diff options
44 files changed, 761 insertions, 484 deletions
diff --git a/Documentation/admin-guide/device-mapper/dm-raid.rst b/Documentation/admin-guide/device-mapper/dm-raid.rst index bb17e26e3c1b..e11f10764770 100644 --- a/Documentation/admin-guide/device-mapper/dm-raid.rst +++ b/Documentation/admin-guide/device-mapper/dm-raid.rst @@ -20,10 +20,10 @@ The target is named "raid" and it accepts the following parameters:: raid0 RAID0 striping (no resilience) raid1 RAID1 mirroring raid4 RAID4 with dedicated last parity disk - raid5_n RAID5 with dedicated last parity disk supporting takeover + raid5_n RAID5 with dedicated last parity disk supporting takeover from/to raid1 Same as raid4 - - Transitory layout + - Transitory layout for takeover from/to raid1 raid5_la RAID5 left asymmetric - rotating parity 0 with data continuation @@ -48,8 +48,8 @@ The target is named "raid" and it accepts the following parameters:: raid6_n_6 RAID6 with dedicate parity disks - parity and Q-syndrome on the last 2 disks; - layout for takeover from/to raid4/raid5_n - raid6_la_6 Same as "raid_la" plus dedicated last Q-syndrome disk + layout for takeover from/to raid0/raid4/raid5_n + raid6_la_6 Same as "raid_la" plus dedicated last Q-syndrome disk supporting takeover from/to raid5 - layout for takeover from raid5_la from/to raid6 raid6_ra_6 Same as "raid5_ra" dedicated last Q-syndrome disk @@ -173,9 +173,9 @@ The target is named "raid" and it accepts the following parameters:: The delta_disks option value (-251 < N < +251) triggers device removal (negative value) or device addition (positive value) to any reshape supporting raid levels 4/5/6 and 10. - RAID levels 4/5/6 allow for addition of devices (metadata - and data device tuple), raid10_near and raid10_offset only - allow for device addition. raid10_far does not support any + RAID levels 4/5/6 allow for addition and removal of devices + (metadata and data device tuple), raid10_near and raid10_offset + only allow for device addition. raid10_far does not support any reshaping at all. A minimum of devices have to be kept to enforce resilience, which is 3 devices for raid4/5 and 4 devices for raid6. @@ -372,6 +372,72 @@ to safely enable discard support for RAID 4/5/6: 'devices_handle_discards_safely' +Takeover/Reshape Support +------------------------ +The target natively supports these two types of MDRAID conversions: + +o Takeover: Converts an array from one RAID level to another + +o Reshape: Changes the internal layout while maintaining the current RAID level + +Each operation is only valid under specific constraints imposed by the existing array's layout and configuration. + + +Takeover: +linear -> raid1 with N >= 2 mirrors +raid0 -> raid4 (add dedicated parity device) +raid0 -> raid5 (add dedicated parity device) +raid0 -> raid10 with near layout and N >= 2 mirror groups (raid0 stripes have to become first member within mirror groups) +raid1 -> linear +raid1 -> raid5 with 2 mirrors +raid4 -> raid5 w/ rotating parity +raid5 with dedicated parity device -> raid4 +raid5 -> raid6 (with dedicated Q-syndrome) +raid6 (with dedicated Q-syndrome) -> raid5 +raid10 with near layout and even number of disks -> raid0 (select any in-sync device from each mirror group) + +Reshape: +linear: not possible +raid0: not possible +raid1: change number of mirrors +raid4: add and remove stripes (minimum 3), change stripesize +raid5: add and remove stripes (minimum 3, special case 2 for raid1 takeover), change rotating parity algorithms, change stripesize +raid6: add and remove stripes (minimum 4), change rotating syndrome algorithms, change stripesize +raid10 near: add stripes (minimum 4), change stripesize, no stripe removal possible, change to offset layout +raid10 offset: add stripes, change stripesize, no stripe removal possible, change to near layout +raid10 far: not possible + +Table line examples: + +### raid1 -> raid5 +# +# 2 devices limitation in raid1. +# raid5 personality is able to just map 2 like raid1. +# Reshape after takeover to change to full raid5 layout + + 0 1960886272 raid raid1 3 0 region_size 2048 2 /dev/dm-0 /dev/dm-1 /dev/dm-2 /dev/dm-3 + +# dm-0 and dm-2 are e.g. 4MiB large metadata devices, dm-1 and dm-3 have to be at least 1960886272 big. +# +# Table line to takeover to raid5 + + 0 1960886272 raid raid5 3 0 region_size 2048 2 /dev/dm-0 /dev/dm-1 /dev/dm-2 /dev/dm-3 + +# Add required out-of-place reshape space to the beginniong of the given 2 data devices, +# allocate another metadata/data device tuple with the same sizes for the parity space +# and zero the first 4K of the metadata device. +# +# Example table of the out-of-place reshape space addition for one data device, e.g. dm-1 + + 0 8192 linear 8:0 0 1960903888 # <- must be free space segment + 8192 1960886272 linear 8:0 0 2048 # previous data segment + +# Mapping table for e.g. raid5_rs reshape causing the size of the raid device to double-fold once the reshape finishes. +# Check the status output (e.g. "dmsetup status $RaidDev") for progess. + + 0 $((2 * 1960886272)) raid raid5 7 0 region_size 2048 data_offset 8192 delta_disk 1 2 /dev/dm-0 /dev/dm-1 /dev/dm-2 /dev/dm-3 + + Version History --------------- diff --git a/Documentation/admin-guide/device-mapper/verity.rst b/Documentation/admin-guide/device-mapper/verity.rst index 8c3f1f967a3c..3ecab1cff9c6 100644 --- a/Documentation/admin-guide/device-mapper/verity.rst +++ b/Documentation/admin-guide/device-mapper/verity.rst @@ -236,8 +236,10 @@ is available at the cryptsetup project's wiki page Status ====== -V (for Valid) is returned if every check performed so far was valid. -If any check failed, C (for Corruption) is returned. +1. V (for Valid) is returned if every check performed so far was valid. + If any check failed, C (for Corruption) is returned. +2. Number of corrected blocks by Forward Error Correction. + '-' if Forward Error Correction is not enabled. Example ======= diff --git a/MAINTAINERS b/MAINTAINERS index aff3e162180d..c9e416ba74c6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7225,6 +7225,7 @@ DEVICE-MAPPER (LVM) M: Alasdair Kergon <agk@redhat.com> M: Mike Snitzer <snitzer@kernel.org> M: Mikulas Patocka <mpatocka@redhat.com> +M: Benjamin Marzinski <bmarzins@redhat.com> L: dm-devel@lists.linux.dev S: Maintained Q: http://patchwork.kernel.org/project/dm-devel/list/ diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig index 104aa5355090..239c1744a926 100644 --- a/drivers/md/Kconfig +++ b/drivers/md/Kconfig @@ -299,6 +299,7 @@ config DM_CRYPT select CRYPTO select CRYPTO_CBC select CRYPTO_ESSIV + select CRYPTO_LIB_MD5 # needed by lmk IV mode help This device-mapper target allows you to create a device that transparently encrypts the data on it. You'll need to activate @@ -546,6 +547,7 @@ config DM_VERITY depends on BLK_DEV_DM select CRYPTO select CRYPTO_HASH + select CRYPTO_LIB_SHA256 select DM_BUFIO help This device-mapper target creates a read-only device that diff --git a/drivers/md/dm-bufio.c b/drivers/md/dm-bufio.c index e6d28be11c5c..5235f3e4924b 100644 --- a/drivers/md/dm-bufio.c +++ b/drivers/md/dm-bufio.c @@ -1374,7 +1374,7 @@ static void submit_io(struct dm_buffer *b, enum req_op op, unsigned short ioprio { unsigned int n_sectors; sector_t sector; - unsigned int offset, end; + unsigned int offset, end, align; b->end_io = end_io; @@ -1388,9 +1388,11 @@ static void submit_io(struct dm_buffer *b, enum req_op op, unsigned short ioprio b->c->write_callback(b); offset = b->write_start; end = b->write_end; - offset &= -DM_BUFIO_WRITE_ALIGN; - end += DM_BUFIO_WRITE_ALIGN - 1; - end &= -DM_BUFIO_WRITE_ALIGN; + align = max(DM_BUFIO_WRITE_ALIGN, + bdev_physical_block_size(b->c->bdev)); + offset &= -align; + end += align - 1; + end &= -align; if (unlikely(end > b->c->block_size)) end = b->c->block_size; diff --git a/drivers/md/dm-core.h b/drivers/md/dm-core.h index a3c9f74fe2dc..1cda8618d74d 100644 --- a/drivers/md/dm-core.h +++ b/drivers/md/dm-core.h @@ -139,7 +139,6 @@ struct mapped_device { struct srcu_struct io_barrier; #ifdef CONFIG_BLK_DEV_ZONED - unsigned int nr_zones; void *zone_revalidate_map; struct task_struct *revalidate_map_task; #endif diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 5ef43231fe77..79704fbc523b 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -21,6 +21,7 @@ #include <linux/mempool.h> #include <linux/slab.h> #include <linux/crypto.h> +#include <linux/fips.h> #include <linux/workqueue.h> #include <linux/kthread.h> #include <linux/backing-dev.h> @@ -120,7 +121,6 @@ struct iv_benbi_private { #define LMK_SEED_SIZE 64 /* hash + 0 */ struct iv_lmk_private { - struct crypto_shash *hash_tfm; u8 *seed; }; @@ -254,22 +254,15 @@ static unsigned int max_write_size = 0; module_param(max_write_size, uint, 0644); MODULE_PARM_DESC(max_write_size, "Maximum size of a write request"); -static unsigned get_max_request_sectors(struct dm_target *ti, struct bio *bio) +static unsigned get_max_request_sectors(struct dm_target *ti, struct bio *bio, bool no_split) { struct crypt_config *cc = ti->private; unsigned val, sector_align; bool wrt = op_is_write(bio_op(bio)); - if (wrt) { - /* - * For zoned devices, splitting write operations creates the - * risk of deadlocking queue freeze operations with zone write - * plugging BIO work when the reminder of a split BIO is - * issued. So always allow the entire BIO to proceed. - */ - if (ti->emulate_zone_append) - return bio_sectors(bio); - + if (no_split) { + val = -1; + } else if (wrt) { val = min_not_zero(READ_ONCE(max_write_size), DM_CRYPT_DEFAULT_MAX_WRITE_SIZE); } else { @@ -465,10 +458,6 @@ static void crypt_iv_lmk_dtr(struct crypt_config *cc) { struct iv_lmk_private *lmk = &cc->iv_gen_private.lmk; - if (lmk->hash_tfm && !IS_ERR(lmk->hash_tfm)) - crypto_free_shash(lmk->hash_tfm); - lmk->hash_tfm = NULL; - kfree_sensitive(lmk->seed); lmk->seed = NULL; } @@ -483,11 +472,10 @@ static int crypt_iv_lmk_ctr(struct crypt_config *cc, struct dm_target *ti, return -EINVAL; } - lmk->hash_tfm = crypto_alloc_shash("md5", 0, - CRYPTO_ALG_ALLOCATES_MEMORY); - if (IS_ERR(lmk->hash_tfm)) { - ti->error = "Error initializing LMK hash"; - return PTR_ERR(lmk->hash_tfm); + if (fips_enabled) { + ti->error = "LMK support is disabled due to FIPS"; + /* ... because it uses MD5. */ + return -EINVAL; } /* No seed in LMK version 2 */ @@ -498,7 +486,6 @@ static int crypt_iv_lmk_ctr(struct crypt_config *cc, struct dm_target *ti, lmk->seed = kzalloc(LMK_SEED_SIZE, GFP_KERNEL); if (!lmk->seed) { - crypt_iv_lmk_dtr(cc); ti->error = "Error kmallocing seed storage in LMK"; return -ENOMEM; } @@ -514,7 +501,7 @@ static int crypt_iv_lmk_init(struct crypt_config *cc) /* LMK seed is on the position of LMK_KEYS + 1 key */ if (lmk->seed) memcpy(lmk->seed, cc->key + (cc->tfms_count * subkey_size), - crypto_shash_digestsize(lmk->hash_tfm)); + MD5_DIGEST_SIZE); return 0; } @@ -529,55 +516,31 @@ static int crypt_iv_lmk_wipe(struct crypt_config *cc) return 0; } -static int crypt_iv_lmk_one(struct crypt_config *cc, u8 *iv, - struct dm_crypt_request *dmreq, - u8 *data) +static void crypt_iv_lmk_one(struct crypt_config *cc, u8 *iv, + struct dm_crypt_request *dmreq, u8 *data) { struct iv_lmk_private *lmk = &cc->iv_gen_private.lmk; - SHASH_DESC_ON_STACK(desc, lmk->hash_tfm); - union { - struct md5_state md5state; - u8 state[CRYPTO_MD5_STATESIZE]; - } u; + struct md5_ctx ctx; __le32 buf[4]; - int i, r; - desc->tfm = lmk->hash_tfm; + md5_init(&ctx); - r = crypto_shash_init(desc); - if (r) - return r; - - if (lmk->seed) { - r = crypto_shash_update(desc, lmk->seed, LMK_SEED_SIZE); - if (r) - return r; - } + if (lmk->seed) + md5_update(&ctx, lmk->seed, LMK_SEED_SIZE); /* Sector is always 512B, block size 16, add data of blocks 1-31 */ - r = crypto_shash_update(desc, data + 16, 16 * 31); - if (r) - return r; + md5_update(&ctx, data + 16, 16 * 31); /* Sector is cropped to 56 bits here */ buf[0] = cpu_to_le32(dmreq->iv_sector & 0xFFFFFFFF); buf[1] = cpu_to_le32((((u64)dmreq->iv_sector >> 32) & 0x00FFFFFF) | 0x80000000); buf[2] = cpu_to_le32(4024); buf[3] = 0; - r = crypto_shash_update(desc, (u8 *)buf, sizeof(buf)); - if (r) - return r; + md5_update(&ctx, (u8 *)buf, sizeof(buf)); /* No MD5 padding here */ - r = crypto_shash_export(desc, &u.md5state); - if (r) - return r; - - for (i = 0; i < MD5_HASH_WORDS; i++) - __cpu_to_le32s(&u.md5state.hash[i]); - memcpy(iv, &u.md5state.hash, cc->iv_size); - - return 0; + cpu_to_le32_array(ctx.state.h, ARRAY_SIZE(ctx.state.h)); + memcpy(iv, ctx.state.h, cc->iv_size); } static int crypt_iv_lmk_gen(struct crypt_config *cc, u8 *iv, @@ -585,17 +548,15 @@ static int crypt_iv_lmk_gen(struct crypt_config *cc, u8 *iv, { struct scatterlist *sg; u8 *src; - int r = 0; if (bio_data_dir(dmreq->ctx->bio_in) == WRITE) { sg = crypt_get_sg_data(cc, dmreq->sg_in); src = kmap_local_page(sg_page(sg)); - r = crypt_iv_lmk_one(cc, iv, dmreq, src + sg->offset); + crypt_iv_lmk_one(cc, iv, dmreq, src + sg->offset); kunmap_local(src); } else memset(iv, 0, cc->iv_size); - - return r; + return 0; } static int crypt_iv_lmk_post(struct crypt_config *cc, u8 *iv, @@ -603,21 +564,19 @@ static int crypt_iv_lmk_post(struct crypt_config *cc, u8 *iv, { struct scatterlist *sg; u8 *dst; - int r; if (bio_data_dir(dmreq->ctx->bio_in) == WRITE) return 0; sg = crypt_get_sg_data(cc, dmreq->sg_out); dst = kmap_local_page(sg_page(sg)); - r = crypt_iv_lmk_one(cc, iv, dmreq, dst + sg->offset); + crypt_iv_lmk_one(cc, iv, dmreq, dst + sg->offset); /* Tweak the first block of plaintext sector */ - if (!r) - crypto_xor(dst + sg->offset, iv, cc->iv_size); + crypto_xor(dst + sg->offset, iv, cc->iv_size); kunmap_local(dst); - return r; + return 0; } static void crypt_iv_tcw_dtr(struct crypt_config *cc) @@ -1781,7 +1740,7 @@ static void crypt_free_buffer_pages(struct crypt_config *cc, struct bio *clone) bio_for_each_folio_all(fi, clone) { if (folio_test_large(fi.folio)) { percpu_counter_sub(&cc->n_allocated_pages, - 1 << folio_order(fi.folio)); + folio_nr_pages(fi.folio)); folio_put(fi.folio); } else { mempool_free(&fi.folio->page, &cc->page_pool); @@ -3496,6 +3455,7 @@ static int crypt_map(struct dm_target *ti, struct bio *bio) struct dm_crypt_io *io; struct crypt_config *cc = ti->private; unsigned max_sectors; + bool no_split; /* * If bio is REQ_PREFLUSH or REQ_OP_DISCARD, just bypass crypt queues. @@ -3513,10 +3473,20 @@ static int crypt_map(struct dm_target *ti, struct bio *bio) /* * Check if bio is too large, split as needed. + * + * For zoned devices, splitting write operations creates the + * risk of deadlocking queue freeze operations with zone write + * plugging BIO work when the reminder of a split BIO is + * issued. So always allow the entire BIO to proceed. */ - max_sectors = get_max_request_sectors(ti, bio); - if (unlikely(bio_sectors(bio) > max_sectors)) + no_split = (ti->emulate_zone_append && op_is_write(bio_op(bio))) || + (bio->bi_opf & REQ_ATOMIC); + max_sectors = get_max_request_sectors(ti, bio, no_split); + if (unlikely(bio_sectors(bio) > max_sectors)) { + if (unlikely(no_split)) + return DM_MAPIO_KILL; dm_accept_partial_bio(bio, max_sectors); + } /* * Ensure that bio is a multiple of internal sector encryption size @@ -3762,15 +3732,20 @@ static void crypt_io_hints(struct dm_target *ti, struct queue_limits *limits) if (ti->emulate_zone_append) limits->max_hw_sectors = min(limits->max_hw_sectors, BIO_MAX_VECS << PAGE_SECTORS_SHIFT); + + limits->atomic_write_hw_unit_max = min(limits->atomic_write_hw_unit_max, + BIO_MAX_VECS << PAGE_SHIFT); + limits->atomic_write_hw_max = min(limits->atomic_write_hw_max, + BIO_MAX_VECS << PAGE_SHIFT); } static struct target_type crypt_target = { .name = "crypt", - .version = {1, 28, 0}, + .version = {1, 29, 0}, .module = THIS_MODULE, .ctr = crypt_ctr, .dtr = crypt_dtr, - .features = DM_TARGET_ZONED_HM, + .features = DM_TARGET_ZONED_HM | DM_TARGET_ATOMIC_WRITES, .report_zones = crypt_report_zones, .map = crypt_map, .status = crypt_status, diff --git a/drivers/md/dm-ebs-target.c b/drivers/md/dm-ebs-target.c index 6abb31ca9662..b354e74a670e 100644 --- a/drivers/md/dm-ebs-target.c +++ b/drivers/md/dm-ebs-target.c @@ -103,7 +103,7 @@ static int __ebs_rw_bvec(struct ebs_c *ec, enum req_op op, struct bio_vec *bv, } else { flush_dcache_page(bv->bv_page); memcpy(ba, pa, cur_len); - dm_bufio_mark_partial_buffer_dirty(b, buf_off, buf_off + cur_len); + dm_bufio_mark_buffer_dirty(b); } dm_bufio_release(b); diff --git a/drivers/md/dm-exception-store.h b/drivers/md/dm-exception-store.h index b67976637538..061b4d310813 100644 --- a/drivers/md/dm-exception-store.h +++ b/drivers/md/dm-exception-store.h @@ -29,7 +29,7 @@ typedef sector_t chunk_t; * chunk within the device. */ struct dm_exception { - struct hlist_bl_node hash_list; + struct hlist_node hash_list; chunk_t old_chunk; chunk_t new_chunk; diff --git a/drivers/md/dm-log-writes.c b/drivers/md/dm-log-writes.c index 7bb7174f8f4f..f0c84e7a5daa 100644 --- a/drivers/md/dm-log-writes.c +++ b/drivers/md/dm-log-writes.c @@ -432,6 +432,7 @@ static int log_writes_kthread(void *arg) struct log_writes_c *lc = arg; sector_t sector = 0; + set_freezable(); while (!kthread_should_stop()) { bool super = false; bool logging_enabled; diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index aaf4a0a4b0eb..c18358271618 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -131,7 +131,7 @@ static void queue_if_no_path_timeout_work(struct timer_list *t); #define MPATHF_QUEUE_IO 0 /* Must we queue all I/O? */ #define MPATHF_QUEUE_IF_NO_PATH 1 /* Queue I/O if last path fails? */ #define MPATHF_SAVED_QUEUE_IF_NO_PATH 2 /* Saved state during suspension */ -#define MPATHF_RETAIN_ATTACHED_HW_HANDLER 3 /* If there's already a hw_handler present, don't change it. */ +/* MPATHF_RETAIN_ATTACHED_HW_HANDLER no longer has any effect */ #define MPATHF_PG_INIT_DISABLED 4 /* pg_init is not currently allowed */ #define MPATHF_PG_INIT_REQUIRED 5 /* pg_init needs calling? */ #define MPATHF_PG_INIT_DELAY_RETRY 6 /* Delay pg_init retry? */ @@ -237,16 +237,10 @@ static struct multipath *alloc_multipath(struct dm_target *ti) static int alloc_multipath_stage2(struct dm_target *ti, struct multipath *m) { - if (m->queue_mode == DM_TYPE_NONE) { + if (m->queue_mode == DM_TYPE_NONE) m->queue_mode = DM_TYPE_REQUEST_BASED; - } else if (m->queue_mode == DM_TYPE_BIO_BASED) { + else if (m->queue_mode == DM_TYPE_BIO_BASED) INIT_WORK(&m->process_queued_bios, process_queued_bios); - /* - * bio-based doesn't support any direct scsi_dh management; - * it just discovers if a scsi_dh is attached. - */ - set_bit(MPATHF_RETAIN_ATTACHED_HW_HANDLER, &m->flags); - } dm_table_set_type(ti->table, m->queue_mode); @@ -887,36 +881,30 @@ static int setup_scsi_dh(struct block_device *bdev, struct multipath *m, struct request_queue *q = bdev_get_queue(bdev); int r; - if (mpath_double_check_test_bit(MPATHF_RETAIN_ATTACHED_HW_HANDLER, m)) { -retain: - if (*attached_handler_name) { - /* - * Clear any hw_handler_params associated with a - * handler that isn't already attached. - */ - if (m->hw_handler_name && strcmp(*attached_handler_name, m->hw_handler_name)) { - kfree(m->hw_handler_params); - m->hw_handler_params = NULL; - } - - /* - * Reset hw_handler_name to match the attached handler - * - * NB. This modifies the table line to show the actual - * handler instead of the original table passed in. - */ - kfree(m->hw_handler_name); - m->hw_handler_name = *attached_handler_name; - *attached_handler_name = NULL; + if (*attached_handler_name) { + /* + * Clear any hw_handler_params associated with a + * handler that isn't already attached. + */ + if (m->hw_handler_name && strcmp(*attached_handler_name, + m->hw_handler_name)) { + kfree(m->hw_handler_params); + m->hw_handler_params = NULL; } + + /* + * Reset hw_handler_name to match the attached handler + * + * NB. This modifies the table line to show the actual + * handler instead of the original table passed in. + */ + kfree(m->hw_handler_name); + m->hw_handler_name = *attached_handler_name; + *attached_handler_name = NULL; } if (m->hw_handler_name) { r = scsi_dh_attach(q, m->hw_handler_name); - if (r == -EBUSY) { - DMINFO("retaining handler on device %pg", bdev); - goto retain; - } if (r < 0) { *error = "error attaching hardware handler"; return r; @@ -1138,7 +1126,7 @@ static int parse_features(struct dm_arg_set *as, struct multipath *m) } if (!strcasecmp(arg_name, "retain_attached_hw_handler")) { - set_bit(MPATHF_RETAIN_ATTACHED_HW_HANDLER, &m->flags); + /* no longer has any effect */ continue; } @@ -1823,7 +1811,6 @@ static void multipath_status(struct dm_target *ti, status_type_t type, DMEMIT("%u ", test_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags) + (m->pg_init_retries > 0) * 2 + (m->pg_init_delay_msecs != DM_PG_INIT_DELAY_DEFAULT) * 2 + - test_bit(MPATHF_RETAIN_ATTACHED_HW_HANDLER, &m->flags) + (m->queue_mode != DM_TYPE_REQUEST_BASED) * 2); if (test_bit(MPATHF_QUEUE_IF_NO_PATH, &m->flags)) @@ -1832,8 +1819,6 @@ static void multipath_status(struct dm_target *ti, status_type_t type, DMEMIT("pg_init_retries %u ", m->pg_init_retries); if (m->pg_init_delay_msecs != DM_PG_INIT_DELAY_DEFAULT) DMEMIT("pg_init_delay_msecs %u ", m->pg_init_delay_msecs); - if (test_bit(MPATHF_RETAIN_ATTACHED_HW_HANDLER, &m->flags)) - DMEMIT("retain_attached_hw_handler "); if (m->queue_mode != DM_TYPE_REQUEST_BASED) { switch (m->queue_mode) { case DM_TYPE_BIO_BASED: @@ -2307,7 +2292,7 @@ static struct target_type multipath_target = { .name = "multipath", .version = {1, 15, 0}, .features = DM_TARGET_SINGLETON | DM_TARGET_IMMUTABLE | - DM_TARGET_PASSES_INTEGRITY, + DM_TARGET_PASSES_INTEGRITY | DM_TARGET_ATOMIC_WRITES, .module = THIS_MODULE, .ctr = multipath_ctr, .dtr = multipath_dtr, diff --git a/drivers/md/dm-pcache/cache.c b/drivers/md/dm-pcache/cache.c index 698697a7a73c..534bf07b794f 100644 --- a/drivers/md/dm-pcache/cache.c +++ b/drivers/md/dm-pcache/cache.c @@ -10,7 +10,8 @@ struct kmem_cache *key_cache; static inline struct pcache_cache_info *get_cache_info_addr(struct pcache_cache *cache) { - return cache->cache_info_addr + cache->info_index; + return (struct pcache_cache_info *)((char *)cache->cache_info_addr + + (size_t)cache->info_index * PCACHE_CACHE_INFO_SIZE); } static void cache_info_write(struct pcache_cache *cache) @@ -21,10 +22,10 @@ static void cache_info_write(struct pcache_cache *cache) cache_info->header.crc = pcache_meta_crc(&cache_info->header, sizeof(struct pcache_cache_info)); + cache->info_index = (cache->info_index + 1) % PCACHE_META_INDEX_MAX; memcpy_flushcache(get_cache_info_addr(cache), cache_info, sizeof(struct pcache_cache_info)); - - cache->info_index = (cache->info_index + 1) % PCACHE_META_INDEX_MAX; + pmem_wmb(); } static void cache_info_init_default(struct pcache_cache *cache); @@ -49,6 +50,8 @@ static int cache_info_init(struct pcache_cache *cache, struct pcache_cache_optio return -EINVAL; } + cache->info_index = ((char *)cache_info_addr - (char *)cache->cache_info_addr) / PCACHE_CACHE_INFO_SIZE; + return 0; } @@ -93,10 +96,10 @@ void cache_pos_encode(struct pcache_cache *cache, pos_onmedia.header.seq = seq; pos_onmedia.header.crc = cache_pos_onmedia_crc(&pos_onmedia); + *index = (*index + 1) % PCACHE_META_INDEX_MAX; + memcpy_flushcache(pos_onmedia_addr, &pos_onmedia, sizeof(struct pcache_cache_pos_onmedia)); pmem_wmb(); - - *index = (*index + 1) % PCACHE_META_INDEX_MAX; } int cach |