linux-kernel - [PATCH 3.10 57/66] md/raid1: restore ability for check and repair to fix read errors.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20140220234911.118067852@linuxfoundation.org>
Date:	Thu, 20 Feb 2014 15:51:56 -0800
From:	Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:	linux-kernel@...r.kernel.org
Cc:	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	stable@...r.kernel.org, NeilBrown <neilb@...e.de>
Subject: [PATCH 3.10 57/66] md/raid1: restore ability for check and repair to fix read errors.

3.10-stable review patch.  If anyone has any objections, please let me know.

------------------

From: NeilBrown <neilb@...e.de>

commit 1877db75589a895bbdc4c4c3f23558e57b521141 upstream.

commit 30bc9b53878a9921b02e3b5bc4283ac1c6de102a
    md/raid1: fix bio handling problems in process_checks()

Move the bio_reset() to a point before where BIO_UPTODATE is checked,
so that check now always report that the bio is uptodate, even if it is not.

This causes process_check() to sometimes treat read-errors as
successful matches so the good data isn't written out.

This patch preserves the flag until it is needed.

Bug was introduced in 3.11, but backported to 3.10-stable (as it fixed
an even worse bug).  So suitable for any -stable since 3.10.

Reported-and-tested-by: Michael Tokarev <mjt@....msk.ru>
Fixed: 30bc9b53878a9921b02e3b5bc4283ac1c6de102a
Signed-off-by: NeilBrown <neilb@...e.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>

---
 drivers/md/raid1.c |   13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1854,11 +1854,15 @@ static int process_checks(struct r1bio *
 	for (i = 0; i < conf->raid_disks * 2; i++) {
 		int j;
 		int size;
+		int uptodate;
 		struct bio *b = r1_bio->bios[i];
 		if (b->bi_end_io != end_sync_read)
 			continue;
-		/* fixup the bio for reuse */
+		/* fixup the bio for reuse, but preserve BIO_UPTODATE */
+		uptodate = test_bit(BIO_UPTODATE, &b->bi_flags);
 		bio_reset(b);
+		if (!uptodate)
+			clear_bit(BIO_UPTODATE, &b->bi_flags);
 		b->bi_vcnt = vcnt;
 		b->bi_size = r1_bio->sectors << 9;
 		b->bi_sector = r1_bio->sector +
@@ -1891,11 +1895,14 @@ static int process_checks(struct r1bio *
 		int j;
 		struct bio *pbio = r1_bio->bios[primary];
 		struct bio *sbio = r1_bio->bios[i];
+		int uptodate = test_bit(BIO_UPTODATE, &sbio->bi_flags);
 
 		if (sbio->bi_end_io != end_sync_read)
 			continue;
+		/* Now we can 'fixup' the BIO_UPTODATE flag */
+		set_bit(BIO_UPTODATE, &sbio->bi_flags);
 
-		if (test_bit(BIO_UPTODATE, &sbio->bi_flags)) {
+		if (uptodate) {
 			for (j = vcnt; j-- ; ) {
 				struct page *p, *s;
 				p = pbio->bi_io_vec[j].bv_page;
@@ -1910,7 +1917,7 @@ static int process_checks(struct r1bio *
 		if (j >= 0)
 			atomic64_add(r1_bio->sectors, &mddev->resync_mismatches);
 		if (j < 0 || (test_bit(MD_RECOVERY_CHECK, &mddev->recovery)
-			      && test_bit(BIO_UPTODATE, &sbio->bi_flags))) {
+			      && uptodate)) {
 			/* No need to write to this device. */
 			sbio->bi_end_io = NULL;
 			rdev_dec_pending(conf->mirrors[i].rdev, mddev);


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/