[PATCH v2 5.4 0/3] dm table: fix iterate_devices based device

[PATCH v2 5.4 0/3] dm table: fix iterate_devices based device

[Jeffle Xu]

Backport notes are detailed in the corresponding commit log.

changes since v1:
- fix build error of patch 1
- fix upstream commit id of patch 2

Jeffle Xu (3):
  dm table: fix iterate_devices based device capability checks
  dm table: fix DAX iterate_devices based device capability checks
  dm table: fix zoned iterate_devices based device capability checks

 drivers/md/dm-table.c | 174 ++++++++++++++++++------------------------
 drivers/md/dm.c       |   2 +-
 drivers/md/dm.h       |   2 +-
 3 files changed, 75 insertions(+), 103 deletions(-)

-- 
2.27.0

[PATCH v2 5.4 1/3] dm table: fix iterate_devices based device capability checks

[Jeffle Xu]

commit a4c8dd9c2d0987cf542a2a0c42684c9c6d78a04e upstream.

According to the definition of dm_iterate_devices_fn:
 * This function must iterate through each section of device used by the
 * target until it encounters a non-zero return code, which it then returns.
 * Returns zero if no callout returned non-zero.

For some target type (e.g. dm-stripe), one call of iterate_devices() may
iterate multiple underlying devices internally, in which case a non-zero
return code returned by iterate_devices_callout_fn will stop the iteration
in advance. No iterate_devices_callout_fn should return non-zero unless
device iteration should stop.

Rename dm_table_requires_stable_pages() to dm_table_any_dev_attr() and
elevate it for reuse to stop iterating (and return non-zero) on the
first device that causes iterate_devices_callout_fn to return non-zero.
Use dm_table_any_dev_attr() to properly iterate through devices.

Rename device_is_nonrot() to device_is_rotational() and invert logic
accordingly to fix improper disposition.

[jeffle: backport notes]
Also convert the partial completion capability check, which is
introduced by commit 22c11858e800 ("dm: introduce
DM_TYPE_NVME_BIO_BASED"), and removed since commit 9c37de297f65 ("dm:
remove special-casing of bio-based immutable singleton target on NVMe")
in v5.10.

Fixes: c3c4555edd10 ("dm table: clear add_random unless all devices have it set")
Fixes: 4693c9668fdc ("dm table: propagate non rotational flag")
Cc: stable@vger.kernel.org
Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
---
 drivers/md/dm-table.c | 103 ++++++++++++++++++++++--------------------
 1 file changed, 54 insertions(+), 49 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 98be040cf958..c470e174e686 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -1376,6 +1376,46 @@ struct dm_target *dm_table_find_target(struct dm_table *t, sector_t sector)
 	return &t->targets[(KEYS_PER_NODE * n) + k];
 }
 
+/*
+ * type->iterate_devices() should be called when the sanity check needs to
+ * iterate and check all underlying data devices. iterate_devices() will
+ * iterate all underlying data devices until it encounters a non-zero return
+ * code, returned by whether the input iterate_devices_callout_fn, or
+ * iterate_devices() itself internally.
+ *
+ * For some target type (e.g. dm-stripe), one call of iterate_devices() may
+ * iterate multiple underlying devices internally, in which case a non-zero
+ * return code returned by iterate_devices_callout_fn will stop the iteration
+ * in advance.
+ *
+ * Cases requiring _any_ underlying device supporting some kind of attribute,
+ * should use the iteration structure like dm_table_any_dev_attr(), or call
+ * it directly. @func should handle semantics of positive examples, e.g.
+ * capable of something.
+ *
+ * Cases requiring _all_ underlying devices supporting some kind of attribute,
+ * should use the iteration structure like dm_table_supports_nowait() or
+ * dm_table_supports_discards(). Or introduce dm_table_all_devs_attr() that
+ * uses an @anti_func that handle semantics of counter examples, e.g. not
+ * capable of something. So: return !dm_table_any_dev_attr(t, anti_func);
+ */
+static bool dm_table_any_dev_attr(struct dm_table *t,
+				  iterate_devices_callout_fn func)
+{
+	struct dm_target *ti;
+	unsigned int i;
+
+	for (i = 0; i < dm_table_get_num_targets(t); i++) {
+		ti = dm_table_get_target(t, i);
+
+		if (ti->type->iterate_devices &&
+		    ti->type->iterate_devices(ti, func, NULL))
+			return true;
+        }
+
+	return false;
+}
+
 static int count_device(struct dm_target *ti, struct dm_dev *dev,
 			sector_t start, sector_t len, void *data)
 {
@@ -1692,12 +1732,12 @@ static int dm_table_supports_dax_write_cache(struct dm_table *t)
 	return false;
 }
 
-static int device_is_nonrot(struct dm_target *ti, struct dm_dev *dev,
-			    sector_t start, sector_t len, void *data)
+static int device_is_rotational(struct dm_target *ti, struct dm_dev *dev,
+				sector_t start, sector_t len, void *data)
 {
 	struct request_queue *q = bdev_get_queue(dev->bdev);
 
-	return q && blk_queue_nonrot(q);
+	return q && !blk_queue_nonrot(q);
 }
 
 static int device_is_not_random(struct dm_target *ti, struct dm_dev *dev,
@@ -1708,35 +1748,18 @@ static int device_is_not_random(struct dm_target *ti, struct dm_dev *dev,
 	return q && !blk_queue_add_random(q);
 }
 
-static bool dm_table_all_devices_attribute(struct dm_table *t,
-					   iterate_devices_callout_fn func)
-{
-	struct dm_target *ti;
-	unsigned i;
-
-	for (i = 0; i < dm_table_get_num_targets(t); i++) {
-		ti = dm_table_get_target(t, i);
-
-		if (!ti->type->iterate_devices ||
-		    !ti->type->iterate_devices(ti, func, NULL))
-			return false;
-	}
-
-	return true;
-}
-
-static int device_no_partial_completion(struct dm_target *ti, struct dm_dev *dev,
+static int device_is_partial_completion(struct dm_target *ti, struct dm_dev *dev,
 					sector_t start, sector_t len, void *data)
 {
 	char b[BDEVNAME_SIZE];
 
 	/* For now, NVMe devices are the only devices of this class */
-	return (strncmp(bdevname(dev->bdev, b), "nvme", 4) == 0);
+	return (strncmp(bdevname(dev->bdev, b), "nvme", 4) != 0);
 }
 
 static bool dm_table_does_not_support_partial_completion(struct dm_table *t)
 {
-	return dm_table_all_devices_attribute(t, device_no_partial_completion);
+	return !dm_table_any_dev_attr(t, device_is_partial_completion);
 }
 
 static int device_not_write_same_capable(struct dm_target *ti, struct dm_dev *dev,
@@ -1863,27 +1886,6 @@ static int device_requires_stable_pages(struct dm_target *ti,
 	return q && bdi_cap_stable_pages_required(q->backing_dev_info);
 }
 
-/*
- * If any underlying device requires stable pages, a table must require
- * them as well.  Only targets that support iterate_devices are considered:
- * don't want error, zero, etc to require stable pages.
- */
-static bool dm_table_requires_stable_pages(struct dm_table *t)
-{
-	struct dm_target *ti;
-	unsigned i;
-
-	for (i = 0; i < dm_table_get_num_targets(t); i++) {
-		ti = dm_table_get_target(t, i);
-
-		if (ti->type->iterate_devices &&
-		    ti->type->iterate_devices(ti, device_requires_stable_pages, NULL))
-			return true;
-	}
-
-	return false;
-}
-
 void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 			       struct queue_limits *limits)
 {
@@ -1928,10 +1930,10 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 		dax_write_cache(t->md->dax_dev, true);
 
 	/* Ensure that all underlying devices are non-rotational. */
-	if (dm_table_all_devices_attribute(t, device_is_nonrot))
-		blk_queue_flag_set(QUEUE_FLAG_NONROT, q);
-	else
+	if (dm_table_any_dev_attr(t, device_is_rotational))
 		blk_queue_flag_clear(QUEUE_FLAG_NONROT, q);
+	else
+		blk_queue_flag_set(QUEUE_FLAG_NONROT, q);
 
 	if (!dm_table_supports_write_same(t))
 		q->limits.max_write_same_sectors = 0;
@@ -1943,8 +1945,11 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 	/*
 	 * Some devices don't use blk_integrity but still want stable pages
 	 * because they do their own checksumming.
+	 * If any underlying device requires stable pages, a table must require
+	 * them as well.  Only targets that support iterate_devices are considered:
+	 * don't want error, zero, etc to require stable pages.
 	 */
-	if (dm_table_requires_stable_pages(t))
+	if (dm_table_any_dev_attr(t, device_requires_stable_pages))
 		q->backing_dev_info->capabilities |= BDI_CAP_STABLE_WRITES;
 	else
 		q->backing_dev_info->capabilities &= ~BDI_CAP_STABLE_WRITES;
@@ -1955,7 +1960,7 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 	 * Clear QUEUE_FLAG_ADD_RANDOM if any underlying device does not
 	 * have it set.
 	 */
-	if (blk_queue_add_random(q) && dm_table_all_devices_attribute(t, device_is_not_random))
+	if (blk_queue_add_random(q) && dm_table_any_dev_attr(t, device_is_not_random))
 		blk_queue_flag_clear(QUEUE_FLAG_ADD_RANDOM, q);
 
 	/*
-- 
2.27.0

[PATCH v2 5.4 2/3] dm table: fix DAX iterate_devices based device capability checks

[Jeffle Xu]

commit 5b0fab508992c2e120971da658ce80027acbc405 upstream.

Fix dm_table_supports_dax() and invert logic of both
iterate_devices_callout_fn so that all devices' DAX capabilities are
properly checked.

Fixes: 545ed20e6df6 ("dm: add infrastructure for DAX support")
Cc: stable@vger.kernel.org
Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
---
 drivers/md/dm-table.c | 37 ++++++++++---------------------------
 drivers/md/dm.c       |  2 +-
 drivers/md/dm.h       |  2 +-
 3 files changed, 12 insertions(+), 29 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index c470e174e686..07530f2aa027 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -888,24 +888,24 @@ void dm_table_set_type(struct dm_table *t, enum dm_queue_mode type)
 EXPORT_SYMBOL_GPL(dm_table_set_type);
 
 /* validate the dax capability of the target device span */
-int device_supports_dax(struct dm_target *ti, struct dm_dev *dev,
+int device_not_dax_capable(struct dm_target *ti, struct dm_dev *dev,
 			sector_t start, sector_t len, void *data)
 {
 	int blocksize = *(int *) data, id;
 	bool rc;
 
 	id = dax_read_lock();
-	rc = dax_supported(dev->dax_dev, dev->bdev, blocksize, start, len);
+	rc = !dax_supported(dev->dax_dev, dev->bdev, blocksize, start, len);
 	dax_read_unlock(id);
 
 	return rc;
 }
 
 /* Check devices support synchronous DAX */
-static int device_dax_synchronous(struct dm_target *ti, struct dm_dev *dev,
-				  sector_t start, sector_t len, void *data)
+static int device_not_dax_synchronous_capable(struct dm_target *ti, struct dm_dev *dev,
+					      sector_t start, sector_t len, void *data)
 {
-	return dev->dax_dev && dax_synchronous(dev->dax_dev);
+	return !dev->dax_dev || !dax_synchronous(dev->dax_dev);
 }
 
 bool dm_table_supports_dax(struct dm_table *t,
@@ -922,7 +922,7 @@ bool dm_table_supports_dax(struct dm_table *t,
 			return false;
 
 		if (!ti->type->iterate_devices ||
-		    !ti->type->iterate_devices(ti, iterate_fn, blocksize))
+		    ti->type->iterate_devices(ti, iterate_fn, blocksize))
 			return false;
 	}
 
@@ -996,7 +996,7 @@ static int dm_table_determine_type(struct dm_table *t)
 verify_bio_based:
 		/* We must use this table as bio-based */
 		t->type = DM_TYPE_BIO_BASED;
-		if (dm_table_supports_dax(t, device_supports_dax, &page_size) ||
+		if (dm_table_supports_dax(t, device_not_dax_capable, &page_size) ||
 		    (list_empty(devices) && live_md_type == DM_TYPE_DAX_BIO_BASED)) {
 			t->type = DM_TYPE_DAX_BIO_BASED;
 		} else {
@@ -1715,23 +1715,6 @@ static int device_dax_write_cache_enabled(struct dm_target *ti,
 	return false;
 }
 
-static int dm_table_supports_dax_write_cache(struct dm_table *t)
-{
-	struct dm_target *ti;
-	unsigned i;
-
-	for (i = 0; i < dm_table_get_num_targets(t); i++) {
-		ti = dm_table_get_target(t, i);
-
-		if (ti->type->iterate_devices &&
-		    ti->type->iterate_devices(ti,
-				device_dax_write_cache_enabled, NULL))
-			return true;
-	}
-
-	return false;
-}
-
 static int device_is_rotational(struct dm_target *ti, struct dm_dev *dev,
 				sector_t start, sector_t len, void *data)
 {
@@ -1918,15 +1901,15 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 	}
 	blk_queue_write_cache(q, wc, fua);
 
-	if (dm_table_supports_dax(t, device_supports_dax, &page_size)) {
+	if (dm_table_supports_dax(t, device_not_dax_capable, &page_size)) {
 		blk_queue_flag_set(QUEUE_FLAG_DAX, q);
-		if (dm_table_supports_dax(t, device_dax_synchronous, NULL))
+		if (dm_table_supports_dax(t, device_not_dax_synchronous_capable, NULL))
 			set_dax_synchronous(t->md->dax_dev);
 	}
 	else
 		blk_queue_flag_clear(QUEUE_FLAG_DAX, q);
 
-	if (dm_table_supports_dax_write_cache(t))
+	if (dm_table_any_dev_attr(t, device_dax_write_cache_enabled))
 		dax_write_cache(t->md->dax_dev, true);
 
 	/* Ensure that all underlying devices are non-rotational. */
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index de32f8553735..530c0fe14229 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1139,7 +1139,7 @@ static bool dm_dax_supported(struct dax_device *dax_dev, struct block_device *bd
 	if (!map)
 		goto out;
 
-	ret = dm_table_supports_dax(map, device_supports_dax, &blocksize);
+	ret = dm_table_supports_dax(map, device_not_dax_capable, &blocksize);
 
 out:
 	dm_put_live_table(md, srcu_idx);
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index d7c4f6606b5f..9fbf87e04019 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -74,7 +74,7 @@ void dm_table_free_md_mempools(struct dm_table *t);
 struct dm_md_mempools *dm_table_get_md_mempools(struct dm_table *t);
 bool dm_table_supports_dax(struct dm_table *t, iterate_devices_callout_fn fn,
 			   int *blocksize);
-int device_supports_dax(struct dm_target *ti, struct dm_dev *dev,
+int device_not_dax_capable(struct dm_target *ti, struct dm_dev *dev,
 			   sector_t start, sector_t len, void *data);
 
 void dm_lock_md_type(struct mapped_device *md);
-- 
2.27.0

[PATCH v2 5.4 3/3] dm table: fix zoned iterate_devices based device capability checks

[Jeffle Xu]

commit 24f6b6036c9eec21191646930ad42808e6180510 upstream.

Fix dm_table_supports_zoned_model() and invert logic of both
iterate_devices_callout_fn so that all devices' zoned capabilities are
properly checked.

Add one more parameter to dm_table_any_dev_attr(), which is actually
used as the @data parameter of iterate_devices_callout_fn, so that
dm_table_matches_zone_sectors() can be replaced by
dm_table_any_dev_attr().

Fixes: dd88d313bef02 ("dm table: add zoned block devices validation")
Cc: stable@vger.kernel.org
Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
[jeffle: also convert partial completion check]
---
 drivers/md/dm-table.c | 50 +++++++++++++++----------------------------
 1 file changed, 17 insertions(+), 33 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 07530f2aa027..06b382304d92 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -1397,10 +1397,10 @@ struct dm_target *dm_table_find_target(struct dm_table *t, sector_t sector)
  * should use the iteration structure like dm_table_supports_nowait() or
  * dm_table_supports_discards(). Or introduce dm_table_all_devs_attr() that
  * uses an @anti_func that handle semantics of counter examples, e.g. not
- * capable of something. So: return !dm_table_any_dev_attr(t, anti_func);
+ * capable of something. So: return !dm_table_any_dev_attr(t, anti_func, data);
  */
 static bool dm_table_any_dev_attr(struct dm_table *t,
-				  iterate_devices_callout_fn func)
+				  iterate_devices_callout_fn func, void *data)
 {
 	struct dm_target *ti;
 	unsigned int i;
@@ -1409,7 +1409,7 @@ static bool dm_table_any_dev_attr(struct dm_table *t,
 		ti = dm_table_get_target(t, i);
 
 		if (ti->type->iterate_devices &&
-		    ti->type->iterate_devices(ti, func, NULL))
+		    ti->type->iterate_devices(ti, func, data))
 			return true;
         }
 
@@ -1452,13 +1452,13 @@ bool dm_table_has_no_data_devices(struct dm_table *table)
 	return true;
 }
 
-static int device_is_zoned_model(struct dm_target *ti, struct dm_dev *dev,
-				 sector_t start, sector_t len, void *data)
+static int device_not_zoned_model(struct dm_target *ti, struct dm_dev *dev,
+				  sector_t start, sector_t len, void *data)
 {
 	struct request_queue *q = bdev_get_queue(dev->bdev);
 	enum blk_zoned_model *zoned_model = data;
 
-	return q && blk_queue_zoned_model(q) == *zoned_model;
+	return !q || blk_queue_zoned_model(q) != *zoned_model;
 }
 
 static bool dm_table_supports_zoned_model(struct dm_table *t,
@@ -1475,37 +1475,20 @@ static bool dm_table_supports_zoned_model(struct dm_table *t,
 			return false;
 
 		if (!ti->type->iterate_devices ||
-		    !ti->type->iterate_devices(ti, device_is_zoned_model, &zoned_model))
+		    ti->type->iterate_devices(ti, device_not_zoned_model, &zoned_model))
 			return false;
 	}
 
 	return true;
 }
 
-static int device_matches_zone_sectors(struct dm_target *ti, struct dm_dev *dev,
-				       sector_t start, sector_t len, void *data)
+static int device_not_matches_zone_sectors(struct dm_target *ti, struct dm_dev *dev,
+					   sector_t start, sector_t len, void *data)
 {
 	struct request_queue *q = bdev_get_queue(dev->bdev);
 	unsigned int *zone_sectors = data;
 
-	return q && blk_queue_zone_sectors(q) == *zone_sectors;
-}
-
-static bool dm_table_matches_zone_sectors(struct dm_table *t,
-					  unsigned int zone_sectors)
-{
-	struct dm_target *ti;
-	unsigned i;
-
-	for (i = 0; i < dm_table_get_num_targets(t); i++) {
-		ti = dm_table_get_target(t, i);
-
-		if (!ti->type->iterate_devices ||
-		    !ti->type->iterate_devices(ti, device_matches_zone_sectors, &zone_sectors))
-			return false;
-	}
-
-	return true;
+	return !q || blk_queue_zone_sectors(q) != *zone_sectors;
 }
 
 static int validate_hardware_zoned_model(struct dm_table *table,
@@ -1525,7 +1508,7 @@ static int validate_hardware_zoned_model(struct dm_table *table,
 	if (!zone_sectors || !is_power_of_2(zone_sectors))
 		return -EINVAL;
 
-	if (!dm_table_matches_zone_sectors(table, zone_sectors)) {
+	if (dm_table_any_dev_attr(table, device_not_matches_zone_sectors, &zone_sectors)) {
 		DMERR("%s: zone sectors is not consistent across all devices",
 		      dm_device_name(table->md));
 		return -EINVAL;
@@ -1742,7 +1725,7 @@ static int device_is_partial_completion(struct dm_target *ti, struct dm_dev *dev
 
 static bool dm_table_does_not_support_partial_completion(struct dm_table *t)
 {
-	return !dm_table_any_dev_attr(t, device_is_partial_completion);
+	return !dm_table_any_dev_attr(t, device_is_partial_completion, NULL);
 }
 
 static int device_not_write_same_capable(struct dm_target *ti, struct dm_dev *dev,
@@ -1909,11 +1892,11 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 	else
 		blk_queue_flag_clear(QUEUE_FLAG_DAX, q);
 
-	if (dm_table_any_dev_attr(t, device_dax_write_cache_enabled))
+	if (dm_table_any_dev_attr(t, device_dax_write_cache_enabled, NULL))
 		dax_write_cache(t->md->dax_dev, true);
 
 	/* Ensure that all underlying devices are non-rotational. */
-	if (dm_table_any_dev_attr(t, device_is_rotational))
+	if (dm_table_any_dev_attr(t, device_is_rotational, NULL))
 		blk_queue_flag_clear(QUEUE_FLAG_NONROT, q);
 	else
 		blk_queue_flag_set(QUEUE_FLAG_NONROT, q);
@@ -1932,7 +1915,7 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 	 * them as well.  Only targets that support iterate_devices are considered:
 	 * don't want error, zero, etc to require stable pages.
 	 */
-	if (dm_table_any_dev_attr(t, device_requires_stable_pages))
+	if (dm_table_any_dev_attr(t, device_requires_stable_pages, NULL))
 		q->backing_dev_info->capabilities |= BDI_CAP_STABLE_WRITES;
 	else
 		q->backing_dev_info->capabilities &= ~BDI_CAP_STABLE_WRITES;
@@ -1943,7 +1926,8 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 	 * Clear QUEUE_FLAG_ADD_RANDOM if any underlying device does not
 	 * have it set.
 	 */
-	if (blk_queue_add_random(q) && dm_table_any_dev_attr(t, device_is_not_random))
+	if (blk_queue_add_random(q) &&
+	    dm_table_any_dev_attr(t, device_is_not_random, NULL))
 		blk_queue_flag_clear(QUEUE_FLAG_ADD_RANDOM, q);
 
 	/*
-- 
2.27.0