Merge tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Don't destroy user event fields when removal fails

   User event fields are destroyed before the event is removed from
   visibility. But that can fail leaving the still visible event with no
   fields. Move the destroying of the fields to after the event is
   successfully removed from visibility.

 - Initialize function graph state is fork before calling
   copy_exec_state()

   For non-CLONE_VM forks, copy_exec_state() allocates a new
   task_exec_state. If that allocation fails, ftrace_graph_exit_task()
   will free the tasks ret_stack pointer. Since that pointer is still
   using the parent's ret_stack, it mistakenly frees the parent's
   pointer too.

   Call ftrace_graph_init() on the task first which will NULL out the
   new tasks's ret_stack and if the copy fails, it will not free
   anything.

 - Remove FGRAPH_MAX_INDEX

   The macro FGRAPH_MAX_INDEX was added but never used. Remove it.

 - Save ent_size in function graph printing of nested functions

   The function graph tracer needs to look at the next event to see if
   the next event is the return of the current function entry. If it is,
   it prints a single line:

	ktime_get();

   Otherwise it prints it like a nested function:

	tick_nohz_irq_exit() {
	    ktime_get();
	    kcpustat_irq_exit();
	}

   In order to look at the next event, it must save the current event so
   that it has the information to print from it. It saves the event in
   the iterator descriptor called "ent". What it doesn't save is the
   ent_size of the event which is now used to know if the function graph
   arguments are to be printed. The peek doesn't save the size so the
   size used happens to be that of the size of the last event that was
   seen.

   Save the entry event size in the iterator descriptor so that the
   correct size is used.

 - Fix several errors with freeing data in the histogram code

   The histogram code had a lot of leaked or or incorrect accounting
   when failures happen. Correct them.

 - Fix histogram regression of .percent and .graph modifiers

   Up until 6.3 histogram values could have "percent" or "graph"
   modifiers that changed how they were printed. But a change that added
   restricting histograms values from being strings, stack traces and
   other modifiers inadvertently prevented them from using the percent
   and graph modifiers, which were legal use cases for values.

   Put back the percent and graph modifiers.

 - Fix various typos in the comments

 - Set the trace_clock before initializing a histogram with clock
   argument

   The histogram API allows the user to specific which trace clock to
   use via a "clock=" string. The histogram is set up first before the
   clock is checked. If the passed in clock is not valid, it exits
   without fully fixing up the histogram leaving it on the list and a
   use-after-free can trigger.

   Update the clock argument first and if it fails then exit gracefully
   before the histogram trigger is placed on any lists.

 - Restore :mod: trailer after parsing in ftrace_set_clr_event

   The function ftrace_set_clr_event() modifies the parse string and
   needs to put it back to what was passed in. It searches for ":mod:"
   via a strsep() but fails to put back the first ':' in the string.

   Add back the ':' in the passed in string.

 - Take trace_array reference when opening a tracer options file

   The options files are dynamically created and some tracers add their
   own options. When a tracer adds their own list of options, the
   trace_array holding them has an array to hold the list of options for
   each tracer. This array increases in size via a krealloc(), and the
   new entry gets a newly allocated array to hold the options of the new
   tracer being added.

   The element in each entry of the tracer's option array holds a
   pointer back to the trace_array, a pointer to the tracer it is
   associated to, a pointer to the flags of the option.

   The issue is that these arrays are freed when the trace_array is
   freed when its instance it represents is removed from the instances
   directory. There's a race that an open of one of these options files
   can happen when the instance is being removed.

   Add a new helper function to be called by the open function of the
   options file to iterate all existing trace_arrays under a lock and
   find the one that has the given option element in one of it's tracer
   arrays. If found, then update the associated trace_array's reference
   counter to keep it from being freed. If not found, have the open call
   return -ENODEV.

 - Disable interrupts when acquiring the lock in rb_wake_up_waiters()

   The function rb_wake_up_waiters() assumes it will be called in
   interrupt context and does not disable irqs when taking
   cpu_buffer->reader_lock, which can be called in hard interrupt
   context. The issue is in PREEMPT_RT, this function is called in
   thread context leaving this lock open to a deadlock.

   Take the lock with interrupts disabled.

 - Use rcu_assign_pointer() for tmp_ops filter hash

   The tmp_ops used in update_ftrace_direct_mod() assigns its
   filter_hash field directly, but that field is annotated as __rcu and
   sparse complains. Assign it with rcu_assign_pointer()

 - Fix use-after-free in enable_trigger_private_data_free()

   The trace_event_call is accessed through the event_trigger_data's
   trace_event_file pointer to put the trace_event_call on freeing. The
   issue is that the trace_event_file data may have been freed already
   causing a use-after-free. Add a field to the event_trigger_data that
   points directly to the trace_event_call so that it can decrement its
   reference directly without needing to go through the
   trace_event_file.

 - Fix accounting of buffer data remote headers

   trace_buffer_desc_size() and trace_remote_alloc_buffer() undercount
   the number of pages is needed for the asked for size as it doesn't
   take into account the meta data on each page. Add a helper function
   to do the calculation properly and use that in these functions.

 - Catch nr_page_va overflow in ring_buffer_desc sizing

   The number of pages per remote ring buffer is capped by
   ring_buffer_desc::nr_page_va (32 bits). A buffer_size large enough to
   overflow that field would silently allocate a descriptor smaller than
   what was asked for.

 - Do not resize the subbuf order if any per_cpu buffer is disabled

   The mmapping of ring buffers disables resizing the subbuffers, but it
   is done per-cpu whereas the subbuf size change is done for all the
   per_cpu buffers under the buffer->mutex. It could change the size of
   some while the mapping is happening on others. Have the resize of the
   subbuf order check all the per_cpu buffers under the lock to see if
   any of them is disabled before starting and causing an inconsistency
   between buffers that are being mapped.

* tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (25 commits)
  ring-buffer: Check resize_disabled before publishing the new subbuf order
  tracing/remotes: Catch nr_page_va overflow in ring_buffer_desc sizing
  tracing/remotes: Account for ring buffer page header in size calculation
  tracing: Don't dereference trace_event_file in deferred trigger free
  ftrace: Use rcu_assign_pointer() for tmp_ops filter hash
  ring-buffer: Acquire the lock with irqsave in rb_wake_up_waiters()
  tracing: Take trace_array reference when opening a tracer options file
  tracing: Fix ring_buffer_read_page_size() kernel-doc
  tracing: Restore :mod: trailer after parsing in ftrace_set_clr_event()
  tracing: Fix memory corruption from a "STACKTRACE" histogram key
  tracing: Fix memory corruption from the histogram stacktrace modifier
  tracing: Undo the registration when enabling the histogram trigger fails
  tracing: Take the reference before publishing the named histogram trigger
  tracing: Set the trace clock before registering the histogram trigger
  tracing: Fix typo "preceeded" in comment
  tracing: Fix typo "availabe" in comment
  tracing: Let histogram values keep the percent and graph modifiers
  tracing: Keep the entry count when the histogram stats allocation fails
  tracing: Free histogram the field rejected for a bad modifier
  tracing: Free histogram the var ref when its initialization fails
  ...

1 files changed, 969 insertions, 0 deletions

diff --git a/drivers/dma/altera-msgdma.c b/drivers/dma/altera-msgdma.c
new file mode 100644
index 000000000..e23e5b441
--- /dev/null
+++ b/drivers/dma/altera-msgdma.c
@@ -0,0 +1,969 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * DMA driver for Altera mSGDMA IP core
+ *
+ * Copyright (C) 2017 Stefan Roese <sr@denx.de>
+ *
+ * Based on drivers/dma/xilinx/zynqmp_dma.c, which is:
+ * Copyright (C) 2016 Xilinx, Inc. All rights reserved.
+ */
+
+#include <linux/bitops.h>
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include <linux/dmapool.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/iopoll.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/of_dma.h>
+
+#include "dmaengine.h"
+
+#define MSGDMA_MAX_TRANS_LEN		U32_MAX
+#define MSGDMA_DESC_NUM			1024
+
+/**
+ * struct msgdma_extended_desc - implements an extended descriptor
+ * @read_addr_lo: data buffer source address low bits
+ * @write_addr_lo: data buffer destination address low bits
+ * @len: the number of bytes to transfer per descriptor
+ * @burst_seq_num: bit 31:24 write burst
+ *		   bit 23:16 read burst
+ *		   bit 15:00 sequence number
+ * @stride: bit 31:16 write stride
+ *	    bit 15:00 read stride
+ * @read_addr_hi: data buffer source address high bits
+ * @write_addr_hi: data buffer destination address high bits
+ * @control: characteristics of the transfer
+ */
+struct msgdma_extended_desc {
+	u32 read_addr_lo;
+	u32 write_addr_lo;
+	u32 len;
+	u32 burst_seq_num;
+	u32 stride;
+	u32 read_addr_hi;
+	u32 write_addr_hi;
+	u32 control;
+};
+
+/* mSGDMA descriptor control field bit definitions */
+#define MSGDMA_DESC_CTL_SET_CH(x)	((x) & 0xff)
+#define MSGDMA_DESC_CTL_GEN_SOP		BIT(8)
+#define MSGDMA_DESC_CTL_GEN_EOP		BIT(9)
+#define MSGDMA_DESC_CTL_PARK_READS	BIT(10)
+#define MSGDMA_DESC_CTL_PARK_WRITES	BIT(11)
+#define MSGDMA_DESC_CTL_END_ON_EOP	BIT(12)
+#define MSGDMA_DESC_CTL_END_ON_LEN	BIT(13)
+#define MSGDMA_DESC_CTL_TR_COMP_IRQ	BIT(14)
+#define MSGDMA_DESC_CTL_EARLY_IRQ	BIT(15)
+#define MSGDMA_DESC_CTL_TR_ERR_IRQ	GENMASK(23, 16)
+#define MSGDMA_DESC_CTL_EARLY_DONE	BIT(24)
+
+/*
+ * Writing "1" the "go" bit commits the entire descriptor into the
+ * descriptor FIFO(s)
+ */
+#define MSGDMA_DESC_CTL_GO		BIT(31)
+
+/* Tx buffer control flags */
+#define MSGDMA_DESC_CTL_TX_FIRST	(MSGDMA_DESC_CTL_GEN_SOP |	\
+					 MSGDMA_DESC_CTL_TR_ERR_IRQ |	\
+					 MSGDMA_DESC_CTL_GO)
+
+#define MSGDMA_DESC_CTL_TX_MIDDLE	(MSGDMA_DESC_CTL_TR_ERR_IRQ |	\
+					 MSGDMA_DESC_CTL_GO)
+
+#define MSGDMA_DESC_CTL_TX_LAST		(MSGDMA_DESC_CTL_GEN_EOP |	\
+					 MSGDMA_DESC_CTL_TR_COMP_IRQ |	\
+					 MSGDMA_DESC_CTL_TR_ERR_IRQ |	\
+					 MSGDMA_DESC_CTL_GO)
+
+#define MSGDMA_DESC_CTL_TX_SINGLE	(MSGDMA_DESC_CTL_GEN_SOP |	\
+					 MSGDMA_DESC_CTL_GEN_EOP |	\
+					 MSGDMA_DESC_CTL_TR_COMP_IRQ |	\
+					 MSGDMA_DESC_CTL_TR_ERR_IRQ |	\
+					 MSGDMA_DESC_CTL_GO)
+
+#define MSGDMA_DESC_CTL_RX_SINGLE	(MSGDMA_DESC_CTL_END_ON_EOP |	\
+					 MSGDMA_DESC_CTL_END_ON_LEN |	\
+					 MSGDMA_DESC_CTL_TR_COMP_IRQ |	\
+					 MSGDMA_DESC_CTL_EARLY_IRQ |	\
+					 MSGDMA_DESC_CTL_TR_ERR_IRQ |	\
+					 MSGDMA_DESC_CTL_GO)
+
+/* mSGDMA extended descriptor stride definitions */
+#define MSGDMA_DESC_STRIDE_RD		0x00000001
+#define MSGDMA_DESC_STRIDE_WR		0x00010000
+#define MSGDMA_DESC_STRIDE_RW		0x00010001
+
+/* mSGDMA dispatcher control and status register map */
+#define MSGDMA_CSR_STATUS		0x00	/* Read / Clear */
+#define MSGDMA_CSR_CONTROL		0x04	/* Read / Write */
+#define MSGDMA_CSR_RW_FILL_LEVEL	0x08	/* 31:16 - write fill level */
+						/* 15:00 - read fill level */
+#define MSGDMA_CSR_RESP_FILL_LEVEL	0x0c	/* response FIFO fill level */
+#define MSGDMA_CSR_RW_SEQ_NUM		0x10	/* 31:16 - write seq number */
+						/* 15:00 - read seq number */
+
+/* mSGDMA CSR status register bit definitions */
+#define MSGDMA_CSR_STAT_BUSY			BIT(0)
+#define MSGDMA_CSR_STAT_DESC_BUF_EMPTY		BIT(1)
+#define MSGDMA_CSR_STAT_DESC_BUF_FULL		BIT(2)
+#define MSGDMA_CSR_STAT_RESP_BUF_EMPTY		BIT(3)
+#define MSGDMA_CSR_STAT_RESP_BUF_FULL		BIT(4)
+#define MSGDMA_CSR_STAT_STOPPED			BIT(5)
+#define MSGDMA_CSR_STAT_RESETTING		BIT(6)
+#define MSGDMA_CSR_STAT_STOPPED_ON_ERR		BIT(7)
+#define MSGDMA_CSR_STAT_STOPPED_ON_EARLY	BIT(8)
+#define MSGDMA_CSR_STAT_IRQ			BIT(9)
+#define MSGDMA_CSR_STAT_MASK			GENMASK(9, 0)
+#define MSGDMA_CSR_STAT_MASK_WITHOUT_IRQ	GENMASK(8, 0)
+
+#define DESC_EMPTY	(MSGDMA_CSR_STAT_DESC_BUF_EMPTY | \
+			 MSGDMA_CSR_STAT_RESP_BUF_EMPTY)
+
+/* mSGDMA CSR control register bit definitions */
+#define MSGDMA_CSR_CTL_STOP			BIT(0)
+#define MSGDMA_CSR_CTL_RESET			BIT(1)
+#define MSGDMA_CSR_CTL_STOP_ON_ERR		BIT(2)
+#define MSGDMA_CSR_CTL_STOP_ON_EARLY		BIT(3)
+#define MSGDMA_CSR_CTL_GLOBAL_INTR		BIT(4)
+#define MSGDMA_CSR_CTL_STOP_DESCS		BIT(5)
+
+/* mSGDMA CSR fill level bits */
+#define MSGDMA_CSR_WR_FILL_LEVEL_GET(v)		(((v) & 0xffff0000) >> 16)
+#define MSGDMA_CSR_RD_FILL_LEVEL_GET(v)		((v) & 0x0000ffff)
+#define MSGDMA_CSR_RESP_FILL_LEVEL_GET(v)	((v) & 0x0000ffff)
+
+#define MSGDMA_CSR_SEQ_NUM_GET(v)		(((v) & 0xffff0000) >> 16)
+
+/* mSGDMA response register map */
+#define MSGDMA_RESP_BYTES_TRANSFERRED	0x00
+#define MSGDMA_RESP_STATUS		0x04
+
+/* mSGDMA response register bit definitions */
+#define MSGDMA_RESP_EARLY_TERM	BIT(8)
+#define MSGDMA_RESP_ERR_MASK	0xff
+
+/**
+ * struct msgdma_sw_desc - implements a sw descriptor
+ * @async_tx: support for the async_tx api
+ * @hw_desc: associated HW descriptor
+ * @node: node to move from the free list to the tx list
+ * @tx_list: transmit list node
+ */
+struct msgdma_sw_desc {
+	struct dma_async_tx_descriptor async_tx;
+	struct msgdma_extended_desc hw_desc;
+	struct list_head node;
+	struct list_head tx_list;
+};
+
+/*
+ * struct msgdma_device - DMA device structure
+ */
+struct msgdma_device {
+	spinlock_t lock;
+	struct device *dev;
+	struct tasklet_struct irq_tasklet;
+	struct list_head pending_list;
+	struct list_head free_list;
+	struct list_head active_list;
+	struct list_head done_list;
+	u32 desc_free_cnt;
+	bool idle;
+
+	struct dma_device dmadev;
+	struct dma_chan	dmachan;
+	dma_addr_t hw_desq;
+	struct msgdma_sw_desc *sw_desq;
+	unsigned int npendings;
+
+	struct dma_slave_config slave_cfg;
+
+	int irq;
+
+	/* mSGDMA controller */
+	void __iomem *csr;
+
+	/* mSGDMA descriptors */
+	void __iomem *desc;
+
+	/* mSGDMA response */
+	void __iomem *resp;
+};
+
+#define to_mdev(chan)	container_of(chan, struct msgdma_device, dmachan)
+#define tx_to_desc(tx)	container_of(tx, struct msgdma_sw_desc, async_tx)
+
+/**
+ * msgdma_get_descriptor - Get the sw descriptor from the pool
+ * @mdev: Pointer to the Altera mSGDMA device structure
+ *
+ * Return: The sw descriptor
+ */
+static struct msgdma_sw_desc *msgdma_get_descriptor(struct msgdma_device *mdev)
+{
+	struct msgdma_sw_desc *desc;
+	unsigned long flags;
+
+	spin_lock_irqsave(&mdev->lock, flags);
+	desc = list_first_entry(&mdev->free_list, struct msgdma_sw_desc, node);
+	list_del(&desc->node);
+	spin_unlock_irqrestore(&mdev->lock, flags);
+
+	INIT_LIST_HEAD(&desc->tx_list);
+
+	return desc;
+}
+
+/**
+ * msgdma_free_descriptor - Issue pending transactions
+ * @mdev: Pointer to the Altera mSGDMA device structure
+ * @desc: Transaction descriptor pointer
+ */
+static void msgdma_free_descriptor(struct msgdma_device *mdev,
+				   struct msgdma_sw_desc *desc)
+{
+	struct msgdma_sw_desc *child, *next;
+
+	mdev->desc_free_cnt++;
+	list_move_tail(&desc->node, &mdev->free_list);
+	list_for_each_entry_safe(child, next, &desc->tx_list, node) {
+		mdev->desc_free_cnt++;
+		list_move_tail(&child->node, &mdev->free_list);
+	}
+}
+
+/**
+ * msgdma_free_desc_list - Free descriptors list
+ * @mdev: Pointer to the Altera mSGDMA device structure
+ * @list: List to parse and delete the descriptor
+ */
+static void msgdma_free_desc_list(struct msgdma_device *mdev,
+				  struct list_head *list)
+{
+	struct msgdma_sw_desc *desc, *next;
+
+	list_for_each_entry_safe(desc, next, list, node)
+		msgdma_free_descriptor(mdev, desc);
+}
+
+/**
+ * msgdma_desc_config - Configure the descriptor
+ * @desc: Hw descriptor pointer
+ * @dst: Destination buffer address
+ * @src: Source buffer address
+ * @len: Transfer length
+ * @stride: Read/write stride value to set
+ */
+static void msgdma_desc_config(struct msgdma_extended_desc *desc,
+			       dma_addr_t dst, dma_addr_t src, size_t len,
+			       u32 stride)
+{
+	/* Set lower 32bits of src & dst addresses in the descriptor */
+	desc->read_addr_lo = lower_32_bits(src);
+	desc->write_addr_lo = lower_32_bits(dst);
+
+	/* Set upper 32bits of src & dst addresses in the descriptor */
+	desc->read_addr_hi = upper_32_bits(src);
+	desc->write_addr_hi = upper_32_bits(dst);
+
+	desc->len = len;
+	desc->stride = stride;
+	desc->burst_seq_num = 0;	/* 0 will result in max burst length */
+
+	/*
+	 * Don't set interrupt on xfer end yet, this will be done later
+	 * for the "last" descriptor
+	 */
+	desc->control = MSGDMA_DESC_CTL_TR_ERR_IRQ | MSGDMA_DESC_CTL_GO |
+		MSGDMA_DESC_CTL_END_ON_LEN;
+}
+
+/**
+ * msgdma_desc_config_eod - Mark the descriptor as end descriptor
+ * @desc: Hw descriptor pointer
+ */
+static void msgdma_desc_config_eod(struct msgdma_extended_desc *desc)
+{
+	desc->control |= MSGDMA_DESC_CTL_TR_COMP_IRQ;
+}
+
+/**
+ * msgdma_tx_submit - Submit DMA transaction
+ * @tx: Async transaction descriptor pointer
+ *
+ * Return: cookie value
+ */
+static dma_cookie_t msgdma_tx_submit(struct dma_async_tx_descriptor *tx)
+{
+	struct msgdma_device *mdev = to_mdev(tx->chan);
+	struct msgdma_sw_desc *new;
+	dma_cookie_t cookie;
+	unsigned long flags;
+
+	new = tx_to_desc(tx);
+	spin_lock_irqsave(&mdev->lock, flags);
+	cookie = dma_cookie_assign(tx);
+
+	list_add_tail(&new->node, &mdev->pending_list);
+	spin_unlock_irqrestore(&mdev->lock, flags);
+
+	return cookie;
+}
+
+/**
+ * msgdma_prep_memcpy - prepare descriptors for memcpy transaction
+ * @dchan: DMA channel
+ * @dma_dst: Destination buffer address
+ * @dma_src: Source buffer address
+ * @len: Transfer length
+ * @flags: transfer ack flags
+ *
+ * Return: Async transaction descriptor on success and NULL on failure
+ */
+static struct dma_async_tx_descriptor *
+msgdma_prep_memcpy(struct dma_chan *dchan, dma_addr_t dma_dst,
+		   dma_addr_t dma_src, size_t len, ulong flags)
+{
+	struct msgdma_device *mdev = to_mdev(dchan);
+	struct msgdma_sw_desc *new, *first = NULL;
+	struct msgdma_extended_desc *desc;
+	size_t copy;
+	u32 desc_cnt;
+	unsigned long irqflags;
+
+	desc_cnt = DIV_ROUND_UP(len, MSGDMA_MAX_TRANS_LEN);
+
+	spin_lock_irqsave(&mdev->lock, irqflags);
+	if (desc_cnt > mdev->desc_free_cnt) {
+		spin_unlock_irqrestore(&mdev->lock, irqflags);
+		dev_dbg(mdev->dev, "mdev %p descs are not available\n", mdev);
+		return NULL;
+	}
+	mdev->desc_free_cnt -= desc_cnt;
+	spin_unlock_irqrestore(&mdev->lock, irqflags);
+
+	do {
+		/* Allocate and populate the descriptor */
+		new = msgdma_get_descriptor(mdev);
+
+		copy = min_t(size_t, len, MSGDMA_MAX_TRANS_LEN);
+		desc = &new->hw_desc;
+		msgdma_desc_config(desc, dma_dst, dma_src, copy,
+				   MSGDMA_DESC_STRIDE_RW);
+		len -= copy;
+		dma_src += copy;
+		dma_dst += copy;
+		if (!first)
+			first = new;
+		else
+			list_add_tail(&new->node, &first->tx_list);
+	} while (len);
+
+	msgdma_desc_config_eod(desc);
+	async_tx_ack(&first->async_tx);
+	first->async_tx.flags = flags;
+
+	return &first->async_tx;
+}
+
+/**
+ * msgdma_prep_slave_sg - prepare descriptors for a slave sg transaction
+ *
+ * @dchan: DMA channel
+ * @sgl: Destination scatter list
+ * @sg_len: Number of entries in destination scatter list
+ * @dir: DMA transfer direction
+ * @flags: transfer ack flags
+ * @context: transfer context (unused)
+ */
+static struct dma_async_tx_descriptor *
+msgdma_prep_slave_sg(struct dma_chan *dchan, struct scatterlist *sgl,
+		     unsigned int sg_len, enum dma_transfer_direction dir,
+		     unsigned long flags, void *context)
+
+{
+	struct msgdma_device *mdev = to_mdev(dchan);
+	struct dma_slave_config *cfg = &mdev->slave_cfg;
+	struct msgdma_sw_desc *new, *first = NULL;
+	void *desc = NULL;
+	size_t len, avail;
+	dma_addr_t dma_dst, dma_src;
+	u32 desc_cnt;
+	u32 stride;
+	unsigned long irqflags;
+
+	desc_cnt = sg_nents_for_dma(sgl, sg_len, MSGDMA_MAX_TRANS_LEN);
+
+	spin_lock_irqsave(&mdev->lock, irqflags);
+	if (desc_cnt > mdev->desc_free_cnt) {
+		spin_unlock_irqrestore(&mdev->lock, irqflags);
+		dev_dbg(mdev->dev, "mdev %p descs are not available\n", mdev);
+		return NULL;
+	}
+	mdev->desc_free_cnt -= desc_cnt;
+	spin_unlock_irqrestore(&mdev->lock, irqflags);
+
+	avail = sg_dma_len(sgl);
+
+	/* Run until we are out of scatterlist entries */
+	while (true) {
+		/* Allocate and populate the descriptor */
+		new = msgdma_get_descriptor(mdev);
+
+		desc = &new->hw_desc;
+		len = min_t(size_t, avail, MSGDMA_MAX_TRANS_LEN);
+
+		if (dir == DMA_MEM_TO_DEV) {
+			dma_src = sg_dma_address(sgl) + sg_dma_len(sgl) - avail;
+			dma_dst = cfg->dst_addr;
+			stride = MSGDMA_DESC_STRIDE_RD;
+		} else {
+			dma_src = cfg->src_addr;
+			dma_dst = sg_dma_address(sgl) + sg_dma_len(sgl) - avail;
+			stride = MSGDMA_DESC_STRIDE_WR;
+		}
+		msgdma_desc_config(desc, dma_dst, dma_src, len, stride);
+		avail -= len;
+
+		if (!first)
+			first = new;
+		else
+			list_add_tail(&new->node, &first->tx_list);
+
+		/* Fetch the next scatterlist entry */
+		if (avail == 0) {
+			if (sg_len == 0)
+				break;
+			sgl = sg_next(sgl);
+			if (sgl == NULL)
+				break;
+			sg_len--;
+			avail = sg_dma_len(sgl);
+		}
+	}
+
+	msgdma_desc_config_eod(desc);
+	first->async_tx.flags = flags;
+
+	return &first->async_tx;
+}
+
+static int msgdma_dma_config(struct dma_chan *dchan,
+			     struct dma_slave_config *config)
+{
+	struct msgdma_device *mdev = to_mdev(dchan);
+
+	memcpy(&mdev->slave_cfg, config, sizeof(*config));
+
+	return 0;
+}
+
+static void msgdma_reset(struct msgdma_device *mdev)
+{
+	u32 val;
+	int ret;
+
+	/* Reset mSGDMA */
+	iowrite32(MSGDMA_CSR_STAT_MASK, mdev->csr + MSGDMA_CSR_STATUS);
+	iowrite32(MSGDMA_CSR_CTL_RESET, mdev->csr + MSGDMA_CSR_CONTROL);
+
+	ret = readl_poll_timeout(mdev->csr + MSGDMA_CSR_STATUS, val,
+				 (val & MSGDMA_CSR_STAT_RESETTING) == 0,
+				 1, 10000);
+	if (ret)
+		dev_err(mdev->dev, "DMA channel did not reset\n");
+
+	/* Clear all status bits */
+	iowrite32(MSGDMA_CSR_STAT_MASK, mdev->csr + MSGDMA_CSR_STATUS);
+
+	/* Enable the DMA controller including interrupts */
+	iowrite32(MSGDMA_CSR_CTL_STOP_ON_ERR | MSGDMA_CSR_CTL_STOP_ON_EARLY |
+		  MSGDMA_CSR_CTL_GLOBAL_INTR, mdev->csr + MSGDMA_CSR_CONTROL);
+
+	mdev->idle = true;
+};
+
+static void msgdma_copy_one(struct msgdma_device *mdev,
+			    struct msgdma_sw_desc *desc)
+{
+	void __iomem *hw_desc = mdev->desc;
+
+	/* Ensure control is the last field — required for correct FIFO flush ordering */
+	static_assert(offsetof(struct msgdma_extended_desc, control) ==
+		      sizeof(struct msgdma_extended_desc) - sizeof(u32),
+		      "control must be the last field in msgdma_extended_desc");
+
+	/*
+	 * Check if the DESC FIFO it not full. If its full, we need to wait
+	 * for at least one entry to become free again
+	 */
+	while (ioread32(mdev->csr + MSGDMA_CSR_STATUS) &
+	       MSGDMA_CSR_STAT_DESC_BUF_FULL)
+		mdelay(1);
+
+	/* Ensure control is the last field — required for correct FIFO flush ordering */
+	static_assert(offsetof(struct msgdma_extended_desc, control) ==
+			sizeof(struct msgdma_extended_desc) - sizeof(u32),
+			"control must be the last field in msgdma_extended_desc");
+
+	/*
+	 * Copy the descriptor into the descriptor FIFO of the DMA controller,
+	 * excluding the control word. The FIFO is flushed and the descriptor
+	 * becomes valid once the control word is written last.
+	 */
+	memcpy_toio(hw_desc, &desc->hw_desc,
+		    offsetof(struct msgdma_extended_desc, control));
+
+	/* Write control word last to flush this descriptor into the FIFO */
+	mdev->idle = false;
+	wmb();
+	iowrite32(desc->hw_desc.control, hw_desc +
+		  offsetof(struct msgdma_extended_desc, control));
+	wmb();
+}
+
+/**
+ * msgdma_copy_desc_to_fifo - copy descriptor(s) into controller FIFO
+ * @mdev: Pointer to the Altera mSGDMA device structure
+ * @desc: Transaction descriptor pointer
+ */
+static void msgdma_copy_desc_to_fifo(struct msgdma_device *mdev,
+				     struct msgdma_sw_desc *desc)
+{
+	struct msgdma_sw_desc *sdesc, *next;
+
+	msgdma_copy_one(mdev, desc);
+
+	list_for_each_entry_safe(sdesc, next, &desc->tx_list, node)
+		msgdma_copy_one(mdev, sdesc);
+}
+
+/**
+ * msgdma_start_transfer - Initiate the new transfer
+ * @mdev: Pointer to the Altera mSGDMA device structure
+ */
+static void msgdma_start_transfer(struct msgdma_device *mdev)
+{
+	struct msgdma_sw_desc *desc;
+
+	if (!mdev->idle)
+		return;
+
+	desc = list_first_entry_or_null(&mdev->pending_list,
+					struct msgdma_sw_desc, node);
+	if (!desc)
+		return;
+
+	list_splice_tail_init(&mdev->pending_list, &mdev->active_list);
+	msgdma_copy_desc_to_fifo(mdev, desc);
+}
+
+/**
+ * msgdma_issue_pending - Issue pending transactions
+ * @chan: DMA channel pointer
+ */
+static void msgdma_issue_pending(struct dma_chan *chan)
+{
+	struct msgdma_device *mdev = to_mdev(chan);
+	unsigned long flags;
+
+	spin_lock_irqsave(&mdev->lock, flags);
+	msgdma_start_transfer(mdev);
+	spin_unlock_irqrestore(&mdev->lock, flags);
+}
+
+/**
+ * msgdma_chan_desc_cleanup - Cleanup the completed descriptors
+ * @mdev: Pointer to the Altera mSGDMA device structure
+ */
+static void msgdma_chan_desc_cleanup(struct msgdma_device *mdev)
+{
+	struct msgdma_sw_desc *desc, *next;
+	unsigned long irqflags;
+
+	spin_lock_irqsave(&mdev->lock, irqflags);
+
+	list_for_each_entry_safe(desc, next, &mdev->done_list, node) {
+		struct dmaengine_desc_callback cb;
+
+		dmaengine_desc_get_callback(&desc->async_tx, &cb);
+		if (dmaengine_desc_callback_valid(&cb)) {
+			spin_unlock_irqrestore(&mdev->lock, irqflags);
+			dmaengine_desc_callback_invoke(&cb, NULL);
+			spin_lock_irqsave(&mdev->lock, irqflags);
+		}
+
+		/* Run any dependencies, then free the descriptor */
+		msgdma_free_descriptor(mdev, desc);
+	}
+
+	spin_unlock_irqrestore(&mdev->lock, irqflags);
+}
+
+/**
+ * msgdma_complete_descriptor - Mark the active descriptor as complete
+ * @mdev: Pointer to the Altera mSGDMA device structure
+ */
+static void msgdma_complete_descriptor(struct msgdma_device *mdev)
+{
+	struct msgdma_sw_desc *desc;
+
+	desc = list_first_entry_or_null(&mdev->active_list,
+					struct msgdma_sw_desc, node);
+	if (!desc)
+		return;
+	list_del(&desc->node);
+	dma_cookie_complete(&desc->async_tx);
+	list_add_tail(&desc->node, &mdev->done_list);
+}
+
+/**
+ * msgdma_free_descriptors - Free channel descriptors
+ * @mdev: Pointer to the Altera mSGDMA device structure
+ */
+static void msgdma_free_descriptors(struct msgdma_device *mdev)
+{
+	msgdma_free_desc_list(mdev, &mdev->active_list);
+	msgdma_free_desc_list(mdev, &mdev->pending_list);
+	msgdma_free_desc_list(mdev, &mdev->done_list);
+}
+
+/**
+ * msgdma_free_chan_resources - Free channel resources
+ * @dchan: DMA channel pointer
+ */
+static void msgdma_free_chan_resources(struct dma_chan *dchan)
+{
+	struct msgdma_device *mdev = to_mdev(dchan);
+	unsigned long flags;
+
+	spin_lock_irqsave(&mdev->lock, flags);
+	msgdma_free_descriptors(mdev);
+	spin_unlock_irqrestore(&mdev->lock, flags);
+	kfree(mdev->sw_desq);
+}
+
+/**
+ * msgdma_alloc_chan_resources - Allocate channel resources
+ * @dchan: DMA channel
+ *
+ * Return: Number of descriptors on success and failure value on error
+ */
+static int msgdma_alloc_chan_resources(struct dma_chan *dchan)
+{
+	struct msgdma_device *mdev = to_mdev(dchan);
+	struct msgdma_sw_desc *desc;
+	int i;
+
+	mdev->sw_desq = kzalloc_objs(*desc, MSGDMA_DESC_NUM, GFP_NOWAIT);
+	if (!mdev->sw_desq)
+		return -ENOMEM;
+
+	mdev->idle = true;
+	mdev->desc_free_cnt = MSGDMA_DESC_NUM;
+
+	INIT_LIST_HEAD(&mdev->free_list);
+
+	for (i = 0; i < MSGDMA_DESC_NUM; i++) {
+		desc = mdev->sw_desq + i;
+		dma_async_tx_descriptor_init(&desc->async_tx, &mdev->dmachan);
+		desc->async_tx.tx_submit = msgdma_tx_submit;
+		list_add_tail(&desc->node, &mdev->free_list);
+	}
+
+	return MSGDMA_DESC_NUM;
+}
+
+/**
+ * msgdma_tasklet - Schedule completion tasklet
+ * @t: Pointer to the Altera sSGDMA channel structure
+ */
+static void msgdma_tasklet(struct tasklet_struct *t)
+{
+	struct msgdma_device *mdev = from_tasklet(mdev, t, irq_tasklet);
+	u32 count;
+	u32 __maybe_unused size;
+	u32 __maybe_unused status;
+	unsigned long flags;
+
+	spin_lock_irqsave(&mdev->lock, flags);
+
+	if (mdev->resp) {
+		/* Read number of responses that are available */
+		count = ioread32(mdev->csr + MSGDMA_CSR_RESP_FILL_LEVEL);
+		dev_dbg(mdev->dev, "%s (%d): response count=%d\n",
+			__func__, __LINE__, count);
+	} else {
+		count = 1;
+	}
+
+	while (count--) {
+		/*
+		 * Read both longwords to purge this response from the FIFO
+		 * On Avalon-MM implementations, size and status do not
+		 * have any real values, like transferred bytes or error
+		 * bits. So we need to just drop these values.
+		 */
+		if (mdev->resp) {
+			size = ioread32(mdev->resp +
+					MSGDMA_RESP_BYTES_TRANSFERRED);
+			status = ioread32(mdev->resp +
+					MSGDMA_RESP_STATUS);
+		}
+
+		msgdma_complete_descriptor(mdev);
+	}
+
+	spin_unlock_irqrestore(&mdev->lock, flags);
+
+	msgdma_chan_desc_cleanup(mdev);
+}
+
+/**
+ * msgdma_irq_handler - Altera mSGDMA Interrupt handler
+ * @irq: IRQ number
+ * @data: Pointer to the Altera mSGDMA device structure
+ *
+ * Return: IRQ_HANDLED/IRQ_NONE
+ */
+static irqreturn_t msgdma_irq_handler(int irq, void *data)
+{
+	struct msgdma_device *mdev = data;
+	u32 status;
+
+	status = ioread32(mdev->csr + MSGDMA_CSR_STATUS);
+	if ((status & MSGDMA_CSR_STAT_BUSY) == 0) {
+		/* Start next transfer if the DMA controller is idle */
+		spin_lock(&mdev->lock);
+		mdev->idle = true;
+		msgdma_start_transfer(mdev);
+		spin_unlock(&mdev->lock);
+	}
+
+	tasklet_schedule(&mdev->irq_tasklet);
+
+	/* Clear interrupt in mSGDMA controller */
+	iowrite32(MSGDMA_CSR_STAT_IRQ, mdev->csr + MSGDMA_CSR_STATUS);
+
+	return IRQ_HANDLED;
+}
+
+/**
+ * msgdma_dev_remove() - Device remove function
+ * @mdev: Pointer to the Altera mSGDMA device structure
+ */
+static void msgdma_dev_remove(struct msgdma_device *mdev)
+{
+	if (!mdev)
+		return;
+
+	devm_free_irq(mdev->dev, mdev->irq, mdev);
+	tasklet_kill(&mdev->irq_tasklet);
+	list_del(&mdev->dmachan.device_node);
+}
+
+static int request_and_map(struct platform_device *pdev, const char *name,
+			   struct resource **res, void __iomem **ptr,
+			   bool optional)
+{
+	struct resource *region;
+	struct device *device = &pdev->dev;
+
+	*res = platform_get_resource_byname(pdev, IORESOURCE_MEM, name);
+	if (*res == NULL) {
+		if (optional) {
+			*ptr = NULL;
+			dev_info(device, "optional resource %s not defined\n",
+				 name);
+			return 0;
+		}
+		dev_err(device, "mandatory resource %s not defined\n", name);
+		return -ENODEV;
+	}
+
+	region = devm_request_mem_region(device, (*res)->start,
+					 resource_size(*res), dev_name(device));
+	if (region == NULL) {
+		dev_err(device, "unable to request %s\n", name);
+		return -EBUSY;
+	}
+
+	*ptr = devm_ioremap(device, region->start,
+				    resource_size(region));
+	if (*ptr == NULL) {
+		dev_err(device, "ioremap of %s failed!", name);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+/**
+ * msgdma_probe - Driver probe function
+ * @pdev: Pointer to the platform_device structure
+ *
+ * Return: '0' on success and failure value on error
+ */
+static int msgdma_probe(struct platform_device *pdev)
+{
+	struct msgdma_device *mdev;
+	struct dma_device *dma_dev;
+	struct resource *dma_res;
+	int ret;
+
+	mdev = devm_kzalloc(&pdev->dev, sizeof(*mdev), GFP_NOWAIT);
+	if (!mdev)
+		return -ENOMEM;
+
+	mdev->dev = &pdev->dev;
+
+	/* Map CSR space */
+	ret = request_and_map(pdev, "csr", &dma_res, &mdev->csr, false);
+	if (ret)
+		return ret;
+
+	/* Map (extended) descriptor space */
+	ret = request_and_map(pdev, "desc", &dma_res, &mdev->desc, false);
+	if (ret)
+		return ret;
+
+	/* Map response space */
+	ret = request_and_map(pdev, "resp", &dma_res, &mdev->resp, true);
+	if (ret)
+		return ret;
+
+	platform_set_drvdata(pdev, mdev);
+
+	/* Get interrupt nr from platform data */
+	mdev->irq = platform_get_irq(pdev, 0);
+	if (mdev->irq < 0)
+		return -ENXIO;
+
+	ret = devm_request_irq(&pdev->dev, mdev->irq, msgdma_irq_handler,
+			       0, dev_name(&pdev->dev), mdev);
+	if (ret)
+		return ret;
+
+	tasklet_setup(&mdev->irq_tasklet, msgdma_tasklet);
+
+	dma_cookie_init(&mdev->dmachan);
+
+	spin_lock_init(&mdev->lock);
+
+	INIT_LIST_HEAD(&mdev->active_list);
+	INIT_LIST_HEAD(&mdev->pending_list);
+	INIT_LIST_HEAD(&mdev->done_list);
+	INIT_LIST_HEAD(&mdev->free_list);
+
+	dma_dev = &mdev->dmadev;
+
+	/* Set DMA capabilities */
+	dma_cap_zero(dma_dev->cap_mask);
+	dma_cap_set(DMA_MEMCPY, dma_dev->cap_mask);
+	dma_cap_set(DMA_SLAVE, dma_dev->cap_mask);
+
+	dma_dev->src_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+	dma_dev->dst_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+	dma_dev->directions = BIT(DMA_MEM_TO_DEV) | BIT(DMA_DEV_TO_MEM) |
+		BIT(DMA_MEM_TO_MEM);
+	dma_dev->residue_granularity = DMA_RESIDUE_GRANULARITY_DESCRIPTOR;
+
+	/* Init DMA link list */
+	INIT_LIST_HEAD(&dma_dev->channels);
+
+	/* Set base routines */
+	dma_dev->device_tx_status = dma_cookie_status;
+	dma_dev->device_issue_pending = msgdma_issue_pending;
+	dma_dev->dev = &pdev->dev;
+
+	dma_dev->copy_align = DMAENGINE_ALIGN_4_BYTES;
+	dma_dev->device_prep_dma_memcpy = msgdma_prep_memcpy;
+	dma_dev->device_prep_slave_sg = msgdma_prep_slave_sg;
+	dma_dev->device_config = msgdma_dma_config;
+
+	dma_dev->device_alloc_chan_resources = msgdma_alloc_chan_resources;
+	dma_dev->device_free_chan_resources = msgdma_free_chan_resources;
+
+	mdev->dmachan.device = dma_dev;
+	list_add_tail(&mdev->dmachan.device_node, &dma_dev->channels);
+
+	/* Set DMA mask to 64 bits */
+	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (ret) {
+		dev_warn(&pdev->dev, "unable to set coherent mask to 64");
+		goto fail;
+	}
+
+	msgdma_reset(mdev);
+
+	ret = dma_async_device_register(dma_dev);
+	if (ret)
+		goto fail;
+
+	ret = of_dma_controller_register(pdev->dev.of_node,
+					 of_dma_xlate_by_chan_id, dma_dev);
+	if (ret == -EINVAL)
+		dev_warn(&pdev->dev, "device was not probed from DT");
+	else if (ret && ret != -ENODEV)
+		goto fail;
+
+	dev_notice(&pdev->dev, "Altera mSGDMA driver probe success\n");
+
+	return 0;
+
+fail:
+	msgdma_dev_remove(mdev);
+
+	return ret;
+}
+
+/**
+ * msgdma_remove() - Driver remove function
+ * @pdev: Pointer to the platform_device structure
+ *
+ * Return: Always '0'
+ */
+static void msgdma_remove(struct platform_device *pdev)
+{
+	struct msgdma_device *mdev = platform_get_drvdata(pdev);
+
+	if (pdev->dev.of_node)
+		of_dma_controller_free(pdev->dev.of_node);
+	dma_async_device_unregister(&mdev->dmadev);
+	msgdma_dev_remove(mdev);
+
+	dev_notice(&pdev->dev, "Altera mSGDMA driver removed\n");
+}
+
+#ifdef CONFIG_OF
+static const struct of_device_id msgdma_match[] = {
+	{ .compatible = "altr,socfpga-msgdma", },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(of, msgdma_match);
+#endif
+
+static struct platform_driver msgdma_driver = {
+	.driver = {
+		.name = "altera-msgdma",
+		.of_match_table = of_match_ptr(msgdma_match),
+	},
+	.probe = msgdma_probe,
+	.remove = msgdma_remove,
+};
+
+module_platform_driver(msgdma_driver);
+
+MODULE_ALIAS("platform:altera-msgdma");
+MODULE_DESCRIPTION("Altera mSGDMA driver");
+MODULE_AUTHOR("Stefan Roese <sr@denx.de>");
+MODULE_LICENSE("GPL");