup_trk;
+	__u8 MDR_ID;
+	__u8 OBR_ID;
+	__u8 director;
+	__u8 rd_trk_set;
+	__u16 max_rec_zero;
+	__u8 reserved1;
+	__u8 RWANY_in_LR;
+	__u8 factor6;
+	__u8 factor7;
+	__u8 factor8;
+	__u8 reserved2[3];
+	__u8 reserved3[6];
+	__u32 long_no_cyl;
+} __attribute__ ((packed));
+
+/* elements of the configuration data */
+struct dasd_ned {
+	struct {
+		__u8 identifier:2;
+		__u8 token_id:1;
+		__u8 sno_valid:1;
+		__u8 subst_sno:1;
+		__u8 recNED:1;
+		__u8 emuNED:1;
+		__u8 reserved:1;
+	} __attribute__ ((packed)) flags;
+	__u8 descriptor;
+	__u8 dev_class;
+	__u8 reserved;
+	__u8 dev_type[6];
+	__u8 dev_model[3];
+	__u8 HDA_manufacturer[3];
+	struct {
+		__u8 HDA_location[2];
+		__u8 HDA_seqno[12];
+	} serial;
+	__u8 ID;
+	__u8 unit_addr;
+} __attribute__ ((packed));
+
+struct dasd_sneq {
+	struct {
+		__u8 identifier:2;
+		__u8 reserved:6;
+	} __attribute__ ((packed)) flags;
+	__u8 res1;
+	__u16 format;
+	__u8 res2[4];		/* byte  4- 7 */
+	__u8 sua_flags;		/* byte  8    */
+	__u8 base_unit_addr;	/* byte  9    */
+	__u8 res3[22];		/* byte 10-31 */
+} __attribute__ ((packed));
+
+struct vd_sneq {
+	struct {
+		__u8 identifier:2;
+		__u8 reserved:6;
+	} __attribute__ ((packed)) flags;
+	__u8 res1;
+	__u16 format;
+	__u8 res2[4];	/* byte  4- 7 */
+	__u8 uit[16];	/* byte  8-23 */
+	__u8 res3[8];	/* byte 24-31 */
+} __attribute__ ((packed));
+
+struct dasd_gneq {
+	struct {
+		__u8 identifier:2;
+		__u8 reserved:6;
+	} __attribute__ ((packed)) flags;
+	__u8 record_selector;
+	__u8 reserved[4];
+	struct {
+		__u8 value:2;
+		__u8 number:6;
+	} __attribute__ ((packed)) timeout;
+	__u8 reserved3;
+	__u16 subsystemID;
+	__u8 reserved2[22];
+} __attribute__ ((packed));
+
+struct dasd_rssd_features {
+	char feature[256];
+} __attribute__((packed));
+
+struct dasd_rssd_messages {
+	__u16 length;
+	__u8 format;
+	__u8 code;
+	__u32 message_id;
+	__u8 flags;
+	char messages[4087];
+} __packed;
+
+/*
+ * Read Subsystem Data - Volume Storage Query
+ */
+struct dasd_rssd_vsq {
+	struct {
+		__u8 tse:1;
+		__u8 space_not_available:1;
+		__u8 ese:1;
+		__u8 unused:5;
+	} __packed vol_info;
+	__u8 unused1;
+	__u16 extent_pool_id;
+	__u8 warn_cap_limit;
+	__u8 warn_cap_guaranteed;
+	__u16 unused2;
+	__u32 limit_capacity;
+	__u32 guaranteed_capacity;
+	__u32 space_allocated;
+	__u32 space_configured;
+	__u32 logical_capacity;
+} __packed;
+
+/*
+ * Extent Pool Summary
+ */
+struct dasd_ext_pool_sum {
+	__u16 pool_id;
+	__u8 repo_warn_thrshld;
+	__u8 warn_thrshld;
+	struct {
+		__u8 type:1;			/* 0 - CKD / 1 - FB */
+		__u8 track_space_efficient:1;
+		__u8 extent_space_efficient:1;
+		__u8 standard_volume:1;
+		__u8 extent_size_valid:1;
+		__u8 capacity_at_warnlevel:1;
+		__u8 pool_oos:1;
+		__u8 unused0:1;
+		__u8 unused1;
+	} __packed flags;
+	struct {
+		__u8 reserved0:1;
+		__u8 size_1G:1;
+		__u8 reserved1:5;
+		__u8 size_16M:1;
+	} __packed extent_size;
+	__u8 unused;
+} __packed;
+
+/*
+ * Read Subsystem Data-Response - Logical Configuration Query - Header
+ */
+struct dasd_rssd_lcq {
+	__u16 data_length;		/* Length of data returned */
+	__u16 pool_count;		/* Count of extent pools returned - Max: 448 */
+	struct {
+		__u8 pool_info_valid:1;	/* Detailed Information valid */
+		__u8 pool_id_volume:1;
+		__u8 pool_id_cec:1;
+		__u8 unused0:5;
+		__u8 unused1;
+	} __packed header_flags;
+	char sfi_type[6];		/* Storage Facility Image Type (EBCDIC) */
+	char sfi_model[3];		/* Storage Facility Image Model (EBCDIC) */
+	__u8 sfi_seq_num[10];		/* Storage Facility Image Sequence Number */
+	__u8 reserved[7];
+	struct dasd_ext_pool_sum ext_pool_sum[448];
+} __packed;
+
+struct dasd_oos_message {
+	__u16 length;
+	__u8 format;
+	__u8 code;
+	__u8 percentage_empty;
+	__u8 reserved;
+	__u16 ext_pool_id;
+	__u16 token;
+	__u8 unused[6];
+} __packed;
+
+struct dasd_cuir_message {
+	__u16 length;
+	__u8 format;
+	__u8 code;
+	__u32 message_id;
+	__u8 flags;
+	__u8 neq_map[3];
+	__u8 ned_map;
+	__u8 record_selector;
+} __packed;
+
+struct dasd_psf_cuir_response {
+	__u8 order;
+	__u8 flags;
+	__u8 cc;
+	__u8 chpid;
+	__u16 device_nr;
+	__u16 reserved;
+	__u32 message_id;
+	__u64 system_id;
+	__u8 cssid;
+	__u8 ssid;
+} __packed;
+
+struct dasd_ckd_path_group_entry {
+	__u8 status_flags;
+	__u8 pgid[11];
+	__u8 sysplex_name[8];
+	__u32 timestamp;
+	__u32 cylinder;
+	__u8 reserved[4];
+} __packed;
+
+struct dasd_ckd_host_information {
+	__u8 access_flags;
+	__u8 entry_size;
+	__u16 entry_count;
+	__u8 entry[16390];
+} __packed;
+
+struct dasd_psf_query_host_access {
+	__u8 access_flag;
+	__u8 version;
+	__u16 CKD_length;
+	__u16 SCSI_length;
+	__u8 unused[10];
+	__u8 host_access_information[16394];
+} __packed;
+
+/*
+ * Perform Subsystem Function - Prepare for Read Subsystem Data
+ */
+struct dasd_psf_prssd_data {
+	unsigned char order;
+	unsigned char flags;
+	unsigned char reserved1;
+	unsigned char reserved2;
+	unsigned char lss;
+	unsigned char volume;
+	unsigned char suborder;
+	unsigned char varies[5];
+} __attribute__ ((packed));
+
+/*
+ * Perform Subsystem Function - Set Subsystem Characteristics
+ */
+struct dasd_psf_ssc_data {
+	unsigned char order;
+	unsigned char flags;
+	unsigned char cu_type[4];
+	unsigned char suborder;
+	unsigned char reserved[59];
+} __attribute__((packed));
+
+/* Maximum number of extents for a single Release Allocated Space command */
+#define DASD_ECKD_RAS_EXTS_MAX		110U
+
+struct dasd_dso_ras_ext_range {
+	struct ch_t beg_ext;
+	struct ch_t end_ext;
+} __packed;
+
+/*
+ * Define Subsystem Operation - Release Allocated Space
+ */
+struct dasd_dso_ras_data {
+	__u8 order;
+	struct {
+		__u8 message:1;		/* Must be zero */
+		__u8 reserved1:2;
+		__u8 vol_type:1;	/* 0 - CKD/FBA, 1 - FB */
+		__u8 reserved2:4;
+	} __packed flags;
+	/* Operation Flags to specify scope */
+	struct {
+		__u8 reserved1:2;
+		/* Release Space by Extent */
+		__u8 by_extent:1;	/* 0 - entire volume, 1 - specified extents */
+		__u8 guarantee_init:1;
+		__u8 force_release:1;	/* Internal - will be ignored */
+		__u16 reserved2:11;
+	} __packed op_flags;
+	__u8 lss;
+	__u8 dev_addr;
+	__u32 reserved1;
+	__u8 reserved2[10];
+	__u16 nr_exts;			/* Defines number of ext_scope - max 110 */
+	__u16 reserved3;
+} __packed;
+
+
+/*
+ * some structures and definitions for alias handling
+ */
+struct dasd_unit_address_configuration {
+	struct {
+		char ua_type;
+		char base_ua;
+	} unit[256];
+} __attribute__((packed));
+
+
+#define MAX_DEVICES_PER_LCU 256
+
+/* flags on the LCU  */
+#define NEED_UAC_UPDATE  0x01
+#define UPDATE_PENDING	0x02
+
+enum pavtype {NO_PAV, BASE_PAV, HYPER_PAV};
+
+
+struct alias_root {
+	struct list_head serverlist;
+	spinlock_t lock;
+};
+
+struct alias_server {
+	struct list_head server;
+	struct dasd_uid uid;
+	struct list_head lculist;
+};
+
+struct summary_unit_check_work_data {
+	char reason;
+	struct dasd_device *device;
+	struct work_struct worker;
+};
+
+struct read_uac_work_data {
+	struct dasd_device *device;
+	struct delayed_work dwork;
+};
+
+struct alias_lcu {
+	struct list_head lcu;
+	struct dasd_uid uid;
+	enum pavtype pav;
+	char flags;
+	spinlock_t lock;
+	struct list_head grouplist;
+	struct list_head active_devices;
+	struct list_head inactive_devices;
+	struct dasd_unit_address_configuration *uac;
+	struct summary_unit_check_work_data suc_data;
+	struct read_uac_work_data ruac_data;
+	struct dasd_ccw_req *rsu_cqr;
+	struct completion lcu_setup;
+};
+
+struct alias_pav_group {
+	struct list_head group;
+	struct dasd_uid uid;
+	struct alias_lcu *lcu;
+	struct list_head baselist;
+	struct list_head aliaslist;
+	struct dasd_device *next;
+};
+
+struct dasd_conf_data {
+	struct dasd_ned neds[5];
+	u8 reserved[64];
+	struct dasd_gneq gneq;
+} __packed;
+
+struct dasd_conf {
+	u8 *data;
+	int len;
+	/* pointers to specific parts in the conf_data */
+	struct dasd_ned *ned;
+	struct dasd_sneq *sneq;
+	struct vd_sneq *vdsneq;
+	struct dasd_gneq *gneq;
+};
+
+struct dasd_eckd_private {
+	struct dasd_eckd_characteristics rdc_data;
+	struct dasd_conf conf;
+
+	struct eckd_count count_area[5];
+	int init_cqr_status;
+	int uses_cdl;
+	struct attrib_data_t attrib;	/* e.g. cache operations */
+	struct dasd_rssd_features features;
+	struct dasd_rssd_vsq vsq;
+	struct dasd_ext_pool_sum eps;
+	u32 real_cyl;
+
+	/* alias management */
+	struct dasd_uid uid;
+	/*
+	 * Cached copies of conf.ned->ID (the LSS) and conf.ned->unit_addr,
+	 * refreshed under ccwdev_lock. Kept outside uid because create_uid()
+	 * memsets uid before repopulating it, which would expose a transient
+	 * zero to the lockless CCW-build readers.
+	 */
+	__u8 ned_lss;
+	__u8 ned_ua;
+	struct alias_pav_group *pavgroup;
+	struct alias_lcu *lcu;
+	int count;
+
+	u32 fcx_max_data;
+	char suc_reason;
+	/*
+	 * Set when the whole volume's space was released (full RAS); consumed by
+	 * the next format to mark the on-disk label as a quick (vs full) format.
+	 */
+	int ese_format_quick;
+	/*
+	 * Cached on-disk format label (R4), read at online and refreshed on
+	 * format. When valid, is_ese() is derived from it; otherwise it falls
+	 * back to the hardware ESE field (vsq.vol_info.ese).
+	 */
+	struct dasd_format_label ese_label;
+	bool ese_label_valid;
+};
+
+
+
+int dasd_alias_make_device_known_to_lcu(struct dasd_device *);
+void dasd_alias_disconnect_device_from_lcu(struct dasd_device *);
+int dasd_alias_add_device(struct dasd_device *);
+int dasd_alias_remove_device(struct dasd_device *);
+struct dasd_device *dasd_alias_get_start_dev(struct dasd_device *);
+void dasd_alias_handle_summary_unit_check(struct work_struct *);
+void dasd_eckd_reset_ccw_to_base_io(struct dasd_ccw_req *);
+int dasd_alias_update_add_device(struct dasd_device *);
+#endif				/* DASD_ECKD_H */
diff --git a/drivers/s390/block/dasd_eer.c b/drivers/s390/block/dasd_eer.c
new file mode 100644
index 000000000..e96d1805b
--- /dev/null
+++ b/drivers/s390/block/dasd_eer.c
@@ -0,0 +1,719 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  Character device driver for extended error reporting.
+ *
+ *  Copyright IBM Corp. 2005
+ *  extended error reporting for DASD ECKD devices
+ *  Author(s): Stefan Weinhuber <wein@de.ibm.com>
+ */
+
+#include <linux/export.h>
+#include <linux/init.h>
+#include <linux/fs.h>
+#include <linux/kernel.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/device.h>
+#include <linux/poll.h>
+#include <linux/mutex.h>
+#include <linux/err.h>
+#include <linux/slab.h>
+
+#include <linux/uaccess.h>
+#include <linux/atomic.h>
+#include <asm/ebcdic.h>
+
+#include "dasd_int.h"
+#include "dasd_eckd.h"
+
+/*
+ * SECTION: the internal buffer
+ */
+
+/*
+ * The internal buffer is meant to store obaque blobs of data, so it does
+ * not know of higher level concepts like triggers.
+ * It consists of a number of pages that are used as a ringbuffer. Each data
+ * blob is stored in a simple record that consists of an integer, which
+ * contains the size of the following data, and the data bytes themselfes.
+ *
+ * To allow for multiple independent readers we create one internal buffer
+ * each time the device is opened and destroy the buffer when the file is
+ * closed again. The number of pages used for this buffer is determined by
+ * the module parmeter eer_pages.
+ *
+ * One record can be written to a buffer by using the functions
+ * - dasd_eer_start_record (one time per record to write the size to the
+ *                          buffer and reserve the space for the data)
+ * - dasd_eer_write_buffer (one or more times per record to write the data)
+ * The data can be written in several steps but you will have to compute
+ * the total size up front for the invocation of dasd_eer_start_record.
+ * If the ringbuffer is full, dasd_eer_start_record will remove the required
+ * number of old records.
+ *
+ * A record is typically read in two steps, first read the integer that
+ * specifies the size of the following data, then read the data.
+ * Both can be done by
+ * - dasd_eer_read_buffer
+ *
+ * For all mentioned functions you need to get the bufferlock first and keep
+ * it until a complete record is written or read.
+ *
+ * All information necessary to keep track of an internal buffer is kept in
+ * a struct eerbuffer. The buffer specific to a file pointer is strored in
+ * the private_data field of that file. To be able to write data to all
+ * existing buffers, each buffer is also added to the bufferlist.
+ * If the user does not want to read a complete record in one go, we have to
+ * keep track of the rest of the record. residual stores the number of bytes
+ * that are still to deliver. If the rest of the record is invalidated between
+ * two reads then residual will be set to -1 so that the next read will fail.
+ * All entries in the eerbuffer structure are protected with the bufferlock.
+ * To avoid races between writing to a buffer on the one side and creating
+ * and destroying buffers on the other side, the bufferlock must also be used
+ * to protect the bufferlist.
+ */
+
+static int eer_pages = 5;
+module_param(eer_pages, int, S_IRUGO|S_IWUSR);
+
+struct eerbuffer {
+	struct list_head list;
+	char **buffer;
+	int buffersize;
+	int buffer_page_count;
+	int head;
+        int tail;
+	int residual;
+};
+
+static LIST_HEAD(bufferlist);
+static DEFINE_SPINLOCK(bufferlock);
+static DECLARE_WAIT_QUEUE_HEAD(dasd_eer_read_wait_queue);
+
+/*
+ * How many free bytes are available on the buffer.
+ * Needs to be called with bufferlock held.
+ */
+static int dasd_eer_get_free_bytes(struct eerbuffer *eerb)
+{
+	if (eerb->head < eerb->tail)
+		return eerb->tail - eerb->head - 1;
+	return eerb->buffersize - eerb->head + eerb->tail -1;
+}
+
+/*
+ * How many bytes of buffer space are used.
+ * Needs to be called with bufferlock held.
+ */
+static int dasd_eer_get_filled_bytes(struct eerbuffer *eerb)
+{
+
+	if (eerb->head >= eerb->tail)
+		return eerb->head - eerb->tail;
+	return eerb->buffersize - eerb->tail + eerb->head;
+}
+
+/*
+ * The dasd_eer_write_buffer function just copies count bytes of data
+ * to the buffer. Make sure to call dasd_eer_start_record first, to
+ * make sure that enough free space is available.
+ * Needs to be called with bufferlock held.
+ */
+static void dasd_eer_write_buffer(struct eerbuffer *eerb,
+				  char *data, int count)
+{
+
+	unsigned long headindex,localhead;
+	unsigned long rest, len;
+	char *nextdata;
+
+	nextdata = data;
+	rest = count;
+	while (rest > 0) {
+ 		headindex = eerb->head / PAGE_SIZE;
+ 		localhead = eerb->head % PAGE_SIZE;
+		len = min(rest, PAGE_SIZE - localhead);
+		memcpy(eerb->buffer[headindex]+localhead, nextdata, len);
+		nextdata += len;
+		rest -= len;
+		eerb->head += len;
+		if (eerb->head == eerb->buffersize)
+			eerb->head = 0; /* wrap around */
+		BUG_ON(eerb->head > eerb->buffersize);
+	}
+}
+
+/*
+ * Needs to be called with bufferlock held.
+ */
+static int dasd_eer_read_buffer(struct eerbuffer *eerb, char *data, int count)
+{
+
+	unsigned long tailindex,localtail;
+	unsigned long rest, len, finalcount;
+	char *nextdata;
+
+	finalcount = min(count, dasd_eer_get_filled_bytes(eerb));
+	nextdata = data;
+	rest = finalcount;
+	while (rest > 0) {
+ 		tailindex = eerb->tail / PAGE_SIZE;
+ 		localtail = eerb->tail % PAGE_SIZE;
+		len = min(rest, PAGE_SIZE - localtail);
+		memcpy(nextdata, eerb->buffer[tailindex] + localtail, len);
+		nextdata += len;
+		rest -= len;
+		eerb->tail += len;
+		if (eerb->tail == eerb->buffersize)
+			eerb->tail = 0; /* wrap around */
+		BUG_ON(eerb->tail > eerb->buffersize);
+	}
+	return finalcount;
+}
+
+/*
+ * Whenever you want to write a blob of data to the internal buffer you
+ * have to start by using this function first. It will write the number
+ * of bytes that will be written to the buffer. If necessary it will remove
+ * old records to make room for the new one.
+ * Needs to be called with bufferlock held.
+ */
+static int dasd_eer_start_record(struct eerbuffer *eerb, int count)
+{
+	int tailcount;
+
+	if (count + sizeof(count) > eerb->buffersize)
+		return -ENOMEM;
+	while (dasd_eer_get_free_bytes(eerb) < count + sizeof(count)) {
+		if (eerb->residual > 0) {
+			eerb->tail += eerb->residual;
+			if (eerb->tail >= eerb->buffersize)
+				eerb->tail -= eerb->buffersize;
+			eerb->residual = -1;
+		}
+		dasd_eer_read_buffer(eerb, (char *) &tailcount,
+				     sizeof(tailcount));
+		eerb->tail += tailcount;
+		if (eerb->tail >= eerb->buffersize)
+			eerb->tail -= eerb->buffersize;
+	}
+	dasd_eer_write_buffer(eerb, (char*) &count, sizeof(count));
+
+	return 0;
+};
+
+/*
+ * Release pages that are not used anymore.
+ */
+static void dasd_eer_free_buffer_pages(char **buf, int no_pages)
+{
+	int i;
+
+	for (i = 0; i < no_pages; i++)
+		kfree(buf[i]);
+}
+
+/*
+ * Allocate a new set of memory pages.
+ */
+static int dasd_eer_allocate_buffer_pages(char **buf, int no_pages)
+{
+	int i;
+
+	for (i = 0; i < no_pages; i++) {
+		buf[i] = kzalloc(PAGE_SIZE, GFP_KERNEL);
+		if (!buf[i]) {
+			dasd_eer_free_buffer_pages(buf, i);
+			return -ENOMEM;
+		}
+	}
+	return 0;
+}
+
+/*
+ * SECTION: The extended error reporting functionality
+ */
+
+/*
+ * When a DASD device driver wants to report an error, it calls the
+ * function dasd_eer_write and gives the respective trigger ID as
+ * parameter. Currently there are four kinds of triggers:
+ *
+ * DASD_EER_FATALERROR:  all kinds of unrecoverable I/O problems
+ * DASD_EER_PPRCSUSPEND: PPRC was suspended
+ * DASD_EER_NOPATH:      There is no path to the device left.
+ * DASD_EER_STATECHANGE: The state of the device has changed.
+ *
+ * For the first three triggers all required information can be supplied by
+ * the caller. For these triggers a record is written by the function
+ * dasd_eer_write_standard_trigger.
+ *
+ * The DASD_EER_STATECHANGE trigger is special since a sense subsystem
+ * status ccw need to be executed to gather the necessary sense data first.
+ * The dasd_eer_snss function will queue the SNSS request and the request
+ * callback will then call dasd_eer_write with the DASD_EER_STATCHANGE
+ * trigger.
+ *
+ * To avoid memory allocations at runtime, the necessary memory is allocated
+ * when the extended error reporting is enabled for a device (by
+ * dasd_eer_probe). There is one sense subsystem status request for each
+ * eer enabled DASD device. The presence of the cqr in device->eer_cqr
+ * indicates that eer is enable for the device. The use of the snss request
+ * is protected by the DASD_FLAG_EER_IN_USE bit. When this flag indicates
+ * that the cqr is currently in use, dasd_eer_snss cannot start a second
+ * request but sets the DASD_FLAG_EER_SNSS flag instead. The callback of
+ * the SNSS request will check the bit and call dasd_eer_snss again.
+ */
+
+#define SNSS_DATA_SIZE 44
+
+#define DASD_EER_BUSID_SIZE 10
+struct dasd_eer_header {
+	__u32 total_size;
+	__u32 trigger;
+	__u64 tv_sec;
+	__u64 tv_usec;
+	char busid[DASD_EER_BUSID_SIZE];
+} __attribute__ ((packed));
+
+/*
+ * The following function can be used for those triggers that have
+ * all necessary data available when the function is called.
+ * If the parameter cqr is not NULL, the chain of requests will be searched
+ * for valid sense data, and all valid sense data sets will be added to
+ * the triggers data.
+ */
+static void dasd_eer_write_standard_trigger(struct dasd_device *device,
+					    struct dasd_ccw_req *cqr,
+					    int trigger)
+{
+	struct dasd_ccw_req *temp_cqr;
+	int data_size;
+	struct timespec64 ts;
+	struct dasd_eer_header header;
+	unsigned long flags;
+	struct eerbuffer *eerb;
+	char *sense;
+
+	/* go through cqr chain and count the valid sense data sets */
+	data_size = 0;
+	for (temp_cqr = cqr; temp_cqr; temp_cqr = temp_cqr->refers)
+		if (dasd_get_sense(&temp_cqr->irb))
+			data_size += 32;
+
+	header.total_size = sizeof(header) + data_size + 4; /* "EOR" */
+	header.trigger = trigger;
+	ktime_get_real_ts64(&ts);
+	header.tv_sec = ts.tv_sec;
+	header.tv_usec = ts.tv_nsec / NSEC_PER_USEC;
+	strscpy(header.busid, dev_name(&device->cdev->dev),
+		DASD_EER_BUSID_SIZE);
+
+	spin_lock_irqsave(&bufferlock, flags);
+	list_for_each_entry(eerb, &bufferlist, list) {
+		dasd_eer_start_record(eerb, header.total_size);
+		dasd_eer_write_buffer(eerb, (char *) &header, sizeof(header));
+		for (temp_cqr = cqr; temp_cqr; temp_cqr = temp_cqr->refers) {
+			sense = dasd_get_sense(&temp_cqr->irb);
+			if (sense)
+				dasd_eer_write_buffer(eerb, sense, 32);
+		}
+		dasd_eer_write_buffer(eerb, "EOR", 4);
+	}
+	spin_unlock_irqrestore(&bufferlock, flags);
+	wake_up_interruptible(&dasd_eer_read_wait_queue);
+}
+
+/*
+ * This function writes a DASD_EER_STATECHANGE trigger.
+ */
+static void dasd_eer_write_snss_trigger(struct dasd_device *device,
+					struct dasd_ccw_req *cqr,
+					int trigger)
+{
+	int data_size;
+	int snss_rc;
+	struct timespec64 ts;
+	struct dasd_eer_header header;
+	unsigned long flags;
+	struct eerbuffer *eerb;
+
+	snss_rc = (cqr->status == DASD_CQR_DONE) ? 0 : -EIO;
+	if (snss_rc)
+		data_size = 0;
+	else
+		data_size = SNSS_DATA_SIZE;
+
+	header.total_size = sizeof(header) + data_size + 4; /* "EOR" */
+	header.trigger = DASD_EER_STATECHANGE;
+	ktime_get_real_ts64(&ts);
+	header.tv_sec = ts.tv_sec;
+	header.tv_usec = ts.tv_nsec / NSEC_PER_USEC;
+	strscpy(header.busid, dev_name(&device->cdev->dev),
+		DASD_EER_BUSID_SIZE);
+
+	spin_lock_irqsave(&bufferlock, flags);
+	list_for_each_entry(eerb, &bufferlist, list) {
+		dasd_eer_start_record(eerb, header.total_size);
+		dasd_eer_write_buffer(eerb, (char *) &header , sizeof(header));
+		if (!snss_rc)
+			dasd_eer_write_buffer(eerb, cqr->data, SNSS_DATA_SIZE);
+		dasd_eer_write_buffer(eerb, "EOR", 4);
+	}
+	spin_unlock_irqrestore(&bufferlock, flags);
+	wake_up_interruptible(&dasd_eer_read_wait_queue);
+}
+
+/*
+ * This function is called for all triggers. It calls the appropriate
+ * function that writes the actual trigger records.
+ */
+void dasd_eer_write(struct dasd_device *device, struct dasd_ccw_req *cqr,
+		    unsigned int id)
+{
+	if (!device->eer_cqr)
+		return;
+	switch (id) {
+	case DASD_EER_FATALERROR:
+	case DASD_EER_PPRCSUSPEND:
+		dasd_eer_write_standard_trigger(device, cqr, id);
+		break;
+	case DASD_EER_NOPATH:
+	case DASD_EER_NOSPC:
+	case DASD_EER_AUTOQUIESCE:
+		dasd_eer_write_standard_trigger(device, NULL, id);
+		break;
+	case DASD_EER_STATECHANGE:
+		dasd_eer_write_snss_trigger(device, cqr, id);
+		break;
+	default: /* unknown trigger, so we write it without any sense data */
+		dasd_eer_write_standard_trigger(device, NULL, id);
+		break;
+	}
+}
+EXPORT_SYMBOL(dasd_eer_write);
+
+/*
+ * Start a sense subsystem status request.
+ * Needs to be called with the device held.
+ */
+void dasd_eer_snss(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr;
+
+	cqr = device->eer_cqr;
+	if (!cqr)	/* Device not eer enabled. */
+		return;
+	if (test_and_set_bit(DASD_FLAG_EER_IN_USE, &device->flags)) {
+		/* Sense subsystem status request in use. */
+		set_bit(DASD_FLAG_EER_SNSS, &device->flags);
+		return;
+	}
+	/* cdev is already locked, can't use dasd_add_request_head */
+	clear_bit(DASD_FLAG_EER_SNSS, &device->flags);
+	cqr->status = DASD_CQR_QUEUED;
+	list_add(&cqr->devlist, &device->ccw_queue);
+	dasd_schedule_device_bh(device);
+}
+
+/*
+ * Callback function for use with sense subsystem status request.
+ */
+static void dasd_eer_snss_cb(struct dasd_ccw_req *cqr, void *data)
+{
+	struct dasd_device *device = cqr->startdev;
+	unsigned long flags;
+
+	dasd_eer_write(device, cqr, DASD_EER_STATECHANGE);
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	if (device->eer_cqr == cqr) {
+		clear_bit(DASD_FLAG_EER_IN_USE, &device->flags);
+		if (test_bit(DASD_FLAG_EER_SNSS, &device->flags))
+			/* Another SNSS has been requested in the meantime. */
+			dasd_eer_snss(device);
+		cqr = NULL;
+	}
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	if (cqr)
+		/*
+		 * Extended error recovery has been switched off while
+		 * the SNSS request was running. It could even have
+		 * been switched off and on again in which case there
+		 * is a new ccw in device->eer_cqr. Free the "old"
+		 * snss request now.
+		 */
+		dasd_sfree_request(cqr, device);
+}
+
+/*
+ * Enable error reporting on a given device.
+ */
+int dasd_eer_enable(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr = NULL;
+	unsigned long flags;
+	struct ccw1 *ccw;
+	int rc = 0;
+
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	if (device->eer_cqr)
+		goto out;
+	else if (!device->discipline ||
+		 strcmp(device->discipline->name, "ECKD"))
+		rc = -EMEDIUMTYPE;
+	else if (test_bit(DASD_FLAG_OFFLINE, &device->flags))
+		rc = -EBUSY;
+
+	if (rc)
+		goto out;
+
+	cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* SNSS */,
+				   SNSS_DATA_SIZE, device, NULL);
+	if (IS_ERR(cqr)) {
+		rc = -ENOMEM;
+		cqr = NULL;
+		goto out;
+	}
+
+	cqr->startdev = device;
+	cqr->retries = 255;
+	cqr->expires = 10 * HZ;
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	set_bit(DASD_CQR_ALLOW_SLOCK, &cqr->flags);
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_SNSS;
+	ccw->count = SNSS_DATA_SIZE;
+	ccw->flags = 0;
+	ccw->cda = virt_to_dma32(cqr->data);
+
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	cqr->callback = dasd_eer_snss_cb;
+
+	if (!device->eer_cqr) {
+		device->eer_cqr = cqr;
+		cqr = NULL;
+	}
+
+out:
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+
+	if (cqr)
+		dasd_sfree_request(cqr, device);
+
+	return rc;
+}
+
+/*
+ * Disable error reporting on a given device.
+ */
+void dasd_eer_disable(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr;
+	unsigned long flags;
+	int in_use;
+
+	if (!device->eer_cqr)
+		return;
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	cqr = device->eer_cqr;
+	device->eer_cqr = NULL;
+	clear_bit(DASD_FLAG_EER_SNSS, &device->flags);
+	in_use = test_and_clear_bit(DASD_FLAG_EER_IN_USE, &device->flags);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	if (cqr && !in_use)
+		dasd_sfree_request(cqr, device);
+}
+
+/*
+ * SECTION: the device operations
+ */
+
+/*
+ * On the one side we need a lock to access our internal buffer, on the
+ * other side a copy_to_user can sleep. So we need to copy the data we have
+ * to transfer in a readbuffer, which is protected by the readbuffer_mutex.
+ */
+static char readbuffer[PAGE_SIZE];
+static DEFINE_MUTEX(readbuffer_mutex);
+
+static int dasd_eer_open(struct inode *inp, struct file *filp)
+{
+	struct eerbuffer *eerb;
+	unsigned long flags;
+
+	eerb = kzalloc_obj(struct eerbuffer);
+	if (!eerb)
+		return -ENOMEM;
+	eerb->buffer_page_count = eer_pages;
+	if (eerb->buffer_page_count < 1 ||
+	    eerb->buffer_page_count > INT_MAX / PAGE_SIZE) {
+		kfree(eerb);
+		DBF_EVENT(DBF_WARNING, "can't open device since module "
+			"parameter eer_pages is smaller than 1 or"
+			" bigger than %d", (int)(INT_MAX / PAGE_SIZE));
+		return -EINVAL;
+	}
+	eerb->buffersize = eerb->buffer_page_count * PAGE_SIZE;
+	eerb->buffer = kmalloc_array(eerb->buffer_page_count, sizeof(char *),
+				     GFP_KERNEL);
+        if (!eerb->buffer) {
+		kfree(eerb);
+                return -ENOMEM;
+	}
+	if (dasd_eer_allocate_buffer_pages(eerb->buffer,
+					   eerb->buffer_page_count)) {
+		kfree(eerb->buffer);
+		kfree(eerb);
+		return -ENOMEM;
+	}
+	filp->private_data = eerb;
+	spin_lock_irqsave(&bufferlock, flags);
+	list_add(&eerb->list, &bufferlist);
+	spin_unlock_irqrestore(&bufferlock, flags);
+
+	return nonseekable_open(inp,filp);
+}
+
+static int dasd_eer_close(struct inode *inp, struct file *filp)
+{
+	struct eerbuffer *eerb;
+	unsigned long flags;
+
+	eerb = (struct eerbuffer *) filp->private_data;
+	spin_lock_irqsave(&bufferlock, flags);
+	list_del(&eerb->list);
+	spin_unlock_irqrestore(&bufferlock, flags);
+	dasd_eer_free_buffer_pages(eerb->buffer, eerb->buffer_page_count);
+	kfree(eerb->buffer);
+	kfree(eerb);
+
+	return 0;
+}
+
+static ssize_t dasd_eer_read(struct file *filp, char __user *buf,
+			     size_t count, loff_t *ppos)
+{
+	int tc,rc;
+	int tailcount,effective_count;
+        unsigned long flags;
+	struct eerbuffer *eerb;
+
+	eerb = (struct eerbuffer *) filp->private_data;
+	if (mutex_lock_interruptible(&readbuffer_mutex))
+		return -ERESTARTSYS;
+
+	spin_lock_irqsave(&bufferlock, flags);
+
+	if (eerb->residual < 0) { /* the remainder of this record */
+		                  /* has been deleted             */
+		eerb->residual = 0;
+		spin_unlock_irqrestore(&bufferlock, flags);
+		mutex_unlock(&readbuffer_mutex);
+		return -EIO;
+	} else if (eerb->residual > 0) {
+		/* OK we still have a second half of a record to deliver */
+		effective_count = min(eerb->residual, (int) count);
+		eerb->residual -= effective_count;
+	} else {
+		tc = 0;
+		while (!tc) {
+			tc = dasd_eer_read_buffer(eerb, (char *) &tailcount,
+						  sizeof(tailcount));
+			if (!tc) {
+				/* no data available */
+				spin_unlock_irqrestore(&bufferlock, flags);
+				mutex_unlock(&readbuffer_mutex);
+				if (filp->f_flags & O_NONBLOCK)
+					return -EAGAIN;
+				rc = wait_event_interruptible(
+					dasd_eer_read_wait_queue,
+					eerb->head != eerb->tail);
+				if (rc)
+					return rc;
+				if (mutex_lock_interruptible(&readbuffer_mutex))
+					return -ERESTARTSYS;
+				spin_lock_irqsave(&bufferlock, flags);
+			}
+		}
+		WARN_ON(tc != sizeof(tailcount));
+		effective_count = min(tailcount,(int)count);
+		eerb->residual = tailcount - effective_count;
+	}
+
+	tc = dasd_eer_read_buffer(eerb, readbuffer, effective_count);
+	WARN_ON(tc != effective_count);
+
+	spin_unlock_irqrestore(&bufferlock, flags);
+
+	if (copy_to_user(buf, readbuffer, effective_count)) {
+		mutex_unlock(&readbuffer_mutex);
+		return -EFAULT;
+	}
+
+	mutex_unlock(&readbuffer_mutex);
+	return effective_count;
+}
+
+static __poll_t dasd_eer_poll(struct file *filp, poll_table *ptable)
+{
+	__poll_t mask;
+	unsigned long flags;
+	struct eerbuffer *eerb;
+
+	eerb = (struct eerbuffer *) filp->private_data;
+	poll_wait(filp, &dasd_eer_read_wait_queue, ptable);
+	spin_lock_irqsave(&bufferlock, flags);
+	if (eerb->head != eerb->tail)
+		mask = EPOLLIN | EPOLLRDNORM ;
+	else
+		mask = 0;
+	spin_unlock_irqrestore(&bufferlock, flags);
+	return mask;
+}
+
+static const struct file_operations dasd_eer_fops = {
+	.open		= &dasd_eer_open,
+	.release	= &dasd_eer_close,
+	.read		= &dasd_eer_read,
+	.poll		= &dasd_eer_poll,
+	.owner		= THIS_MODULE,
+	.llseek		= noop_llseek,
+};
+
+static struct miscdevice *dasd_eer_dev = NULL;
+
+int __init dasd_eer_init(void)
+{
+	int rc;
+
+	dasd_eer_dev = kzalloc_obj(*dasd_eer_dev);
+	if (!dasd_eer_dev)
+		return -ENOMEM;
+
+	dasd_eer_dev->minor = MISC_DYNAMIC_MINOR;
+	dasd_eer_dev->name  = "dasd_eer";
+	dasd_eer_dev->fops  = &dasd_eer_fops;
+
+	rc = misc_register(dasd_eer_dev);
+	if (rc) {
+		kfree(dasd_eer_dev);
+		dasd_eer_dev = NULL;
+		DBF_EVENT(DBF_ERR, "%s", "dasd_eer_init could not "
+		       "register misc device");
+		return rc;
+	}
+
+	return 0;
+}
+
+void dasd_eer_exit(void)
+{
+	if (dasd_eer_dev) {
+		misc_deregister(dasd_eer_dev);
+		kfree(dasd_eer_dev);
+		dasd_eer_dev = NULL;
+	}
+}
diff --git a/drivers/s390/block/dasd_erp.c b/drivers/s390/block/dasd_erp.c
new file mode 100644
index 000000000..05d536648
--- /dev/null
+++ b/drivers/s390/block/dasd_erp.c
@@ -0,0 +1,206 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Carsten Otte <Cotte@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2001
+ *
+ */
+
+#include <linux/export.h>
+#include <linux/ctype.h>
+#include <linux/init.h>
+
+#include <asm/debug.h>
+#include <asm/ebcdic.h>
+#include <linux/uaccess.h>
+
+#include "dasd_int.h"
+
+struct dasd_ccw_req *
+dasd_alloc_erp_request(unsigned int magic, int cplength, int datasize,
+		       struct dasd_device * device)
+{
+	unsigned long flags;
+	struct dasd_ccw_req *cqr;
+	char *data;
+	int size;
+
+	/* Sanity checks */
+	BUG_ON(datasize > PAGE_SIZE ||
+	       (cplength*sizeof(struct ccw1)) > PAGE_SIZE);
+
+	size = (sizeof(struct dasd_ccw_req) + 7L) & -8L;
+	if (cplength > 0)
+		size += cplength * sizeof(struct ccw1);
+	if (datasize > 0)
+		size += datasize;
+	spin_lock_irqsave(&device->mem_lock, flags);
+	cqr = (struct dasd_ccw_req *)
+		dasd_alloc_chunk(&device->erp_chunks, size);
+	spin_unlock_irqrestore(&device->mem_lock, flags);
+	if (cqr == NULL)
+		return ERR_PTR(-ENOMEM);
+	memset(cqr, 0, sizeof(struct dasd_ccw_req));
+	INIT_LIST_HEAD(&cqr->devlist);
+	INIT_LIST_HEAD(&cqr->blocklist);
+	data = (char *) cqr + ((sizeof(struct dasd_ccw_req) + 7L) & -8L);
+	cqr->cpaddr = NULL;
+	if (cplength > 0) {
+		cqr->cpaddr = (struct ccw1 *) data;
+		data += cplength*sizeof(struct ccw1);
+		memset(cqr->cpaddr, 0, cplength*sizeof(struct ccw1));
+	}
+	cqr->data = NULL;
+	if (datasize > 0) {
+		cqr->data = data;
+ 		memset(cqr->data, 0, datasize);
+	}
+	cqr->magic = magic;
+	ASCEBC((char *) &cqr->magic, 4);
+	set_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	dasd_get_device(device);
+	return cqr;
+}
+
+void
+dasd_free_erp_request(struct dasd_ccw_req *cqr, struct dasd_device * device)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&device->mem_lock, flags);
+	dasd_free_chunk(&device->erp_chunks, cqr);
+	spin_unlock_irqrestore(&device->mem_lock, flags);
+	atomic_dec(&device->ref_count);
+}
+
+
+/*
+ * dasd_default_erp_action just retries the current cqr
+ */
+struct dasd_ccw_req *
+dasd_default_erp_action(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device;
+
+	device = cqr->startdev;
+
+        /* just retry - there is nothing to save ... I got no sense data.... */
+        if (cqr->retries > 0) {
+		DBF_DEV_EVENT(DBF_DEBUG, device,
+                             "default ERP called (%i retries left)",
+                             cqr->retries);
+		if (!test_bit(DASD_CQR_VERIFY_PATH, &cqr->flags))
+			cqr->lpm = dasd_path_get_opm(device);
+		cqr->status = DASD_CQR_FILLED;
+        } else {
+		pr_err("%s: default ERP has run out of retries and failed\n",
+		       dev_name(&device->cdev->dev));
+		cqr->status = DASD_CQR_FAILED;
+		cqr->stopclk = get_tod_clock();
+        }
+        return cqr;
+}				/* end dasd_default_erp_action */
+
+/*
+ * DESCRIPTION
+ *   Frees all ERPs of the current ERP Chain and set the status
+ *   of the original CQR either to DASD_CQR_DONE if ERP was successful
+ *   or to DASD_CQR_FAILED if ERP was NOT successful.
+ *   NOTE: This function is only called if no discipline postaction
+ *	   is available
+ *
+ * PARAMETER
+ *   erp		current erp_head
+ *
+ * RETURN VALUES
+ *   cqr		pointer to the original CQR
+ */
+struct dasd_ccw_req *dasd_default_erp_postaction(struct dasd_ccw_req *cqr)
+{
+	int success, aborted;
+	unsigned long startclk, stopclk;
+	struct dasd_device *startdev;
+	unsigned int proc_bytes;
+
+	BUG_ON(cqr->refers == NULL || cqr->function == NULL);
+
+	success = cqr->status == DASD_CQR_DONE;
+	aborted = cqr->status == DASD_CQR_ABORTED;
+	startclk = cqr->startclk;
+	stopclk = cqr->stopclk;
+	startdev = cqr->startdev;
+	proc_bytes = cqr->proc_bytes;
+
+	/* free all ERPs - but NOT the original cqr */
+	while (cqr->refers != NULL) {
+		struct dasd_ccw_req *refers;
+
+		refers = cqr->refers;
+		/* remove the request from the block queue */
+		list_del(&cqr->blocklist);
+		/* free the finished erp request */
+		dasd_free_erp_request(cqr, cqr->memdev);
+		cqr = refers;
+	}
+
+	/* set corresponding status to original cqr */
+	cqr->startclk = startclk;
+	cqr->stopclk = stopclk;
+	cqr->startdev = startdev;
+	cqr->proc_bytes = proc_bytes;
+	if (aborted)
+		/* base request is owned by the ESE format replacement CQR */
+		cqr->status = DASD_CQR_ABORTED;
+	else if (success)
+		cqr->status = DASD_CQR_DONE;
+	else {
+		cqr->status = DASD_CQR_FAILED;
+		cqr->stopclk = get_tod_clock();
+	}
+
+	return cqr;
+
+}				/* end default_erp_postaction */
+
+void
+dasd_log_sense(struct dasd_ccw_req *cqr, struct irb *irb)
+{
+	struct dasd_device *device;
+
+	device = cqr->startdev;
+	if (cqr->intrc == -ETIMEDOUT) {
+		dev_err(&device->cdev->dev,
+			"A timeout error occurred for cqr %px\n", cqr);
+		return;
+	}
+	if (cqr->intrc == -ENOLINK) {
+		dev_err(&device->cdev->dev,
+			"A transport error occurred for cqr %px\n", cqr);
+		return;
+	}
+	/* dump sense data */
+	if (device->discipline && device->discipline->dump_sense)
+		device->discipline->dump_sense(device, cqr, irb);
+}
+
+void
+dasd_log_sense_dbf(struct dasd_ccw_req *cqr, struct irb *irb)
+{
+	struct dasd_device *device;
+
+	device = cqr->startdev;
+	/* dump sense data to s390 debugfeature*/
+	if (device->discipline && device->discipline->dump_sense_dbf)
+		device->discipline->dump_sense_dbf(device, irb, "log");
+}
+EXPORT_SYMBOL(dasd_log_sense_dbf);
+
+EXPORT_SYMBOL(dasd_default_erp_action);
+EXPORT_SYMBOL(dasd_default_erp_postaction);
+EXPORT_SYMBOL(dasd_alloc_erp_request);
+EXPORT_SYMBOL(dasd_free_erp_request);
+EXPORT_SYMBOL(dasd_log_sense);
+
diff --git a/drivers/s390/block/dasd_fba.c b/drivers/s390/block/dasd_fba.c
new file mode 100644
index 000000000..029acb83e
--- /dev/null
+++ b/drivers/s390/block/dasd_fba.c
@@ -0,0 +1,811 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2009
+ */
+
+
+#include <linux/stddef.h>
+#include <linux/kernel.h>
+#include <asm/debug.h>
+
+#include <linux/slab.h>
+#include <linux/hdreg.h>	/* HDIO_GETGEO			    */
+#include <linux/bio.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/io.h>
+
+#include <asm/idals.h>
+#include <asm/ebcdic.h>
+#include <asm/ccwdev.h>
+
+#include "dasd_int.h"
+#include "dasd_fba.h"
+
+#define FBA_DEFAULT_RETRIES 32
+
+#define DASD_FBA_CCW_WRITE 0x41
+#define DASD_FBA_CCW_READ 0x42
+#define DASD_FBA_CCW_LOCATE 0x43
+#define DASD_FBA_CCW_DEFINE_EXTENT 0x63
+
+MODULE_DESCRIPTION("S/390 DASD FBA Disks device driver");
+MODULE_LICENSE("GPL");
+
+static struct dasd_discipline dasd_fba_discipline;
+static void *dasd_fba_zero_page;
+
+struct dasd_fba_private {
+	struct dasd_fba_characteristics rdc_data;
+};
+
+static struct ccw_device_id dasd_fba_ids[] = {
+	{ CCW_DEVICE_DEVTYPE (0x6310, 0, 0x9336, 0), .driver_info = 0x1},
+	{ CCW_DEVICE_DEVTYPE (0x3880, 0, 0x3370, 0), .driver_info = 0x2},
+	{ /* end of list */ },
+};
+
+MODULE_DEVICE_TABLE(ccw, dasd_fba_ids);
+
+static int
+dasd_fba_set_online(struct ccw_device *cdev)
+{
+	return dasd_generic_set_online(cdev, &dasd_fba_discipline);
+}
+
+static struct ccw_driver dasd_fba_driver = {
+	.driver = {
+		.name	= "dasd-fba",
+		.owner	= THIS_MODULE,
+		.dev_groups = dasd_dev_groups,
+	},
+	.ids         = dasd_fba_ids,
+	.probe       = dasd_generic_probe,
+	.remove      = dasd_generic_remove,
+	.set_offline = dasd_generic_set_offline,
+	.set_online  = dasd_fba_set_online,
+	.notify      = dasd_generic_notify,
+	.path_event  = dasd_generic_path_event,
+	.int_class   = IRQIO_DAS,
+};
+
+static void
+define_extent(struct ccw1 * ccw, struct DE_fba_data *data, int rw,
+	      int blksize, int beg, int nr)
+{
+	ccw->cmd_code = DASD_FBA_CCW_DEFINE_EXTENT;
+	ccw->flags = 0;
+	ccw->count = 16;
+	ccw->cda = virt_to_dma32(data);
+	memset(data, 0, sizeof (struct DE_fba_data));
+	if (rw == WRITE)
+		(data->mask).perm = 0x0;
+	else if (rw == READ)
+		(data->mask).perm = 0x1;
+	else
+		data->mask.perm = 0x2;
+	data->blk_size = blksize;
+	data->ext_loc = beg;
+	data->ext_end = nr - 1;
+}
+
+static void
+locate_record(struct ccw1 * ccw, struct LO_fba_data *data, int rw,
+	      int block_nr, int block_ct)
+{
+	ccw->cmd_code = DASD_FBA_CCW_LOCATE;
+	ccw->flags = 0;
+	ccw->count = 8;
+	ccw->cda = virt_to_dma32(data);
+	memset(data, 0, sizeof (struct LO_fba_data));
+	if (rw == WRITE)
+		data->operation.cmd = 0x5;
+	else if (rw == READ)
+		data->operation.cmd = 0x6;
+	else
+		data->operation.cmd = 0x8;
+	data->blk_nr = block_nr;
+	data->blk_ct = block_ct;
+}
+
+static int
+dasd_fba_check_characteristics(struct dasd_device *device)
+{
+	struct dasd_fba_private *private = device->private;
+	struct ccw_device *cdev = device->cdev;
+	struct dasd_block *block;
+	int readonly, rc;
+
+	if (!private) {
+		private = kzalloc_obj(*private, GFP_KERNEL | GFP_DMA);
+		if (!private) {
+			dev_warn(&device->cdev->dev,
+				 "Allocating memory for private DASD "
+				 "data failed\n");
+			return -ENOMEM;
+		}
+		device->private = private;
+	} else {
+		memset(private, 0, sizeof(*private));
+	}
+	block = dasd_alloc_block();
+	if (IS_ERR(block)) {
+		DBF_EVENT_DEVID(DBF_WARNING, cdev, "%s", "could not allocate "
+				"dasd block structure");
+		device->private = NULL;
+		kfree(private);
+		return PTR_ERR(block);
+	}
+	device->block = block;
+	block->base = device;
+
+	/* Read Device Characteristics */
+	rc = dasd_generic_read_dev_chars(device, DASD_FBA_MAGIC,
+					 &private->rdc_data, 32);
+	if (rc) {
+		DBF_EVENT_DEVID(DBF_WARNING, cdev, "Read device "
+				"characteristics returned error %d", rc);
+		device->block = NULL;
+		dasd_free_block(block);
+		device->private = NULL;
+		kfree(private);
+		return rc;
+	}
+
+	device->default_expires = DASD_EXPIRES;
+	device->default_retries = FBA_DEFAULT_RETRIES;
+	dasd_path_set_opm(device, LPM_ANYPATH);
+
+	readonly = dasd_device_is_ro(device);
+	if (readonly)
+		set_bit(DASD_FLAG_DEVICE_RO, &device->flags);
+
+	/* FBA supports discard, set the according feature bit */
+	dasd_set_feature(cdev, DASD_FEATURE_DISCARD, 1);
+
+	dev_info(&device->cdev->dev,
+		 "New FBA DASD %04X/%02X (CU %04X/%02X) with %d MB "
+		 "and %d B/blk%s\n",
+		 cdev->id.dev_type,
+		 cdev->id.dev_model,
+		 cdev->id.cu_type,
+		 cdev->id.cu_model,
+		 ((private->rdc_data.blk_bdsa *
+		   (private->rdc_data.blk_size >> 9)) >> 11),
+		 private->rdc_data.blk_size,
+		 readonly ? ", read-only device" : "");
+	return 0;
+}
+
+static int dasd_fba_do_analysis(struct dasd_block *block)
+{
+	struct dasd_fba_private *private = block->base->private;
+	int sb, rc;
+
+	rc = dasd_check_blocksize(private->rdc_data.blk_size);
+	if (rc) {
+		DBF_DEV_EVENT(DBF_WARNING, block->base, "unknown blocksize %d",
+			    private->rdc_data.blk_size);
+		return rc;
+	}
+	block->blocks = private->rdc_data.blk_bdsa;
+	block->bp_block = private->rdc_data.blk_size;
+	block->s2b_shift = 0;	/* bits to shift 512 to get a block */
+	for (sb = 512; sb < private->rdc_data.blk_size; sb = sb << 1)
+		block->s2b_shift++;
+	return 0;
+}
+
+static int dasd_fba_fill_geometry(struct dasd_block *block,
+				  struct hd_geometry *geo)
+{
+	if (dasd_check_blocksize(block->bp_block) != 0)
+		return -EINVAL;
+	geo->cylinders = (block->blocks << block->s2b_shift) >> 10;
+	geo->heads = 16;
+	geo->sectors = 128 >> block->s2b_shift;
+	return 0;
+}
+
+static dasd_erp_fn_t
+dasd_fba_erp_action(struct dasd_ccw_req * cqr)
+{
+	return dasd_default_erp_action;
+}
+
+static dasd_erp_fn_t
+dasd_fba_erp_postaction(struct dasd_ccw_req * cqr)
+{
+	if (cqr->function == dasd_default_erp_action)
+		return dasd_default_erp_postaction;
+
+	DBF_DEV_EVENT(DBF_WARNING, cqr->startdev, "unknown ERP action %p",
+		    cqr->function);
+	return NULL;
+}
+
+static void dasd_fba_check_for_device_change(struct dasd_device *device,
+					     struct dasd_ccw_req *cqr,
+					     struct irb *irb)
+{
+	char mask;
+
+	/* first of all check for state change pending interrupt */
+	mask = DEV_STAT_ATTENTION | DEV_STAT_DEV_END | DEV_STAT_UNIT_EXCEP;
+	if ((irb->scsw.cmd.dstat & mask) == mask)
+		dasd_generic_handle_state_change(device);
+};
+
+
+/*
+ * Builds a CCW with no data payload
+ */
+static void ccw_write_no_data(struct ccw1 *ccw)
+{
+	ccw->cmd_code = DASD_FBA_CCW_WRITE;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->count = 0;
+}
+
+/*
+ * Builds a CCW that writes only zeroes.
+ */
+static void ccw_write_zero(struct ccw1 *ccw, int count)
+{
+	ccw->cmd_code = DASD_FBA_CCW_WRITE;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->count = count;
+	ccw->cda = virt_to_dma32(dasd_fba_zero_page);
+}
+
+/*
+ * Helper function to count the amount of necessary CCWs within a given range
+ * with 4k alignment and command chaining in mind.
+ */
+static int count_ccws(sector_t first_rec, sector_t last_rec,
+		      unsigned int blocks_per_page)
+{
+	sector_t wz_stop = 0, d_stop = 0;
+	int cur_pos = 0;
+	int count = 0;
+
+	if (first_rec % blocks_per_page != 0) {
+		wz_stop = first_rec + blocks_per_page -
+			(first_rec % blocks_per_page) - 1;
+		if (wz_stop > last_rec)
+			wz_stop = last_rec;
+		cur_pos = wz_stop - first_rec + 1;
+		count++;
+	}
+
+	if (last_rec - (first_rec + cur_pos) + 1 >= blocks_per_page) {
+		if ((last_rec - blocks_per_page + 1) % blocks_per_page != 0)
+			d_stop = last_rec - ((last_rec - blocks_per_page + 1) %
+					     blocks_per_page);
+		else
+			d_stop = last_rec;
+
+		cur_pos += d_stop - (first_rec + cur_pos) + 1;
+		count++;
+	}
+
+	if (cur_pos == 0 || first_rec + cur_pos - 1 < last_rec)
+		count++;
+
+	return count;
+}
+
+/*
+ * This function builds a CCW request for block layer discard requests.
+ * Each page in the z/VM hypervisor that represents certain records of an FBA
+ * device will be padded with zeros. This is a special behaviour of the WRITE
+ * command which is triggered when no data payload is added to the CCW.
+ *
+ * Note: Due to issues in some z/VM versions, we can't fully utilise this
+ * special behaviour. We have to keep a 4k (or 8 block) alignment in mind to
+ * work around those issues and write actual zeroes to the unaligned parts in
+ * the request. This workaround might be removed in the future.
+ */
+static struct dasd_ccw_req *dasd_fba_build_cp_discard(
+						struct dasd_device *memdev,
+						struct dasd_block *block,
+						struct request *req)
+{
+	struct LO_fba_data *LO_data;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+
+	sector_t wz_stop = 0, d_stop = 0;
+	sector_t first_rec, last_rec;
+
+	unsigned int blksize = block->bp_block;
+	unsigned int blocks_per_page;
+	int wz_count = 0;
+	int d_count = 0;
+	int cur_pos = 0; /* Current position within the extent */
+	int count = 0;
+	int cplength;
+	int datasize;
+	int nr_ccws;
+
+	first_rec = blk_rq_pos(req) >> block->s2b_shift;
+	last_rec =
+		(blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift;
+	count = last_rec - first_rec + 1;
+
+	blocks_per_page = BLOCKS_PER_PAGE(blksize);
+	nr_ccws = count_ccws(first_rec, last_rec, blocks_per_page);
+
+	/* define extent + nr_ccws * locate record + nr_ccws * single CCW */
+	cplength = 1 + 2 * nr_ccws;
+	datasize = sizeof(struct DE_fba_data) +
+		nr_ccws * (sizeof(struct LO_fba_data) + sizeof(struct ccw1));
+
+	cqr = dasd_smalloc_request(DASD_FBA_MAGIC, cplength, datasize, memdev,
+				   blk_mq_rq_to_pdu(req));
+	if (IS_ERR(cqr))
+		return cqr;
+
+	ccw = cqr->cpaddr;
+
+	define_extent(ccw++, cqr->data, WRITE, blksize, first_rec, count);
+	LO_data = cqr->data + sizeof(struct DE_fba_data);
+
+	/* First part is not aligned. Calculate range to write zeroes. */
+	if (first_rec % blocks_per_page != 0) {
+		wz_stop = first_rec + blocks_per_page -
+			(first_rec % blocks_per_page) - 1;
+		if (wz_stop > last_rec)
+			wz_stop = last_rec;
+		wz_count = wz_stop - first_rec + 1;
+
+		ccw[-1].flags |= CCW_FLAG_CC;
+		locate_record(ccw++, LO_data++, WRITE, cur_pos, wz_count);
+
+		ccw[-1].flags |= CCW_FLAG_CC;
+		ccw_write_zero(ccw++, wz_count * blksize);
+
+		cur_pos = wz_count;
+	}
+
+	/* We can do proper discard when we've got at least blocks_per_page blocks. */
+	if (last_rec - (first_rec + cur_pos) + 1 >= blocks_per_page) {
+		/* is last record at page boundary? */
+		if ((last_rec - blocks_per_page + 1) % blocks_per_page != 0)
+			d_stop = last_rec - ((last_rec - blocks_per_page + 1) %
+					     blocks_per_page);
+		else
+			d_stop = last_rec;
+
+		d_count = d_stop - (first_rec + cur_pos) + 1;
+
+		ccw[-1].flags |= CCW_FLAG_CC;
+		locate_record(ccw++, LO_data++, WRITE, cur_pos, d_count);
+
+		ccw[-1].flags |= CCW_FLAG_CC;
+		ccw_write_no_data(ccw++);
+
+		cur_pos += d_count;
+	}
+
+	/* We might still have some bits left which need to be zeroed. */
+	if (cur_pos == 0 || first_rec + cur_pos - 1 < last_rec) {
+		if (d_stop != 0)
+			wz_count = last_rec - d_stop;
+		else if (wz_stop != 0)
+			wz_count = last_rec - wz_stop;
+		else
+			wz_count = count;
+
+		ccw[-1].flags |= CCW_FLAG_CC;
+		locate_record(ccw++, LO_data++, WRITE, cur_pos, wz_count);
+
+		ccw[-1].flags |= CCW_FLAG_CC;
+		ccw_write_zero(ccw++, wz_count * blksize);
+	}
+
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+
+	cqr->startdev = memdev;
+	cqr->memdev = memdev;
+	cqr->block = block;
+	cqr->expires = memdev->default_expires * HZ;	/* default 5 minutes */
+	cqr->retries = memdev->default_retries;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	return cqr;
+}
+
+static struct dasd_ccw_req *dasd_fba_build_cp_regular(
+						struct dasd_device *memdev,
+						struct dasd_block *block,
+						struct request *req)
+{
+	struct dasd_fba_private *private = block->base->private;
+	dma64_t *idaws;
+	struct LO_fba_data *LO_data;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	struct req_iterator iter;
+	struct bio_vec bv;
+	char *dst;
+	int count, cidaw, cplength, datasize;
+	sector_t recid, first_rec, last_rec;
+	unsigned int blksize, off;
+	unsigned char cmd;
+
+	if (rq_data_dir(req) == READ) {
+		cmd = DASD_FBA_CCW_READ;
+	} else if (rq_data_dir(req) == WRITE) {
+		cmd = DASD_FBA_CCW_WRITE;
+	} else
+		return ERR_PTR(-EINVAL);
+	blksize = block->bp_block;
+	/* Calculate record id of first and last block. */
+	first_rec = blk_rq_pos(req) >> block->s2b_shift;
+	last_rec =
+		(blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift;
+	/* Check struct bio and count the number of blocks for the request. */
+	count = 0;
+	cidaw = 0;
+	rq_for_each_segment(bv, req, iter) {
+		if (bv.bv_len & (blksize - 1))
+			/* Fba can only do full blocks. */
+			return ERR_PTR(-EINVAL);
+		count += bv.bv_len >> (block->s2b_shift + 9);
+		if (idal_is_needed (page_address(bv.bv_page), bv.bv_len))
+			cidaw += bv.bv_len / blksize;
+	}
+	/* Paranoia. */
+	if (count != last_rec - first_rec + 1)
+		return ERR_PTR(-EINVAL);
+	/* 1x define extent + 1x locate record + number of blocks */
+	cplength = 2 + count;
+	/* 1x define extent + 1x locate record */
+	datasize = sizeof(struct DE_fba_data) + sizeof(struct LO_fba_data) +
+		cidaw * sizeof(unsigned long);
+	/*
+	 * Find out number of additional locate record ccws if the device
+	 * can't do data chaining.
+	 */
+	if (private->rdc_data.mode.bits.data_chain == 0) {
+		cplength += count - 1;
+		datasize += (count - 1)*sizeof(struct LO_fba_data);
+	}
+	/* Allocate the ccw request. */
+	cqr = dasd_smalloc_request(DASD_FBA_MAGIC, cplength, datasize, memdev,
+				   blk_mq_rq_to_pdu(req));
+	if (IS_ERR(cqr))
+		return cqr;
+	ccw = cqr->cpaddr;
+	/* First ccw is define extent. */
+	define_extent(ccw++, cqr->data, rq_data_dir(req),
+		      block->bp_block, blk_rq_pos(req), blk_rq_sectors(req));
+	/* Build locate_record + read/write ccws. */
+	idaws = (dma64_t *)(cqr->data + sizeof(struct DE_fba_data));
+	LO_data = (struct LO_fba_data *) (idaws + cidaw);
+	/* Locate record for all blocks for smart devices. */
+	if (private->rdc_data.mode.bits.data_chain != 0) {
+		ccw[-1].flags |= CCW_FLAG_CC;
+		locate_record(ccw++, LO_data++, rq_data_dir(req), 0, count);
+	}
+	recid = first_rec;
+	rq_for_each_segment(bv, req, iter) {
+		dst = bvec_virt(&bv);
+		if (dasd_page_cache) {
+			char *copy = kmem_cache_alloc(dasd_page_cache,
+						      GFP_DMA | __GFP_NOWARN);
+			if (copy && rq_data_dir(req) == WRITE)
+				memcpy(copy + bv.bv_offset, dst, bv.bv_len);
+			if (copy)
+				dst = copy + bv.bv_offset;
+		}
+		for (off = 0; off < bv.bv_len; off += blksize) {
+			/* Locate record for stupid devices. */
+			if (private->rdc_data.mode.bits.data_chain == 0) {
+				ccw[-1].flags |= CCW_FLAG_CC;
+				locate_record(ccw, LO_data++,
+					      rq_data_dir(req),
+					      recid - first_rec, 1);
+				ccw->flags = CCW_FLAG_CC;
+				ccw++;
+			} else {
+				if (recid > first_rec)
+					ccw[-1].flags |= CCW_FLAG_DC;
+				else
+					ccw[-1].flags |= CCW_FLAG_CC;
+			}
+			ccw->cmd_code = cmd;
+			ccw->count = block->bp_block;
+			if (idal_is_needed(dst, blksize)) {
+				ccw->cda = virt_to_dma32(idaws);
+				ccw->flags = CCW_FLAG_IDA;
+				idaws = idal_create_words(idaws, dst, blksize);
+			} else {
+				ccw->cda = virt_to_dma32(dst);
+				ccw->flags = 0;
+			}
+			ccw++;
+			dst += blksize;
+			recid++;
+		}
+	}
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->startdev = memdev;
+	cqr->memdev = memdev;
+	cqr->block = block;
+	cqr->expires = memdev->default_expires * HZ;	/* default 5 minutes */
+	cqr->retries = memdev->default_retries;
+	cqr->buildclk = get_tod_clock();
+	cqr->status = DASD_CQR_FILLED;
+	return cqr;
+}
+
+static struct dasd_ccw_req *dasd_fba_build_cp(struct dasd_device *memdev,
+					      struct dasd_block *block,
+					      struct request *req)
+{
+	if (req_op(req) == REQ_OP_DISCARD || req_op(req) == REQ_OP_WRITE_ZEROES)
+		return dasd_fba_build_cp_discard(memdev, block, req);
+	else
+		return dasd_fba_build_cp_regular(memdev, block, req);
+}
+
+static int
+dasd_fba_free_cp(struct dasd_ccw_req *cqr, struct request *req)
+{
+	struct dasd_fba_private *private = cqr->block->base->private;
+	struct ccw1 *ccw;
+	struct req_iterator iter;
+	struct bio_vec bv;
+	char *dst, *cda;
+	unsigned int blksize, off;
+	int status;
+
+	if (!dasd_page_cache)
+		goto out;
+	blksize = cqr->block->bp_block;
+	ccw = cqr->cpaddr;
+	/* Skip over define extent & locate record. */
+	ccw++;
+	if (private->rdc_data.mode.bits.data_chain != 0)
+		ccw++;
+	rq_for_each_segment(bv, req, iter) {
+		dst = bvec_virt(&bv);
+		for (off = 0; off < bv.bv_len; off += blksize) {
+			/* Skip locate record. */
+			if (private->rdc_data.mode.bits.data_chain == 0)
+				ccw++;
+			if (dst) {
+				if (ccw->flags & CCW_FLAG_IDA)
+					cda = dma64_to_virt(*((dma64_t *)dma32_to_virt(ccw->cda)));
+				else
+					cda = dma32_to_virt(ccw->cda);
+				if (dst != cda) {
+					if (rq_data_dir(req) == READ)
+						memcpy(dst, cda, bv.bv_len);
+					kmem_cache_free(dasd_page_cache,
+					    (void *)((addr_t)cda & PAGE_MASK));
+				}
+				dst = NULL;
+			}
+			ccw++;
+		}
+	}
+out:
+	status = cqr->status == DASD_CQR_DONE;
+	dasd_sfree_request(cqr, cqr->memdev);
+	return status;
+}
+
+static void dasd_fba_handle_terminated_request(struct dasd_ccw_req *cqr)
+{
+	if (cqr->retries < 0)
+		cqr->status = DASD_CQR_FAILED;
+	else
+		cqr->status = DASD_CQR_FILLED;
+};
+
+static int
+dasd_fba_fill_info(struct dasd_device * device,
+		   struct dasd_information2_t * info)
+{
+	struct dasd_fba_private *private = device->private;
+
+	info->label_block = 1;
+	info->FBA_layout = 1;
+	info->format = DASD_FORMAT_LDL;
+	info->characteristics_size = sizeof(private->rdc_data);
+	memcpy(info->characteristics, &private->rdc_data,
+	       sizeof(private->rdc_data));
+	info->confdata_size = 0;
+	return 0;
+}
+
+static void
+dasd_fba_dump_sense_dbf(struct dasd_device *device, struct irb *irb,
+			char *reason)
+{
+	u64 *sense;
+
+	sense = (u64 *) dasd_get_sense(irb);
+	if (sense) {
+		DBF_DEV_EVENT(DBF_EMERG, device,
+			      "%s: %s %02x%02x%02x %016llx %016llx %016llx "
+			      "%016llx", reason,
+			      scsw_is_tm(&irb->scsw) ? "t" : "c",
+			      scsw_cc(&irb->scsw), scsw_cstat(&irb->scsw),
+			      scsw_dstat(&irb->scsw), sense[0], sense[1],
+			      sense[2], sense[3]);
+	} else {
+		DBF_DEV_EVENT(DBF_EMERG, device, "%s",
+			      "SORRY - NO VALID SENSE AVAILABLE\n");
+	}
+}
+
+
+static void
+dasd_fba_dump_sense(struct dasd_device *device, struct dasd_ccw_req * req,
+		    struct irb *irb)
+{
+	struct ccw1 *act, *end, *last;
+	int len, sl, sct, count;
+	struct device *dev;
+	char *page;
+
+	dev = &device->cdev->dev;
+
+	page = (char *) get_zeroed_page(GFP_ATOMIC);
+	if (page == NULL) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "No memory to dump sense data");
+		return;
+	}
+	len = sprintf(page, "I/O status report:\n");
+	len += sprintf(page + len, "in req: %px CS: 0x%02X DS: 0x%02X\n",
+		       req, irb->scsw.cmd.cstat, irb->scsw.cmd.dstat);
+	len += sprintf(page + len, "Failing CCW: %px\n",
+		       (void *)(u64)dma32_to_u32(irb->scsw.cmd.cpa));
+	if (irb->esw.esw0.erw.cons) {
+		for (sl = 0; sl < 4; sl++) {
+			len += sprintf(page + len, "Sense(hex) %2d-%2d:",
+				       (8 * sl), ((8 * sl) + 7));
+
+			for (sct = 0; sct < 8; sct++) {
+				len += sprintf(page + len, " %02x",
+					       irb->ecw[8 * sl + sct]);
+			}
+			len += sprintf(page + len, "\n");
+		}
+	} else {
+		len += sprintf(page + len, "SORRY - NO VALID SENSE AVAILABLE\n");
+	}
+	dev_err(dev, "%s", page);
+
+	/* dump the Channel Program */
+	/* print first CCWs (maximum 8) */
+	act = req->cpaddr;
+	for (last = act; last->flags & (CCW_FLAG_CC | CCW_FLAG_DC); last++);
+	end = min(act + 8, last);
+	len = sprintf(page, "Related CP in req: %px\n", req);
+	while (act <= end) {
+		len += sprintf(page + len, "CCW %px: %08X %08X DAT:",
+			       act, ((int *) act)[0], ((int *) act)[1]);
+		for (count = 0; count < 32 && count < act->count;
+		     count += sizeof(int))
+			len += sprintf(page + len, " %08X",
+				       ((int *)dma32_to_virt(act->cda))
+				       [(count>>2)]);
+		len += sprintf(page + len, "\n");
+		act++;
+	}
+	dev_err(dev, "%s", page);
+
+	/* print failing CCW area */
+	len = 0;
+	if (act < ((struct ccw1 *)dma32_to_virt(irb->scsw.cmd.cpa)) - 2) {
+		act = ((struct ccw1 *)dma32_to_virt(irb->scsw.cmd.cpa)) - 2;
+		len += sprintf(page + len, "......\n");
+	}
+	end = min((struct ccw1 *)dma32_to_virt(irb->scsw.cmd.cpa) + 2, last);
+	while (act <= end) {
+		len += sprintf(page + len, "CCW %px: %08X %08X DAT:",
+			       act, ((int *) act)[0], ((int *) act)[1]);
+		for (count = 0; count < 32 && count < act->count;
+		     count += sizeof(int))
+			len += sprintf(page + len, " %08X",
+				       ((int *)dma32_to_virt(act->cda))
+				       [(count>>2)]);
+		len += sprintf(page + len, "\n");
+		act++;
+	}
+
+	/* print last CCWs */
+	if (act <  last - 2) {
+		act = last - 2;
+		len += sprintf(page + len, "......\n");
+	}
+	while (act <= last) {
+		len += sprintf(page + len, "CCW %px: %08X %08X DAT:",
+			       act, ((int *) act)[0], ((int *) act)[1]);
+		for (count = 0; count < 32 && count < act->count;
+		     count += sizeof(int))
+			len += sprintf(page + len, " %08X",
+				       ((int *)dma32_to_virt(act->cda))
+				       [(count>>2)]);
+		len += sprintf(page + len, "\n");
+		act++;
+	}
+	if (len > 0)
+		dev_err(dev, "%s", page);
+	free_page((unsigned long) page);
+}
+
+static unsigned int dasd_fba_max_sectors(struct dasd_block *block)
+{
+	return DASD_FBA_MAX_BLOCKS << block->s2b_shift;
+}
+
+static int dasd_fba_pe_handler(struct dasd_device *device,
+			       __u8 tbvpm, __u8 fcsecpm)
+{
+	return dasd_generic_verify_path(device, tbvpm);
+}
+
+static struct dasd_discipline dasd_fba_discipline = {
+	.owner = THIS_MODULE,
+	.name = "FBA ",
+	.ebcname = "FBA ",
+	.has_discard = true,
+	.check_device = dasd_fba_check_characteristics,
+	.do_analysis = dasd_fba_do_analysis,
+	.pe_handler = dasd_fba_pe_handler,
+	.max_sectors = dasd_fba_max_sectors,
+	.fill_geometry = dasd_fba_fill_geometry,
+	.start_IO = dasd_start_IO,
+	.term_IO = dasd_term_IO,
+	.handle_terminated_request = dasd_fba_handle_terminated_request,
+	.erp_action = dasd_fba_erp_action,
+	.erp_postaction = dasd_fba_erp_postaction,
+	.check_for_device_change = dasd_fba_check_for_device_change,
+	.build_cp = dasd_fba_build_cp,
+	.free_cp = dasd_fba_free_cp,
+	.dump_sense = dasd_fba_dump_sense,
+	.dump_sense_dbf = dasd_fba_dump_sense_dbf,
+	.fill_info = dasd_fba_fill_info,
+};
+
+static int __init
+dasd_fba_init(void)
+{
+	int ret;
+
+	ASCEBC(dasd_fba_discipline.ebcname, 4);
+
+	dasd_fba_zero_page = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!dasd_fba_zero_page)
+		return -ENOMEM;
+
+	ret = ccw_driver_register(&dasd_fba_driver);
+	if (!ret)
+		wait_for_device_probe();
+
+	return ret;
+}
+
+static void __exit
+dasd_fba_cleanup(void)
+{
+	ccw_driver_unregister(&dasd_fba_driver);
+	free_page((unsigned long)dasd_fba_zero_page);
+}
+
+module_init(dasd_fba_init);
+module_exit(dasd_fba_cleanup);
diff --git a/drivers/s390/block/dasd_fba.h b/drivers/s390/block/dasd_fba.h
new file mode 100644
index 000000000..45ddabec4
--- /dev/null
+++ b/drivers/s390/block/dasd_fba.h
@@ -0,0 +1,77 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2000
+ *
+ */
+
+#ifndef DASD_FBA_H
+#define DASD_FBA_H
+
+/*
+ * Maximum number of blocks to be chained
+ */
+#define DASD_FBA_MAX_BLOCKS		96
+
+struct DE_fba_data {
+	struct {
+		unsigned char perm:2;	/* Permissions on this extent */
+		unsigned char zero:2;	/* Must be zero */
+		unsigned char da:1;	/* usually zero */
+		unsigned char diag:1;	/* allow diagnose */
+		unsigned char zero2:2;	/* zero */
+	} __attribute__ ((packed)) mask;
+	__u8 zero;		/* Must be zero */
+	__u16 blk_size;		/* Blocksize */
+	__u32 ext_loc;		/* Extent locator */
+	__u32 ext_beg;		/* logical number of block 0 in extent */
+	__u32 ext_end;		/* logocal number of last block in extent */
+} __attribute__ ((packed));
+
+struct LO_fba_data {
+	struct {
+		unsigned char zero:4;
+		unsigned char cmd:4;
+	} __attribute__ ((packed)) operation;
+	__u8 auxiliary;
+	__u16 blk_ct;
+	__u32 blk_nr;
+} __attribute__ ((packed));
+
+struct dasd_fba_characteristics {
+	union {
+		__u8 c;
+		struct {
+			unsigned char reserved:1;
+			unsigned char overrunnable:1;
+			unsigned char burst_byte:1;
+			unsigned char data_chain:1;
+			unsigned char zeros:4;
+		} __attribute__ ((packed)) bits;
+	} __attribute__ ((packed)) mode;
+	union {
+		__u8 c;
+		struct {
+			unsigned char zero0:1;
+			unsigned char removable:1;
+			unsigned char shared:1;
+			unsigned char zero1:1;
+			unsigned char mam:1;
+			unsigned char zeros:3;
+		} __attribute__ ((packed)) bits;
+	} __attribute__ ((packed)) features;
+	__u8 dev_class;
+	__u8 unit_type;
+	__u16 blk_size;
+	__u32 blk_per_cycl;
+	__u32 blk_per_bound;
+	__u32 blk_bdsa;
+	__u32 reserved0;
+	__u16 reserved1;
+	__u16 blk_ce;
+	__u32 reserved2;
+	__u16 reserved3;
+} __attribute__ ((packed));
+
+#endif				/* DASD_FBA_H */
diff --git a/drivers/s390/block/dasd_genhd.c b/drivers/s390/block/dasd_genhd.c
new file mode 100644
index 000000000..6ee3d9524
--- /dev/null
+++ b/drivers/s390/block/dasd_genhd.c
@@ -0,0 +1,233 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Carsten Otte <Cotte@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2001
+ *
+ * gendisk related functions for the dasd driver.
+ *
+ */
+
+#include <linux/interrupt.h>
+#include <linux/major.h>
+#include <linux/fs.h>
+#include <linux/blkpg.h>
+
+#include <linux/uaccess.h>
+
+#include "dasd_int.h"
+
+static unsigned int queue_depth = 32;
+static unsigned int nr_hw_queues = 4;
+static void dasd_gd_free(struct gendisk *gdp);
+
+module_param(queue_depth, uint, 0444);
+MODULE_PARM_DESC(queue_depth, "Default queue depth for new DASD devices");
+
+module_param(nr_hw_queues, uint, 0444);
+MODULE_PARM_DESC(nr_hw_queues, "Default number of hardware queues for new DASD devices");
+
+/*
+ * Set device name.
+ *   dasda - dasdz : 26 devices
+ *   dasdaa - dasdzz : 676 devices, added up = 702
+ *   dasdaaa - dasdzzz : 17576 devices, added up = 18278
+ *   dasdaaaa - dasdzzzz : 456976 devices, added up = 475252
+ */
+static int dasd_name_format(char *prefix, int index, char *buf, int buflen)
+{
+	const int base = 'z' - 'a' + 1;
+	char *begin = buf + strlen(prefix);
+	char *end = buf + buflen;
+	char *p;
+	int unit;
+
+	p = end - 1;
+	*p = '\0';
+	unit = base;
+	do {
+		if (p == begin)
+			return -EINVAL;
+		*--p = 'a' + (index % unit);
+		index = (index / unit) - 1;
+	} while (index >= 0);
+
+	memmove(begin, p, end - p);
+	memcpy(buf, prefix, strlen(prefix));
+
+	return 0;
+}
+
+/*
+ * Allocate and register gendisk structure for device.
+ */
+int dasd_gendisk_alloc(struct dasd_block *block)
+{
+	struct queue_limits lim = {
+		/*
+		 * With page sized segments, each segment can be translated into
+		 * one idaw/tidaw.
+		 */
+		.max_segment_size = PAGE_SIZE,
+		.seg_boundary_mask = PAGE_SIZE - 1,
+		.max_segments = USHRT_MAX,
+	};
+	struct gendisk *gdp;
+	struct dasd_device *base;
+	unsigned int devindex;
+	int rc;
+
+	/* Make sure the minor for this device exists. */
+	base = block->base;
+	devindex = base->devindex;
+	if (devindex >= DASD_PER_MAJOR)
+		return -EBUSY;
+
+	block->tag_set.ops = &dasd_mq_ops;
+	block->tag_set.cmd_size = sizeof(struct dasd_ccw_req);
+	block->tag_set.nr_hw_queues = nr_hw_queues;
+	block->tag_set.queue_depth = queue_depth;
+	block->tag_set.numa_node = NUMA_NO_NODE;
+	rc = blk_mq_alloc_tag_set(&block->tag_set);
+	if (rc)
+		return rc;
+
+	gdp = blk_mq_alloc_disk(&block->tag_set, &lim, block);
+	if (IS_ERR(gdp)) {
+		blk_mq_free_tag_set(&block->tag_set);
+		return PTR_ERR(gdp);
+	}
+
+	/* Initialize gendisk structure. */
+	gdp->major = DASD_MAJOR;
+	gdp->first_minor = devindex << DASD_PARTN_BITS;
+	gdp->minors = 1 << DASD_PARTN_BITS;
+	gdp->fops = &dasd_device_operations;
+
+	rc = dasd_name_format("dasd", devindex, gdp->disk_name, sizeof(gdp->disk_name));
+	if (rc) {
+		DBF_DEV_EVENT(DBF_ERR, block->base,
+			      "setting disk name failed, rc %d", rc);
+		dasd_gd_free(gdp);
+		return rc;
+	}
+
+	if (base->features & DASD_FEATURE_READONLY ||
+	    test_bit(DASD_FLAG_DEVICE_RO, &base->flags))
+		set_disk_ro(gdp, 1);
+	dasd_add_link_to_gendisk(gdp, base);
+	block->gdp = gdp;
+	set_capacity(block->gdp, 0);
+
+	rc = device_add_disk(&base->cdev->dev, block->gdp, NULL);
+	if (rc) {
+		dasd_gendisk_free(block);
+		return rc;
+	}
+
+	return 0;
+}
+
+/*
+ * Free gendisk structure
+ */
+static void dasd_gd_free(struct gendisk *gd)
+{
+	del_gendisk(gd);
+	gd->private_data = NULL;
+	put_disk(gd);
+}
+
+/*
+ * Unregister and free gendisk structure for device.
+ */
+void dasd_gendisk_free(struct dasd_block *block)
+{
+	if (block->gdp) {
+		dasd_gd_free(block->gdp);
+		block->gdp = NULL;
+		blk_mq_free_tag_set(&block->tag_set);
+	}
+}
+
+/*
+ * Trigger a partition detection.
+ */
+int dasd_scan_partitions(struct dasd_block *block)
+{
+	struct file *bdev_file;
+	int rc;
+
+	bdev_file = bdev_file_open_by_dev(disk_devt(block->gdp), BLK_OPEN_READ,
+				       NULL, NULL);
+	if (IS_ERR(bdev_file)) {
+		DBF_DEV_EVENT(DBF_ERR, block->base,
+			      "scan partitions error, blkdev_get returned %ld",
+			      PTR_ERR(bdev_file));
+		return -ENODEV;
+	}
+
+	mutex_lock(&block->gdp->open_mutex);
+	rc = bdev_disk_changed(block->gdp, false);
+	mutex_unlock(&block->gdp->open_mutex);
+	if (rc)
+		DBF_DEV_EVENT(DBF_ERR, block->base,
+				"scan partitions error, rc %d", rc);
+
+	/*
+	 * Since the matching fput() call to the
+	 * bdev_file_open_by_path() in this function is not called before
+	 * dasd_destroy_partitions the offline open_count limit needs to be
+	 * increased from 0 to 1. This is done by setting device->bdev_file
+	 * (see dasd_generic_set_offline). As long as the partition detection
+	 * is running no offline should be allowed. That is why the assignment
+	 * to block->bdev_file is done AFTER the BLKRRPART ioctl.
+	 */
+	block->bdev_file = bdev_file;
+	return 0;
+}
+
+/*
+ * Remove all inodes in the system for a device, delete the
+ * partitions and make device unusable by setting its size to zero.
+ */
+void dasd_destroy_partitions(struct dasd_block *block)
+{
+	struct file *bdev_file;
+
+	/*
+	 * Get the bdev_file pointer from the device structure and clear
+	 * device->bdev_file to lower the offline open_count limit again.
+	 */
+	bdev_file = block->bdev_file;
+	block->bdev_file = NULL;
+
+	mutex_lock(&file_bdev(bdev_file)->bd_disk->open_mutex);
+	bdev_disk_changed(file_bdev(bdev_file)->bd_disk, true);
+	mutex_unlock(&file_bdev(bdev_file)->bd_disk->open_mutex);
+
+	/* Matching blkdev_put to the blkdev_get in dasd_scan_partitions. */
+	fput(bdev_file);
+}
+
+int dasd_gendisk_init(void)
+{
+	int rc;
+
+	/* Register to static dasd major 94 */
+	rc = register_blkdev(DASD_MAJOR, "dasd");
+	if (rc != 0) {
+		pr_warn("Registering the device driver with major number %d failed\n",
+			DASD_MAJOR);
+		return rc;
+	}
+	return 0;
+}
+
+void dasd_gendisk_exit(void)
+{
+	unregister_blkdev(DASD_MAJOR, "dasd");
+}
diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h
new file mode 100644
index 000000000..ef4930432
--- /dev/null
+++ b/drivers/s390/block/dasd_int.h
@@ -0,0 +1,1606 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2009
+ */
+
+#ifndef DASD_INT_H
+#define DASD_INT_H
+
+/* we keep old device allocation scheme; IOW, minors are still in 0..255 */
+#define DASD_PER_MAJOR (1U << (MINORBITS - DASD_PARTN_BITS))
+#define DASD_PARTN_MASK ((1 << DASD_PARTN_BITS) - 1)
+
+/*
+ * States a dasd device can have:
+ *   new: the dasd_device structure is allocated.
+ *   known: the discipline for the device is identified.
+ *   basic: the device can do basic i/o.
+ *   unfmt: the device could not be analyzed (format is unknown).
+ *   ready: partition detection is done and the device is can do block io.
+ *   online: the device accepts requests from the block device queue.
+ *
+ * Things to do for startup state transitions:
+ *   new -> known: find discipline for the device and create devfs entries.
+ *   known -> basic: request irq line for the device.
+ *   basic -> ready: do the initial analysis, e.g. format detection,
+ *                   do block device setup and detect partitions.
+ *   ready -> online: schedule the device tasklet.
+ * Things to do for shutdown state transitions:
+ *   online -> ready: just set the new device state.
+ *   ready -> basic: flush requests from the block device layer, clear
+ *                   partition information and reset format information.
+ *   basic -> known: terminate all requests and free irq.
+ *   known -> new: remove devfs entries and forget discipline.
+ */
+
+#define DASD_STATE_NEW	  0
+#define DASD_STATE_KNOWN  1
+#define DASD_STATE_BASIC  2
+#define DASD_STATE_UNFMT  3
+#define DASD_STATE_READY  4
+#define DASD_STATE_ONLINE 5
+
+#include <linux/module.h>
+#include <linux/wait.h>
+#include <linux/blkdev.h>
+#include <linux/hdreg.h>
+#include <linux/interrupt.h>
+#include <linux/log2.h>
+#include <asm/ccwdev.h>
+#include <linux/workqueue.h>
+#include <asm/debug.h>
+#include <asm/dasd.h>
+#include <asm/idals.h>
+#include <linux/bitops.h>
+#include <linux/blk-mq.h>
+
+/* DASD discipline magic */
+#define DASD_ECKD_MAGIC 0xC5C3D2C4
+#define DASD_DIAG_MAGIC 0xC4C9C1C7
+#define DASD_FBA_MAGIC 0xC6C2C140
+
+/*
+ * SECTION: Type definitions
+ */
+struct dasd_device;
+struct dasd_block;
+
+/* BIT DEFINITIONS FOR SENSE DATA */
+#define DASD_SENSE_BIT_0 0x80
+#define DASD_SENSE_BIT_1 0x40
+#define DASD_SENSE_BIT_2 0x20
+#define DASD_SENSE_BIT_3 0x10
+
+/* BIT DEFINITIONS FOR SIM SENSE */
+#define DASD_SIM_SENSE 0x0F
+#define DASD_SIM_MSG_TO_OP 0x03
+#define DASD_SIM_LOG 0x0C
+
+/* lock class for nested cdev lock */
+#define CDEV_NESTED_FIRST 1
+#define CDEV_NESTED_SECOND 2
+
+/*
+ * SECTION: MACROs for klogd and s390 debug feature (dbf)
+ */
+#define DBF_DEV_EVENT(d_level, d_device, d_str, d_data...) \
+do { \
+	debug_sprintf_event(d_device->debug_area, \
+			    d_level, \
+			    d_str "\n", \
+			    d_data); \
+} while(0)
+
+#define DBF_EVENT(d_level, d_str, d_data...)\
+do { \
+	debug_sprintf_event(dasd_debug_area, \
+			    d_level,\
+			    d_str "\n", \
+			    d_data); \
+} while(0)
+
+#define DBF_EVENT_DEVID(d_level, d_cdev, d_str, d_data...)	\
+do { \
+	struct ccw_dev_id __dev_id;			\
+	ccw_device_get_id(d_cdev, &__dev_id);		\
+	debug_sprintf_event(dasd_debug_area,		\
+			    d_level,					\
+			    "0.%x.%04x " d_str "\n",			\
+			    __dev_id.ssid, __dev_id.devno, d_data);	\
+} while (0)
+
+/* definition of dbf debug levels */
+#define	DBF_EMERG	0	/* system is unusable			*/
+#define	DBF_ALERT	1	/* action must be taken immediately	*/
+#define	DBF_CRIT	2	/* critical conditions			*/
+#define	DBF_ERR		3	/* error conditions			*/
+#define	DBF_WARNING	4	/* warning conditions			*/
+#define	DBF_NOTICE	5	/* normal but significant condition	*/
+#define	DBF_INFO	6	/* informational			*/
+#define	DBF_DEBUG	6	/* debug-level messages			*/
+
+/* Macro to calculate number of blocks per page */
+#define BLOCKS_PER_PAGE(blksize) (PAGE_SIZE / blksize)
+
+struct dasd_ccw_req {
+	unsigned int magic;		/* Eye catcher */
+	int intrc;			/* internal error, e.g. from start_IO */
+	struct list_head devlist;	/* for dasd_device request queue */
+	struct list_head blocklist;	/* for dasd_block request queue */
+	struct dasd_block *block;	/* the originating block device */
+	struct dasd_device *memdev;	/* the device used to allocate this */
+	struct dasd_device *startdev;	/* device the request is started on */
+	struct dasd_device *basedev;	/* base device if no block->base */
+	void *cpaddr;			/* address of ccw or tcw */
+	short retries;			/* A retry counter */
+	unsigned char cpmode;		/* 0 = cmd mode, 1 = itcw */
+	char status;			/* status of this request */
+	char lpm;			/* logical path mask */
+	unsigned long flags;        	/* flags of this request */
+	struct dasd_queue *dq;
+	unsigned long starttime;	/* jiffies time of request start */
+	unsigned long expires;		/* expiration period in jiffies */
+	void *data;			/* pointer to data area */
+	struct irb irb;			/* device status in case of an error */
+	struct dasd_ccw_req *refers;	/* ERP-chain queueing. */
+	void *function; 		/* originating ERP action */
+	void *mem_chunk;
+
+	unsigned long buildclk;		/* TOD-clock of request generation */
+	unsigned long startclk;		/* TOD-clock of request start */
+	unsigned long stopclk;		/* TOD-clock of request interrupt */
+	unsigned long endclk;		/* TOD-clock of request termination */
+
+	void (*callback)(struct dasd_ccw_req *, void *data);
+	void *callback_data;
+	unsigned int proc_bytes;	/* bytes for partial completion */
+	unsigned int trkcount;		/* count formatted tracks */
+	void *filldata;			/* address of filler data */
+	struct dasd_format_entry *format;
+	sector_t start_trk;
+	sector_t end_trk;
+	bool collision;
+};
+
+/*
+ * dasd_ccw_req -> status can be:
+ */
+#define DASD_CQR_FILLED 	0x00	/* request is ready to be processed */
+#define DASD_CQR_DONE		0x01	/* request is completed successfully */
+#define DASD_CQR_NEED_ERP	0x02	/* request needs recovery action */
+#define DASD_CQR_IN_ERP 	0x03	/* request is in recovery */
+#define DASD_CQR_FAILED 	0x04	/* request is finally failed */
+#define DASD_CQR_TERMINATED	0x05	/* request was stopped by driver */
+#define DASD_CQR_ABORTED	0x06	/* request was replaced and will be deleted */
+
+#define DASD_CQR_QUEUED 	0x80	/* request is queued to be processed */
+#define DASD_CQR_IN_IO		0x81	/* request is currently in IO */
+#define DASD_CQR_ERROR		0x82	/* request is completed with error */
+#define DASD_CQR_CLEAR_PENDING	0x83	/* request is clear pending */
+#define DASD_CQR_CLEARED	0x84	/* request was cleared */
+#define DASD_CQR_SUCCESS	0x85	/* request was successful */
+#define DASD_CQR_ABORT		0x86	/* request was replaced and will not be handled */
+
+/* default expiration time*/
+#define DASD_EXPIRES	  300
+#define DASD_EXPIRES_MAX  40000000
+#define DASD_RETRIES	  256
+#define DASD_RETRIES_MAX  32768
+
+/* per dasd_ccw_req flags */
+#define DASD_CQR_FLAGS_USE_ERP   0	/* use ERP for this request */
+#define DASD_CQR_FLAGS_FAILFAST  1	/* FAILFAST */
+#define DASD_CQR_VERIFY_PATH	 2	/* path verification request */
+#define DASD_CQR_ALLOW_SLOCK	 3	/* Try this request even when lock was
+					 * stolen. Should not be combined with
+					 * DASD_CQR_FLAGS_USE_ERP
+					 */
+/*
+ * The following flags are used to suppress output of certain errors.
+ */
+#define DASD_CQR_SUPPRESS_NRF	4	/* Suppress 'No Record Found' error */
+#define DASD_CQR_SUPPRESS_IT	5	/* Suppress 'Invalid Track' error*/
+#define DASD_CQR_SUPPRESS_IL	6	/* Suppress 'Incorrect Length' error */
+#define DASD_CQR_SUPPRESS_CR	7	/* Suppress 'Command Reject' error */
+
+#define DASD_REQ_PER_DEV 4
+
+/* Signature for error recovery functions. */
+typedef struct dasd_ccw_req *(*dasd_erp_fn_t) (struct dasd_ccw_req *);
+
+/*
+ * A single CQR can only contain a maximum of 255 CCWs. It is limited by
+ * the locate record and locate record extended count value which can only hold
+ * 1 Byte max.
+ */
+#define DASD_CQR_MAX_CCW 255
+
+/*
+ * Unique identifier for dasd device.
+ */
+#define UA_NOT_CONFIGURED  0x00
+#define UA_BASE_DEVICE	   0x01
+#define UA_BASE_PAV_ALIAS  0x02
+#define UA_HYPER_PAV_ALIAS 0x03
+
+struct dasd_uid {
+	__u8 type;
+	char vendor[4];
+	char serial[15];
+	__u16 ssid;
+	__u8 real_unit_addr;
+	__u8 base_unit_addr;
+	char vduit[33];
+};
+
+#define DASD_UID_STRLEN ( /* vendor */ 3 + 1 + /* serial    */ 14 + 1 +	\
+			  /* SSID   */ 4 + 1 + /* unit addr */ 2 + 1 +	\
+			  /* vduit */ 32 + 1)
+
+/*
+ * PPRC Status data
+ */
+struct dasd_pprc_header {
+	__u8 entries;		/* 0     Number of device entries */
+	__u8 unused;		/* 1     unused */
+	__u16 entry_length;	/* 2-3   Length of device entry */
+	__u32 unused2;		/* 4-7   unused */
+} __packed;
+
+struct dasd_pprc_dev_info {
+	__u8 state;		/* 0       Copy State */
+	__u8 flags;		/* 1       Flags */
+	__u8 reserved1[2];	/* 2-3     reserved */
+	__u8 prim_lss;		/* 4       Primary device LSS */
+	__u8 primary;		/* 5       Primary device address */
+	__u8 sec_lss;		/* 6       Secondary device LSS */
+	__u8 secondary;		/* 7       Secondary device address */
+	__u16 pprc_id;		/* 8-9     Peer-to-Peer Remote Copy ID */
+	__u8 reserved2[12];	/* 10-21   reserved */
+	__u16 prim_cu_ssid;	/* 22-23   Primary Control Unit SSID */
+	__u8 reserved3[12];	/* 24-35   reserved */
+	__u16 sec_cu_ssid;	/* 36-37   Secondary Control Unit SSID */
+	__u8 reserved4[90];	/* 38-127  reserved */
+} __packed;
+
+struct dasd_pprc_data_sc4 {
+	struct dasd_pprc_header header;
+	struct dasd_pprc_dev_info dev_info[5];
+} __packed;
+
+#define DASD_BUS_ID_SIZE 20
+#define DASD_CP_ENTRIES 5
+
+struct dasd_copy_entry {
+	char busid[DASD_BUS_ID_SIZE];
+	struct dasd_device *device;
+	bool primary;
+	bool configured;
+};
+
+struct dasd_copy_relation {
+	struct dasd_copy_entry entry[DASD_CP_ENTRIES];
+	struct dasd_copy_entry *active;
+};
+
+int dasd_devmap_set_device_copy_relation(struct ccw_device *,
+					 bool pprc_enabled);
+
+/*
+ * the struct dasd_discipline is
+ * sth like a table of virtual functions, if you think of dasd_eckd
+ * inheriting dasd...
+ * no, currently we are not planning to reimplement the driver in C++
+ */
+struct dasd_discipline {
+	struct module *owner;
+	char ebcname[8];	/* a name used for tagging and printks */
+	char name[8];		/* a name used for tagging and printks */
+	bool has_discard;
+
+	struct list_head list;	/* used for list of disciplines */
+
+	/*
+	 * Device recognition functions. check_device is used to verify
+	 * the sense data and the information returned by read device
+	 * characteristics. It returns 0 if the discipline can be used
+	 * for the device in question. uncheck_device is called during
+	 * device shutdown to deregister a device from its discipline.
+	 */
+	int (*check_device) (struct dasd_device *);
+	void (*uncheck_device) (struct dasd_device *);
+
+	/*
+	 * do_analysis is used in the step from device state "basic" to
+	 * state "accept". It returns 0 if the device can be made ready,
+	 * it returns -EMEDIUMTYPE if the device can't be made ready or
+	 * -EAGAIN if do_analysis started a ccw that needs to complete
+	 * before the analysis may be repeated.
+	 */
+	int (*do_analysis) (struct dasd_block *);
+
+	/*
+	 * This function is called, when new paths become available.
+	 * Disciplins may use this callback to do necessary setup work,
+	 * e.g. verify that new path is compatible with the current
+	 * configuration.
+	 */
+	int (*pe_handler)(struct dasd_device *, __u8, __u8);
+
+	/*
+	 * Last things to do when a device is set online, and first things
+	 * when it is set offline.
+	 */
+	int (*basic_to_ready) (struct dasd_device *);
+	int (*online_to_ready) (struct dasd_device *);
+	int (*basic_to_known)(struct dasd_device *);
+
+	unsigned int (*max_sectors)(struct dasd_block *);
+	/* (struct dasd_device *);
+	 * Device operation functions. build_cp creates a ccw chain for
+	 * a block device request, start_io starts the request and
+	 * term_IO cancels it (e.g. in case of a timeout). format_device
+	 * formats the device and check_device_format compares the format of
+	 * a device with the expected format_data.
+	 * handle_terminated_request allows to examine a cqr and prepare
+	 * it for retry.
+	 */
+	struct dasd_ccw_req *(*build_cp) (struct dasd_device *,
+					  struct dasd_block *,
+					  struct request *);
+	int (*start_IO) (struct dasd_ccw_req *);
+	int (*term_IO) (struct dasd_ccw_req *);
+	void (*handle_terminated_request) (struct dasd_ccw_req *);
+	int (*format_device) (struct dasd_device *,
+			      struct format_data_t *, int);
+	int (*check_device_format)(struct dasd_device *,
+				   struct format_check_t *, int);
+	int (*free_cp) (struct dasd_ccw_req *, struct request *);
+
+	/*
+	 * Error recovery functions. examine_error() returns a value that
+	 * indicates what to do for an error condition. If examine_error()
+	 * returns 'dasd_era_recover' erp_action() is called to create a
+	 * special error recovery ccw. erp_postaction() is called after
+	 * an error recovery ccw has finished its execution. dump_sense
+	 * is called for every error condition to print the sense data
+	 * to the console.
+	 */
+	dasd_erp_fn_t(*erp_action) (struct dasd_ccw_req *);
+	dasd_erp_fn_t(*erp_postaction) (struct dasd_ccw_req *);
+	void (*dump_sense) (struct dasd_device *, struct dasd_ccw_req *,
+			    struct irb *);
+	void (*dump_sense_dbf) (struct dasd_device *, struct irb *, char *);
+	void (*check_for_device_change) (struct dasd_device *,
+					 struct dasd_ccw_req *,
+					 struct irb *);
+
+        /* i/o control functions. */
+	int (*fill_geometry) (struct dasd_block *, struct hd_geometry *);
+	int (*fill_info) (struct dasd_device *, struct dasd_information2_t *);
+	int (*ioctl) (struct dasd_block *, unsigned int, void __user *);
+
+	/* reload device after state change */
+	int (*reload) (struct dasd_device *);
+
+	int (*get_uid) (struct dasd_device *, struct dasd_uid *);
+	void (*kick_validate) (struct dasd_device *);
+	int (*check_attention)(struct dasd_device *, __u8);
+	int (*host_access_count)(struct dasd_device *);
+	int (*hosts_print)(struct dasd_device *, struct seq_file *);
+	void (*handle_hpf_error)(struct dasd_device *, struct irb *);
+	void (*disable_hpf)(struct dasd_device *);
+	int (*hpf_enabled)(struct dasd_device *);
+	void (*reset_path)(struct dasd_device *, __u8);
+
+	/*
+	 * Extent Space Efficient (ESE) relevant functions
+	 */
+	int (*is_ese)(struct dasd_device *);
+	int (*ese_capable)(struct dasd_device *);
+	/* Whether the volume is formatted on demand (thin), from the label */
+	int (*on_demand_format)(struct dasd_device *);
+	/* Fill discard queue limits */
+	void (*disc_limits)(struct dasd_block *, struct queue_limits *);
+	/* Capacity */
+	int (*space_allocated)(struct dasd_device *);
+	int (*space_configured)(struct dasd_device *);
+	int (*logical_capacity)(struct dasd_device *);
+	int (*release_space)(struct dasd_device *, struct format_data_t *);
+	/* Extent Pool */
+	int (*ext_pool_id)(struct dasd_device *);
+	int (*ext_size)(struct dasd_device *);
+	int (*ext_pool_cap_at_warnlevel)(struct dasd_device *);
+	int (*ext_pool_warn_thrshld)(struct dasd_device *);
+	int (*ext_pool_oos)(struct dasd_device *);
+	int (*ext_pool_exhaust)(struct dasd_device *, struct dasd_ccw_req *);
+	void (*ese_format)(struct dasd_device *, struct dasd_ccw_req *, struct irb *);
+	int (*ese_read)(struct dasd_ccw_req *, struct irb *);
+	int (*pprc_status)(struct dasd_device *, struct	dasd_pprc_data_sc4 *);
+	bool (*pprc_enabled)(struct dasd_device *);
+	int (*copy_pair_swap)(struct dasd_device *, char *, char *);
+	int (*device_ping)(struct dasd_device *);
+};
+
+extern struct dasd_discipline *dasd_diag_discipline_pointer;
+
+/* Trigger IDs for extended error reporting DASD EER and autoquiesce */
+enum eer_trigger {
+	DASD_EER_FATALERROR = 1,
+	DASD_EER_NOPATH,
+	DASD_EER_STATECHANGE,
+	DASD_EER_PPRCSUSPEND,
+	DASD_EER_NOSPC,
+	DASD_EER_TIMEOUTS,
+	DASD_EER_STARTIO,
+
+	/* enum end marker, only add new trigger above */
+	DASD_EER_MAX,
+	DASD_EER_AUTOQUIESCE = 31, /* internal only */
+};
+
+#define DASD_EER_VALID ((1U << DASD_EER_MAX) - 1)
+
+/* DASD path handling */
+
+#define DASD_PATH_OPERATIONAL  1
+#define DASD_PATH_TBV	       2
+#define DASD_PATH_PP	       3
+#define DASD_PATH_NPP	       4
+#define DASD_PATH_MISCABLED    5
+#define DASD_PATH_NOHPF        6
+#define DASD_PATH_CUIR	       7
+#define DASD_PATH_IFCC	       8
+#define DASD_PATH_FCSEC	       9
+
+#define DASD_THRHLD_MAX		4294967295U
+#define DASD_INTERVAL_MAX	4294967295U
+
+/* FC Endpoint Security Capabilities */
+#define DASD_FC_SECURITY_UNSUP		0
+#define DASD_FC_SECURITY_AUTH		1
+#define DASD_FC_SECURITY_ENC_FCSP2	2
+#define DASD_FC_SECURITY_ENC_ERAS	3
+
+#define DASD_FC_SECURITY_ENC_STR	"Encryption"
+static const struct {
+	u8 value;
+	char *name;
+} dasd_path_fcs_mnemonics[] = {
+	{ DASD_FC_SECURITY_UNSUP,	"Unsupported" },
+	{ DASD_FC_SECURITY_AUTH,	"Authentication" },
+	{ DASD_FC_SECURITY_ENC_FCSP2,	DASD_FC_SECURITY_ENC_STR },
+	{ DASD_FC_SECURITY_ENC_ERAS,	DASD_FC_SECURITY_ENC_STR },
+};
+
+static inline char *dasd_path_get_fcs_str(int val)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(dasd_path_fcs_mnemonics); i++) {
+		if (dasd_path_fcs_mnemonics[i].value == val)
+			return dasd_path_fcs_mnemonics[i].name;
+	}
+
+	return dasd_path_fcs_mnemonics[0].name;
+}
+
+struct dasd_path {
+	unsigned long flags;
+	u8 cssid;
+	u8 ssid;
+	u8 chpid;
+	struct dasd_conf_data *conf_data;
+	atomic_t error_count;
+	unsigned long errorclk;
+	u8 fc_security;
+	struct kobject kobj;
+	bool in_sysfs;
+};
+
+#define to_dasd_path(path) container_of(path, struct dasd_path, kobj)
+
+static inline void dasd_path_release(struct kobject *kobj)
+{
+/* Memory for the dasd_path kobject is freed when dasd_free_device() is called */
+}
+
+
+struct dasd_profile_info {
+	/* legacy part of profile data, as in dasd_profile_info_t */
+	unsigned int dasd_io_reqs;	 /* number of requests processed */
+	unsigned int dasd_io_sects;	 /* number of sectors processed */
+	unsigned int dasd_io_secs[32];	 /* histogram of request's sizes */
+	unsigned int dasd_io_times[32];	 /* histogram of requests's times */
+	unsigned int dasd_io_timps[32];	 /* h. of requests's times per sector */
+	unsigned int dasd_io_time1[32];	 /* hist. of time from build to start */
+	unsigned int dasd_io_time2[32];	 /* hist. of time from start to irq */
+	unsigned int dasd_io_time2ps[32]; /* hist. of time from start to irq */
+	unsigned int dasd_io_time3[32];	 /* hist. of time from irq to end */
+	unsigned int dasd_io_nr_req[32]; /* hist. of # of requests in chanq */
+
+	/* new data */
+	struct timespec64 starttod;	   /* time of start or last reset */
+	unsigned int dasd_io_alias;	   /* requests using an alias */
+	unsigned int dasd_io_tpm;	   /* requests using transport mode */
+	unsigned int dasd_read_reqs;	   /* total number of read  requests */
+	unsigned int dasd_read_sects;	   /* total number read sectors */
+	unsigned int dasd_read_alias;	   /* read request using an alias */
+	unsigned int dasd_read_tpm;	   /* read requests in transport mode */
+	unsigned int dasd_read_secs[32];   /* histogram of request's sizes */
+	unsigned int dasd_read_times[32];  /* histogram of requests's times */
+	unsigned int dasd_read_time1[32];  /* hist. time from build to start */
+	unsigned int dasd_read_time2[32];  /* hist. of time from start to irq */
+	unsigned int dasd_read_time3[32];  /* hist. of time from irq to end */
+	unsigned int dasd_read_nr_req[32]; /* hist. of # of requests in chanq */
+	unsigned long dasd_sum_times;	   /* sum of request times */
+	unsigned long dasd_sum_time_str;   /* sum of time from build to start */
+	unsigned long dasd_sum_time_irq;   /* sum of time from start to irq */
+	unsigned long dasd_sum_time_end;   /* sum of time from irq to end */
+};
+
+struct dasd_profile {
+	struct dentry *dentry;
+	struct dasd_profile_info *data;
+	spinlock_t lock;
+};
+
+/*
+ * concurrent ESE format ranges in flight; also caps a WRITE_FULL_TRACK's
+ * track count, which the LRE track bitmask limits to 16
+ */
+#define DASD_NR_FORMAT_ENTRIES	16
+
+struct dasd_format_entry {
+	struct list_head list;
+	struct dasd_ccw_req *cqr;
+	sector_t start_trk;
+	sector_t end_trk;
+};
+
+struct dasd_device {
+	/* Block device stuff. */
+	struct dasd_block *block;
+
+        unsigned int devindex;
+	unsigned long flags;	   /* per device flags */
+	unsigned short features;   /* copy of devmap-features (read-only!) */
+
+	/* extended error reporting stuff (eer) */
+	struct dasd_ccw_req *eer_cqr;
+
+	/* Device discipline stuff. */
+	struct dasd_discipline *discipline;
+	struct dasd_discipline *base_discipline;
+	void *private;
+	struct dasd_path path[8];
+	__u8 opm;
+
+	/* Device state and target state. */
+	int state, target;
+	struct mutex state_mutex;
+	int stopped;		/* device (ccw_device_start) was stopped */
+
+	/* reference count. */
+        atomic_t ref_count;
+
+	/* ccw queue and memory for static ccw/erp buffers. */
+	struct list_head ccw_queue;
+	spinlock_t mem_lock;
+	void *ccw_mem;
+	void *fill_mem;
+	void *erp_mem;
+	void *ese_mem;
+	void *nulldata;
+	struct list_head ccw_chunks;
+	struct list_head fill_chunks;
+	struct list_head erp_chunks;
+	struct list_head ese_chunks;
+
+	atomic_t tasklet_scheduled;
+        struct tasklet_struct tasklet;
+	struct work_struct kick_work;
+	struct work_struct reload_device;
+	struct work_struct kick_validate;
+	struct work_struct suc_work;
+	struct work_struct requeue_requests;
+	struct timer_list timer;
+
+	debug_info_t *debug_area;
+
+	struct ccw_device *cdev;
+
+	/* hook for alias management */
+	struct list_head alias_list;
+
+	/* default expiration time in s */
+	unsigned long default_expires;
+	unsigned long default_retries;
+
+	unsigned long blk_timeout;
+
+	unsigned long path_thrhld;
+	unsigned long path_interval;
+
+	struct dentry *debugfs_dentry;
+	struct dentry *hosts_dentry;
+	struct dasd_profile profile;
+	struct dasd_format_entry format_entry[DASD_NR_FORMAT_ENTRIES];
+	struct kset *paths_info;
+	struct dasd_copy_relation *copy;
+	unsigned long aq_mask;
+	unsigned int aq_timeouts;
+
+	/* ESE fulltrack write control (see full_track_bias sysfs attribute) */
+	unsigned int ft_bias;	/* aggressiveness 0..100: 0=off, 100=always */
+	unsigned int fulltrack;	/* internal: use WRITE_FULL_TRACK for aligned writes */
+	/* adaptive heuristic (active for ft_bias 1..99), derived from ft_bias */
+	unsigned int ese_probe_state;        /* heuristic FSM state */
+	unsigned int ese_probe_interval;     /* IOs between evaluations */
+	atomic_t ese_io_cnt;                 /* IO counter for current window */
+	atomic_t ese_nrf_window;             /* NRF/INV_TRACK_FORMAT events in window */
+	unsigned int ese_heu_start_interval; /* IOs before first probe */
+	unsigned int ese_heu_probe_window;   /* IOs in probe window */
+	unsigned int ese_heu_max_interval;   /* max IOs between probes (backoff cap) */
+	unsigned int ese_heu_nrf_high;       /* NRF per-mille threshold → activate ft1 */
+};
+
+struct dasd_block {
+	/* Block device stuff. */
+	struct gendisk *gdp;
+	spinlock_t request_queue_lock;
+	struct blk_mq_tag_set tag_set;
+	struct file *bdev_file;
+	atomic_t open_count;
+
+	unsigned long blocks;	   /* size of volume in blocks */
+	unsigned int bp_block;	   /* bytes per block */
+	unsigned int s2b_shift;	   /* log2 (bp_block/512) */
+
+	struct dasd_device *base;
+	struct list_head ccw_queue;
+	spinlock_t queue_lock;
+
+	atomic_t tasklet_scheduled;
+	struct tasklet_struct tasklet;
+	struct timer_list timer;
+
+	struct dentry *debugfs_dentry;
+	struct dasd_profile profile;
+
+	struct list_head format_list;
+	spinlock_t format_lock;
+	atomic_t trkcount;
+
+	/*
+	 * ESE format CQRs staged from hardirq, spliced into
+	 * ccw_queue in dasd_block_tasklet under queue_lock. Direct enqueue from
+	 * the IRQ handler would invert the queue_lock / ccwdev_lock order.
+	 */
+	struct list_head ese_staging;
+	/* lock for ese_staging */
+	spinlock_t ese_lock;
+};
+
+struct dasd_attention_data {
+	struct dasd_device *device;
+	__u8 lpum;
+};
+
+struct dasd_queue {
+	spinlock_t lock;
+};
+
+/* reasons why device (ccw_device_start) was stopped */
+#define DASD_STOPPED_NOT_ACC 1         /* not accessible */
+#define DASD_STOPPED_QUIESCE 2         /* Quiesced */
+#define DASD_STOPPED_PENDING 4         /* long busy */
+#define DASD_STOPPED_DC_WAIT 8         /* disconnected, wait */
+#define DASD_STOPPED_SU      16        /* summary unit check handling */
+#define DASD_STOPPED_PPRC    32        /* PPRC swap */
+#define DASD_STOPPED_NOSPC   128       /* no space left */
+
+/*
+ * ESE fulltrack write aggressiveness (full_track_bias sysfs attribute), 0..100:
+ *   0   - never use proactively WRITE_FULL_TRACK
+ *   100 - always use proactively WRITE_FULL_TRACK, no probing
+ *   1..99 - adaptive; higher means switch to ft more eagerly
+ * WRITE_FULL_TRACK has an advantage on sparse formatted ESE devices
+ * but it has an overall penalty for maximum throughput for fully
+ * formatted devices.
+ * The default of 50 tries to balance both and do some probing in between
+ * to choose the best mode for default IO.
+ */
+#define DASD_FT_BIAS_MAX	100
+#define DASD_FT_BIAS_DEFAULT	50
+
+/* ESE fulltrack heuristic FSM states (adaptive range, ft_bias 1..99) */
+#define DASD_ESE_HEU_FT1_ACTIVE   0   /* fulltrack write active */
+#define DASD_ESE_HEU_PROBING      1   /* ft0 probe window, measuring NRF rate */
+#define DASD_ESE_HEU_FT0_STABLE   2   /* device formatted, ft0 active */
+
+/*
+ * Heuristic parameters are derived from ft_bias by linear interpolation,
+ * anchored so that ft_bias == 50 reproduces the previously shipped defaults
+ * and ft_bias == 100 is the most aggressive end of the range.
+ * probe_window is constant.
+ */
+#define DASD_ESE_HEU_PROBE_WINDOW	   100
+#define DASD_ESE_HEU_NRF_HIGH_A50	    10	/* NRF per-mille threshold */
+#define DASD_ESE_HEU_NRF_HIGH_A100	     1
+#define DASD_ESE_HEU_START_A50		  2000	/* IOs before first probe */
+#define DASD_ESE_HEU_START_A100		   500
+#define DASD_ESE_HEU_MAX_A50		500000	/* backoff cap */
+#define DASD_ESE_HEU_MAX_A100		 20000
+
+/* per device flags */
+#define DASD_FLAG_OFFLINE	3	/* device is in offline processing */
+#define DASD_FLAG_EER_SNSS	4	/* A SNSS is required */
+#define DASD_FLAG_EER_IN_USE	5	/* A SNSS request is running */
+#define DASD_FLAG_DEVICE_RO	6	/* The device itself is read-only. Don't
+					 * confuse this with the user specified
+					 * read-only feature.
+					 */
+#define DASD_FLAG_IS_RESERVED	7	/* The device is reserved */
+#define DASD_FLAG_LOCK_STOLEN	8	/* The device lock was stolen */
+#define DASD_FLAG_SUSPENDED	9	/* The device was suspended */
+#define DASD_FLAG_SAFE_OFFLINE	10	/* safe offline processing requested*/
+#define DASD_FLAG_SAFE_OFFLINE_RUNNING	11	/* safe offline running */
+#define DASD_FLAG_ABORTALL	12	/* Abort all noretry requests */
+#define DASD_FLAG_PATH_VERIFY	13	/* Path verification worker running */
+#define DASD_FLAG_SUC		14	/* unhandled summary unit check */
+
+#define DASD_SLEEPON_START_TAG	((void *) 1)
+#define DASD_SLEEPON_END_TAG	((void *) 2)
+
+void dasd_put_device_wake(struct dasd_device *);
+
+/*
+ * return values to be returned from the copy pair swap function
+ * 0x00: swap successful
+ * 0x01: swap data invalid
+ * 0x02: no active device found
+ * 0x03: wrong primary specified
+ * 0x04: secondary device not found
+ * 0x05: swap already running
+ */
+#define DASD_COPYPAIRSWAP_SUCCESS	0
+#define DASD_COPYPAIRSWAP_INVALID	1
+#define DASD_COPYPAIRSWAP_NOACTIVE	2
+#define DASD_COPYPAIRSWAP_PRIMARY	3
+#define DASD_COPYPAIRSWAP_SECONDARY	4
+#define DASD_COPYPAIRSWAP_MULTIPLE	5
+
+/*
+ * Reference count inliners
+ */
+static inline void
+dasd_get_device(struct dasd_device *device)
+{
+	atomic_inc(&device->ref_count);
+}
+
+static inline void
+dasd_put_device(struct dasd_device *device)
+{
+	if (atomic_dec_return(&device->ref_count) == 0)
+		dasd_put_device_wake(device);
+}
+
+/*
+ * The static memory in ccw_mem and erp_mem is managed by a sorted
+ * list of free memory chunks.
+ */
+struct dasd_mchunk
+{
+	struct list_head list;
+	unsigned long size;
+} __attribute__ ((aligned(8)));
+
+static inline void
+dasd_init_chunklist(struct list_head *chunk_list, void *mem,
+		    unsigned long size)
+{
+	struct dasd_mchunk *chunk;
+
+	INIT_LIST_HEAD(chunk_list);
+	chunk = (struct dasd_mchunk *) mem;
+	chunk->size = size - sizeof(struct dasd_mchunk);
+	list_add(&chunk->list, chunk_list);
+}
+
+static inline void *
+dasd_alloc_chunk(struct list_head *chunk_list, unsigned long size)
+{
+	struct dasd_mchunk *chunk, *tmp;
+
+	size = (size + 7L) & -8L;
+	list_for_each_entry(chunk, chunk_list, list) {
+		if (chunk->size < size)
+			continue;
+		if (chunk->size > size + sizeof(struct dasd_mchunk)) {
+			char *endaddr = (char *) (chunk + 1) + chunk->size;
+			tmp = (struct dasd_mchunk *) (endaddr - size) - 1;
+			tmp->size = size;
+			chunk->size -= size + sizeof(struct dasd_mchunk);
+			chunk = tmp;
+		} else
+			list_del(&chunk->list);
+		return (void *) (chunk + 1);
+	}
+	return NULL;
+}
+
+static inline void
+dasd_free_chunk(struct list_head *chunk_list, void *mem)
+{
+	struct dasd_mchunk *chunk, *tmp;
+	struct list_head *p, *left;
+
+	chunk = (struct dasd_mchunk *)
+		((char *) mem - sizeof(struct dasd_mchunk));
+	/* Find out the left neighbour in chunk_list. */
+	left = chunk_list;
+	list_for_each(p, chunk_list) {
+		if (list_entry(p, struct dasd_mchunk, list) > chunk)
+			break;
+		left = p;
+	}
+	/* Try to merge with right neighbour = next element from left. */
+	if (left->next != chunk_list) {
+		tmp = list_entry(left->next, struct dasd_mchunk, list);
+		if ((char *) (chunk + 1) + chunk->size == (char *) tmp) {
+			list_del(&tmp->list);
+			chunk->size += tmp->size + sizeof(struct dasd_mchunk);
+		}
+	}
+	/* Try to merge with left neighbour. */
+	if (left != chunk_list) {
+		tmp = list_entry(left, struct dasd_mchunk, list);
+		if ((char *) (tmp + 1) + tmp->size == (char *) chunk) {
+			tmp->size += chunk->size + sizeof(struct dasd_mchunk);
+			return;
+		}
+	}
+	__list_add(&chunk->list, left, left->next);
+}
+
+/*
+ * Check if bsize is in { 512, 1024, 2048, 4096 }
+ */
+static inline int
+dasd_check_blocksize(int bsize)
+{
+	if (bsize < 512 || bsize > 4096 || !is_power_of_2(bsize))
+		return -EMEDIUMTYPE;
+	return 0;
+}
+
+/*
+ * return the callback data of the original request in case there are
+ * ERP requests build on top of it
+ */
+static inline void *dasd_get_callback_data(struct dasd_ccw_req *cqr)
+{
+	while (cqr->refers)
+		cqr = cqr->refers;
+
+	return cqr->callback_data;
+}
+
+static inline bool dasd_req_conflict(struct dasd_ccw_req *cqr1,
+				     struct dasd_ccw_req *cqr2)
+{
+	return !(cqr1->format->end_trk < cqr2->start_trk ||
+		 cqr2->end_trk < cqr1->format->start_trk);
+}
+
+/*
+ * true when device is ese device and ft_bias selects the adaptive
+ * heuristic (neither hard endpoint)
+ */
+static inline bool dasd_ese_adaptive(struct dasd_device *device)
+{
+	return device->discipline &&
+		device->discipline->is_ese &&
+		device->discipline->is_ese(device) &&
+		device->ft_bias > 0 &&
+		device->ft_bias < DASD_FT_BIAS_MAX;
+}
+
+/*
+ * Linear interpolation of a heuristic parameter between its value at aggr==50
+ * (v50) and its value at aggr==100 (v100).
+ */
+static inline unsigned int dasd_ese_lerp(unsigned int v50, unsigned int v100,
+					 unsigned int aggr)
+{
+	return (unsigned int)((int)v50 +
+		((int)v100 - (int)v50) * ((int)aggr - 50) / 50);
+}
+
+/*
+ * Apply the ft_bias knob. For the hard endpoints just pin the mode; for the
+ * adaptive range derive the heuristic parameters from ft_bias and (re)start
+ * the FSM in ft1 so a freshly sparse device avoids the NRF penalty right away.
+ */
+static inline void dasd_ft_bias_apply(struct dasd_device *device)
+{
+	unsigned int a = device->ft_bias;
+
+	if (!dasd_ese_adaptive(device)) {
+		device->fulltrack = (a >= DASD_FT_BIAS_MAX) ? 1 : 0;
+		device->ese_probe_state = DASD_ESE_HEU_FT1_ACTIVE;
+		return;
+	}
+
+	device->ese_heu_nrf_high =
+		dasd_ese_lerp(DASD_ESE_HEU_NRF_HIGH_A50,
+			      DASD_ESE_HEU_NRF_HIGH_A100, a);
+	device->ese_heu_start_interval =
+		dasd_ese_lerp(DASD_ESE_HEU_START_A50,
+			      DASD_ESE_HEU_START_A100, a);
+	device->ese_heu_max_interval =
+		dasd_ese_lerp(DASD_ESE_HEU_MAX_A50,
+			      DASD_ESE_HEU_MAX_A100, a);
+	device->ese_heu_probe_window = DASD_ESE_HEU_PROBE_WINDOW;
+
+	device->ese_probe_state    = DASD_ESE_HEU_FT1_ACTIVE;
+	device->ese_probe_interval = device->ese_heu_start_interval;
+	device->fulltrack          = 1;
+	atomic_set(&device->ese_io_cnt, 0);
+	atomic_set(&device->ese_nrf_window, 0);
+}
+
+/* externals in dasd.c */
+#define DASD_PROFILE_OFF	 0
+#define DASD_PROFILE_ON 	 1
+#define DASD_PROFILE_GLOBAL_ONLY 2
+
+extern debug_info_t *dasd_debug_area;
+extern struct dasd_profile dasd_global_profile;
+extern unsigned int dasd_global_profile_level;
+extern const struct block_device_operations dasd_device_operations;
+extern struct blk_mq_ops dasd_mq_ops;
+
+extern struct kmem_cache *dasd_page_cache;
+
+struct dasd_ccw_req *
+dasd_smalloc_request(int, int, int, struct dasd_device *, struct dasd_ccw_req *);
+struct dasd_ccw_req *dasd_fmalloc_request(int, int, int, struct dasd_device *);
+void dasd_sfree_request(struct dasd_ccw_req *, struct dasd_device *);
+void dasd_ffree_request(struct dasd_ccw_req *, struct dasd_device *);
+void dasd_wakeup_cb(struct dasd_ccw_req *, void *);
+
+struct dasd_device *dasd_alloc_device(void);
+void dasd_free_device(struct dasd_device *);
+
+struct dasd_block *dasd_alloc_block(void);
+void dasd_free_block(struct dasd_block *);
+
+enum blk_eh_timer_return dasd_times_out(struct request *req);
+
+void dasd_enable_device(struct dasd_device *);
+void dasd_set_target_state(struct dasd_device *, int);
+void dasd_kick_device(struct dasd_device *);
+void dasd_reload_device(struct dasd_device *);
+void dasd_schedule_requeue(struct dasd_device *);
+
+void dasd_add_request_head(struct dasd_ccw_req *);
+void dasd_add_request_tail(struct dasd_ccw_req *);
+int  dasd_start_IO(struct dasd_ccw_req *);
+int  dasd_term_IO(struct dasd_ccw_req *);
+void dasd_schedule_device_bh(struct dasd_device *);
+void dasd_schedule_block_bh(struct dasd_block *);
+int  dasd_sleep_on(struct dasd_ccw_req *);
+int  dasd_sleep_on_queue(struct list_head *);
+int  dasd_sleep_on_immediatly(struct dasd_ccw_req *);
+int  dasd_sleep_on_queue_interruptible(struct list_head *);
+int  dasd_sleep_on_interruptible(struct dasd_ccw_req *);
+void dasd_device_set_timer(struct dasd_device *, int);
+void dasd_device_clear_timer(struct dasd_device *);
+void dasd_block_set_timer(struct dasd_block *, int);
+void dasd_block_clear_timer(struct dasd_block *);
+int  dasd_cancel_req(struct dasd_ccw_req *);
+int dasd_flush_device_queue(struct dasd_device *);
+int dasd_generic_probe(struct ccw_device *);
+void dasd_generic_free_discipline(struct dasd_device *);
+void dasd_generic_remove (struct ccw_device *cdev);
+int dasd_generic_set_online(struct ccw_device *, struct dasd_discipline *);
+int dasd_generic_set_offline (struct ccw_device *cdev);
+int dasd_generic_notify(struct ccw_device *, int);
+int dasd_generic_last_path_gone(struct dasd_device *);
+int dasd_generic_path_operational(struct dasd_device *);
+void dasd_generic_shutdown(struct ccw_device *);
+
+void dasd_generic_handle_state_change(struct dasd_device *);
+enum uc_todo dasd_generic_uc_handler(struct ccw_device *, struct irb *);
+void dasd_generic_path_event(struct ccw_device *, int *);
+int dasd_generic_verify_path(struct dasd_device *, __u8);
+void dasd_generic_space_exhaust(struct dasd_device *, struct dasd_ccw_req *);
+void dasd_generic_space_avail(struct dasd_device *);
+
+int dasd_generic_requeue_all_requests(struct dasd_device *);
+
+int dasd_generic_read_dev_chars(struct dasd_device *, int, void *, int);
+char *dasd_get_sense(struct irb *);
+
+void dasd_device_set_stop_bits(struct dasd_device *, int);
+void dasd_device_remove_stop_bits(struct dasd_device *, int);
+
+int dasd_device_is_ro(struct dasd_device *);
+
+void dasd_profile_reset(struct dasd_profile *);
+int dasd_profile_on(struct dasd_profile *);
+void dasd_profile_off(struct dasd_profile *);
+char *dasd_get_user_string(const char __user *, size_t);
+
+/* externals in dasd_devmap.c */
+extern int dasd_max_devindex;
+extern int dasd_probeonly;
+extern int dasd_autodetect;
+extern int dasd_nopav;
+extern int dasd_nofcx;
+
+int dasd_devmap_init(void);
+void dasd_devmap_exit(void);
+
+struct dasd_device *dasd_create_device(struct ccw_device *);
+void dasd_delete_device(struct dasd_device *);
+
+int dasd_get_feature(struct ccw_device *, int);
+int dasd_set_feature(struct ccw_device *, int, int);
+
+extern const struct attribute_group *dasd_dev_groups[];
+void dasd_path_create_kobj(struct dasd_device *, int);
+void dasd_path_create_kobjects(struct dasd_device *);
+void dasd_path_remove_kobjects(struct dasd_device *);
+
+struct dasd_device *dasd_device_from_cdev(struct ccw_device *);
+struct dasd_device *dasd_device_from_cdev_locked(struct ccw_device *);
+struct dasd_device *dasd_device_from_devindex(int);
+
+void dasd_add_link_to_gendisk(struct gendisk *, struct dasd_device *);
+struct dasd_device *dasd_device_from_gendisk(struct gendisk *);
+
+int dasd_parse(void) __init;
+int dasd_busid_known(const char *);
+
+/* externals in dasd_gendisk.c */
+int  dasd_gendisk_init(void);
+void dasd_gendisk_exit(void);
+int dasd_gendisk_alloc(struct dasd_block *);
+void dasd_gendisk_free(struct dasd_block *);
+int dasd_scan_partitions(struct dasd_block *);
+void dasd_destroy_partitions(struct dasd_block *);
+
+/* externals in dasd_ioctl.c */
+int dasd_ioctl(struct block_device *bdev, blk_mode_t mode, unsigned int cmd,
+		unsigned long arg);
+int dasd_set_read_only(struct block_device *bdev, bool ro);
+
+/* externals in dasd_proc.c */
+int dasd_proc_init(void);
+void dasd_proc_exit(void);
+
+/* externals in dasd_erp.c */
+struct dasd_ccw_req *dasd_default_erp_action(struct dasd_ccw_req *);
+struct dasd_ccw_req *dasd_default_erp_postaction(struct dasd_ccw_req *);
+struct dasd_ccw_req *dasd_alloc_erp_request(unsigned int, int, int,
+					    struct dasd_device *);
+void dasd_free_erp_request(struct dasd_ccw_req *, struct dasd_device *);
+void dasd_log_sense(struct dasd_ccw_req *, struct irb *);
+void dasd_log_sense_dbf(struct dasd_ccw_req *cqr, struct irb *irb);
+
+/* externals in dasd_3990_erp.c */
+struct dasd_ccw_req *dasd_3990_erp_action(struct dasd_ccw_req *);
+void dasd_3990_erp_handle_sim(struct dasd_device *, char *);
+
+/* externals in dasd_eer.c */
+#ifdef CONFIG_DASD_EER
+int dasd_eer_init(void);
+void dasd_eer_exit(void);
+int dasd_eer_enable(struct dasd_device *);
+void dasd_eer_disable(struct dasd_device *);
+void dasd_eer_write(struct dasd_device *, struct dasd_ccw_req *cqr,
+		    unsigned int id);
+void dasd_eer_snss(struct dasd_device *);
+
+static inline int dasd_eer_enabled(struct dasd_device *device)
+{
+	return device->eer_cqr != NULL;
+}
+#else
+#define dasd_eer_init()		(0)
+#define dasd_eer_exit()		do { } while (0)
+#define dasd_eer_enable(d)	(0)
+#define dasd_eer_disable(d)	do { } while (0)
+#define dasd_eer_write(d,c,i)	do { } while (0)
+#define dasd_eer_snss(d)	do { } while (0)
+#define dasd_eer_enabled(d)	(0)
+#endif	/* CONFIG_DASD_ERR */
+
+
+/* DASD path handling functions */
+
+/*
+ * helper functions to modify bit masks for a given channel path for a device
+ */
+static inline int dasd_path_is_operational(struct dasd_device *device, int chp)
+{
+	return test_bit(DASD_PATH_OPERATIONAL, &device->path[chp].flags);
+}
+
+static inline int dasd_path_need_verify(struct dasd_device *device, int chp)
+{
+	return test_bit(DASD_PATH_TBV, &device->path[chp].flags);
+}
+
+static inline void dasd_path_verify(struct dasd_device *device, int chp)
+{
+	__set_bit(DASD_PATH_TBV, &device->path[chp].flags);
+}
+
+static inline void dasd_path_clear_verify(struct dasd_device *device, int chp)
+{
+	__clear_bit(DASD_PATH_TBV, &device->path[chp].flags);
+}
+
+static inline void dasd_path_clear_all_verify(struct dasd_device *device)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		dasd_path_clear_verify(device, chp);
+}
+
+static inline void dasd_path_fcsec(struct dasd_device *device, int chp)
+{
+	__set_bit(DASD_PATH_FCSEC, &device->path[chp].flags);
+}
+
+static inline void dasd_path_clear_fcsec(struct dasd_device *device, int chp)
+{
+	__clear_bit(DASD_PATH_FCSEC, &device->path[chp].flags);
+}
+
+static inline int dasd_path_need_fcsec(struct dasd_device *device, int chp)
+{
+	return test_bit(DASD_PATH_FCSEC, &device->path[chp].flags);
+}
+
+static inline void dasd_path_clear_all_fcsec(struct dasd_device *device)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		dasd_path_clear_fcsec(device, chp);
+}
+
+static inline void dasd_path_operational(struct dasd_device *device, int chp)
+{
+	__set_bit(DASD_PATH_OPERATIONAL, &device->path[chp].flags);
+	device->opm |= (0x80 >> chp);
+}
+
+static inline void dasd_path_nonpreferred(struct dasd_device *device, int chp)
+{
+	__set_bit(DASD_PATH_NPP, &device->path[chp].flags);
+}
+
+static inline int dasd_path_is_nonpreferred(struct dasd_device *device, int chp)
+{
+	return test_bit(DASD_PATH_NPP, &device->path[chp].flags);
+}
+
+static inline void dasd_path_clear_nonpreferred(struct dasd_device *device,
+						int chp)
+{
+	__clear_bit(DASD_PATH_NPP, &device->path[chp].flags);
+}
+
+static inline void dasd_path_preferred(struct dasd_device *device, int chp)
+{
+	__set_bit(DASD_PATH_PP, &device->path[chp].flags);
+}
+
+static inline int dasd_path_is_preferred(struct dasd_device *device, int chp)
+{
+	return test_bit(DASD_PATH_PP, &device->path[chp].flags);
+}
+
+static inline void dasd_path_clear_preferred(struct dasd_device *device,
+					     int chp)
+{
+	__clear_bit(DASD_PATH_PP, &device->path[chp].flags);
+}
+
+static inline void dasd_path_clear_oper(struct dasd_device *device, int chp)
+{
+	__clear_bit(DASD_PATH_OPERATIONAL, &device->path[chp].flags);
+	device->opm &= ~(0x80 >> chp);
+}
+
+static inline void dasd_path_clear_cable(struct dasd_device *device, int chp)
+{
+	__clear_bit(DASD_PATH_MISCABLED, &device->path[chp].flags);
+}
+
+static inline void dasd_path_cuir(struct dasd_device *device, int chp)
+{
+	__set_bit(DASD_PATH_CUIR, &device->path[chp].flags);
+}
+
+static inline int dasd_path_is_cuir(struct dasd_device *device, int chp)
+{
+	return test_bit(DASD_PATH_CUIR, &device->path[chp].flags);
+}
+
+static inline void dasd_path_clear_cuir(struct dasd_device *device, int chp)
+{
+	__clear_bit(DASD_PATH_CUIR, &device->path[chp].flags);
+}
+
+static inline void dasd_path_ifcc(struct dasd_device *device, int chp)
+{
+	set_bit(DASD_PATH_IFCC, &device->path[chp].flags);
+}
+
+static inline int dasd_path_is_ifcc(struct dasd_device *device, int chp)
+{
+	return test_bit(DASD_PATH_IFCC, &device->path[chp].flags);
+}
+
+static inline void dasd_path_clear_ifcc(struct dasd_device *device, int chp)
+{
+	clear_bit(DASD_PATH_IFCC, &device->path[chp].flags);
+}
+
+static inline void dasd_path_clear_nohpf(struct dasd_device *device, int chp)
+{
+	__clear_bit(DASD_PATH_NOHPF, &device->path[chp].flags);
+}
+
+static inline void dasd_path_miscabled(struct dasd_device *device, int chp)
+{
+	__set_bit(DASD_PATH_MISCABLED, &device->path[chp].flags);
+}
+
+static inline int dasd_path_is_miscabled(struct dasd_device *device, int chp)
+{
+	return test_bit(DASD_PATH_MISCABLED, &device->path[chp].flags);
+}
+
+static inline void dasd_path_nohpf(struct dasd_device *device, int chp)
+{
+	__set_bit(DASD_PATH_NOHPF, &device->path[chp].flags);
+}
+
+static inline int dasd_path_is_nohpf(struct dasd_device *device, int chp)
+{
+	return test_bit(DASD_PATH_NOHPF, &device->path[chp].flags);
+}
+
+/*
+ * get functions for path masks
+ * will return a path masks for the given device
+ */
+
+static inline __u8 dasd_path_get_opm(struct dasd_device *device)
+{
+	return device->opm;
+}
+
+static inline __u8 dasd_path_get_tbvpm(struct dasd_device *device)
+{
+	int chp;
+	__u8 tbvpm = 0x00;
+
+	for (chp = 0; chp < 8; chp++)
+		if (dasd_path_need_verify(device, chp))
+			tbvpm |= 0x80 >> chp;
+	return tbvpm;
+}
+
+static inline int dasd_path_get_fcsecpm(struct dasd_device *device)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		if (dasd_path_need_fcsec(device, chp))
+			return 1;
+
+	return 0;
+}
+
+static inline __u8 dasd_path_get_nppm(struct dasd_device *device)
+{
+	int chp;
+	__u8 npm = 0x00;
+
+	for (chp = 0; chp < 8; chp++) {
+		if (dasd_path_is_nonpreferred(device, chp))
+			npm |= 0x80 >> chp;
+	}
+	return npm;
+}
+
+static inline __u8 dasd_path_get_ppm(struct dasd_device *device)
+{
+	int chp;
+	__u8 ppm = 0x00;
+
+	for (chp = 0; chp < 8; chp++)
+		if (dasd_path_is_preferred(device, chp))
+			ppm |= 0x80 >> chp;
+	return ppm;
+}
+
+static inline __u8 dasd_path_get_cablepm(struct dasd_device *device)
+{
+	int chp;
+	__u8 cablepm = 0x00;
+
+	for (chp = 0; chp < 8; chp++)
+		if (dasd_path_is_miscabled(device, chp))
+			cablepm |= 0x80 >> chp;
+	return cablepm;
+}
+
+static inline __u8 dasd_path_get_cuirpm(struct dasd_device *device)
+{
+	int chp;
+	__u8 cuirpm = 0x00;
+
+	for (chp = 0; chp < 8; chp++)
+		if (dasd_path_is_cuir(device, chp))
+			cuirpm |= 0x80 >> chp;
+	return cuirpm;
+}
+
+static inline __u8 dasd_path_get_ifccpm(struct dasd_device *device)
+{
+	int chp;
+	__u8 ifccpm = 0x00;
+
+	for (chp = 0; chp < 8; chp++)
+		if (dasd_path_is_ifcc(device, chp))
+			ifccpm |= 0x80 >> chp;
+	return ifccpm;
+}
+
+static inline __u8 dasd_path_get_hpfpm(struct dasd_device *device)
+{
+	int chp;
+	__u8 hpfpm = 0x00;
+
+	for (chp = 0; chp < 8; chp++)
+		if (dasd_path_is_nohpf(device, chp))
+			hpfpm |= 0x80 >> chp;
+	return hpfpm;
+}
+
+static inline u8 dasd_path_get_fcs_path(struct dasd_device *device, int chp)
+{
+	return device->path[chp].fc_security;
+}
+
+static inline int dasd_path_get_fcs_device(struct dasd_device *device)
+{
+	u8 fc_sec = 0;
+	int chp;
+
+	for (chp = 0; chp < 8; chp++) {
+		if (device->opm & (0x80 >> chp)) {
+			fc_sec = device->path[chp].fc_security;
+			break;
+		}
+	}
+	for (; chp < 8; chp++) {
+		if (device->opm & (0x80 >> chp))
+			if (device->path[chp].fc_security != fc_sec)
+				return -EINVAL;
+	}
+
+	return fc_sec;
+}
+
+/*
+ * add functions for path masks
+ * the existing path mask will be extended by the given path mask
+ */
+static inline void dasd_path_add_tbvpm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		if (pm & (0x80 >> chp))
+			dasd_path_verify(device, chp);
+}
+
+static inline __u8 dasd_path_get_notoperpm(struct dasd_device *device)
+{
+	int chp;
+	__u8 nopm = 0x00;
+
+	for (chp = 0; chp < 8; chp++)
+		if (dasd_path_is_nohpf(device, chp) ||
+		    dasd_path_is_ifcc(device, chp) ||
+		    dasd_path_is_cuir(device, chp) ||
+		    dasd_path_is_miscabled(device, chp))
+			nopm |= 0x80 >> chp;
+	return nopm;
+}
+
+static inline void dasd_path_add_opm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		if (pm & (0x80 >> chp)) {
+			dasd_path_operational(device, chp);
+			/*
+			 * if the path is used
+			 * it should not be in one of the negative lists
+			 */
+			dasd_path_clear_nohpf(device, chp);
+			dasd_path_clear_cuir(device, chp);
+			dasd_path_clear_cable(device, chp);
+			dasd_path_clear_ifcc(device, chp);
+		}
+}
+
+static inline void dasd_path_add_cablepm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		if (pm & (0x80 >> chp))
+			dasd_path_miscabled(device, chp);
+}
+
+static inline void dasd_path_add_cuirpm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		if (pm & (0x80 >> chp))
+			dasd_path_cuir(device, chp);
+}
+
+static inline void dasd_path_add_ifccpm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		if (pm & (0x80 >> chp))
+			dasd_path_ifcc(device, chp);
+}
+
+static inline void dasd_path_add_nppm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		if (pm & (0x80 >> chp))
+			dasd_path_nonpreferred(device, chp);
+}
+
+static inline void dasd_path_add_nohpfpm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		if (pm & (0x80 >> chp))
+			dasd_path_nohpf(device, chp);
+}
+
+static inline void dasd_path_add_ppm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		if (pm & (0x80 >> chp))
+			dasd_path_preferred(device, chp);
+}
+
+static inline void dasd_path_add_fcsecpm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		if (pm & (0x80 >> chp))
+			dasd_path_fcsec(device, chp);
+}
+
+/*
+ * set functions for path masks
+ * the existing path mask will be replaced by the given path mask
+ */
+static inline void dasd_path_set_tbvpm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		if (pm & (0x80 >> chp))
+			dasd_path_verify(device, chp);
+		else
+			dasd_path_clear_verify(device, chp);
+}
+
+static inline void dasd_path_set_opm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++) {
+		dasd_path_clear_oper(device, chp);
+		if (pm & (0x80 >> chp)) {
+			dasd_path_operational(device, chp);
+			/*
+			 * if the path is used
+			 * it should not be in one of the negative lists
+			 */
+			dasd_path_clear_nohpf(device, chp);
+			dasd_path_clear_cuir(device, chp);
+			dasd_path_clear_cable(device, chp);
+			dasd_path_clear_ifcc(device, chp);
+		}
+	}
+}
+
+/*
+ * remove functions for path masks
+ * the existing path mask will be cleared with the given path mask
+ */
+static inline void dasd_path_remove_opm(struct dasd_device *device, __u8 pm)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++) {
+		if (pm & (0x80 >> chp))
+			dasd_path_clear_oper(device, chp);
+	}
+}
+
+/*
+ * add the newly available path to the to be verified pm and remove it from
+ * normal operation until it is verified
+ */
+static inline void dasd_path_available(struct dasd_device *device, int chp)
+{
+	dasd_path_clear_oper(device, chp);
+	dasd_path_verify(device, chp);
+}
+
+static inline void dasd_path_notoper(struct dasd_device *device, int chp)
+{
+	dasd_path_clear_oper(device, chp);
+	dasd_path_clear_preferred(device, chp);
+	dasd_path_clear_nonpreferred(device, chp);
+}
+
+static inline void dasd_path_fcsec_update(struct dasd_device *device, int chp)
+{
+	dasd_path_fcsec(device, chp);
+}
+
+/*
+ * remove all paths from normal operation
+ */
+static inline void dasd_path_no_path(struct dasd_device *device)
+{
+	int chp;
+
+	for (chp = 0; chp < 8; chp++)
+		dasd_path_notoper(device, chp);
+
+	dasd_path_clear_all_verify(device);
+}
+
+/* end - path handling */
+
+#endif				/* DASD_H */
diff --git a/drivers/s390/block/dasd_ioctl.c b/drivers/s390/block/dasd_ioctl.c
new file mode 100644
index 000000000..e5b8b413f
--- /dev/null
+++ b/drivers/s390/block/dasd_ioctl.c
@@ -0,0 +1,726 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Carsten Otte <Cotte@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2001
+ *
+ * i/o controls for the dasd driver.
+ */
+
+#include <linux/interrupt.h>
+#include <linux/export.h>
+#include <linux/major.h>
+#include <linux/fs.h>
+#include <linux/blkpg.h>
+#include <linux/slab.h>
+#include <asm/ccwdev.h>
+#include <asm/schid.h>
+#include <asm/cmb.h>
+#include <linux/uaccess.h>
+#include <linux/dasd_mod.h>
+
+#include "dasd_int.h"
+
+static int
+dasd_ioctl_api_version(void __user *argp)
+{
+	int ver = DASD_API_VERSION;
+	return put_user(ver, (int __user *)argp);
+}
+
+/*
+ * Enable device.
+ * used by dasdfmt after BIODASDDISABLE to retrigger blocksize detection
+ */
+static int
+dasd_ioctl_enable(struct block_device *bdev)
+{
+	struct dasd_device *base;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	base = dasd_device_from_gendisk(bdev->bd_disk);
+	if (!base)
+		return -ENODEV;
+
+	dasd_enable_device(base);
+	dasd_put_device(base);
+	return 0;
+}
+
+/*
+ * Disable device.
+ * Used by dasdfmt. Disable I/O operations but allow ioctls.
+ */
+static int
+dasd_ioctl_disable(struct block_device *bdev)
+{
+	struct dasd_device *base;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	base = dasd_device_from_gendisk(bdev->bd_disk);
+	if (!base)
+		return -ENODEV;
+	/*
+	 * Man this is sick. We don't do a real disable but only downgrade
+	 * the device to DASD_STATE_BASIC. The reason is that dasdfmt uses
+	 * BIODASDDISABLE to disable accesses to the device via the block
+	 * device layer but it still wants to do i/o on the device by
+	 * using the BIODASDFMT ioctl. Therefore the correct state for the
+	 * device is DASD_STATE_BASIC that allows to do basic i/o.
+	 */
+	dasd_set_target_state(base, DASD_STATE_BASIC);
+	/*
+	 * Set i_size to zero, since read, write, etc. check against this
+	 * value.
+	 */
+	set_capacity(bdev->bd_disk, 0);
+	dasd_put_device(base);
+	return 0;
+}
+
+/*
+ * Quiesce device.
+ */
+static int dasd_ioctl_quiesce(struct dasd_block *block)
+{
+	unsigned long flags;
+	struct dasd_device *base;
+
+	base = block->base;
+	if (!capable (CAP_SYS_ADMIN))
+		return -EACCES;
+
+	pr_info("%s: The DASD has been put in the quiesce "
+		"state\n", dev_name(&base->cdev->dev));
+	spin_lock_irqsave(get_ccwdev_lock(base->cdev), flags);
+	dasd_device_set_stop_bits(base, DASD_STOPPED_QUIESCE);
+	spin_unlock_irqrestore(get_ccwdev_lock(base->cdev), flags);
+	return 0;
+}
+
+
+/*
+ * Resume device.
+ */
+static int dasd_ioctl_resume(struct dasd_block *block)
+{
+	unsigned long flags;
+	struct dasd_device *base;
+
+	base = block->base;
+	if (!capable (CAP_SYS_ADMIN))
+		return -EACCES;
+
+	pr_info("%s: I/O operations have been resumed "
+		"on the DASD\n", dev_name(&base->cdev->dev));
+	spin_lock_irqsave(get_ccwdev_lock(base->cdev), flags);
+	dasd_device_remove_stop_bits(base, DASD_STOPPED_QUIESCE);
+	spin_unlock_irqrestore(get_ccwdev_lock(base->cdev), flags);
+
+	dasd_schedule_block_bh(block);
+	dasd_schedule_device_bh(base);
+	return 0;
+}
+
+/*
+ * Abort all failfast I/O on a device.
+ */
+static int dasd_ioctl_abortio(struct dasd_block *block)
+{
+	unsigned long flags;
+	struct dasd_device *base;
+	struct dasd_ccw_req *cqr, *n;
+
+	base = block->base;
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	if (test_and_set_bit(DASD_FLAG_ABORTALL, &base->flags))
+		return 0;
+	DBF_DEV_EVENT(DBF_NOTICE, base, "%s", "abortall flag set");
+
+	spin_lock_irqsave(&block->request_queue_lock, flags);
+	spin_lock(&block->queue_lock);
+	list_for_each_entry_safe(cqr, n, &block->ccw_queue, blocklist) {
+		if (test_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags) &&
+		    cqr->callback_data &&
+		    cqr->callback_data != DASD_SLEEPON_START_TAG &&
+		    cqr->callback_data != DASD_SLEEPON_END_TAG) {
+			spin_unlock(&block->queue_lock);
+			blk_abort_request(cqr->callback_data);
+			spin_lock(&block->queue_lock);
+		}
+	}
+	spin_unlock(&block->queue_lock);
+	spin_unlock_irqrestore(&block->request_queue_lock, flags);
+
+	dasd_schedule_block_bh(block);
+	return 0;
+}
+
+/*
+ * Allow I/O on a device
+ */
+static int dasd_ioctl_allowio(struct dasd_block *block)
+{
+	struct dasd_device *base;
+
+	base = block->base;
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	if (test_and_clear_bit(DASD_FLAG_ABORTALL, &base->flags))
+		DBF_DEV_EVENT(DBF_NOTICE, base, "%s", "abortall flag unset");
+
+	return 0;
+}
+
+/*
+ * performs formatting of _device_ according to _fdata_
+ * Note: The discipline's format_function is assumed to deliver formatting
+ * commands to format multiple units of the device. In terms of the ECKD
+ * devices this means CCWs are generated to format multiple tracks.
+ */
+static int
+dasd_format(struct dasd_block *block, struct format_data_t *fdata)
+{
+	struct dasd_device *base;
+	int rc;
+
+	base = block->base;
+	if (base->discipline->format_device == NULL)
+		return -EPERM;
+
+	if (base->state != DASD_STATE_BASIC) {
+		pr_warn("%s: The DASD cannot be formatted while it is enabled\n",
+			dev_name(&base->cdev->dev));
+		return -EBUSY;
+	}
+
+	DBF_DEV_EVENT(DBF_NOTICE, base,
+		      "formatting units %u to %u (%u B blocks) flags %u",
+		      fdata->start_unit,
+		      fdata->stop_unit, fdata->blksize, fdata->intensity);
+
+	/* Since dasdfmt keeps the device open after it was disabled,
+	 * there still exists an inode for this device.
+	 * We must update i_blkbits, otherwise we might get errors when
+	 * enabling the device later.
+	 */
+	if (fdata->start_unit == 0) {
+		block->gdp->part0->bd_mapping->host->i_blkbits =
+			blksize_bits(fdata->blksize);
+	}
+
+	rc = base->discipline->format_device(base, fdata, 1);
+	if (rc == -EAGAIN)
+		rc = base->discipline->format_device(base, fdata, 0);
+
+	return rc;
+}
+
+static int dasd_check_format(struct dasd_block *block,
+			     struct format_check_t *cdata)
+{
+	struct dasd_device *base;
+	int rc;
+
+	base = block->base;
+	if (!base->discipline->check_device_format)
+		return -ENOTTY;
+
+	rc = base->discipline->check_device_format(base, cdata, 1);
+	if (rc == -EAGAIN)
+		rc = base->discipline->check_device_format(base, cdata, 0);
+
+	return rc;
+}
+
+/*
+ * Format device.
+ */
+static int
+dasd_ioctl_format(struct block_device *bdev, void __user *argp)
+{
+	struct dasd_device *base;
+	struct format_data_t fdata;
+	int rc;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+	if (!argp)
+		return -EINVAL;
+	base = dasd_device_from_gendisk(bdev->bd_disk);
+	if (!base)
+		return -ENODEV;
+	if (base->features & DASD_FEATURE_READONLY ||
+	    test_bit(DASD_FLAG_DEVICE_RO, &base->flags)) {
+		dasd_put_device(base);
+		return -EROFS;
+	}
+	if (copy_from_user(&fdata, argp, sizeof(struct format_data_t))) {
+		dasd_put_device(base);
+		return -EFAULT;
+	}
+	if (bdev_is_partition(bdev)) {
+		pr_warn("%s: The specified DASD is a partition and cannot be formatted\n",
+			dev_name(&base->cdev->dev));
+		dasd_put_device(base);
+		return -EINVAL;
+	}
+	rc = dasd_format(base->block, &fdata);
+	dasd_put_device(base);
+
+	return rc;
+}
+
+/*
+ * Check device format
+ */
+static int dasd_ioctl_check_format(struct block_device *bdev, void __user *argp)
+{
+	struct format_check_t cdata;
+	struct dasd_device *base;
+	int rc = 0;
+
+	if (!argp)
+		return -EINVAL;
+
+	base = dasd_device_from_gendisk(bdev->bd_disk);
+	if (!base)
+		return -ENODEV;
+	if (bdev_is_partition(bdev)) {
+		pr_warn("%s: The specified DASD is a partition and cannot be checked\n",
+			dev_name(&base->cdev->dev));
+		rc = -EINVAL;
+		goto out_err;
+	}
+
+	if (copy_from_user(&cdata, argp, sizeof(cdata))) {
+		rc = -EFAULT;
+		goto out_err;
+	}
+
+	rc = dasd_check_format(base->block, &cdata);
+	if (rc)
+		goto out_err;
+
+	if (copy_to_user(argp, &cdata, sizeof(cdata)))
+		rc = -EFAULT;
+
+out_err:
+	dasd_put_device(base);
+
+	return rc;
+}
+
+static int dasd_release_space(struct dasd_device *device,
+			      struct format_data_t *rdata)
+{
+	if (!device->discipline->is_ese || !device->discipline->is_ese(device))
+		return -ENOTSUPP;
+	if (!device->discipline->release_space)
+		return -ENOTSUPP;
+
+	return device->discipline->release_space(device, rdata);
+}
+
+/*
+ * Release allocated space
+ */
+static int dasd_ioctl_release_space(struct block_device *bdev, void __user *argp)
+{
+	struct format_data_t rdata;
+	struct dasd_device *base;
+	int rc = 0;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+	if (!argp)
+		return -EINVAL;
+
+	base = dasd_device_from_gendisk(bdev->bd_disk);
+	if (!base)
+		return -ENODEV;
+	if (base->features & DASD_FEATURE_READONLY ||
+	    test_bit(DASD_FLAG_DEVICE_RO, &base->flags)) {
+		rc = -EROFS;
+		goto out_err;
+	}
+	if (bdev_is_partition(bdev)) {
+		pr_warn("%s: The specified DASD is a partition and tracks cannot be released\n",
+			dev_name(&base->cdev->dev));
+		rc = -EINVAL;
+		goto out_err;
+	}
+
+	if (copy_from_user(&rdata, argp, sizeof(rdata))) {
+		rc = -EFAULT;
+		goto out_err;
+	}
+
+	rc = dasd_release_space(base, &rdata);
+
+out_err:
+	dasd_put_device(base);
+
+	return rc;
+}
+
+/*
+ * Swap driver iternal copy relation.
+ */
+static int
+dasd_ioctl_copy_pair_swap(struct block_device *bdev, void __user *argp)
+{
+	struct dasd_copypair_swap_data_t data;
+	struct dasd_device *device;
+	int rc;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	device = dasd_device_from_gendisk(bdev->bd_disk);
+	if (!device)
+		return -ENODEV;
+
+	if (copy_from_user(&data, argp, sizeof(struct dasd_copypair_swap_data_t))) {
+		dasd_put_device(device);
+		return -EFAULT;
+	}
+	if (memchr_inv(data.reserved, 0, sizeof(data.reserved))) {
+		pr_warn("%s: Invalid swap data specified\n",
+			dev_name(&device->cdev->dev));
+		dasd_put_device(device);
+		return DASD_COPYPAIRSWAP_INVALID;
+	}
+	if (bdev_is_partition(bdev)) {
+		pr_warn("%s: The specified DASD is a partition and cannot be swapped\n",
+			dev_name(&device->cdev->dev));
+		dasd_put_device(device);
+		return DASD_COPYPAIRSWAP_INVALID;
+	}
+	if (!device->copy) {
+		pr_warn("%s: The specified DASD has no copy pair set up\n",
+			dev_name(&device->cdev->dev));
+		dasd_put_device(device);
+		return -ENODEV;
+	}
+	if (!device->discipline->copy_pair_swap) {
+		dasd_put_device(device);
+		return -EOPNOTSUPP;
+	}
+	rc = device->discipline->copy_pair_swap(device, data.primary,
+						data.secondary);
+	dasd_put_device(device);
+
+	return rc;
+}
+
+#ifdef CONFIG_DASD_PROFILE
+/*
+ * Reset device profile information
+ */
+static int dasd_ioctl_reset_profile(struct dasd_block *block)
+{
+	dasd_profile_reset(&block->profile);
+	return 0;
+}
+
+/*
+ * Return device profile information
+ */
+static int dasd_ioctl_read_profile(struct dasd_block *block, void __user *argp)
+{
+	struct dasd_profile_info_t *data;
+	int rc = 0;
+
+	data = kmalloc_obj(*data);
+	if (!data)
+		return -ENOMEM;
+
+	spin_lock_bh(&block->profile.lock);
+	if (block->profile.data) {
+		data->dasd_io_reqs = block->profile.data->dasd_io_reqs;
+		data->dasd_io_sects = block->profile.data->dasd_io_sects;
+		memcpy(data->dasd_io_secs, block->profile.data->dasd_io_secs,
+		       sizeof(data->dasd_io_secs));
+		memcpy(data->dasd_io_times, block->profile.data->dasd_io_times,
+		       sizeof(data->dasd_io_times));
+		memcpy(data->dasd_io_timps, block->profile.data->dasd_io_timps,
+		       sizeof(data->dasd_io_timps));
+		memcpy(data->dasd_io_time1, block->profile.data->dasd_io_time1,
+		       sizeof(data->dasd_io_time1));
+		memcpy(data->dasd_io_time2, block->profile.data->dasd_io_time2,
+		       sizeof(data->dasd_io_time2));
+		memcpy(data->dasd_io_time2ps,
+		       block->profile.data->dasd_io_time2ps,
+		       sizeof(data->dasd_io_time2ps));
+		memcpy(data->dasd_io_time3, block->profile.data->dasd_io_time3,
+		       sizeof(data->dasd_io_time3));
+		memcpy(data->dasd_io_nr_req,
+		       block->profile.data->dasd_io_nr_req,
+		       sizeof(data->dasd_io_nr_req));
+		spin_unlock_bh(&block->profile.lock);
+	} else {
+		spin_unlock_bh(&block->profile.lock);
+		rc = -EIO;
+		goto out;
+	}
+	if (copy_to_user(argp, data, sizeof(*data)))
+		rc = -EFAULT;
+out:
+	kfree(data);
+	return rc;
+}
+#else
+static int dasd_ioctl_reset_profile(struct dasd_block *block)
+{
+	return -ENOTTY;
+}
+
+static int dasd_ioctl_read_profile(struct dasd_block *block, void __user *argp)
+{
+	return -ENOTTY;
+}
+#endif
+
+/*
+ * Return dasd information. Used for BIODASDINFO and BIODASDINFO2.
+ */
+static int __dasd_ioctl_information(struct dasd_block *block,
+		struct dasd_information2_t *dasd_info)
+{
+	struct subchannel_id sch_id;
+	struct ccw_dev_id dev_id;
+	struct dasd_device *base;
+	struct ccw_device *cdev;
+	struct list_head *l;
+	unsigned long flags;
+	int rc;
+
+	base = block->base;
+	if (!base->discipline || !base->discipline->fill_info)
+		return -EINVAL;
+
+	rc = base->discipline->fill_info(base, dasd_info);
+	if (rc)
+		return rc;
+
+	cdev = base->cdev;
+	ccw_device_get_id(cdev, &dev_id);
+	ccw_device_get_schid(cdev, &sch_id);
+
+	dasd_info->devno = dev_id.devno;
+	dasd_info->schid = sch_id.sch_no;
+	dasd_info->cu_type = cdev->id.cu_type;
+	dasd_info->cu_model = cdev->id.cu_model;
+	dasd_info->dev_type = cdev->id.dev_type;
+	dasd_info->dev_model = cdev->id.dev_model;
+	dasd_info->status = base->state;
+	/*
+	 * The open_count is increased for every opener, that includes
+	 * the blkdev_get in dasd_scan_partitions.
+	 * This must be hidden from user-space.
+	 */
+	dasd_info->open_count = atomic_read(&block->open_count);
+	if (!block->bdev_file)
+		dasd_info->open_count++;
+
+	/*
+	 * check if device is really formatted
+	 * LDL / CDL was returned by 'fill_info'
+	 */
+	if ((base->state < DASD_STATE_READY) ||
+	    (dasd_check_blocksize(block->bp_block)))
+		dasd_info->format = DASD_FORMAT_NONE;
+
+	dasd_info->features |=
+		((base->features & DASD_FEATURE_READONLY) != 0);
+
+	memcpy(dasd_info->type, base->discipline->name, 4);
+
+	spin_lock_irqsave(get_ccwdev_lock(base->cdev), flags);
+	list_for_each(l, &base->ccw_queue)
+		dasd_info->chanq_len++;
+	spin_unlock_irqrestore(get_ccwdev_lock(base->cdev), flags);
+	return 0;
+}
+
+static int dasd_ioctl_information(struct dasd_block *block, void __user *argp,
+		size_t copy_size)
+{
+	struct dasd_information2_t *dasd_info;
+	int error;
+
+	dasd_info = kzalloc_obj(*dasd_info);
+	if (!dasd_info)
+		return -ENOMEM;
+
+	error = __dasd_ioctl_information(block, dasd_info);
+	if (!error && copy_to_user(argp, dasd_info, copy_size))
+		error = -EFAULT;
+	kfree(dasd_info);
+	return error;
+}
+
+/*
+ * Set read only
+ */
+int dasd_set_read_only(struct block_device *bdev, bool ro)
+{
+	struct dasd_device *base;
+	int rc;
+
+	/* do not manipulate hardware state for partitions */
+	if (bdev_is_partition(bdev))
+		return 0;
+
+	base = dasd_device_from_gendisk(bdev->bd_disk);
+	if (!base)
+		return -ENODEV;
+	if (!ro && test_bit(DASD_FLAG_DEVICE_RO, &base->flags))
+		rc = -EROFS;
+	else
+		rc = dasd_set_feature(base->cdev, DASD_FEATURE_READONLY, ro);
+	dasd_put_device(base);
+	return rc;
+}
+
+static int dasd_ioctl_readall_cmb(struct dasd_block *block, unsigned int cmd,
+				  struct cmbdata __user *argp)
+{
+	size_t size = _IOC_SIZE(cmd);
+	struct cmbdata data;
+	int ret;
+
+	ret = cmf_readall(block->base->cdev, &data);
+	if (!ret && copy_to_user(argp, &data, min(size, sizeof(*argp))))
+		return -EFAULT;
+	return ret;
+}
+
+int dasd_ioctl(struct block_device *bdev, blk_mode_t mode,
+	       unsigned int cmd, unsigned long arg)
+{
+	struct dasd_block *block;
+	struct dasd_device *base;
+	void __user *argp;
+	int rc;
+
+	argp = (void __user *)arg;
+
+	if ((_IOC_DIR(cmd) != _IOC_NONE) && !arg)
+		return -EINVAL;
+
+	base = dasd_device_from_gendisk(bdev->bd_disk);
+	if (!base)
+		return -ENODEV;
+	block = base->block;
+	rc = 0;
+	switch (cmd) {
+	case BIODASDDISABLE:
+		rc = dasd_ioctl_disable(bdev);
+		break;
+	case BIODASDENABLE:
+		rc = dasd_ioctl_enable(bdev);
+		break;
+	case BIODASDQUIESCE:
+		rc = dasd_ioctl_quiesce(block);
+		break;
+	case BIODASDRESUME:
+		rc = dasd_ioctl_resume(block);
+		break;
+	case BIODASDABORTIO:
+		rc = dasd_ioctl_abortio(block);
+		break;
+	case BIODASDALLOWIO:
+		rc = dasd_ioctl_allowio(block);
+		break;
+	case BIODASDFMT:
+		rc = dasd_ioctl_format(bdev, argp);
+		break;
+	case BIODASDCHECKFMT:
+		rc = dasd_ioctl_check_format(bdev, argp);
+		break;
+	case BIODASDINFO:
+		rc = dasd_ioctl_information(block, argp,
+				sizeof(struct dasd_information_t));
+		break;
+	case BIODASDINFO2:
+		rc = dasd_ioctl_information(block, argp,
+				sizeof(struct dasd_information2_t));
+		break;
+	case BIODASDPRRD:
+		rc = dasd_ioctl_read_profile(block, argp);
+		break;
+	case BIODASDPRRST:
+		rc = dasd_ioctl_reset_profile(block);
+		break;
+	case DASDAPIVER:
+		rc = dasd_ioctl_api_version(argp);
+		break;
+	case BIODASDCMFENABLE:
+		rc = enable_cmf(base->cdev);
+		break;
+	case BIODASDCMFDISABLE:
+		rc = disable_cmf(base->cdev);
+		break;
+	case BIODASDREADALLCMB:
+		rc = dasd_ioctl_readall_cmb(block, cmd, argp);
+		break;
+	case BIODASDRAS:
+		rc = dasd_ioctl_release_space(bdev, argp);
+		break;
+	case BIODASDCOPYPAIRSWAP:
+		rc = dasd_ioctl_copy_pair_swap(bdev, argp);
+		break;
+	default:
+		/* if the discipline has an ioctl method try it. */
+		rc = -ENOTTY;
+		if (base->discipline->ioctl)
+			rc = base->discipline->ioctl(block, cmd, argp);
+	}
+	dasd_put_device(base);
+	return rc;
+}
+
+
+/**
+ * dasd_biodasdinfo() - fill out the dasd information structure
+ * @disk: [in] pointer to gendisk structure that references a DASD
+ * @info: [out] pointer to the dasd_information2_t structure
+ *
+ * Provide access to DASD specific information.
+ * The gendisk structure is checked if it belongs to the DASD driver by
+ * comparing the gendisk->fops pointer.
+ * If it does not belong to the DASD driver -EINVAL is returned.
+ * Otherwise the provided dasd_information2_t structure is filled out.
+ *
+ * Returns:
+ *   %0 on success and a negative error value on failure.
+ */
+int dasd_biodasdinfo(struct gendisk *disk, struct dasd_information2_t *info)
+{
+	struct dasd_device *base;
+	int error;
+
+	if (disk->fops != &dasd_device_operations)
+		return -EINVAL;
+
+	base = dasd_device_from_gendisk(disk);
+	if (!base)
+		return -ENODEV;
+	error = __dasd_ioctl_information(base->block, info);
+	dasd_put_device(base);
+	return error;
+}
+/* export that symbol_get in partition detection is possible */
+EXPORT_SYMBOL_GPL(dasd_biodasdinfo);
diff --git a/drivers/s390/block/dasd_proc.c b/drivers/s390/block/dasd_proc.c
new file mode 100644
index 000000000..48e12e81d
--- /dev/null
+++ b/drivers/s390/block/dasd_proc.c
@@ -0,0 +1,368 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Carsten Otte <Cotte@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2002
+ *
+ * /proc interface for the dasd driver.
+ *
+ */
+
+#include <linux/ctype.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/seq_file.h>
+#include <linux/vmalloc.h>
+#include <linux/proc_fs.h>
+
+#include <asm/debug.h>
+#include <linux/uaccess.h>
+
+#include "dasd_int.h"
+
+static struct proc_dir_entry *dasd_proc_root_entry = NULL;
+static struct proc_dir_entry *dasd_devices_entry = NULL;
+static struct proc_dir_entry *dasd_statistics_entry = NULL;
+
+static int
+dasd_devices_show(struct seq_file *m, void *v)
+{
+	struct dasd_device *device;
+	struct dasd_block *block;
+	char *substr;
+
+	device = dasd_device_from_devindex((unsigned long) v - 1);
+	if (IS_ERR(device))
+		return 0;
+	if (device->block)
+		block = device->block;
+	else {
+		dasd_put_device(device);
+		return 0;
+	}
+	/* Print device number. */
+	seq_printf(m, "%s", dev_name(&device->cdev->dev));
+	/* Print discipline string. */
+	if (device->discipline != NULL)
+		seq_printf(m, "(%s)", device->discipline->name);
+	else
+		seq_printf(m, "(none)");
+	/* Print kdev. */
+	if (block->gdp)
+		seq_printf(m, " at (%3d:%6d)",
+			   MAJOR(disk_devt(block->gdp)),
+			   MINOR(disk_devt(block->gdp)));
+	else
+		seq_printf(m, "  at (???:??????)");
+	/* Print device name. */
+	if (block->gdp)
+		seq_printf(m, " is %-8s", block->gdp->disk_name);
+	else
+		seq_printf(m, " is ????????");
+	/* Print devices features. */
+	substr = (device->features & DASD_FEATURE_READONLY) ? "(ro)" : " ";
+	seq_printf(m, "%4s: ", substr);
+	/* Print device status information. */
+	switch (device->state) {
+	case DASD_STATE_NEW:
+		seq_printf(m, "new");
+		break;
+	case DASD_STATE_KNOWN:
+		seq_printf(m, "detected");
+		break;
+	case DASD_STATE_BASIC:
+		seq_printf(m, "basic");
+		break;
+	case DASD_STATE_UNFMT:
+		seq_printf(m, "unformatted");
+		break;
+	case DASD_STATE_READY:
+	case DASD_STATE_ONLINE:
+		seq_printf(m, "active ");
+		if (dasd_check_blocksize(block->bp_block))
+			seq_printf(m, "n/f	 ");
+		else
+			seq_printf(m,
+				   "at blocksize: %u, %lu blocks, %lu MB",
+				   block->bp_block, block->blocks,
+				   ((block->bp_block >> 9) *
+				    block->blocks) >> 11);
+		break;
+	default:
+		seq_printf(m, "no stat");
+		break;
+	}
+	dasd_put_device(device);
+	if (dasd_probeonly)
+		seq_printf(m, "(probeonly)");
+	seq_printf(m, "\n");
+	return 0;
+}
+
+static void *dasd_devices_start(struct seq_file *m, loff_t *pos)
+{
+	if (*pos >= dasd_max_devindex)
+		return NULL;
+	return (void *)((unsigned long) *pos + 1);
+}
+
+static void *dasd_devices_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	++*pos;
+	return dasd_devices_start(m, pos);
+}
+
+static void dasd_devices_stop(struct seq_file *m, void *v)
+{
+}
+
+static const struct seq_operations dasd_devices_seq_ops = {
+	.start		= dasd_devices_start,
+	.next		= dasd_devices_next,
+	.stop		= dasd_devices_stop,
+	.show		= dasd_devices_show,
+};
+
+#ifdef CONFIG_DASD_PROFILE
+static int dasd_stats_all_block_on(void)
+{
+	int i, rc;
+	struct dasd_device *device;
+
+	rc = 0;
+	for (i = 0; i < dasd_max_devindex; ++i) {
+		device = dasd_device_from_devindex(i);
+		if (IS_ERR(device))
+			continue;
+		if (device->block)
+			rc = dasd_profile_on(&device->block->profile);
+		dasd_put_device(device);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
+
+static void dasd_stats_all_block_off(void)
+{
+	int i;
+	struct dasd_device *device;
+
+	for (i = 0; i < dasd_max_devindex; ++i) {
+		device = dasd_device_from_devindex(i);
+		if (IS_ERR(device))
+			continue;
+		if (device->block)
+			dasd_profile_off(&device->block->profile);
+		dasd_put_device(device);
+	}
+}
+
+static void dasd_stats_all_block_reset(void)
+{
+	int i;
+	struct dasd_device *device;
+
+	for (i = 0; i < dasd_max_devindex; ++i) {
+		device = dasd_device_from_devindex(i);
+		if (IS_ERR(device))
+			continue;
+		if (device->block)
+			dasd_profile_reset(&device->block->profile);
+		dasd_put_device(device);
+	}
+}
+
+static void dasd_statistics_array(struct seq_file *m, unsigned int *array, int factor)
+{
+	int i;
+
+	for (i = 0; i < 32; i++) {
+		seq_printf(m, "%7d ", array[i] / factor);
+		if (i == 15)
+			seq_putc(m, '\n');
+	}
+	seq_putc(m, '\n');
+}
+#endif /* CONFIG_DASD_PROFILE */
+
+static int dasd_stats_proc_show(struct seq_file *m, void *v)
+{
+#ifdef CONFIG_DASD_PROFILE
+	struct dasd_profile_info *prof;
+	int factor;
+
+	spin_lock_bh(&dasd_global_profile.lock);
+	prof = dasd_global_profile.data;
+	if (!prof) {
+		spin_unlock_bh(&dasd_global_profile.lock);
+		seq_printf(m, "Statistics are off - they might be "
+				    "switched on using 'echo set on > "
+				    "/proc/dasd/statistics'\n");
+		return 0;
+	}
+
+	/* prevent counter 'overflow' on output */
+	for (factor = 1; (prof->dasd_io_reqs / factor) > 9999999;
+	     factor *= 10);
+
+	seq_printf(m, "%d dasd I/O requests\n", prof->dasd_io_reqs);
+	seq_printf(m, "with %u sectors(512B each)\n",
+		       prof->dasd_io_sects);
+	seq_printf(m, "Scale Factor is  %d\n", factor);
+	seq_printf(m,
+		       "   __<4	   ___8	   __16	   __32	   __64	   _128	"
+		       "   _256	   _512	   __1k	   __2k	   __4k	   __8k	"
+		       "   _16k	   _32k	   _64k	   128k\n");
+	seq_printf(m,
+		       "   _256	   _512	   __1M	   __2M	   __4M	   __8M	"
+		       "   _16M	   _32M	   _64M	   128M	   256M	   512M	"
+		       "   __1G	   __2G	   __4G " "   _>4G\n");
+
+	seq_printf(m, "Histogram of sizes (512B secs)\n");
+	dasd_statistics_array(m, prof->dasd_io_secs, factor);
+	seq_printf(m, "Histogram of I/O times (microseconds)\n");
+	dasd_statistics_array(m, prof->dasd_io_times, factor);
+	seq_printf(m, "Histogram of I/O times per sector\n");
+	dasd_statistics_array(m, prof->dasd_io_timps, factor);
+	seq_printf(m, "Histogram of I/O time till ssch\n");
+	dasd_statistics_array(m, prof->dasd_io_time1, factor);
+	seq_printf(m, "Histogram of I/O time between ssch and irq\n");
+	dasd_statistics_array(m, prof->dasd_io_time2, factor);
+	seq_printf(m, "Histogram of I/O time between ssch "
+			    "and irq per sector\n");
+	dasd_statistics_array(m, prof->dasd_io_time2ps, factor);
+	seq_printf(m, "Histogram of I/O time between irq and end\n");
+	dasd_statistics_array(m, prof->dasd_io_time3, factor);
+	seq_printf(m, "# of req in chanq at enqueuing (1..32) \n");
+	dasd_statistics_array(m, prof->dasd_io_nr_req, factor);
+	spin_unlock_bh(&dasd_global_profile.lock);
+#else
+	seq_printf(m, "Statistics are not activated in this kernel\n");
+#endif
+	return 0;
+}
+
+static int dasd_stats_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, dasd_stats_proc_show, NULL);
+}
+
+static ssize_t dasd_stats_proc_write(struct file *file,
+		const char __user *user_buf, size_t user_len, loff_t *pos)
+{
+#ifdef CONFIG_DASD_PROFILE
+	char *buffer, *str;
+	int rc;
+
+	if (user_len > 65536)
+		user_len = 65536;
+	buffer = dasd_get_user_string(user_buf, user_len);
+	if (IS_ERR(buffer))
+		return PTR_ERR(buffer);
+
+	/* check for valid verbs */
+	str = skip_spaces(buffer);
+	if (strncmp(str, "set", 3) == 0 && isspace(str[3])) {
+		/* 'set xxx' was given */
+		str = skip_spaces(str + 4);
+		if (strcmp(str, "on") == 0) {
+			/* switch on statistics profiling */
+			rc = dasd_stats_all_block_on();
+			if (rc) {
+				dasd_stats_all_block_off();
+				goto out_error;
+			}
+			rc = dasd_profile_on(&dasd_global_profile);
+			if (rc) {
+				dasd_stats_all_block_off();
+				goto out_error;
+			}
+			dasd_profile_reset(&dasd_global_profile);
+			dasd_global_profile_level = DASD_PROFILE_ON;
+			pr_info("The statistics feature has been switched "
+				"on\n");
+		} else if (strcmp(str, "off") == 0) {
+			/* switch off statistics profiling */
+			dasd_global_profile_level = DASD_PROFILE_OFF;
+			dasd_profile_off(&dasd_global_profile);
+			dasd_stats_all_block_off();
+			pr_info("The statistics feature has been switched "
+				"off\n");
+		} else
+			goto out_parse_error;
+	} else if (strncmp(str, "reset", 5) == 0) {
+		/* reset the statistics */
+		dasd_profile_reset(&dasd_global_profile);
+		dasd_stats_all_block_reset();
+		pr_info("The statistics have been reset\n");
+	} else
+		goto out_parse_error;
+	vfree(buffer);
+	return user_len;
+out_parse_error:
+	rc = -EINVAL;
+	pr_warn("%s is not a supported value for /proc/dasd/statistics\n", str);
+out_error:
+	vfree(buffer);
+	return rc;
+#else
+	pr_warn("/proc/dasd/statistics: is not activated in this kernel\n");
+	return user_len;
+#endif				/* CONFIG_DASD_PROFILE */
+}
+
+static const struct proc_ops dasd_stats_proc_ops = {
+	.proc_open	= dasd_stats_proc_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
+	.proc_write	= dasd_stats_proc_write,
+};
+
+/*
+ * Create dasd proc-fs entries.
+ * In case creation failed, cleanup and return -ENOENT.
+ */
+int
+dasd_proc_init(void)
+{
+	dasd_proc_root_entry = proc_mkdir("dasd", NULL);
+	if (!dasd_proc_root_entry)
+		goto out_nodasd;
+	dasd_devices_entry = proc_create_seq("devices", 0444,
+					 dasd_proc_root_entry,
+					 &dasd_devices_seq_ops);
+	if (!dasd_devices_entry)
+		goto out_nodevices;
+	dasd_statistics_entry = proc_create("statistics",
+					    S_IFREG | S_IRUGO | S_IWUSR,
+					    dasd_proc_root_entry,
+					    &dasd_stats_proc_ops);
+	if (!dasd_statistics_entry)
+		goto out_nostatistics;
+	return 0;
+
+ out_nostatistics:
+	remove_proc_entry("devices", dasd_proc_root_entry);
+ out_nodevices:
+	remove_proc_entry("dasd", NULL);
+	dasd_proc_root_entry = NULL;
+ out_nodasd:
+	return -ENOENT;
+}
+
+void
+dasd_proc_exit(void)
+{
+	if (!dasd_proc_root_entry)
+		return;
+
+	remove_proc_entry("devices", dasd_proc_root_entry);
+	remove_proc_entry("statistics", dasd_proc_root_entry);
+	remove_proc_entry("dasd", NULL);
+	dasd_proc_root_entry = NULL;
+}
diff --git a/drivers/s390/block/dcssblk.c b/drivers/s390/block/dcssblk.c
new file mode 100644
index 000000000..e5a02c299
--- /dev/null
+++ b/drivers/s390/block/dcssblk.c
@@ -0,0 +1,1054 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * dcssblk.c -- the S/390 block driver for dcss memory
+ *
+ * Authors: Carsten Otte, Stefan Weinhuber, Gerald Schaefer
+ */
+
+#define pr_fmt(fmt) "dcssblk: " fmt
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/ctype.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/blkdev.h>
+#include <linux/completion.h>
+#include <linux/interrupt.h>
+#include <linux/uio.h>
+#include <linux/dax.h>
+#include <linux/io.h>
+#include <asm/extmem.h>
+
+#define DCSSBLK_NAME "dcssblk"
+#define DCSSBLK_MINORS_PER_DISK 1
+#define DCSSBLK_PARM_LEN 400
+#define DCSS_BUS_ID_SIZE 20
+
+static int dcssblk_open(struct gendisk *disk, blk_mode_t mode);
+static void dcssblk_release(struct gendisk *disk);
+static void dcssblk_submit_bio(struct bio *bio);
+static long dcssblk_dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff,
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		unsigned long *pfn);
+
+static char dcssblk_segments[DCSSBLK_PARM_LEN] = "\0";
+
+static int dcssblk_major;
+static const struct block_device_operations dcssblk_devops = {
+	.owner   	= THIS_MODULE,
+	.submit_bio	= dcssblk_submit_bio,
+	.open    	= dcssblk_open,
+	.release 	= dcssblk_release,
+};
+
+static int dcssblk_dax_zero_page_range(struct dax_device *dax_dev,
+				       pgoff_t pgoff, size_t nr_pages)
+{
+	long rc;
+	void *kaddr;
+
+	rc = dax_direct_access(dax_dev, pgoff, nr_pages, DAX_ACCESS,
+			&kaddr, NULL);
+	if (rc < 0)
+		return dax_mem2blk_err(rc);
+
+	memset(kaddr, 0, nr_pages << PAGE_SHIFT);
+	dax_flush(dax_dev, kaddr, nr_pages << PAGE_SHIFT);
+	return 0;
+}
+
+static const struct dax_operations dcssblk_dax_ops = {
+	.direct_access = dcssblk_dax_direct_access,
+	.zero_page_range = dcssblk_dax_zero_page_range,
+};
+
+struct dcssblk_dev_info {
+	struct list_head lh;
+	struct device dev;
+	char segment_name[DCSS_BUS_ID_SIZE];
+	atomic_t use_count;
+	struct gendisk *gd;
+	unsigned long start;
+	unsigned long end;
+	int segment_type;
+	unsigned char save_pending;
+	unsigned char is_shared;
+	int num_of_segments;
+	struct list_head seg_list;
+	struct dax_device *dax_dev;
+	struct dev_pagemap pgmap;
+	void *pgmap_addr;
+};
+
+struct segment_info {
+	struct list_head lh;
+	char segment_name[DCSS_BUS_ID_SIZE];
+	unsigned long start;
+	unsigned long end;
+	int segment_type;
+};
+
+static ssize_t dcssblk_add_store(struct device * dev, struct device_attribute *attr, const char * buf,
+				  size_t count);
+static ssize_t dcssblk_remove_store(struct device * dev, struct device_attribute *attr, const char * buf,
+				  size_t count);
+
+static DEVICE_ATTR(add, S_IWUSR, NULL, dcssblk_add_store);
+static DEVICE_ATTR(remove, S_IWUSR, NULL, dcssblk_remove_store);
+
+static struct device *dcssblk_root_dev;
+
+static LIST_HEAD(dcssblk_devices);
+static struct rw_semaphore dcssblk_devices_sem;
+
+/*
+ * release function for segment device.
+ */
+static void
+dcssblk_release_segment(struct device *dev)
+{
+	struct dcssblk_dev_info *dev_info;
+	struct segment_info *entry, *temp;
+
+	dev_info = container_of(dev, struct dcssblk_dev_info, dev);
+	list_for_each_entry_safe(entry, temp, &dev_info->seg_list, lh) {
+		list_del(&entry->lh);
+		kfree(entry);
+	}
+	kfree(dev_info);
+	module_put(THIS_MODULE);
+}
+
+/*
+ * get a minor number. needs to be called with
+ * down_write(&dcssblk_devices_sem) and the
+ * device needs to be enqueued before the semaphore is
+ * freed.
+ */
+static int
+dcssblk_assign_free_minor(struct dcssblk_dev_info *dev_info)
+{
+	int minor, found;
+	struct dcssblk_dev_info *entry;
+
+	if (dev_info == NULL)
+		return -EINVAL;
+	for (minor = 0; minor < (1<<MINORBITS); minor++) {
+		found = 0;
+		// test if minor available
+		list_for_each_entry(entry, &dcssblk_devices, lh)
+			if (minor == entry->gd->first_minor)
+				found++;
+		if (!found) break; // got unused minor
+	}
+	if (found)
+		return -EBUSY;
+	dev_info->gd->first_minor = minor;
+	return 0;
+}
+
+/*
+ * get the struct dcssblk_dev_info from dcssblk_devices
+ * for the given name.
+ * down_read(&dcssblk_devices_sem) must be held.
+ */
+static struct dcssblk_dev_info *
+dcssblk_get_device_by_name(char *name)
+{
+	struct dcssblk_dev_info *entry;
+
+	list_for_each_entry(entry, &dcssblk_devices, lh) {
+		if (!strcmp(name, entry->segment_name)) {
+			return entry;
+		}
+	}
+	return NULL;
+}
+
+/*
+ * get the struct segment_info from seg_list
+ * for the given name.
+ * down_read(&dcssblk_devices_sem) must be held.
+ */
+static struct segment_info *
+dcssblk_get_segment_by_name(char *name)
+{
+	struct dcssblk_dev_info *dev_info;
+	struct segment_info *entry;
+
+	list_for_each_entry(dev_info, &dcssblk_devices, lh) {
+		list_for_each_entry(entry, &dev_info->seg_list, lh) {
+			if (!strcmp(name, entry->segment_name))
+				return entry;
+		}
+	}
+	return NULL;
+}
+
+/*
+ * get the highest address of the multi-segment block.
+ */
+static unsigned long
+dcssblk_find_highest_addr(struct dcssblk_dev_info *dev_info)
+{
+	unsigned long highest_addr;
+	struct segment_info *entry;
+
+	highest_addr = 0;
+	list_for_each_entry(entry, &dev_info->seg_list, lh) {
+		if (highest_addr < entry->end)
+			highest_addr = entry->end;
+	}
+	return highest_addr;
+}
+
+/*
+ * get the lowest address of the multi-segment block.
+ */
+static unsigned long
+dcssblk_find_lowest_addr(struct dcssblk_dev_info *dev_info)
+{
+	int set_first;
+	unsigned long lowest_addr;
+	struct segment_info *entry;
+
+	set_first = 0;
+	lowest_addr = 0;
+	list_for_each_entry(entry, &dev_info->seg_list, lh) {
+		if (set_first == 0) {
+			lowest_addr = entry->start;
+			set_first = 1;
+		} else {
+			if (lowest_addr > entry->start)
+				lowest_addr = entry->start;
+		}
+	}
+	return lowest_addr;
+}
+
+/*
+ * Check continuity of segments.
+ */
+static int
+dcssblk_is_continuous(struct dcssblk_dev_info *dev_info)
+{
+	int i, j, rc;
+	struct segment_info *sort_list, *entry, temp;
+
+	if (dev_info->num_of_segments <= 1)
+		return 0;
+
+	sort_list = kzalloc_objs(struct segment_info, dev_info->num_of_segments);
+	if (sort_list == NULL)
+		return -ENOMEM;
+	i = 0;
+	list_for_each_entry(entry, &dev_info->seg_list, lh) {
+		memcpy(&sort_list[i], entry, sizeof(struct segment_info));
+		i++;
+	}
+
+	/* sort segments */
+	for (i = 0; i < dev_info->num_of_segments; i++)
+		for (j = 0; j < dev_info->num_of_segments; j++)
+			if (sort_list[j].start > sort_list[i].start) {
+				memcpy(&temp, &sort_list[i],
+					sizeof(struct segment_info));
+				memcpy(&sort_list[i], &sort_list[j],
+					sizeof(struct segment_info));
+				memcpy(&sort_list[j], &temp,
+					sizeof(struct segment_info));
+			}
+
+	/* check continuity */
+	for (i = 0; i < dev_info->num_of_segments - 1; i++) {
+		if ((sort_list[i].end + 1) != sort_list[i+1].start) {
+			pr_err("Adjacent DCSSs %s and %s are not "
+			       "contiguous\n", sort_list[i].segment_name,
+			       sort_list[i+1].segment_name);
+			rc = -EINVAL;
+			goto out;
+		}
+		/* EN and EW are allowed in a block device */
+		if (sort_list[i].segment_type != sort_list[i+1].segment_type) {
+			if (!(sort_list[i].segment_type & SEGMENT_EXCLUSIVE) ||
+				(sort_list[i].segment_type == SEG_TYPE_ER) ||
+				!(sort_list[i+1].segment_type &
+				SEGMENT_EXCLUSIVE) ||
+				(sort_list[i+1].segment_type == SEG_TYPE_ER)) {
+				pr_err("DCSS %s and DCSS %s have "
+				       "incompatible types\n",
+				       sort_list[i].segment_name,
+				       sort_list[i+1].segment_name);
+				rc = -EINVAL;
+				goto out;
+			}
+		}
+	}
+	rc = 0;
+out:
+	kfree(sort_list);
+	return rc;
+}
+
+/*
+ * Load a segment
+ */
+static int
+dcssblk_load_segment(char *name, struct segment_info **seg_info)
+{
+	int rc;
+
+	/* already loaded? */
+	down_read(&dcssblk_devices_sem);
+	*seg_info = dcssblk_get_segment_by_name(name);
+	up_read(&dcssblk_devices_sem);
+	if (*seg_info != NULL)
+		return -EEXIST;
+
+	/* get a struct segment_info */
+	*seg_info = kzalloc_obj(struct segment_info);
+	if (*seg_info == NULL)
+		return -ENOMEM;
+
+	strscpy((*seg_info)->segment_name, name);
+
+	/* load the segment */
+	rc = segment_load(name, SEGMENT_SHARED,
+			&(*seg_info)->start, &(*seg_info)->end);
+	if (rc < 0) {
+		segment_warning(rc, (*seg_info)->segment_name);
+		kfree(*seg_info);
+	} else {
+		INIT_LIST_HEAD(&(*seg_info)->lh);
+		(*seg_info)->segment_type = rc;
+	}
+	return rc;
+}
+
+/*
+ * device attribute for switching shared/nonshared (exclusive)
+ * operation (show + store)
+ */
+static ssize_t
+dcssblk_shared_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct dcssblk_dev_info *dev_info;
+
+	dev_info = container_of(dev, struct dcssblk_dev_info, dev);
+	return sysfs_emit(buf, dev_info->is_shared ? "1\n" : "0\n");
+}
+
+static ssize_t
+dcssblk_shared_store(struct device *dev, struct device_attribute *attr, const char *inbuf, size_t count)
+{
+	struct dcssblk_dev_info *dev_info;
+	struct segment_info *entry, *temp;
+	int rc;
+
+	if ((count > 1) && (inbuf[1] != '\n') && (inbuf[1] != '\0'))
+		return -EINVAL;
+	down_write(&dcssblk_devices_sem);
+	dev_info = container_of(dev, struct dcssblk_dev_info, dev);
+	if (atomic_read(&dev_info->use_count)) {
+		rc = -EBUSY;
+		goto out;
+	}
+	if (inbuf[0] == '1') {
+		/* reload segments in shared mode */
+		list_for_each_entry(entry, &dev_info->seg_list, lh) {
+			rc = segment_modify_shared(entry->segment_name,
+						SEGMENT_SHARED);
+			if (rc < 0) {
+				BUG_ON(rc == -EINVAL);
+				if (rc != -EAGAIN)
+					goto removeseg;
+			}
+		}
+		dev_info->is_shared = 1;
+		switch (dev_info->segment_type) {
+		case SEG_TYPE_SR:
+		case SEG_TYPE_ER:
+		case SEG_TYPE_SC:
+			set_disk_ro(dev_info->gd, 1);
+		}
+	} else if (inbuf[0] == '0') {
+		/* reload segments in exclusive mode */
+		if (dev_info->segment_type == SEG_TYPE_SC) {
+			pr_err("DCSS %s is of type SC and cannot be "
+			       "loaded as exclusive-writable\n",
+			       dev_info->segment_name);
+			rc = -EINVAL;
+			goto out;
+		}
+		list_for_each_entry(entry, &dev_info->seg_list, lh) {
+			rc = segment_modify_shared(entry->segment_name,
+						   SEGMENT_EXCLUSIVE);
+			if (rc < 0) {
+				BUG_ON(rc == -EINVAL);
+				if (rc != -EAGAIN)
+					goto removeseg;
+			}
+		}
+		dev_info->is_shared = 0;
+		set_disk_ro(dev_info->gd, 0);
+	} else {
+		rc = -EINVAL;
+		goto out;
+	}
+	rc = count;
+	goto out;
+
+removeseg:
+	pr_err("DCSS device %s is removed after a failed access mode "
+	       "change\n", dev_info->segment_name);
+	temp = entry;
+	list_for_each_entry(entry, &dev_info->seg_list, lh) {
+		if (entry != temp)
+			segment_unload(entry->segment_name);
+	}
+	list_del(&dev_info->lh);
+	up_write(&dcssblk_devices_sem);
+
+	dax_remove_host(dev_info->gd);
+	kill_dax(dev_info->dax_dev);
+	put_dax(dev_info->dax_dev);
+	if (dev_info->pgmap_addr)
+		devm_memunmap_pages(&dev_info->dev, &dev_info->pgmap);
+	del_gendisk(dev_info->gd);
+	put_disk(dev_info->gd);
+
+	if (device_remove_file_self(dev, attr)) {
+		device_unregister(dev);
+		put_device(dev);
+	}
+	return rc;
+out:
+	up_write(&dcssblk_devices_sem);
+	return rc;
+}
+static DEVICE_ATTR(shared, S_IWUSR | S_IRUSR, dcssblk_shared_show,
+		   dcssblk_shared_store);
+
+/*
+ * device attribute for save operation on current copy
+ * of the segment. If the segment is busy, saving will
+ * become pending until it gets released, which can be
+ * undone by storing a non-true value to this entry.
+ * (show + store)
+ */
+static ssize_t
+dcssblk_save_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct dcssblk_dev_info *dev_info;
+
+	dev_info = container_of(dev, struct dcssblk_dev_info, dev);
+	return sysfs_emit(buf, dev_info->save_pending ? "1\n" : "0\n");
+}
+
+static ssize_t
+dcssblk_save_store(struct device *dev, struct device_attribute *attr, const char *inbuf, size_t count)
+{
+	struct dcssblk_dev_info *dev_info;
+	struct segment_info *entry;
+
+	if ((count > 1) && (inbuf[1] != '\n') && (inbuf[1] != '\0'))
+		return -EINVAL;
+	dev_info = container_of(dev, struct dcssblk_dev_info, dev);
+
+	down_write(&dcssblk_devices_sem);
+	if (inbuf[0] == '1') {
+		if (atomic_read(&dev_info->use_count) == 0) {
+			// device is idle => we save immediately
+			pr_info("All DCSSs that map to device %s are "
+				"saved\n", dev_info->segment_name);
+			list_for_each_entry(entry, &dev_info->seg_list, lh) {
+				if (entry->segment_type == SEG_TYPE_EN ||
+				    entry->segment_type == SEG_TYPE_SN)
+					pr_warn("DCSS %s is of type SN or EN"
+						" and cannot be saved\n",
+						entry->segment_name);
+				else
+					segment_save(entry->segment_name);
+			}
+		}  else {
+			// device is busy => we save it when it becomes
+			// idle in dcssblk_release
+			pr_info("Device %s is in use, its DCSSs will be "
+				"saved when it becomes idle\n",
+				dev_info->segment_name);
+			dev_info->save_pending = 1;
+		}
+	} else if (inbuf[0] == '0') {
+		if (dev_info->save_pending) {
+			// device is busy & the user wants to undo his save
+			// request
+			dev_info->save_pending = 0;
+			pr_info("A pending save request for device %s "
+				"has been canceled\n",
+				dev_info->segment_name);
+		}
+	} else {
+		up_write(&dcssblk_devices_sem);
+		return -EINVAL;
+	}
+	up_write(&dcssblk_devices_sem);
+	return count;
+}
+static DEVICE_ATTR(save, S_IWUSR | S_IRUSR, dcssblk_save_show,
+		   dcssblk_save_store);
+
+/*
+ * device attribute for showing all segments in a device
+ */
+static ssize_t
+dcssblk_seglist_show(struct device *dev, struct device_attribute *attr,
+		char *buf)
+{
+	struct dcssblk_dev_info *dev_info;
+	struct segment_info *entry;
+	int i;
+
+	i = 0;
+	down_read(&dcssblk_devices_sem);
+	dev_info = container_of(dev, struct dcssblk_dev_info, dev);
+	list_for_each_entry(entry, &dev_info->seg_list, lh)
+		i += sysfs_emit_at(buf, i, "%s\n", entry->segment_name);
+	up_read(&dcssblk_devices_sem);
+	return i;
+}
+static DEVICE_ATTR(seglist, S_IRUSR, dcssblk_seglist_show, NULL);
+
+static struct attribute *dcssblk_dev_attrs[] = {
+	&dev_attr_shared.attr,
+	&dev_attr_save.attr,
+	&dev_attr_seglist.attr,
+	NULL,
+};
+static struct attribute_group dcssblk_dev_attr_group = {
+	.attrs = dcssblk_dev_attrs,
+};
+static const struct attribute_group *dcssblk_dev_attr_groups[] = {
+	&dcssblk_dev_attr_group,
+	NULL,
+};
+
+static int dcssblk_setup_dax(struct dcssblk_dev_info *dev_info)
+{
+	struct dax_device *dax_dev;
+
+	dax_dev = alloc_dax(dev_info, &dcssblk_dax_ops);
+	if (IS_ERR(dax_dev))
+		return PTR_ERR(dax_dev);
+	set_dax_synchronous(dax_dev);
+	dev_info->dax_dev = dax_dev;
+	return dax_add_host(dev_info->dax_dev, dev_info->gd);
+}
+
+/*
+ * device attribute for adding devices
+ */
+static ssize_t
+dcssblk_add_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count)
+{
+	struct queue_limits lim = {
+		.logical_block_size	= 4096,
+		.features		= BLK_FEAT_DAX,
+	};
+	int rc, i, j, num_of_segments;
+	struct dcssblk_dev_info *dev_info;
+	struct segment_info *seg_info, *temp;
+	char *local_buf;
+	void *addr;
+	unsigned long seg_byte_size;
+
+	dev_info = NULL;
+	seg_info = NULL;
+	if (dev != dcssblk_root_dev) {
+		rc = -EINVAL;
+		goto out_nobuf;
+	}
+	if ((count < 1) || (buf[0] == '\0') || (buf[0] == '\n')) {
+		rc = -ENAMETOOLONG;
+		goto out_nobuf;
+	}
+
+	local_buf = kmalloc(count + 1, GFP_KERNEL);
+	if (local_buf == NULL) {
+		rc = -ENOMEM;
+		goto out_nobuf;
+	}
+
+	/*
+	 * parse input
+	 */
+	num_of_segments = 0;
+	for (i = 0; (i < count && (buf[i] != '\0') && (buf[i] != '\n')); i++) {
+		for (j = i; j < count &&
+			(buf[j] != ':') &&
+			(buf[j] != '\0') &&
+			(buf[j] != '\n'); j++) {
+			local_buf[j-i] = toupper(buf[j]);
+		}
+		local_buf[j-i] = '\0';
+		if (((j - i) == 0) || ((j - i) > 8)) {
+			rc = -ENAMETOOLONG;
+			goto seg_list_del;
+		}
+
+		rc = dcssblk_load_segment(local_buf, &seg_info);
+		if (rc < 0)
+			goto seg_list_del;
+		/*
+		 * get a struct dcssblk_dev_info
+		 */
+		if (num_of_segments == 0) {
+			dev_info = kzalloc_obj(struct dcssblk_dev_info);
+			if (dev_info == NULL) {
+				rc = -ENOMEM;
+				goto out;
+			}
+			strscpy(dev_info->segment_name, local_buf);
+			dev_info->segment_type = seg_info->segment_type;
+			INIT_LIST_HEAD(&dev_info->seg_list);
+		}
+		list_add_tail(&seg_info->lh, &dev_info->seg_list);
+		num_of_segments++;
+		i = j;
+
+		if ((buf[j] == '\0') || (buf[j] == '\n'))
+			break;
+	}
+
+	/* no trailing colon at the end of the input */
+	if ((i > 0) && (buf[i-1] == ':')) {
+		rc = -ENAMETOOLONG;
+		goto seg_list_del;
+	}
+	strscpy(local_buf, buf, i + 1);
+	dev_info->num_of_segments = num_of_segments;
+	rc = dcssblk_is_continuous(dev_info);
+	if (rc < 0)
+		goto seg_list_del;
+
+	dev_info->start = dcssblk_find_lowest_addr(dev_info);
+	dev_info->end = dcssblk_find_highest_addr(dev_info);
+
+	dev_set_name(&dev_info->dev, "%s", dev_info->segment_name);
+	dev_info->dev.release = dcssblk_release_segment;
+	dev_info->dev.groups = dcssblk_dev_attr_groups;
+	INIT_LIST_HEAD(&dev_info->lh);
+	dev_info->gd = blk_alloc_disk(&lim, NUMA_NO_NODE);
+	if (IS_ERR(dev_info->gd)) {
+		rc = PTR_ERR(dev_info->gd);
+		goto seg_list_del;
+	}
+	dev_info->gd->major = dcssblk_major;
+	dev_info->gd->minors = DCSSBLK_MINORS_PER_DISK;
+	dev_info->gd->fops = &dcssblk_devops;
+	dev_info->gd->private_data = dev_info;
+	dev_info->gd->flags |= GENHD_FL_NO_PART;
+
+	seg_byte_size = (dev_info->end - dev_info->start + 1);
+	set_capacity(dev_info->gd, seg_byte_size >> 9); // size in sectors
+	pr_info("Loaded %s with total size %lu bytes and capacity %lu "
+		"sectors\n", local_buf, seg_byte_size, seg_byte_size >> 9);
+
+	dev_info->save_pending = 0;
+	dev_info->is_shared = 1;
+	dev_info->dev.parent = dcssblk_root_dev;
+
+	/*
+	 *get minor, add to list
+	 */
+	down_write(&dcssblk_devices_sem);
+	if (dcssblk_get_segment_by_name(local_buf)) {
+		rc = -EEXIST;
+		goto release_gd;
+	}
+	rc = dcssblk_assign_free_minor(dev_info);
+	if (rc)
+		goto release_gd;
+	scnprintf(dev_info->gd->disk_name, sizeof(dev_info->gd->disk_name),
+		  "dcssblk%d", dev_info->gd->first_minor);
+	list_add_tail(&dev_info->lh, &dcssblk_devices);
+
+	if (!try_module_get(THIS_MODULE)) {
+		rc = -ENODEV;
+		goto dev_list_del;
+	}
+	/*
+	 * register the device
+	 */
+	rc = device_register(&dev_info->dev);
+	if (rc)
+		goto put_dev;
+
+	if (!IS_ALIGNED(dev_info->start, SUBSECTION_SIZE) ||
+	    !IS_ALIGNED(dev_info->end + 1, SUBSECTION_SIZE)) {
+		pr_info("DCSS %s is not aligned to %lu bytes, DAX support disabled\n",
+			local_buf, SUBSECTION_SIZE);
+	} else {
+		dev_info->pgmap.type		= MEMORY_DEVICE_FS_DAX;
+		dev_info->pgmap.range.start	= dev_info->start;
+		dev_info->pgmap.range.end	= dev_info->end;
+		dev_info->pgmap.nr_range	= 1;
+		addr = devm_memremap_pages(&dev_info->dev, &dev_info->pgmap);
+		if (IS_ERR(addr)) {
+			rc = PTR_ERR(addr);
+			goto put_dev;
+		}
+		dev_info->pgmap_addr = addr;
+		rc = dcssblk_setup_dax(dev_info);
+		if (rc)
+			goto out_dax;
+		pr_info("DAX support enabled for DCSS %s\n", local_buf);
+	}
+
+	get_device(&dev_info->dev);
+	rc = device_add_disk(&dev_info->dev, dev_info->gd, NULL);
+	if (rc)
+		goto out_dax_host;
+
+	switch (dev_info->segment_type) {
+		case SEG_TYPE_SR:
+		case SEG_TYPE_ER:
+		case SEG_TYPE_SC:
+			set_disk_ro(dev_info->gd,1);
+			break;
+		default:
+			set_disk_ro(dev_info->gd,0);
+			break;
+	}
+	up_write(&dcssblk_devices_sem);
+	rc = count;
+	goto out;
+
+out_dax_host:
+	put_device(&dev_info->dev);
+	dax_remove_host(dev_info->gd);
+out_dax:
+	kill_dax(dev_info->dax_dev);
+	put_dax(dev_info->dax_dev);
+	if (dev_info->pgmap_addr)
+		devm_memunmap_pages(&dev_info->dev, &dev_info->pgmap);
+put_dev:
+	list_del(&dev_info->lh);
+	put_disk(dev_info->gd);
+	list_for_each_entry(seg_info, &dev_info->seg_list, lh) {
+		segment_unload(seg_info->segment_name);
+	}
+	put_device(&dev_info->dev);
+	up_write(&dcssblk_devices_sem);
+	goto out;
+dev_list_del:
+	list_del(&dev_info->lh);
+release_gd:
+	put_disk(dev_info->gd);
+	up_write(&dcssblk_devices_sem);
+seg_list_del:
+	if (dev_info == NULL)
+		goto out;
+	list_for_each_entry_safe(seg_info, temp, &dev_info->seg_list, lh) {
+		list_del(&seg_info->lh);
+		segment_unload(seg_info->segment_name);
+		kfree(seg_info);
+	}
+	kfree(dev_info);
+out:
+	kfree(local_buf);
+out_nobuf:
+	return rc;
+}
+
+/*
+ * device attribute for removing devices
+ */
+static ssize_t
+dcssblk_remove_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count)
+{
+	struct dcssblk_dev_info *dev_info;
+	struct segment_info *entry;
+	int rc, i;
+	char *local_buf;
+
+	if (dev != dcssblk_root_dev) {
+		return -EINVAL;
+	}
+	local_buf = kmalloc(count + 1, GFP_KERNEL);
+	if (local_buf == NULL) {
+		return -ENOMEM;
+	}
+	/*
+	 * parse input
+	 */
+	for (i = 0; (i < count && (*(buf+i)!='\0') && (*(buf+i)!='\n')); i++) {
+		local_buf[i] = toupper(buf[i]);
+	}
+	local_buf[i] = '\0';
+	if ((i == 0) || (i > 8)) {
+		rc = -ENAMETOOLONG;
+		goto out_buf;
+	}
+
+	down_write(&dcssblk_devices_sem);
+	dev_info = dcssblk_get_device_by_name(local_buf);
+	if (dev_info == NULL) {
+		up_write(&dcssblk_devices_sem);
+		pr_warn("Device %s cannot be removed because it is not a known device\n",
+			local_buf);
+		rc = -ENODEV;
+		goto out_buf;
+	}
+	if (atomic_read(&dev_info->use_count) != 0) {
+		up_write(&dcssblk_devices_sem);
+		pr_warn("Device %s cannot be removed while it is in use\n",
+			local_buf);
+		rc = -EBUSY;
+		goto out_buf;
+	}
+
+	list_del(&dev_info->lh);
+	/* unload all related segments */
+	list_for_each_entry(entry, &dev_info->seg_list, lh)
+		segment_unload(entry->segment_name);
+	up_write(&dcssblk_devices_sem);
+
+	dax_remove_host(dev_info->gd);
+	kill_dax(dev_info->dax_dev);
+	put_dax(dev_info->dax_dev);
+	if (dev_info->pgmap_addr)
+		devm_memunmap_pages(&dev_info->dev, &dev_info->pgmap);
+	del_gendisk(dev_info->gd);
+	put_disk(dev_info->gd);
+
+	device_unregister(&dev_info->dev);
+	put_device(&dev_info->dev);
+
+	rc = count;
+out_buf:
+	kfree(local_buf);
+	return rc;
+}
+
+static int
+dcssblk_open(struct gendisk *disk, blk_mode_t mode)
+{
+	struct dcssblk_dev_info *dev_info = disk->private_data;
+	int rc;
+
+	if (NULL == dev_info) {
+		rc = -ENODEV;
+		goto out;
+	}
+	atomic_inc(&dev_info->use_count);
+	rc = 0;
+out:
+	return rc;
+}
+
+static void
+dcssblk_release(struct gendisk *disk)
+{
+	struct dcssblk_dev_info *dev_info = disk->private_data;
+	struct segment_info *entry;
+
+	if (!dev_info) {
+		WARN_ON(1);
+		return;
+	}
+	down_write(&dcssblk_devices_sem);
+	if (atomic_dec_and_test(&dev_info->use_count)
+	    && (dev_info->save_pending)) {
+		pr_info("Device %s has become idle and is being saved "
+			"now\n", dev_info->segment_name);
+		list_for_each_entry(entry, &dev_info->seg_list, lh) {
+			if (entry->segment_type == SEG_TYPE_EN ||
+			    entry->segment_type == SEG_TYPE_SN)
+				pr_warn("DCSS %s is of type SN or EN and cannot"
+					" be saved\n", entry->segment_name);
+			else
+				segment_save(entry->segment_name);
+		}
+		dev_info->save_pending = 0;
+	}
+	up_write(&dcssblk_devices_sem);
+}
+
+static void
+dcssblk_submit_bio(struct bio *bio)
+{
+	struct dcssblk_dev_info *dev_info;
+	struct bio_vec bvec;
+	struct bvec_iter iter;
+	unsigned long index;
+	void *page_addr;
+	unsigned long source_addr;
+	unsigned long bytes_done;
+
+	bytes_done = 0;
+	dev_info = bio->bi_bdev->bd_disk->private_data;
+	if (dev_info == NULL)
+		goto fail;
+	if (!IS_ALIGNED(bio->bi_iter.bi_sector, 8) ||
+	    !IS_ALIGNED(bio->bi_iter.bi_size, PAGE_SIZE))
+		/* Request is not page-aligned. */
+		goto fail;
+	/* verify data transfer direction */
+	if (dev_info->is_shared) {
+		switch (dev_info->segment_type) {
+		case SEG_TYPE_SR:
+		case SEG_TYPE_ER:
+		case SEG_TYPE_SC:
+			/* cannot write to these segments */
+			if (bio_data_dir(bio) == WRITE) {
+				pr_warn("Writing to %s failed because it is a read-only device\n",
+					dev_name(&dev_info->dev));
+				goto fail;
+			}
+		}
+	}
+
+	index = (bio->bi_iter.bi_sector >> 3);
+	bio_for_each_segment(bvec, bio, iter) {
+		page_addr = bvec_virt(&bvec);
+		source_addr = dev_info->start + (index<<12) + bytes_done;
+		if (unlikely(!IS_ALIGNED((unsigned long)page_addr, PAGE_SIZE) ||
+			     !IS_ALIGNED(bvec.bv_len, PAGE_SIZE)))
+			// More paranoia.
+			goto fail;
+		if (bio_data_dir(bio) == READ)
+			memcpy(page_addr, __va(source_addr), bvec.bv_len);
+		else
+			memcpy(__va(source_addr), page_addr, bvec.bv_len);
+		bytes_done += bvec.bv_len;
+	}
+	bio_endio(bio);
+	return;
+fail:
+	bio_io_error(bio);
+}
+
+static long
+__dcssblk_direct_access(struct dcssblk_dev_info *dev_info, pgoff_t pgoff,
+		long nr_pages, void **kaddr, unsigned long *pfn)
+{
+	resource_size_t offset = pgoff * PAGE_SIZE;
+	unsigned long dev_sz;
+
+	dev_sz = dev_info->end - dev_info->start + 1;
+	if (kaddr)
+		*kaddr = __va(dev_info->start + offset);
+	if (pfn)
+		*pfn = PFN_DOWN(dev_info->start + offset);
+
+	return (dev_sz - offset) / PAGE_SIZE;
+}
+
+static long
+dcssblk_dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff,
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		unsigned long *pfn)
+{
+	struct dcssblk_dev_info *dev_info = dax_get_private(dax_dev);
+
+	return __dcssblk_direct_access(dev_info, pgoff, nr_pages, kaddr, pfn);
+}
+
+static void
+dcssblk_check_params(void)
+{
+	int rc, i, j, k;
+	char buf[DCSSBLK_PARM_LEN + 1];
+	struct dcssblk_dev_info *dev_info;
+
+	for (i = 0; (i < DCSSBLK_PARM_LEN) && (dcssblk_segments[i] != '\0');
+	     i++) {
+		for (j = i; (j < DCSSBLK_PARM_LEN) &&
+			    (dcssblk_segments[j] != ',')  &&
+			    (dcssblk_segments[j] != '\0') &&
+			    (dcssblk_segments[j] != '('); j++)
+		{
+			buf[j-i] = dcssblk_segments[j];
+		}
+		buf[j-i] = '\0';
+		rc = dcssblk_add_store(dcssblk_root_dev, NULL, buf, j-i);
+		if ((rc >= 0) && (dcssblk_segments[j] == '(')) {
+			for (k = 0; (buf[k] != ':') && (buf[k] != '\0'); k++)
+				buf[k] = toupper(buf[k]);
+			buf[k] = '\0';
+			if (!strncmp(&dcssblk_segments[j], "(local)", 7)) {
+				down_read(&dcssblk_devices_sem);
+				dev_info = dcssblk_get_device_by_name(buf);
+				up_read(&dcssblk_devices_sem);
+				if (dev_info)
+					dcssblk_shared_store(&dev_info->dev,
+							     NULL, "0\n", 2);
+			}
+		}
+		while ((dcssblk_segments[j] != ',') &&
+		       (dcssblk_segments[j] != '\0'))
+		{
+			j++;
+		}
+		if (dcssblk_segments[j] == '\0')
+			break;
+		i = j;
+	}
+}
+
+/*
+ * The init/exit functions.
+ */
+static void __exit
+dcssblk_exit(void)
+{
+	root_device_unregister(dcssblk_root_dev);
+	unregister_blkdev(dcssblk_major, DCSSBLK_NAME);
+}
+
+static int __init
+dcssblk_init(void)
+{
+	int rc;
+
+	dcssblk_root_dev = root_device_register("dcssblk");
+	if (IS_ERR(dcssblk_root_dev))
+		return PTR_ERR(dcssblk_root_dev);
+	rc = device_create_file(dcssblk_root_dev, &dev_attr_add);
+	if (rc)
+		goto out_root;
+	rc = device_create_file(dcssblk_root_dev, &dev_attr_remove);
+	if (rc)
+		goto out_root;
+	rc = register_blkdev(0, DCSSBLK_NAME);
+	if (rc < 0)
+		goto out_root;
+	dcssblk_major = rc;
+	init_rwsem(&dcssblk_devices_sem);
+
+	dcssblk_check_params();
+	return 0;
+
+out_root:
+	root_device_unregister(dcssblk_root_dev);
+
+	return rc;
+}
+
+module_init(dcssblk_init);
+module_exit(dcssblk_exit);
+
+module_param_string(segments, dcssblk_segments, DCSSBLK_PARM_LEN, 0444);
+MODULE_PARM_DESC(segments, "Name of DCSS segment(s) to be loaded, "
+		 "comma-separated list, names in each set separated "
+		 "by commas are separated by colons, each set contains "
+		 "names of contiguous segments and each name max. 8 chars.\n"
+		 "Adding \"(local)\" to the end of each set equals echoing 0 "
+		 "to /sys/devices/dcssblk/<device name>/shared after loading "
+		 "the contiguous segments - \n"
+		 "e.g. segments=\"mydcss1,mydcss2:mydcss3,mydcss4(local)\"");
+
+MODULE_DESCRIPTION("S/390 block driver for DCSS memory");
+MODULE_LICENSE("GPL");
diff --git a/drivers/s390/block/scm_blk.c b/drivers/s390/block/scm_blk.c
new file mode 100644
index 000000000..fa223949e
--- /dev/null
+++ b/drivers/s390/block/scm_blk.c
@@ -0,0 +1,581 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Block driver for s390 storage class memory.
+ *
+ * Copyright IBM Corp. 2012
+ * Author(s): Sebastian Ott <sebott@linux.vnet.ibm.com>
+ */
+
+#define pr_fmt(fmt) "scm_block: " fmt
+
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/mempool.h>
+#include <linux/module.h>
+#include <linux/blkdev.h>
+#include <linux/blk-mq.h>
+#include <linux/slab.h>
+#include <linux/list.h>
+#include <linux/io.h>
+#include <asm/eadm.h>
+#include "scm_blk.h"
+
+debug_info_t *scm_debug;
+static int scm_major;
+static mempool_t *aidaw_pool;
+static DEFINE_SPINLOCK(list_lock);
+static LIST_HEAD(inactive_requests);
+static unsigned int nr_requests = 64;
+static unsigned int nr_requests_per_io = 8;
+static atomic_t nr_devices = ATOMIC_INIT(0);
+module_param(nr_requests, uint, S_IRUGO);
+MODULE_PARM_DESC(nr_requests, "Number of parallel requests.");
+
+module_param(nr_requests_per_io, uint, S_IRUGO);
+MODULE_PARM_DESC(nr_requests_per_io, "Number of requests per IO.");
+
+MODULE_DESCRIPTION("Block driver for s390 storage class memory.");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("scm:scmdev*");
+
+static void __scm_free_rq(struct scm_request *scmrq)
+{
+	struct aob_rq_header *aobrq = to_aobrq(scmrq);
+
+	free_page((unsigned long) scmrq->aob);
+	kfree(scmrq->request);
+	kfree(aobrq);
+}
+
+static void scm_free_rqs(void)
+{
+	struct list_head *iter, *safe;
+	struct scm_request *scmrq;
+
+	spin_lock_irq(&list_lock);
+	list_for_each_safe(iter, safe, &inactive_requests) {
+		scmrq = list_entry(iter, struct scm_request, list);
+		list_del(&scmrq->list);
+		__scm_free_rq(scmrq);
+	}
+	spin_unlock_irq(&list_lock);
+
+	mempool_destroy(aidaw_pool);
+}
+
+static int __scm_alloc_rq(void)
+{
+	struct aob_rq_header *aobrq;
+	struct scm_request *scmrq;
+
+	aobrq = kzalloc(sizeof(*aobrq) + sizeof(*scmrq), GFP_KERNEL);
+	if (!aobrq)
+		return -ENOMEM;
+
+	scmrq = (void *) aobrq->data;
+	scmrq->aob = (void *) get_zeroed_page(GFP_DMA);
+	if (!scmrq->aob)
+		goto free;
+
+	scmrq->request = kzalloc_objs(scmrq->request[0], nr_requests_per_io);
+	if (!scmrq->request)
+		goto free;
+
+	INIT_LIST_HEAD(&scmrq->list);
+	spin_lock_irq(&list_lock);
+	list_add(&scmrq->list, &inactive_requests);
+	spin_unlock_irq(&list_lock);
+
+	return 0;
+free:
+	__scm_free_rq(scmrq);
+	return -ENOMEM;
+}
+
+static int scm_alloc_rqs(unsigned int nrqs)
+{
+	int ret = 0;
+
+	aidaw_pool = mempool_create_page_pool(max(nrqs/8, 1U), 0);
+	if (!aidaw_pool)
+		return -ENOMEM;
+
+	while (nrqs-- && !ret)
+		ret = __scm_alloc_rq();
+
+	return ret;
+}
+
+static struct scm_request *scm_request_fetch(void)
+{
+	struct scm_request *scmrq = NULL;
+
+	spin_lock_irq(&list_lock);
+	if (list_empty(&inactive_requests))
+		goto out;
+	scmrq = list_first_entry(&inactive_requests, struct scm_request, list);
+	list_del(&scmrq->list);
+out:
+	spin_unlock_irq(&list_lock);
+	return scmrq;
+}
+
+static void scm_request_done(struct scm_request *scmrq)
+{
+	unsigned long flags;
+	struct msb *msb;
+	u64 aidaw;
+	int i;
+
+	for (i = 0; i < nr_requests_per_io && scmrq->request[i]; i++) {
+		msb = &scmrq->aob->msb[i];
+		aidaw = (u64)dma64_to_virt(msb->data_addr);
+
+		if ((msb->flags & MSB_FLAG_IDA) && aidaw &&
+		    IS_ALIGNED(aidaw, PAGE_SIZE))
+			mempool_free(virt_to_page((void *)aidaw), aidaw_pool);
+	}
+
+	spin_lock_irqsave(&list_lock, flags);
+	list_add(&scmrq->list, &inactive_requests);
+	spin_unlock_irqrestore(&list_lock, flags);
+}
+
+static bool scm_permit_request(struct scm_blk_dev *bdev, struct request *req)
+{
+	return rq_data_dir(req) != WRITE || bdev->state != SCM_WR_PROHIBIT;
+}
+
+static inline struct aidaw *scm_aidaw_alloc(void)
+{
+	struct page *page = mempool_alloc(aidaw_pool, GFP_ATOMIC);
+
+	return page ? page_address(page) : NULL;
+}
+
+static inline unsigned long scm_aidaw_bytes(struct aidaw *aidaw)
+{
+	unsigned long _aidaw = (unsigned long) aidaw;
+	unsigned long bytes = ALIGN(_aidaw, PAGE_SIZE) - _aidaw;
+
+	return (bytes / sizeof(*aidaw)) * PAGE_SIZE;
+}
+
+struct aidaw *scm_aidaw_fetch(struct scm_request *scmrq, unsigned int bytes)
+{
+	struct aidaw *aidaw;
+
+	if (scm_aidaw_bytes(scmrq->next_aidaw) >= bytes)
+		return scmrq->next_aidaw;
+
+	aidaw = scm_aidaw_alloc();
+	if (aidaw)
+		memset(aidaw, 0, PAGE_SIZE);
+	return aidaw;
+}
+
+static int scm_request_prepare(struct scm_request *scmrq)
+{
+	struct scm_blk_dev *bdev = scmrq->bdev;
+	struct scm_device *scmdev = bdev->gendisk->private_data;
+	int pos = scmrq->aob->request.msb_count;
+	struct msb *msb = &scmrq->aob->msb[pos];
+	struct request *req = scmrq->request[pos];
+	struct req_iterator iter;
+	struct aidaw *aidaw;
+	struct bio_vec bv;
+
+	aidaw = scm_aidaw_fetch(scmrq, blk_rq_bytes(req));
+	if (!aidaw)
+		return -ENOMEM;
+
+	msb->bs = MSB_BS_4K;
+	scmrq->aob->request.msb_count++;
+	msb->scm_addr = scmdev->address + ((u64) blk_rq_pos(req) << 9);
+	msb->oc = (rq_data_dir(req) == READ) ? MSB_OC_READ : MSB_OC_WRITE;
+	msb->flags |= MSB_FLAG_IDA;
+	msb->data_addr = virt_to_dma64(aidaw);
+
+	rq_for_each_segment(bv, req, iter) {
+		WARN_ON(bv.bv_offset);
+		msb->blk_count += bv.bv_len >> 12;
+		aidaw->data_addr = virt_to_dma64(page_address(bv.bv_page));
+		aidaw++;
+	}
+
+	scmrq->next_aidaw = aidaw;
+	return 0;
+}
+
+static inline void scm_request_set(struct scm_request *scmrq,
+				   struct request *req)
+{
+	scmrq->request[scmrq->aob->request.msb_count] = req;
+}
+
+static inline void scm_request_init(struct scm_blk_dev *bdev,
+				    struct scm_request *scmrq)
+{
+	struct aob_rq_header *aobrq = to_aobrq(scmrq);
+	struct aob *aob = scmrq->aob;
+
+	memset(scmrq->request, 0,
+	       nr_requests_per_io * sizeof(scmrq->request[0]));
+	memset(aob, 0, sizeof(*aob));
+	aobrq->scmdev = bdev->scmdev;
+	aob->request.cmd_code = ARQB_CMD_MOVE;
+	aob->request.data = (u64) aobrq;
+	scmrq->bdev = bdev;
+	scmrq->retries = 4;
+	scmrq->error = BLK_STS_OK;
+	/* We don't use all msbs - place aidaws at the end of the aob page. */
+	scmrq->next_aidaw = (void *) &aob->msb[nr_requests_per_io];
+}
+
+static void scm_request_requeue(struct scm_request *scmrq)
+{
+	struct scm_blk_dev *bdev = scmrq->bdev;
+	int i;
+
+	for (i = 0; i < nr_requests_per_io && scmrq->request[i]; i++)
+		blk_mq_requeue_request(scmrq->request[i], false);
+
+	atomic_dec(&bdev->queued_reqs);
+	scm_request_done(scmrq);
+	blk_mq_kick_requeue_list(bdev->rq);
+}
+
+static void scm_request_finish(struct scm_request *scmrq)
+{
+	struct scm_blk_dev *bdev = scmrq->bdev;
+	blk_status_t *error;
+	int i;
+
+	for (i = 0; i < nr_requests_per_io && scmrq->request[i]; i++) {
+		error = blk_mq_rq_to_pdu(scmrq->request[i]);
+		*error = scmrq->error;
+		if (likely(!blk_should_fake_timeout(scmrq->request[i]->q)))
+			blk_mq_complete_request(scmrq->request[i]);
+	}
+
+	atomic_dec(&bdev->queued_reqs);
+	scm_request_done(scmrq);
+}
+
+static void scm_request_start(struct scm_request *scmrq)
+{
+	struct scm_blk_dev *bdev = scmrq->bdev;
+
+	atomic_inc(&bdev->queued_reqs);
+	if (eadm_start_aob(scmrq->aob)) {
+		SCM_LOG(5, "no subchannel");
+		scm_request_requeue(scmrq);
+	}
+}
+
+struct scm_queue {
+	struct scm_request *scmrq;
+	spinlock_t lock;
+};
+
+static blk_status_t scm_blk_request(struct blk_mq_hw_ctx *hctx,
+			   const struct blk_mq_queue_data *qd)
+{
+	struct scm_device *scmdev = hctx->queue->queuedata;
+	struct scm_blk_dev *bdev = dev_get_drvdata(&scmdev->dev);
+	struct scm_queue *sq = hctx->driver_data;
+	struct request *req = qd->rq;
+	struct scm_request *scmrq;
+
+	spin_lock(&sq->lock);
+	if (!scm_permit_request(bdev, req)) {
+		spin_unlock(&sq->lock);
+		return BLK_STS_RESOURCE;
+	}
+
+	scmrq = sq->scmrq;
+	if (!scmrq) {
+		scmrq = scm_request_fetch();
+		if (!scmrq) {
+			SCM_LOG(5, "no request");
+			spin_unlock(&sq->lock);
+			return BLK_STS_RESOURCE;
+		}
+		scm_request_init(bdev, scmrq);
+		sq->scmrq = scmrq;
+	}
+	scm_request_set(scmrq, req);
+
+	if (scm_request_prepare(scmrq)) {
+		SCM_LOG(5, "aidaw alloc failed");
+		scm_request_set(scmrq, NULL);
+
+		if (scmrq->aob->request.msb_count)
+			scm_request_start(scmrq);
+
+		sq->scmrq = NULL;
+		spin_unlock(&sq->lock);
+		return BLK_STS_RESOURCE;
+	}
+	blk_mq_start_request(req);
+
+	if (qd->last || scmrq->aob->request.msb_count == nr_requests_per_io) {
+		scm_request_start(scmrq);
+		sq->scmrq = NULL;
+	}
+	spin_unlock(&sq->lock);
+	return BLK_STS_OK;
+}
+
+static int scm_blk_init_hctx(struct blk_mq_hw_ctx *hctx, void *data,
+			     unsigned int idx)
+{
+	struct scm_queue *qd = kzalloc_obj(*qd);
+
+	if (!qd)
+		return -ENOMEM;
+
+	spin_lock_init(&qd->lock);
+	hctx->driver_data = qd;
+
+	return 0;
+}
+
+static void scm_blk_exit_hctx(struct blk_mq_hw_ctx *hctx, unsigned int idx)
+{
+	struct scm_queue *qd = hctx->driver_data;
+
+	WARN_ON(qd->scmrq);
+	kfree(hctx->driver_data);
+	hctx->driver_data = NULL;
+}
+
+static void __scmrq_log_error(struct scm_request *scmrq)
+{
+	struct aob *aob = scmrq->aob;
+
+	if (scmrq->error == BLK_STS_TIMEOUT)
+		SCM_LOG(1, "Request timeout");
+	else {
+		SCM_LOG(1, "Request error");
+		SCM_LOG_HEX(1, &aob->response, sizeof(aob->response));
+	}
+	if (scmrq->retries)
+		SCM_LOG(1, "Retry request");
+	else
+		pr_err("An I/O operation to SCM failed with rc=%d\n",
+		       scmrq->error);
+}
+
+static void scm_blk_handle_error(struct scm_request *scmrq)
+{
+	struct scm_blk_dev *bdev = scmrq->bdev;
+	unsigned long flags;
+
+	if (scmrq->error != BLK_STS_IOERR)
+		goto restart;
+
+	/* For -EIO the response block is valid. */
+	switch (scmrq->aob->response.eqc) {
+	case EQC_WR_PROHIBIT:
+		spin_lock_irqsave(&bdev->lock, flags);
+		if (bdev->state != SCM_WR_PROHIBIT)
+			pr_info("%lx: Write access to the SCM increment is suspended\n",
+				(unsigned long) bdev->scmdev->address);
+		bdev->state = SCM_WR_PROHIBIT;
+		spin_unlock_irqrestore(&bdev->lock, flags);
+		goto requeue;
+	default:
+		break;
+	}
+
+restart:
+	if (!eadm_start_aob(scmrq->aob))
+		return;
+
+requeue:
+	scm_request_requeue(scmrq);
+}
+
+void scm_blk_irq(struct scm_device *scmdev, void *data, blk_status_t error)
+{
+	struct scm_request *scmrq = data;
+
+	scmrq->error = error;
+	if (error) {
+		__scmrq_log_error(scmrq);
+		if (scmrq->retries-- > 0) {
+			scm_blk_handle_error(scmrq);
+			return;
+		}
+	}
+
+	scm_request_finish(scmrq);
+}
+
+static void scm_blk_request_done(struct request *req)
+{
+	blk_status_t *error = blk_mq_rq_to_pdu(req);
+
+	blk_mq_end_request(req, *error);
+}
+
+static const struct block_device_operations scm_blk_devops = {
+	.owner = THIS_MODULE,
+};
+
+static const struct blk_mq_ops scm_mq_ops = {
+	.queue_rq = scm_blk_request,
+	.complete = scm_blk_request_done,
+	.init_hctx = scm_blk_init_hctx,
+	.exit_hctx = scm_blk_exit_hctx,
+};
+
+int scm_blk_dev_setup(struct scm_blk_dev *bdev, struct scm_device *scmdev)
+{
+	struct queue_limits lim = {
+		.logical_block_size	= 1 << 12,
+	};
+	unsigned int devindex;
+	int len, ret;
+
+	lim.max_segments = min(scmdev->nr_max_block,
+		(unsigned int) (PAGE_SIZE / sizeof(struct aidaw)));
+	lim.max_hw_sectors = lim.max_segments << 3; /* 8 * 512 = blk_size */
+
+	devindex = atomic_inc_return(&nr_devices) - 1;
+	/* scma..scmz + scmaa..scmzz */
+	if (devindex > 701) {
+		ret = -ENODEV;
+		goto out;
+	}
+
+	bdev->scmdev = scmdev;
+	bdev->state = SCM_OPER;
+	spin_lock_init(&bdev->lock);
+	atomic_set(&bdev->queued_reqs, 0);
+
+	bdev->tag_set.ops = &scm_mq_ops;
+	bdev->tag_set.cmd_size = sizeof(blk_status_t);
+	bdev->tag_set.nr_hw_queues = nr_requests;
+	bdev->tag_set.queue_depth = nr_requests_per_io * nr_requests;
+	bdev->tag_set.numa_node = NUMA_NO_NODE;
+
+	ret = blk_mq_alloc_tag_set(&bdev->tag_set);
+	if (ret)
+		goto out;
+
+	bdev->gendisk = blk_mq_alloc_disk(&bdev->tag_set, &lim, scmdev);
+	if (IS_ERR(bdev->gendisk)) {
+		ret = PTR_ERR(bdev->gendisk);
+		goto out_tag;
+	}
+	bdev->gendisk->private_data = scmdev;
+	bdev->gendisk->fops = &scm_blk_devops;
+	bdev->gendisk->major = scm_major;
+	bdev->gendisk->first_minor = devindex * SCM_NR_PARTS;
+	bdev->gendisk->minors = SCM_NR_PARTS;
+
+	len = snprintf(bdev->gendisk->disk_name, DISK_NAME_LEN, "scm");
+	if (devindex > 25) {
+		len += snprintf(bdev->gendisk->disk_name + len,
+				DISK_NAME_LEN - len, "%c",
+				'a' + (devindex / 26) - 1);
+		devindex = devindex % 26;
+	}
+	snprintf(bdev->gendisk->disk_name + len, DISK_NAME_LEN - len, "%c",
+		 'a' + devindex);
+
+	/* 512 byte sectors */
+	set_capacity(bdev->gendisk, scmdev->size >> 9);
+	ret = device_add_disk(&scmdev->dev, bdev->gendisk, NULL);
+	if (ret)
+		goto out_cleanup_disk;
+
+	return 0;
+
+out_cleanup_disk:
+	put_disk(bdev->gendisk);
+out_tag:
+	blk_mq_free_tag_set(&bdev->tag_set);
+out:
+	atomic_dec(&nr_devices);
+	return ret;
+}
+
+void scm_blk_dev_cleanup(struct scm_blk_dev *bdev)
+{
+	del_gendisk(bdev->gendisk);
+	put_disk(bdev->gendisk);
+	blk_mq_free_tag_set(&bdev->tag_set);
+}
+
+void scm_blk_set_available(struct scm_blk_dev *bdev)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&bdev->lock, flags);
+	if (bdev->state == SCM_WR_PROHIBIT)
+		pr_info("%lx: Write access to the SCM increment is restored\n",
+			(unsigned long) bdev->scmdev->address);
+	bdev->state = SCM_OPER;
+	spin_unlock_irqrestore(&bdev->lock, flags);
+}
+
+static bool __init scm_blk_params_valid(void)
+{
+	if (!nr_requests_per_io || nr_requests_per_io > 64)
+		return false;
+
+	return true;
+}
+
+static int __init scm_blk_init(void)
+{
+	int ret = -EINVAL;
+
+	if (!scm_blk_params_valid())
+		goto out;
+
+	ret = register_blkdev(0, "scm");
+	if (ret < 0)
+		goto out;
+
+	scm_major = ret;
+	ret = scm_alloc_rqs(nr_requests);
+	if (ret)
+		goto out_free;
+
+	scm_debug = debug_register("scm_log", 16, 1, 16);
+	if (!scm_debug) {
+		ret = -ENOMEM;
+		goto out_free;
+	}
+
+	debug_register_view(scm_debug, &debug_hex_ascii_view);
+	debug_set_level(scm_debug, 2);
+
+	ret = scm_drv_init();
+	if (ret)
+		goto out_dbf;
+
+	return ret;
+
+out_dbf:
+	debug_unregister(scm_debug);
+out_free:
+	scm_free_rqs();
+	unregister_blkdev(scm_major, "scm");
+out:
+	return ret;
+}
+module_init(scm_blk_init);
+
+static void __exit scm_blk_cleanup(void)
+{
+	scm_drv_cleanup();
+	debug_unregister(scm_debug);
+	scm_free_rqs();
+	unregister_blkdev(scm_major, "scm");
+}
+module_exit(scm_blk_cleanup);
diff --git a/drivers/s390/block/scm_blk.h b/drivers/s390/block/scm_blk.h
new file mode 100644
index 000000000..af82b3214
--- /dev/null
+++ b/drivers/s390/block/scm_blk.h
@@ -0,0 +1,76 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef SCM_BLK_H
+#define SCM_BLK_H
+
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/blkdev.h>
+#include <linux/blk-mq.h>
+#include <linux/list.h>
+
+#include <asm/debug.h>
+#include <asm/eadm.h>
+
+#define SCM_NR_PARTS 8
+#define SCM_QUEUE_DELAY 5
+
+struct scm_blk_dev {
+	struct request_queue *rq;
+	struct gendisk *gendisk;
+	struct blk_mq_tag_set tag_set;
+	struct scm_device *scmdev;
+	spinlock_t lock;
+	atomic_t queued_reqs;
+	enum {SCM_OPER, SCM_WR_PROHIBIT} state;
+	struct list_head finished_requests;
+};
+
+struct scm_request {
+	struct scm_blk_dev *bdev;
+	struct aidaw *next_aidaw;
+	struct request **request;
+	struct aob *aob;
+	struct list_head list;
+	u8 retries;
+	blk_status_t error;
+};
+
+#define to_aobrq(rq) container_of((void *) rq, struct aob_rq_header, data)
+
+int scm_blk_dev_setup(struct scm_blk_dev *, struct scm_device *);
+void scm_blk_dev_cleanup(struct scm_blk_dev *);
+void scm_blk_set_available(struct scm_blk_dev *);
+void scm_blk_irq(struct scm_device *, void *, blk_status_t);
+
+struct aidaw *scm_aidaw_fetch(struct scm_request *scmrq, unsigned int bytes);
+
+int scm_drv_init(void);
+void scm_drv_cleanup(void);
+
+extern debug_info_t *scm_debug;
+
+#define SCM_LOG(imp, txt) do {					\
+		debug_text_event(scm_debug, imp, txt);		\
+	} while (0)
+
+static inline void SCM_LOG_HEX(int level, void *data, int length)
+{
+	debug_event(scm_debug, level, data, length);
+}
+
+static inline void SCM_LOG_STATE(int level, struct scm_device *scmdev)
+{
+	struct {
+		u64 address;
+		u8 oper_state;
+		u8 rank;
+	} __packed data = {
+		.address = scmdev->address,
+		.oper_state = scmdev->attrs.oper_state,
+		.rank = scmdev->attrs.rank,
+	};
+
+	SCM_LOG_HEX(level, &data, sizeof(data));
+}
+
+#endif /* SCM_BLK_H */
diff --git a/drivers/s390/block/scm_drv.c b/drivers/s390/block/scm_drv.c
new file mode 100644
index 000000000..095d06eea
--- /dev/null
+++ b/drivers/s390/block/scm_drv.c
@@ -0,0 +1,90 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Device driver for s390 storage class memory.
+ *
+ * Copyright IBM Corp. 2012
+ * Author(s): Sebastian Ott <sebott@linux.vnet.ibm.com>
+ */
+
+#define pr_fmt(fmt) "scm_block: " fmt
+
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <asm/eadm.h>
+#include "scm_blk.h"
+
+static void scm_notify(struct scm_device *scmdev, enum scm_event event)
+{
+	struct scm_blk_dev *bdev = dev_get_drvdata(&scmdev->dev);
+
+	switch (event) {
+	case SCM_CHANGE:
+		pr_info("%lx: The capabilities of the SCM increment changed\n",
+			(unsigned long) scmdev->address);
+		SCM_LOG(2, "State changed");
+		SCM_LOG_STATE(2, scmdev);
+		break;
+	case SCM_AVAIL:
+		SCM_LOG(2, "Increment available");
+		SCM_LOG_STATE(2, scmdev);
+		scm_blk_set_available(bdev);
+		break;
+	}
+}
+
+static int scm_probe(struct scm_device *scmdev)
+{
+	struct scm_blk_dev *bdev;
+	int ret;
+
+	SCM_LOG(2, "probe");
+	SCM_LOG_STATE(2, scmdev);
+
+	if (scmdev->attrs.oper_state != OP_STATE_GOOD)
+		return -EINVAL;
+
+	bdev = kzalloc_obj(*bdev);
+	if (!bdev)
+		return -ENOMEM;
+
+	dev_set_drvdata(&scmdev->dev, bdev);
+	ret = scm_blk_dev_setup(bdev, scmdev);
+	if (ret) {
+		dev_set_drvdata(&scmdev->dev, NULL);
+		kfree(bdev);
+		goto out;
+	}
+
+out:
+	return ret;
+}
+
+static void scm_remove(struct scm_device *scmdev)
+{
+	struct scm_blk_dev *bdev = dev_get_drvdata(&scmdev->dev);
+
+	scm_blk_dev_cleanup(bdev);
+	dev_set_drvdata(&scmdev->dev, NULL);
+	kfree(bdev);
+}
+
+static struct scm_driver scm_drv = {
+	.drv = {
+		.name = "scm_block",
+		.owner = THIS_MODULE,
+	},
+	.notify = scm_notify,
+	.probe = scm_probe,
+	.remove = scm_remove,
+	.handler = scm_blk_irq,
+};
+
+int __init scm_drv_init(void)
+{
+	return scm_driver_register(&scm_drv);
+}
+
+void scm_drv_cleanup(void)
+{
+	scm_driver_unregister(&scm_drv);
+}