| Message ID | cfc81e25919ac1abcee5cbbbf3462c147a838d02.1307991539.git.list.btrfs@jan-o-sch.net (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Mon, Jun 13, 2011 at 09:10:35PM +0200, Jan Schmidt wrote: > In normal operation, scrub is reading data sequentially in large portions. > In case of an i/o error, we try to find the corrupted area(s) by issuing > page sized read requests. With this commit we increment the > unverified_errors counter if all of the small size requests succeed. > > Userland patches carrying such conspicous events to the administrator should > already be around. > > Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net> > --- > fs/btrfs/scrub.c | 37 ++++++++++++++++++++++++++----------- > 1 files changed, 26 insertions(+), 11 deletions(-) > > diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c > index d5a4108..00e4e58 100644 > --- a/fs/btrfs/scrub.c > +++ b/fs/btrfs/scrub.c > @@ -207,18 +207,25 @@ nomem: > * recheck_error gets called for every page in the bio, even though only > * one may be bad > */ > -static void scrub_recheck_error(struct scrub_bio *sbio, int ix) > +static int scrub_recheck_error(struct scrub_bio *sbio, int ix) > { > + struct scrub_dev *sdev = sbio->sdev; > + u64 sector = (sbio->physical + ix * PAGE_SIZE) >> 9; > + > if (sbio->err) { > - if (scrub_fixup_io(READ, sbio->sdev->dev->bdev, > - (sbio->physical + ix * PAGE_SIZE) >> 9, > + if (scrub_fixup_io(READ, sbio->sdev->dev->bdev, sector, > sbio->bio->bi_io_vec[ix].bv_page) == 0) { > if (scrub_fixup_check(sbio, ix) == 0) > - return; > + return 0; > } > } > > + spin_lock(&sdev->stat_lock); > + ++sdev->stat.read_errors; > + spin_unlock(&sdev->stat_lock); > + > scrub_fixup(sbio, ix); > + return 1; > } > > static int scrub_fixup_check(struct scrub_bio *sbio, int ix) > @@ -388,8 +395,14 @@ static void scrub_checksum(struct btrfs_work *work) > int ret; > > if (sbio->err) { > + ret = 0; > for (i = 0; i < sbio->count; ++i) > - scrub_recheck_error(sbio, i); > + ret |= scrub_recheck_error(sbio, i); > + if (!ret) { > + spin_lock(&sdev->stat_lock); > + ++sdev->stat.unverified_errors; > + spin_unlock(&sdev->stat_lock); > + } > > sbio->bio->bi_flags &= ~(BIO_POOL_MASK - 1); > sbio->bio->bi_flags |= 1 << BIO_UPTODATE; > @@ -402,10 +415,6 @@ static void scrub_checksum(struct btrfs_work *work) > bi->bv_offset = 0; > bi->bv_len = PAGE_SIZE; > } > - > - spin_lock(&sdev->stat_lock); > - ++sdev->stat.read_errors; > - spin_unlock(&sdev->stat_lock); > goto out; > } > for (i = 0; i < sbio->count; ++i) { > @@ -426,8 +435,14 @@ static void scrub_checksum(struct btrfs_work *work) > WARN_ON(1); > } > kunmap_atomic(buffer, KM_USER0); > - if (ret) > - scrub_recheck_error(sbio, i); > + if (ret) { > + ret = scrub_recheck_error(sbio, i); > + if (!ret) { > + spin_lock(&sdev->stat_lock); > + ++sdev->stat.unverified_errors; > + spin_unlock(&sdev->stat_lock); > + } > + } > } > > out: -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi, sorry for the noise, this has no comments from me, david On Thu, Jun 16, 2011 at 11:25:21PM +0200, David Sterba wrote: > On Mon, Jun 13, 2011 at 09:10:35PM +0200, Jan Schmidt wrote: > > In normal operation, scrub is reading data sequentially in large portions. > > In case of an i/o error, we try to find the corrupted area(s) by issuing > > page sized read requests. With this commit we increment the > > unverified_errors counter if all of the small size requests succeed. > > > > Userland patches carrying such conspicous events to the administrator should > > already be around. > > > > Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net> > > --- > > fs/btrfs/scrub.c | 37 ++++++++++++++++++++++++++----------- > > 1 files changed, 26 insertions(+), 11 deletions(-) > > > > diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c > > index d5a4108..00e4e58 100644 > > --- a/fs/btrfs/scrub.c > > +++ b/fs/btrfs/scrub.c > > @@ -207,18 +207,25 @@ nomem: > > * recheck_error gets called for every page in the bio, even though only > > * one may be bad > > */ > > -static void scrub_recheck_error(struct scrub_bio *sbio, int ix) > > +static int scrub_recheck_error(struct scrub_bio *sbio, int ix) > > { > > + struct scrub_dev *sdev = sbio->sdev; > > + u64 sector = (sbio->physical + ix * PAGE_SIZE) >> 9; > > + > > if (sbio->err) { > > - if (scrub_fixup_io(READ, sbio->sdev->dev->bdev, > > - (sbio->physical + ix * PAGE_SIZE) >> 9, > > + if (scrub_fixup_io(READ, sbio->sdev->dev->bdev, sector, > > sbio->bio->bi_io_vec[ix].bv_page) == 0) { > > if (scrub_fixup_check(sbio, ix) == 0) > > - return; > > + return 0; > > } > > } > > > > + spin_lock(&sdev->stat_lock); > > + ++sdev->stat.read_errors; > > + spin_unlock(&sdev->stat_lock); > > + > > scrub_fixup(sbio, ix); > > + return 1; > > } > > > > static int scrub_fixup_check(struct scrub_bio *sbio, int ix) > > @@ -388,8 +395,14 @@ static void scrub_checksum(struct btrfs_work *work) > > int ret; > > > > if (sbio->err) { > > + ret = 0; > > for (i = 0; i < sbio->count; ++i) > > - scrub_recheck_error(sbio, i); > > + ret |= scrub_recheck_error(sbio, i); > > + if (!ret) { > > + spin_lock(&sdev->stat_lock); > > + ++sdev->stat.unverified_errors; > > + spin_unlock(&sdev->stat_lock); > > + } > > > > sbio->bio->bi_flags &= ~(BIO_POOL_MASK - 1); > > sbio->bio->bi_flags |= 1 << BIO_UPTODATE; > > @@ -402,10 +415,6 @@ static void scrub_checksum(struct btrfs_work *work) > > bi->bv_offset = 0; > > bi->bv_len = PAGE_SIZE; > > } > > - > > - spin_lock(&sdev->stat_lock); > > - ++sdev->stat.read_errors; > > - spin_unlock(&sdev->stat_lock); > > goto out; > > } > > for (i = 0; i < sbio->count; ++i) { > > @@ -426,8 +435,14 @@ static void scrub_checksum(struct btrfs_work *work) > > WARN_ON(1); > > } > > kunmap_atomic(buffer, KM_USER0); > > - if (ret) > > - scrub_recheck_error(sbio, i); > > + if (ret) { > > + ret = scrub_recheck_error(sbio, i); > > + if (!ret) { > > + spin_lock(&sdev->stat_lock); > > + ++sdev->stat.unverified_errors; > > + spin_unlock(&sdev->stat_lock); > > + } > > + } > > } > > > > out: > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index d5a4108..00e4e58 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -207,18 +207,25 @@ nomem: * recheck_error gets called for every page in the bio, even though only * one may be bad */ -static void scrub_recheck_error(struct scrub_bio *sbio, int ix) +static int scrub_recheck_error(struct scrub_bio *sbio, int ix) { + struct scrub_dev *sdev = sbio->sdev; + u64 sector = (sbio->physical + ix * PAGE_SIZE) >> 9; + if (sbio->err) { - if (scrub_fixup_io(READ, sbio->sdev->dev->bdev, - (sbio->physical + ix * PAGE_SIZE) >> 9, + if (scrub_fixup_io(READ, sbio->sdev->dev->bdev, sector, sbio->bio->bi_io_vec[ix].bv_page) == 0) { if (scrub_fixup_check(sbio, ix) == 0) - return; + return 0; } } + spin_lock(&sdev->stat_lock); + ++sdev->stat.read_errors; + spin_unlock(&sdev->stat_lock); + scrub_fixup(sbio, ix); + return 1; } static int scrub_fixup_check(struct scrub_bio *sbio, int ix) @@ -388,8 +395,14 @@ static void scrub_checksum(struct btrfs_work *work) int ret; if (sbio->err) { + ret = 0; for (i = 0; i < sbio->count; ++i) - scrub_recheck_error(sbio, i); + ret |= scrub_recheck_error(sbio, i); + if (!ret) { + spin_lock(&sdev->stat_lock); + ++sdev->stat.unverified_errors; + spin_unlock(&sdev->stat_lock); + } sbio->bio->bi_flags &= ~(BIO_POOL_MASK - 1); sbio->bio->bi_flags |= 1 << BIO_UPTODATE; @@ -402,10 +415,6 @@ static void scrub_checksum(struct btrfs_work *work) bi->bv_offset = 0; bi->bv_len = PAGE_SIZE; } - - spin_lock(&sdev->stat_lock); - ++sdev->stat.read_errors; - spin_unlock(&sdev->stat_lock); goto out; } for (i = 0; i < sbio->count; ++i) { @@ -426,8 +435,14 @@ static void scrub_checksum(struct btrfs_work *work) WARN_ON(1); } kunmap_atomic(buffer, KM_USER0); - if (ret) - scrub_recheck_error(sbio, i); + if (ret) { + ret = scrub_recheck_error(sbio, i); + if (!ret) { + spin_lock(&sdev->stat_lock); + ++sdev->stat.unverified_errors; + spin_unlock(&sdev->stat_lock); + } + } } out:
In normal operation, scrub is reading data sequentially in large portions. In case of an i/o error, we try to find the corrupted area(s) by issuing page sized read requests. With this commit we increment the unverified_errors counter if all of the small size requests succeed. Userland patches carrying such conspicous events to the administrator should already be around. Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net> --- fs/btrfs/scrub.c | 37 ++++++++++++++++++++++++++----------- 1 files changed, 26 insertions(+), 11 deletions(-)