From: Brian Foster <bfoster@redhat.com>
To: linux-xfs@vger.kernel.org
Subject: [PATCH 4/4] xfs: add log item pinning error injection tag
Date: Tue, 27 Jun 2017 10:40:36 -0400 [thread overview]
Message-ID: <1498574436-57561-5-git-send-email-bfoster@redhat.com> (raw)
In-Reply-To: <1498574436-57561-1-git-send-email-bfoster@redhat.com>
Add an error injection tag to force log items in the AIL to the
pinned state. This option can be used by test infrastructure to
induce head behind tail conditions. Specifically, this is intended
to be used by xfstests to reproduce log recovery problems after
failed/corrupted log writes overwrite the last good tail LSN in the
log.
When enabled, AIL push attempts see log items in the AIL in the
pinned state. This stalls metadata writeback and thus prevents the
current tail of the log from moving forward. When disabled,
subsequent AIL pushes observe the log items in their appropriate
state and filesystem operation continues as normal.
Signed-off-by: Brian Foster <bfoster@redhat.com>
---
fs/xfs/xfs_error.c | 3 +++
fs/xfs/xfs_error.h | 4 +++-
fs/xfs/xfs_trans_ail.c | 17 ++++++++++++++++-
3 files changed, 22 insertions(+), 2 deletions(-)
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index 26c32bc..f2f22a8 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -57,6 +57,7 @@ static unsigned int xfs_errortag_random_default[] = {
XFS_RANDOM_AG_RESV_CRITICAL,
XFS_RANDOM_DROP_WRITES,
XFS_RANDOM_LOG_BAD_CRC,
+ XFS_RANDOM_LOG_ITEM_PIN,
};
struct xfs_errortag_attr {
@@ -161,6 +162,7 @@ XFS_ERRORTAG_ATTR_RW(bmap_finish_one, XFS_ERRTAG_BMAP_FINISH_ONE);
XFS_ERRORTAG_ATTR_RW(ag_resv_critical, XFS_ERRTAG_AG_RESV_CRITICAL);
XFS_ERRORTAG_ATTR_RW(drop_writes, XFS_ERRTAG_DROP_WRITES);
XFS_ERRORTAG_ATTR_RW(log_bad_crc, XFS_ERRTAG_LOG_BAD_CRC);
+XFS_ERRORTAG_ATTR_RW(log_item_pin, XFS_ERRTAG_LOG_ITEM_PIN);
static struct attribute *xfs_errortag_attrs[] = {
XFS_ERRORTAG_ATTR_LIST(noerror),
@@ -193,6 +195,7 @@ static struct attribute *xfs_errortag_attrs[] = {
XFS_ERRORTAG_ATTR_LIST(ag_resv_critical),
XFS_ERRORTAG_ATTR_LIST(drop_writes),
XFS_ERRORTAG_ATTR_LIST(log_bad_crc),
+ XFS_ERRORTAG_ATTR_LIST(log_item_pin),
NULL,
};
diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h
index 0874d6b..43d349c 100644
--- a/fs/xfs/xfs_error.h
+++ b/fs/xfs/xfs_error.h
@@ -107,8 +107,9 @@ extern void xfs_verifier_error(struct xfs_buf *bp);
*/
#define XFS_ERRTAG_DROP_WRITES 28
#define XFS_ERRTAG_LOG_BAD_CRC 29
+#define XFS_ERRTAG_LOG_ITEM_PIN 30
-#define XFS_ERRTAG_MAX 30
+#define XFS_ERRTAG_MAX 31
/*
* Random factors for above tags, 1 means always, 2 means 1/2 time, etc.
@@ -143,6 +144,7 @@ extern void xfs_verifier_error(struct xfs_buf *bp);
#define XFS_RANDOM_AG_RESV_CRITICAL 4
#define XFS_RANDOM_DROP_WRITES 1
#define XFS_RANDOM_LOG_BAD_CRC 1
+#define XFS_RANDOM_LOG_ITEM_PIN 1
#ifdef DEBUG
extern int xfs_errortag_init(struct xfs_mount *mp);
diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c
index 9056c0f..563b315 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -325,6 +325,21 @@ xfs_ail_delete(
xfs_trans_ail_cursor_clear(ailp, lip);
}
+static inline uint
+xfsaild_push_item(
+ struct xfs_ail *ailp,
+ struct xfs_log_item *lip)
+{
+ /*
+ * If log item pinning is enabled, skip the push and track the item as
+ * pinned. This can help induce head-behind-tail conditions.
+ */
+ if (XFS_TEST_ERROR(false, ailp->xa_mount, XFS_ERRTAG_LOG_ITEM_PIN))
+ return XFS_ITEM_PINNED;
+
+ return lip->li_ops->iop_push(lip, &ailp->xa_buf_list);
+}
+
static long
xfsaild_push(
struct xfs_ail *ailp)
@@ -382,7 +397,7 @@ xfsaild_push(
* rely on the AIL cursor implementation to be able to deal with
* the dropped lock.
*/
- lock_result = lip->li_ops->iop_push(lip, &ailp->xa_buf_list);
+ lock_result = xfsaild_push_item(ailp, lip);
switch (lock_result) {
case XFS_ITEM_SUCCESS:
XFS_STATS_INC(mp, xs_push_ail_success);
--
2.7.5
next prev parent reply other threads:[~2017-06-27 14:40 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-27 14:40 [PATCH 0/4] xfs: log recovery wrap and tail overwrite fixes Brian Foster
2017-06-27 14:40 ` [PATCH 1/4] xfs: fix recovery failure when log record header wraps log end Brian Foster
2017-07-01 4:38 ` Darrick J. Wong
2017-07-03 12:11 ` Brian Foster
2017-06-27 14:40 ` [PATCH 2/4] xfs: always verify the log tail during recovery Brian Foster
2017-07-01 4:43 ` Darrick J. Wong
2017-07-03 12:11 ` Brian Foster
2017-06-27 14:40 ` [PATCH 3/4] xfs: fix log recovery corruption error due to tail overwrite Brian Foster
2017-07-01 5:06 ` Darrick J. Wong
2017-07-03 12:13 ` Brian Foster
2017-07-03 16:27 ` Brian Foster
2017-07-03 16:39 ` Darrick J. Wong
2017-06-27 14:40 ` Brian Foster [this message]
2017-07-01 3:03 ` [PATCH 4/4] xfs: add log item pinning error injection tag Darrick J. Wong
2017-06-27 14:50 ` [PATCH] tests/xfs: test for log recovery failure after tail overwrite Brian Foster
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1498574436-57561-5-git-send-email-bfoster@redhat.com \
--to=bfoster@redhat.com \
--cc=linux-xfs@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).