From 7bf7eac8d648057519adb6fce1e31458c902212c Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Thu, 16 May 2019 13:26:29 -0700 Subject: dax: Arrange for dax_supported check to span multiple devices Pankaj reports that starting with commit ad428cdb525a "dax: Check the end of the block-device capacity with dax_direct_access()" device-mapper no longer allows dax operation. This results from the stricter checks in __bdev_dax_supported() that validate that the start and end of a block-device map to the same 'pagemap' instance. Teach the dax-core and device-mapper to validate the 'pagemap' on a per-target basis. This is accomplished by refactoring the bdev_dax_supported() internals into generic_fsdax_supported() which takes a sector range to validate. Consequently generic_fsdax_supported() is suitable to be used in a device-mapper ->iterate_devices() callback. A new ->dax_supported() operation is added to allow composite devices to split and route upper-level bdev_dax_supported() requests. Fixes: ad428cdb525a ("dax: Check the end of the block-device...") Cc: Cc: Ira Weiny Cc: Dave Jiang Cc: Keith Busch Cc: Matthew Wilcox Cc: Vishal Verma Cc: Heiko Carstens Cc: Martin Schwidefsky Reviewed-by: Jan Kara Reported-by: Pankaj Gupta Reviewed-by: Pankaj Gupta Tested-by: Pankaj Gupta Tested-by: Vaibhav Jain Reviewed-by: Mike Snitzer Signed-off-by: Dan Williams --- drivers/dax/super.c | 88 ++++++++++++++++++++++++++++---------------- drivers/md/dm-table.c | 17 ++++++--- drivers/md/dm.c | 20 ++++++++++ drivers/md/dm.h | 1 + drivers/nvdimm/pmem.c | 1 + drivers/s390/block/dcssblk.c | 1 + 6 files changed, 91 insertions(+), 37 deletions(-) (limited to 'drivers') diff --git a/drivers/dax/super.c b/drivers/dax/super.c index bbd57ca0634a..3a7b0a0bf469 100644 --- a/drivers/dax/super.c +++ b/drivers/dax/super.c @@ -73,22 +73,12 @@ struct dax_device *fs_dax_get_by_bdev(struct block_device *bdev) EXPORT_SYMBOL_GPL(fs_dax_get_by_bdev); #endif -/** - * __bdev_dax_supported() - Check if the device supports dax for filesystem - * @bdev: block device to check - * @blocksize: The block size of the device - * - * This is a library function for filesystems to check if the block device - * can be mounted with dax option. - * - * Return: true if supported, false if unsupported - */ -bool __bdev_dax_supported(struct block_device *bdev, int blocksize) +bool __generic_fsdax_supported(struct dax_device *dax_dev, + struct block_device *bdev, int blocksize, sector_t start, + sector_t sectors) { - struct dax_device *dax_dev; bool dax_enabled = false; pgoff_t pgoff, pgoff_end; - struct request_queue *q; char buf[BDEVNAME_SIZE]; void *kaddr, *end_kaddr; pfn_t pfn, end_pfn; @@ -102,21 +92,14 @@ bool __bdev_dax_supported(struct block_device *bdev, int blocksize) return false; } - q = bdev_get_queue(bdev); - if (!q || !blk_queue_dax(q)) { - pr_debug("%s: error: request queue doesn't support dax\n", - bdevname(bdev, buf)); - return false; - } - - err = bdev_dax_pgoff(bdev, 0, PAGE_SIZE, &pgoff); + err = bdev_dax_pgoff(bdev, start, PAGE_SIZE, &pgoff); if (err) { pr_debug("%s: error: unaligned partition for dax\n", bdevname(bdev, buf)); return false; } - last_page = PFN_DOWN(i_size_read(bdev->bd_inode) - 1) * 8; + last_page = PFN_DOWN((start + sectors - 1) * 512) * PAGE_SIZE / 512; err = bdev_dax_pgoff(bdev, last_page, PAGE_SIZE, &pgoff_end); if (err) { pr_debug("%s: error: unaligned partition for dax\n", @@ -124,20 +107,11 @@ bool __bdev_dax_supported(struct block_device *bdev, int blocksize) return false; } - dax_dev = dax_get_by_host(bdev->bd_disk->disk_name); - if (!dax_dev) { - pr_debug("%s: error: device does not support dax\n", - bdevname(bdev, buf)); - return false; - } - id = dax_read_lock(); len = dax_direct_access(dax_dev, pgoff, 1, &kaddr, &pfn); len2 = dax_direct_access(dax_dev, pgoff_end, 1, &end_kaddr, &end_pfn); dax_read_unlock(id); - put_dax(dax_dev); - if (len < 1 || len2 < 1) { pr_debug("%s: error: dax access failed (%ld)\n", bdevname(bdev, buf), len < 1 ? len : len2); @@ -178,6 +152,49 @@ bool __bdev_dax_supported(struct block_device *bdev, int blocksize) } return true; } +EXPORT_SYMBOL_GPL(__generic_fsdax_supported); + +/** + * __bdev_dax_supported() - Check if the device supports dax for filesystem + * @bdev: block device to check + * @blocksize: The block size of the device + * + * This is a library function for filesystems to check if the block device + * can be mounted with dax option. + * + * Return: true if supported, false if unsupported + */ +bool __bdev_dax_supported(struct block_device *bdev, int blocksize) +{ + struct dax_device *dax_dev; + struct request_queue *q; + char buf[BDEVNAME_SIZE]; + bool ret; + int id; + + q = bdev_get_queue(bdev); + if (!q || !blk_queue_dax(q)) { + pr_debug("%s: error: request queue doesn't support dax\n", + bdevname(bdev, buf)); + return false; + } + + dax_dev = dax_get_by_host(bdev->bd_disk->disk_name); + if (!dax_dev) { + pr_debug("%s: error: device does not support dax\n", + bdevname(bdev, buf)); + return false; + } + + id = dax_read_lock(); + ret = dax_supported(dax_dev, bdev, blocksize, 0, + i_size_read(bdev->bd_inode) / 512); + dax_read_unlock(id); + + put_dax(dax_dev); + + return ret; +} EXPORT_SYMBOL_GPL(__bdev_dax_supported); #endif @@ -303,6 +320,15 @@ long dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff, long nr_pages, } EXPORT_SYMBOL_GPL(dax_direct_access); +bool dax_supported(struct dax_device *dax_dev, struct block_device *bdev, + int blocksize, sector_t start, sector_t len) +{ + if (!dax_alive(dax_dev)) + return false; + + return dax_dev->ops->dax_supported(dax_dev, bdev, blocksize, start, len); +} + size_t dax_copy_from_iter(struct dax_device *dax_dev, pgoff_t pgoff, void *addr, size_t bytes, struct iov_iter *i) { diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index cde3b49b2a91..350cf0451456 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -880,13 +880,17 @@ 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 */ static int device_supports_dax(struct dm_target *ti, struct dm_dev *dev, - sector_t start, sector_t len, void *data) + sector_t start, sector_t len, void *data) { - return bdev_dax_supported(dev->bdev, PAGE_SIZE); + int blocksize = *(int *) data; + + return generic_fsdax_supported(dev->dax_dev, dev->bdev, blocksize, + start, len); } -static bool dm_table_supports_dax(struct dm_table *t) +bool dm_table_supports_dax(struct dm_table *t, int blocksize) { struct dm_target *ti; unsigned i; @@ -899,7 +903,8 @@ static bool dm_table_supports_dax(struct dm_table *t) return false; if (!ti->type->iterate_devices || - !ti->type->iterate_devices(ti, device_supports_dax, NULL)) + !ti->type->iterate_devices(ti, device_supports_dax, + &blocksize)) return false; } @@ -979,7 +984,7 @@ 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) || + if (dm_table_supports_dax(t, PAGE_SIZE) || (list_empty(devices) && live_md_type == DM_TYPE_DAX_BIO_BASED)) { t->type = DM_TYPE_DAX_BIO_BASED; } else { @@ -1905,7 +1910,7 @@ 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)) + if (dm_table_supports_dax(t, PAGE_SIZE)) blk_queue_flag_set(QUEUE_FLAG_DAX, q); else blk_queue_flag_clear(QUEUE_FLAG_DAX, q); diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 1fb1333fefec..b7c0ad01084d 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1107,6 +1107,25 @@ static long dm_dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff, return ret; } +static bool dm_dax_supported(struct dax_device *dax_dev, struct block_device *bdev, + int blocksize, sector_t start, sector_t len) +{ + struct mapped_device *md = dax_get_private(dax_dev); + struct dm_table *map; + int srcu_idx; + bool ret; + + map = dm_get_live_table(md, &srcu_idx); + if (!map) + return false; + + ret = dm_table_supports_dax(map, blocksize); + + dm_put_live_table(md, srcu_idx); + + return ret; +} + static size_t dm_dax_copy_from_iter(struct dax_device *dax_dev, pgoff_t pgoff, void *addr, size_t bytes, struct iov_iter *i) { @@ -3192,6 +3211,7 @@ static const struct block_device_operations dm_blk_dops = { static const struct dax_operations dm_dax_ops = { .direct_access = dm_dax_direct_access, + .dax_supported = dm_dax_supported, .copy_from_iter = dm_dax_copy_from_iter, .copy_to_iter = dm_dax_copy_to_iter, }; diff --git a/drivers/md/dm.h b/drivers/md/dm.h index 2d539b82ec08..17e3db54404c 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -72,6 +72,7 @@ bool dm_table_bio_based(struct dm_table *t); bool dm_table_request_based(struct dm_table *t); 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, int blocksize); void dm_lock_md_type(struct mapped_device *md); void dm_unlock_md_type(struct mapped_device *md); diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c index 0279eb1da3ef..845c5b430cdd 100644 --- a/drivers/nvdimm/pmem.c +++ b/drivers/nvdimm/pmem.c @@ -295,6 +295,7 @@ static size_t pmem_copy_to_iter(struct dax_device *dax_dev, pgoff_t pgoff, static const struct dax_operations pmem_dax_ops = { .direct_access = pmem_dax_direct_access, + .dax_supported = generic_fsdax_supported, .copy_from_iter = pmem_copy_from_iter, .copy_to_iter = pmem_copy_to_iter, }; diff --git a/drivers/s390/block/dcssblk.c b/drivers/s390/block/dcssblk.c index 4e8aedd50cb0..d04d4378ca50 100644 --- a/drivers/s390/block/dcssblk.c +++ b/drivers/s390/block/dcssblk.c @@ -59,6 +59,7 @@ static size_t dcssblk_dax_copy_to_iter(struct dax_device *dax_dev, static const struct dax_operations dcssblk_dax_ops = { .direct_access = dcssblk_dax_direct_access, + .dax_supported = generic_fsdax_supported, .copy_from_iter = dcssblk_dax_copy_from_iter, .copy_to_iter = dcssblk_dax_copy_to_iter, }; -- cgit v1.2.3-59-g8ed1b