virtio_blk: set the zone write granularitygrafted

virtblk_read_zoned_limits() reads the write granularity that the device
reports in virtio_blk_zoned_characteristics and assigns it to the
physical block size and to io_min, but never to the limit that is named
after it. queue_limits.zone_write_granularity is left at zero, so
blk_validate_zoned_limits() raises it to the logical block size:

	if (lim->zone_write_granularity < lim->logical_block_size)
		lim->zone_write_granularity = lim->logical_block_size;

A device that reports a granularity coarser than its logical block size,
which is what the field exists to express, therefore has it silently
reduced. A 512e host managed disk passed through to a guest reports a
logical block size of 512 and a write granularity of 4096, and the guest
ends up with a zone write granularity of 512.

bio_split_alignment() returns lim->zone_write_granularity if it is non-zero
and bio_split_io_at() may split a bio with as per bio_split_alignment().
This can real to the write getting rejected by the host drive, as the write
is not aligned to the physical block size.

zonefs also takes its block size from bdev_zone_write_granularity(), so it
would incorrectly use 512 on a disk that requires 4096.

sd_zbc_read_zones() sets the limit from the physical block size for the
same reason. NVMe ZNS and null_blk leave it unset, but the fallback
gives the right answer for them, as their write granularity is the
logical block size. virtio carries a separate value that may exceed it.

Set the zone write granularity from the value that the device reports.

Fixes: 95bfec41bd3d ("virtio-blk: add support for zoned block devices")
Signed-off-by: Niklas Cassel <cassel@kernel.org>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Link: https://patch.msgid.link/20260918140641.2031075-2-cassel@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

1 files changed, 651 insertions, 0 deletions

diff --git a/drivers/acpi/device_sysfs.c b/drivers/acpi/device_sysfs.c
new file mode 100644
index 000000000..c5434b998
--- /dev/null
+++ b/drivers/acpi/device_sysfs.c
@@ -0,0 +1,651 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * drivers/acpi/device_sysfs.c - ACPI device sysfs attributes and modalias.
+ *
+ * Copyright (C) 2015, Intel Corp.
+ * Author: Mika Westerberg <mika.westerberg@linux.intel.com>
+ * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+
+#include <linux/acpi.h>
+#include <linux/device.h>
+#include <linux/export.h>
+#include <linux/nls.h>
+
+#include "internal.h"
+
+static ssize_t acpi_object_path(acpi_handle handle, char *buf)
+{
+	struct acpi_buffer path = {ACPI_ALLOCATE_BUFFER, NULL};
+	int result;
+
+	result = acpi_get_name(handle, ACPI_FULL_PATHNAME, &path);
+	if (result)
+		return result;
+
+	result = sysfs_emit(buf, "%s\n", (char *)path.pointer);
+	kfree(path.pointer);
+	return result;
+}
+
+struct acpi_data_node_attr {
+	struct attribute attr;
+	ssize_t (*show)(struct acpi_data_node *, char *);
+	ssize_t (*store)(struct acpi_data_node *, const char *, size_t count);
+};
+
+#define DATA_NODE_ATTR(_name)			\
+	static struct acpi_data_node_attr data_node_##_name =	\
+		__ATTR(_name, 0444, data_node_show_##_name, NULL)
+
+static ssize_t data_node_show_path(struct acpi_data_node *dn, char *buf)
+{
+	return dn->handle ? acpi_object_path(dn->handle, buf) : 0;
+}
+
+DATA_NODE_ATTR(path);
+
+static struct attribute *acpi_data_node_default_attrs[] = {
+	&data_node_path.attr,
+	NULL
+};
+ATTRIBUTE_GROUPS(acpi_data_node_default);
+
+#define to_data_node(k) container_of(k, struct acpi_data_node, kobj)
+#define to_attr(a) container_of(a, struct acpi_data_node_attr, attr)
+
+static ssize_t acpi_data_node_attr_show(struct kobject *kobj,
+					struct attribute *attr, char *buf)
+{
+	struct acpi_data_node *dn = to_data_node(kobj);
+	struct acpi_data_node_attr *dn_attr = to_attr(attr);
+
+	return dn_attr->show ? dn_attr->show(dn, buf) : -ENXIO;
+}
+
+static const struct sysfs_ops acpi_data_node_sysfs_ops = {
+	.show	= acpi_data_node_attr_show,
+};
+
+static void acpi_data_node_release(struct kobject *kobj)
+{
+	struct acpi_data_node *dn = to_data_node(kobj);
+
+	complete(&dn->kobj_done);
+}
+
+static const struct kobj_type acpi_data_node_ktype = {
+	.sysfs_ops = &acpi_data_node_sysfs_ops,
+	.default_groups = acpi_data_node_default_groups,
+	.release = acpi_data_node_release,
+};
+
+static void acpi_expose_nondev_subnodes(struct kobject *kobj,
+					struct acpi_device_data *data)
+{
+	struct list_head *list = &data->subnodes;
+	struct acpi_data_node *dn;
+
+	if (list_empty(list))
+		return;
+
+	list_for_each_entry(dn, list, sibling) {
+		int ret;
+
+		init_completion(&dn->kobj_done);
+		ret = kobject_init_and_add(&dn->kobj, &acpi_data_node_ktype,
+					   kobj, "%s", dn->name);
+		if (!ret)
+			acpi_expose_nondev_subnodes(&dn->kobj, &dn->data);
+		else if (dn->handle)
+			acpi_handle_err(dn->handle, "Failed to expose (%d)\n", ret);
+	}
+}
+
+static void acpi_hide_nondev_subnodes(struct acpi_device_data *data)
+{
+	struct list_head *list = &data->subnodes;
+	struct acpi_data_node *dn;
+
+	if (list_empty(list))
+		return;
+
+	list_for_each_entry_reverse(dn, list, sibling) {
+		acpi_hide_nondev_subnodes(&dn->data);
+		kobject_put(&dn->kobj);
+	}
+}
+
+/**
+ * create_pnp_modalias - Create hid/cid(s) string for modalias and uevent
+ * @acpi_dev: ACPI device object.
+ * @modalias: Buffer to print into.
+ * @size: Size of the buffer.
+ *
+ * Creates hid/cid(s) string needed for modalias and uevent
+ * e.g. on a device with hid:IBM0001 and cid:ACPI0001 you get:
+ * char *modalias: "acpi:IBM0001:ACPI0001"
+ * Return: 0: no _HID and no _CID
+ *         -EINVAL: output error
+ *         -ENOMEM: output is truncated
+ */
+static int create_pnp_modalias(const struct acpi_device *acpi_dev, char *modalias,
+			       int size)
+{
+	int len;
+	int count;
+	struct acpi_hardware_id *id;
+
+	/* Avoid unnecessarily loading modules for non present devices. */
+	if (!acpi_device_is_present(acpi_dev))
+		return 0;
+
+	/*
+	 * Since we skip ACPI_DT_NAMESPACE_HID from the modalias below, 0 should
+	 * be returned if ACPI_DT_NAMESPACE_HID is the only ACPI/PNP ID in the
+	 * device's list.
+	 */
+	count = 0;
+	list_for_each_entry(id, &acpi_dev->pnp.ids, list)
+		if (strcmp(id->id, ACPI_DT_NAMESPACE_HID))
+			count++;
+
+	if (!count)
+		return 0;
+
+	len = snprintf(modalias, size, "acpi:");
+	if (len >= size)
+		return -ENOMEM;
+
+	size -= len;
+
+	list_for_each_entry(id, &acpi_dev->pnp.ids, list) {
+		if (!strcmp(id->id, ACPI_DT_NAMESPACE_HID))
+			continue;
+
+		count = snprintf(&modalias[len], size, "%s:", id->id);
+
+		if (count >= size)
+			return -ENOMEM;
+
+		len += count;
+		size -= count;
+	}
+
+	return len;
+}
+
+/**
+ * create_of_modalias - Creates DT compatible string for modalias and uevent
+ * @acpi_dev: ACPI device object.
+ * @modalias: Buffer to print into.
+ * @size: Size of the buffer.
+ *
+ * Expose DT compatible modalias as of:NnameTCcompatible.  This function should
+ * only be called for devices having ACPI_DT_NAMESPACE_HID in their list of
+ * ACPI/PNP IDs.
+ */
+static int create_of_modalias(const struct acpi_device *acpi_dev, char *modalias,
+			      int size)
+{
+	struct acpi_buffer buf = { ACPI_ALLOCATE_BUFFER };
+	const union acpi_object *of_compatible, *obj;
+	acpi_status status;
+	int len, count;
+	int i, nval;
+	char *c;
+
+	status = acpi_get_name(acpi_dev->handle, ACPI_SINGLE_NAME, &buf);
+	if (ACPI_FAILURE(status))
+		return -ENODEV;
+
+	/* DT strings are all in lower case */
+	for (c = buf.pointer; *c != '\0'; c++)
+		*c = tolower(*c);
+
+	len = snprintf(modalias, size, "of:N%sT", (char *)buf.pointer);
+	ACPI_FREE(buf.pointer);
+
+	if (len >= size)
+		return -ENOMEM;
+
+	size -= len;
+
+	of_compatible = acpi_dev->data.of_compatible;
+	if (of_compatible->type == ACPI_TYPE_PACKAGE) {
+		nval = of_compatible->package.count;
+		obj = of_compatible->package.elements;
+	} else { /* Must be ACPI_TYPE_STRING. */
+		nval = 1;
+		obj = of_compatible;
+	}
+	for (i = 0; i < nval; i++, obj++) {
+		count = snprintf(&modalias[len], size, "C%s",
+				 obj->string.pointer);
+
+		if (count >= size)
+			return -ENOMEM;
+
+		len += count;
+		size -= count;
+	}
+
+	return len;
+}
+
+int __acpi_device_uevent_modalias(const struct acpi_device *adev,
+				  struct kobj_uevent_env *env)
+{
+	int len;
+
+	if (!adev)
+		return -ENODEV;
+
+	if (list_empty(&adev->pnp.ids))
+		return 0;
+
+	if (add_uevent_var(env, "MODALIAS="))
+		return -ENOMEM;
+
+	if (adev->data.of_compatible)
+		len = create_of_modalias(adev, &env->buf[env->buflen - 1],
+					 sizeof(env->buf) - env->buflen);
+	else
+		len = create_pnp_modalias(adev, &env->buf[env->buflen - 1],
+					  sizeof(env->buf) - env->buflen);
+	if (len < 0)
+		return len;
+
+	env->buflen += len;
+
+	return 0;
+}
+
+/**
+ * acpi_device_uevent_modalias - uevent modalias for ACPI-enumerated devices.
+ * @dev: Struct device to get ACPI device node.
+ * @env: Environment variables of the kobject uevent.
+ *
+ * Create the uevent modalias field for ACPI-enumerated devices.
+ *
+ * Because other buses do not support ACPI HIDs & CIDs, e.g. for a device with
+ * hid:IBM0001 and cid:ACPI0001 you get: "acpi:IBM0001:ACPI0001".
+ */
+int acpi_device_uevent_modalias(const struct device *dev, struct kobj_uevent_env *env)
+{
+	return __acpi_device_uevent_modalias(acpi_companion_match(dev), env);
+}
+EXPORT_SYMBOL_GPL(acpi_device_uevent_modalias);
+
+static int __acpi_device_modalias(const struct acpi_device *adev, char *buf, int size)
+{
+	int len, count;
+
+	if (!adev)
+		return -ENODEV;
+
+	if (list_empty(&adev->pnp.ids))
+		return 0;
+
+	len = create_pnp_modalias(adev, buf, size - 1);
+	if (len < 0) {
+		return len;
+	} else if (len > 0) {
+		buf[len++] = '\n';
+		size -= len;
+	}
+	if (!adev->data.of_compatible)
+		return len;
+
+	count = create_of_modalias(adev, buf + len, size - 1);
+	if (count < 0) {
+		return count;
+	} else if (count > 0) {
+		len += count;
+		buf[len++] = '\n';
+	}
+
+	return len;
+}
+
+/**
+ * acpi_device_modalias - modalias sysfs attribute for ACPI-enumerated devices.
+ * @dev: Struct device to get ACPI device node.
+ * @buf: The buffer to save pnp_modalias and of_modalias.
+ * @size: Size of buffer.
+ *
+ * Create the modalias sysfs attribute for ACPI-enumerated devices.
+ *
+ * Because other buses do not support ACPI HIDs & CIDs, e.g. for a device with
+ * hid:IBM0001 and cid:ACPI0001 you get: "acpi:IBM0001:ACPI0001".
+ */
+int acpi_device_modalias(struct device *dev, char *buf, int size)
+{
+	return __acpi_device_modalias(acpi_companion_match(dev), buf, size);
+}
+EXPORT_SYMBOL_GPL(acpi_device_modalias);
+
+static ssize_t
+modalias_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	return __acpi_device_modalias(to_acpi_device(dev), buf, 1024);
+}
+static DEVICE_ATTR_RO(modalias);
+
+static ssize_t real_power_state_show(struct device *dev,
+				     struct device_attribute *attr, char *buf)
+{
+	struct acpi_device *adev = to_acpi_device(dev);
+	int state;
+	int ret;
+
+	ret = acpi_device_get_power(adev, &state);
+	if (ret)
+		return ret;
+
+	return sysfs_emit(buf, "%s\n", acpi_power_state_string(state));
+}
+
+static DEVICE_ATTR_RO(real_power_state);
+
+static ssize_t power_state_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct acpi_device *adev = to_acpi_device(dev);
+
+	return sysfs_emit(buf, "%s\n", acpi_power_state_string(adev->power.state));
+}
+
+static DEVICE_ATTR_RO(power_state);
+
+static ssize_t
+eject_store(struct device *d, struct device_attribute *attr,
+	    const char *buf, size_t count)
+{
+	struct acpi_device *acpi_device = to_acpi_device(d);
+	acpi_object_type not_used;
+	acpi_status status;
+
+	if (!count || buf[0] != '1')
+		return -EINVAL;
+
+	if ((!acpi_device->handler || !acpi_device->handler->hotplug.enabled)
+	    && !d->driver)
+		return -ENODEV;
+
+	status = acpi_get_type(acpi_device->handle, &not_used);
+	if (ACPI_FAILURE(status) || !acpi_device->flags.ejectable)
+		return -ENODEV;
+
+	acpi_dev_get(acpi_device);
+	status = acpi_hotplug_schedule(acpi_device, ACPI_OST_EC_OSPM_EJECT);
+	if (ACPI_SUCCESS(status))
+		return count;
+
+	acpi_dev_put(acpi_device);
+	acpi_evaluate_ost(acpi_device->handle, ACPI_OST_EC_OSPM_EJECT,
+			  ACPI_OST_SC_NON_SPECIFIC_FAILURE, NULL);
+	return status == AE_NO_MEMORY ? -ENOMEM : -EAGAIN;
+}
+
+static DEVICE_ATTR_WO(eject);
+
+static ssize_t
+hid_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct acpi_device *acpi_dev = to_acpi_device(dev);
+
+	return sysfs_emit(buf, "%s\n", acpi_device_hid(acpi_dev));
+}
+static DEVICE_ATTR_RO(hid);
+
+static ssize_t cid_show(struct device *dev, struct device_attribute *attr,
+			char *buf)
+{
+	struct acpi_device *acpi_dev = to_acpi_device(dev);
+	struct acpi_device_info *info = NULL;
+	ssize_t len = 0;
+
+	acpi_get_object_info(acpi_dev->handle, &info);
+	if (!info)
+		return 0;
+
+	if (info->valid & ACPI_VALID_CID) {
+		struct acpi_pnp_device_id_list *cid_list = &info->compatible_id_list;
+		int i;
+
+		for (i = 0; i < cid_list->count - 1; i++)
+			len += sysfs_emit_at(buf, len, "%s,", cid_list->ids[i].string);
+
+		len += sysfs_emit_at(buf, len, "%s\n", cid_list->ids[i].string);
+	}
+
+	kfree(info);
+
+	return len;
+}
+static DEVICE_ATTR_RO(cid);
+
+static ssize_t uid_show(struct device *dev,
+			struct device_attribute *attr, char *buf)
+{
+	struct acpi_device *acpi_dev = to_acpi_device(dev);
+
+	return sysfs_emit(buf, "%s\n", acpi_device_uid(acpi_dev));
+}
+static DEVICE_ATTR_RO(uid);
+
+static ssize_t adr_show(struct device *dev,
+			struct device_attribute *attr, char *buf)
+{
+	struct acpi_device *acpi_dev = to_acpi_device(dev);
+
+	if (acpi_dev->pnp.bus_address > U32_MAX)
+		return sysfs_emit(buf, "0x%016llx\n", acpi_dev->pnp.bus_address);
+	else
+		return sysfs_emit(buf, "0x%08llx\n", acpi_dev->pnp.bus_address);
+}
+static DEVICE_ATTR_RO(adr);
+
+static ssize_t path_show(struct device *dev,
+			 struct device_attribute *attr, char *buf)
+{
+	struct acpi_device *acpi_dev = to_acpi_device(dev);
+
+	return acpi_object_path(acpi_dev->handle, buf);
+}
+static DEVICE_ATTR_RO(path);
+
+/* sysfs file that shows description text from the ACPI _STR method */
+static ssize_t description_show(struct device *dev,
+				struct device_attribute *attr,
+				char *buf)
+{
+	struct acpi_device *acpi_dev = to_acpi_device(dev);
+	struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL};
+	union acpi_object *str_obj;
+	acpi_status status;
+	int result;
+
+	status = acpi_evaluate_object_typed(acpi_dev->handle, "_STR",
+					    NULL, &buffer,
+					    ACPI_TYPE_BUFFER);
+	if (ACPI_FAILURE(status))
+		return -EIO;
+
+	str_obj = buffer.pointer;
+
+	/*
+	 * The _STR object contains a Unicode identifier for a device.
+	 * We need to convert to utf-8 so it can be displayed.
+	 */
+	result = utf16s_to_utf8s(
+		(wchar_t *)str_obj->buffer.pointer,
+		str_obj->buffer.length,
+		UTF16_LITTLE_ENDIAN, buf,
+		PAGE_SIZE - 1);
+
+	buf[result++] = '\n';
+
+	ACPI_FREE(str_obj);
+
+	return result;
+}
+static DEVICE_ATTR_RO(description);
+
+static ssize_t
+sun_show(struct device *dev, struct device_attribute *attr,
+	 char *buf)
+{
+	struct acpi_device *acpi_dev = to_acpi_device(dev);
+	acpi_status status;
+	unsigned long long sun;
+
+	status = acpi_evaluate_integer(acpi_dev->handle, "_SUN", NULL, &sun);
+	if (ACPI_FAILURE(status))
+		return -EIO;
+
+	return sysfs_emit(buf, "%llu\n", sun);
+}
+static DEVICE_ATTR_RO(sun);
+
+static ssize_t
+hrv_show(struct device *dev, struct device_attribute *attr,
+	 char *buf)
+{
+	struct acpi_device *acpi_dev = to_acpi_device(dev);
+	acpi_status status;
+	unsigned long long hrv;
+
+	status = acpi_evaluate_integer(acpi_dev->handle, "_HRV", NULL, &hrv);
+	if (ACPI_FAILURE(status))
+		return -EIO;
+
+	return sysfs_emit(buf, "%llu\n", hrv);
+}
+static DEVICE_ATTR_RO(hrv);
+
+static ssize_t status_show(struct device *dev, struct device_attribute *attr,
+				char *buf)
+{
+	struct acpi_device *acpi_dev = to_acpi_device(dev);
+	acpi_status status;
+	unsigned long long sta;
+
+	status = acpi_evaluate_integer(acpi_dev->handle, "_STA", NULL, &sta);
+	if (ACPI_FAILURE(status))
+		return -EIO;
+
+	return sysfs_emit(buf, "%llu\n", sta);
+}
+static DEVICE_ATTR_RO(status);
+
+static struct attribute *acpi_attrs[] = {
+	&dev_attr_path.attr,
+	&dev_attr_hid.attr,
+	&dev_attr_cid.attr,
+	&dev_attr_modalias.attr,
+	&dev_attr_description.attr,
+	&dev_attr_adr.attr,
+	&dev_attr_uid.attr,
+	&dev_attr_sun.attr,
+	&dev_attr_hrv.attr,
+	&dev_attr_status.attr,
+	&dev_attr_eject.attr,
+	&dev_attr_power_state.attr,
+	&dev_attr_real_power_state.attr,
+	NULL
+};
+
+static bool acpi_show_attr(struct acpi_device *dev, const struct device_attribute *attr)
+{
+	/*
+	 * Devices gotten from FADT don't have a "path" attribute
+	 */
+	if (attr == &dev_attr_path)
+		return dev->handle;
+
+	if (attr == &dev_attr_hid || attr == &dev_attr_modalias)
+		return !list_empty(&dev->pnp.ids);
+
+	if (attr == &dev_attr_description)
+		return acpi_has_method(dev->handle, "_STR");
+
+	if (attr == &dev_attr_adr)
+		return dev->pnp.type.bus_address;
+
+	if (attr == &dev_attr_uid)
+		return acpi_device_uid(dev);
+
+	if (attr == &dev_attr_sun)
+		return acpi_has_method(dev->handle, "_SUN");
+
+	if (attr == &dev_attr_hrv)
+		return acpi_has_method(dev->handle, "_HRV");
+
+	if (attr == &dev_attr_status)
+		return acpi_has_method(dev->handle, "_STA");
+
+	if (attr == &dev_attr_cid)
+		return acpi_has_method(dev->handle, "_CID");
+
+	/*
+	 * If device has _EJ0, 'eject' file is created that is used to trigger
+	 * hot-removal function from userland.
+	 */
+	if (attr == &dev_attr_eject)
+		return acpi_has_method(dev->handle, "_EJ0");
+
+	if (attr == &dev_attr_power_state)
+		return dev->flags.power_manageable;
+
+	if (attr == &dev_attr_real_power_state)
+		return dev->flags.power_manageable && dev->power.flags.power_resources;
+
+	dev_warn_once(&dev->dev, "Unexpected attribute: %s\n", attr->attr.name);
+	return false;
+}
+
+static umode_t acpi_attr_is_visible(struct kobject *kobj,
+				    struct attribute *attr,
+				    int attrno)
+{
+	struct acpi_device *dev = to_acpi_device(kobj_to_dev(kobj));
+
+	if (acpi_show_attr(dev, container_of(attr, struct device_attribute, attr)))
+		return attr->mode;
+	else
+		return 0;
+}
+
+static const struct attribute_group acpi_group = {
+	.attrs = acpi_attrs,
+	.is_visible = acpi_attr_is_visible,
+};
+
+const struct attribute_group *acpi_groups[] = {
+	&acpi_group,
+	NULL
+};
+
+/**
+ * acpi_device_setup_files - Create sysfs attributes of an ACPI device.
+ * @dev: ACPI device object.
+ */
+void acpi_device_setup_files(struct acpi_device *dev)
+{
+	acpi_expose_nondev_subnodes(&dev->dev.kobj, &dev->data);
+}
+
+/**
+ * acpi_device_remove_files - Remove sysfs attributes of an ACPI device.
+ * @dev: ACPI device object.
+ */
+void acpi_device_remove_files(struct acpi_device *dev)
+{
+	acpi_hide_nondev_subnodes(&dev->data);
+}