From 62e1e389f87a8839ad83b08c44691d1df8320846 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 09:40:59 +1000 Subject: md: always update level / chunk_size / layout when writing v1.x metadata. We previously didn't update these fields when writing the metadata because they could never change. They can now, so we better write them. v0.90 metadata always updated these fields. Signed-off-by: NeilBrown --- drivers/md/md.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/md/md.c b/drivers/md/md.c index fccc8343a250..aa79d55875f0 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -1375,6 +1375,9 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev) sb->raid_disks = cpu_to_le32(mddev->raid_disks); sb->size = cpu_to_le64(mddev->dev_sectors); + sb->chunksize = cpu_to_le32(mddev->chunk_size >> 9); + sb->level = cpu_to_le32(mddev->level); + sb->layout = cpu_to_le32(mddev->layout); if (mddev->bitmap && mddev->bitmap_file == NULL) { sb->bitmap_offset = cpu_to_le32((__u32)mddev->bitmap_offset); -- cgit v1.2.2 From 2b69c83924396ad1eda36fdd267c9d2f360f5555 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 09:41:17 +1000 Subject: md: improve errno return when setting array_size Instead of always returns EINVAL if anything goes wrong when setting the array size, add the option of E2BIG if the size requested is too large. This makes it easier for user-space to be sure what went wrong. Signed-off-by: NeilBrown --- drivers/md/md.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index aa79d55875f0..58e0b02a74c2 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -3683,7 +3683,7 @@ array_size_store(mddev_t *mddev, const char *buf, size_t len) if (strict_blocks_to_sectors(buf, §ors) < 0) return -EINVAL; if (mddev->pers && mddev->pers->size(mddev, 0, 0) < sectors) - return -EINVAL; + return -E2BIG; mddev->external_size = 1; } -- cgit v1.2.2 From be512691036cc989c11d0f418187efbbf14468e6 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 09:41:17 +1000 Subject: md: bitmap: improve bitmap maintenance code. The code for checking which bits in the bitmap can be cleared has 2 problems: 1/ it repeatedly takes and drops a spinlock, where it would make more sense to just hold on to it most of the time. 2/ it doesn't make use of some opportunities to skip large sections of the bitmap This patch fixes those. It will only affect CPU consumption, not correctness. Signed-off-by: NeilBrown --- drivers/md/bitmap.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index 47c68bc75a17..56df1cee8fb3 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -1097,14 +1097,12 @@ void bitmap_daemon_work(struct bitmap *bitmap) } bitmap->allclean = 1; + spin_lock_irqsave(&bitmap->lock, flags); for (j = 0; j < bitmap->chunks; j++) { bitmap_counter_t *bmc; - spin_lock_irqsave(&bitmap->lock, flags); - if (!bitmap->filemap) { + if (!bitmap->filemap) /* error or shutdown */ - spin_unlock_irqrestore(&bitmap->lock, flags); break; - } page = filemap_get_page(bitmap, j); @@ -1121,6 +1119,8 @@ void bitmap_daemon_work(struct bitmap *bitmap) write_page(bitmap, page, 0); bitmap->allclean = 0; } + spin_lock_irqsave(&bitmap->lock, flags); + j |= (PAGE_BITS - 1); continue; } @@ -1181,9 +1181,10 @@ void bitmap_daemon_work(struct bitmap *bitmap) ext2_clear_bit(file_page_offset(j), paddr); kunmap_atomic(paddr, KM_USER0); } - } - spin_unlock_irqrestore(&bitmap->lock, flags); + } else + j |= PAGE_COUNTER_MASK; } + spin_unlock_irqrestore(&bitmap->lock, flags); /* now sync the final page */ if (lastpage != NULL) { -- cgit v1.2.2 From b6a9ce688f613e2ee5f15e6720e0bb8520efc36e Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 09:41:17 +1000 Subject: md: export 'frozen' resync state through sysfs The md resync engine has a 'frozen' state which ensures that no resync/recovery. This is used to avoid races. Export this state through the 'sync_action' sysfs attribute so that user-space can benefit and also avoid some races. Signed-off-by: NeilBrown --- drivers/md/md.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 58e0b02a74c2..384e4f0904c8 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -3306,7 +3306,9 @@ static ssize_t action_show(mddev_t *mddev, char *page) { char *type = "idle"; - if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) || + if (test_bit(MD_RECOVERY_FROZEN, &mddev->recovery)) + type = "frozen"; + else if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) || (!mddev->ro && test_bit(MD_RECOVERY_NEEDED, &mddev->recovery))) { if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery)) type = "reshape"; @@ -3329,7 +3331,12 @@ action_store(mddev_t *mddev, const char *page, size_t len) if (!mddev->pers || !mddev->pers->sync_request) return -EINVAL; - if (cmd_match(page, "idle")) { + if (cmd_match(page, "frozen")) + set_bit(MD_RECOVERY_FROZEN, &mddev->recovery); + else + clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery); + + if (cmd_match(page, "idle") || cmd_match(page, "frozen")) { if (mddev->sync_thread) { set_bit(MD_RECOVERY_INTR, &mddev->recovery); md_unregister_thread(mddev->sync_thread); -- cgit v1.2.2 From 848b3182365fdf5a05bcd5ed949071cac2c894b3 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 12:41:08 +1000 Subject: md: raid5: avoid sector values going negative when testing reshape progress. As sector_t in unsigned, we cannot afford to let 'safepos' etc go negative. So replace a -= b; by a -= min(b,a); Signed-off-by: NeilBrown --- drivers/md/raid5.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 4616bc3a6e71..3c3626d2a1f9 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3811,13 +3811,13 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped safepos = conf->reshape_safe; sector_div(safepos, data_disks); if (mddev->delta_disks < 0) { - writepos -= reshape_sectors; + writepos -= min(reshape_sectors, writepos); readpos += reshape_sectors; safepos += reshape_sectors; } else { writepos += reshape_sectors; - readpos -= reshape_sectors; - safepos -= reshape_sectors; + readpos -= min(reshape_sectors, readpos); + safepos -= min(reshape_sectors, safepos); } /* 'writepos' is the most advanced device address we might write. -- cgit v1.2.2 From 7a91ee1f628ef6bfe3f13067c0ddf9db520cb86b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 12:57:21 +1000 Subject: md: don't update curr_resync_completed without also updating reshape_position. In order for the metadata to always be consistent, we mustn't updated curr_resync_completed without also updating reshape_position. The reshape code updates both at the same time. However since commit 97e4f42d62badb0f9fbc27c013e89bc1336a03bc the common md_do_sync will sometimes update curr_resync_completed but is not in a position to update reshape_position. So if MD_RECOVERY_RESHAPE is set (indicating that a reshape is happening, so reshape_position might change), don't update curr_resync_completed in md_do_sync, leave it to the per-personality reshape code. Signed-off-by: NeilBrown --- drivers/md/md.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 384e4f0904c8..954456532acc 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -6362,12 +6362,13 @@ void md_do_sync(mddev_t *mddev) skipped = 0; - if ((mddev->curr_resync > mddev->curr_resync_completed && - (mddev->curr_resync - mddev->curr_resync_completed) - > (max_sectors >> 4)) || - (j - mddev->curr_resync_completed)*2 - >= mddev->resync_max - mddev->curr_resync_completed - ) { + if (!test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) && + ((mddev->curr_resync > mddev->curr_resync_completed && + (mddev->curr_resync - mddev->curr_resync_completed) + > (max_sectors >> 4)) || + (j - mddev->curr_resync_completed)*2 + >= mddev->resync_max - mddev->curr_resync_completed + )) { /* time to update curr_resync_completed */ blk_unplug(mddev->queue); wait_event(mddev->recovery_wait, -- cgit v1.2.2 From b492b852cd8c99505708152c29a5e09a787af9de Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 12:57:36 +1000 Subject: md: don't use locked_ioctl. md has no need for the BKL - it does its own locking. So md_ioctl doesn't need to be a locked_ioctl. Signed-off-by: NeilBrown --- drivers/md/md.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 954456532acc..641b211fe3fe 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -5567,7 +5567,7 @@ static struct block_device_operations md_fops = .owner = THIS_MODULE, .open = md_open, .release = md_release, - .locked_ioctl = md_ioctl, + .ioctl = md_ioctl, .getgeo = md_getgeo, .media_changed = md_media_changed, .revalidate_disk= md_revalidate, -- cgit v1.2.2