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  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]
Date:   Thu, 28 Feb 2019 10:08:13 -0500
From:   Sasha Levin <sashal@...nel.org>
To:     linux-kernel@...r.kernel.org, stable@...r.kernel.org
Cc:     Liu Bo <bo.liu@...ux.alibaba.com>, Jens Axboe <axboe@...nel.dk>,
        Sasha Levin <sashal@...nel.org>, linux-block@...r.kernel.org
Subject: [PATCH AUTOSEL 4.20 81/81] blk-iolatency: fix IO hang due to negative inflight counter

From: Liu Bo <bo.liu@...ux.alibaba.com>

[ Upstream commit 8c772a9bfc7c07c76f4a58b58910452fbb20843b ]

Our test reported the following stack, and vmcore showed that
->inflight counter is -1.

[ffffc9003fcc38d0] __schedule at ffffffff8173d95d
[ffffc9003fcc3958] schedule at ffffffff8173de26
[ffffc9003fcc3970] io_schedule at ffffffff810bb6b6
[ffffc9003fcc3988] blkcg_iolatency_throttle at ffffffff813911cb
[ffffc9003fcc3a20] rq_qos_throttle at ffffffff813847f3
[ffffc9003fcc3a48] blk_mq_make_request at ffffffff8137468a
[ffffc9003fcc3b08] generic_make_request at ffffffff81368b49
[ffffc9003fcc3b68] submit_bio at ffffffff81368d7d
[ffffc9003fcc3bb8] ext4_io_submit at ffffffffa031be00 [ext4]
[ffffc9003fcc3c00] ext4_writepages at ffffffffa03163de [ext4]
[ffffc9003fcc3d68] do_writepages at ffffffff811c49ae
[ffffc9003fcc3d78] __filemap_fdatawrite_range at ffffffff811b6188
[ffffc9003fcc3e30] filemap_write_and_wait_range at ffffffff811b6301
[ffffc9003fcc3e60] ext4_sync_file at ffffffffa030cee8 [ext4]
[ffffc9003fcc3ea8] vfs_fsync_range at ffffffff8128594b
[ffffc9003fcc3ee8] do_fsync at ffffffff81285abd
[ffffc9003fcc3f18] sys_fsync at ffffffff81285d50
[ffffc9003fcc3f28] do_syscall_64 at ffffffff81003c04
[ffffc9003fcc3f50] entry_SYSCALL_64_after_swapgs at ffffffff81742b8e

The ->inflight counter may be negative (-1) if

1) blk-iolatency was disabled when the IO was issued,

2) blk-iolatency was enabled before this IO reached its endio,

3) the ->inflight counter is decreased from 0 to -1 in endio()

In fact the hang can be easily reproduced by the below script,

H=/sys/fs/cgroup/unified/
P=/sys/fs/cgroup/unified/test

echo "+io" > $H/cgroup.subtree_control
mkdir -p $P

echo $$ > $P/cgroup.procs

xfs_io -f -d -c "pwrite 0 4k" /dev/sdg

echo "`cat /sys/block/sdg/dev` target=1000000" > $P/io.latency

xfs_io -f -d -c "pwrite 0 4k" /dev/sdg

This fixes the problem by freezing the queue so that while
enabling/disabling iolatency, there is no inflight rq running.

Note that quiesce_queue is not needed as this only updating iolatency
configuration about which dispatching request_queue doesn't care.

Signed-off-by: Liu Bo <bo.liu@...ux.alibaba.com>
Signed-off-by: Jens Axboe <axboe@...nel.dk>
Signed-off-by: Sasha Levin <sashal@...nel.org>
---
 block/blk-iolatency.c | 52 +++++++++++++++++++++++++++++++++++++------
 1 file changed, 45 insertions(+), 7 deletions(-)

diff --git a/block/blk-iolatency.c b/block/blk-iolatency.c
index 38c35c32aff2d..c1c72b42dda0a 100644
--- a/block/blk-iolatency.c
+++ b/block/blk-iolatency.c
@@ -72,6 +72,7 @@
 #include <linux/sched/loadavg.h>
 #include <linux/sched/signal.h>
 #include <trace/events/block.h>
+#include <linux/blk-mq.h>
 #include "blk-rq-qos.h"
 #include "blk-stat.h"
 
@@ -648,6 +649,9 @@ static void blkcg_iolatency_done_bio(struct rq_qos *rqos, struct bio *bio)
 		return;
 
 	enabled = blk_iolatency_enabled(iolat->blkiolat);
+	if (!enabled)
+		return;
+
 	while (blkg && blkg->parent) {
 		iolat = blkg_to_lat(blkg);
 		if (!iolat) {
@@ -657,7 +661,7 @@ static void blkcg_iolatency_done_bio(struct rq_qos *rqos, struct bio *bio)
 		rqw = &iolat->rq_wait;
 
 		atomic_dec(&rqw->inflight);
-		if (!enabled || iolat->min_lat_nsec == 0)
+		if (iolat->min_lat_nsec == 0)
 			goto next;
 		iolatency_record_time(iolat, &bio->bi_issue, now,
 				      issue_as_root);
@@ -801,10 +805,13 @@ int blk_iolatency_init(struct request_queue *q)
 	return 0;
 }
 
-static void iolatency_set_min_lat_nsec(struct blkcg_gq *blkg, u64 val)
+/*
+ * return 1 for enabling iolatency, return -1 for disabling iolatency, otherwise
+ * return 0.
+ */
+static int iolatency_set_min_lat_nsec(struct blkcg_gq *blkg, u64 val)
 {
 	struct iolatency_grp *iolat = blkg_to_lat(blkg);
-	struct blk_iolatency *blkiolat = iolat->blkiolat;
 	u64 oldval = iolat->min_lat_nsec;
 
 	iolat->min_lat_nsec = val;
@@ -813,9 +820,10 @@ static void iolatency_set_min_lat_nsec(struct blkcg_gq *blkg, u64 val)
 				    BLKIOLATENCY_MAX_WIN_SIZE);
 
 	if (!oldval && val)
-		atomic_inc(&blkiolat->enabled);
+		return 1;
 	if (oldval && !val)
-		atomic_dec(&blkiolat->enabled);
+		return -1;
+	return 0;
 }
 
 static void iolatency_clear_scaling(struct blkcg_gq *blkg)
@@ -847,6 +855,7 @@ static ssize_t iolatency_set_limit(struct kernfs_open_file *of, char *buf,
 	u64 lat_val = 0;
 	u64 oldval;
 	int ret;
+	int enable = 0;
 
 	ret = blkg_conf_prep(blkcg, &blkcg_policy_iolatency, buf, &ctx);
 	if (ret)
@@ -881,7 +890,12 @@ static ssize_t iolatency_set_limit(struct kernfs_open_file *of, char *buf,
 	blkg = ctx.blkg;
 	oldval = iolat->min_lat_nsec;
 
-	iolatency_set_min_lat_nsec(blkg, lat_val);
+	enable = iolatency_set_min_lat_nsec(blkg, lat_val);
+	if (enable) {
+		WARN_ON_ONCE(!blk_get_queue(blkg->q));
+		blkg_get(blkg);
+	}
+
 	if (oldval != iolat->min_lat_nsec) {
 		iolatency_clear_scaling(blkg);
 	}
@@ -889,6 +903,24 @@ static ssize_t iolatency_set_limit(struct kernfs_open_file *of, char *buf,
 	ret = 0;
 out:
 	blkg_conf_finish(&ctx);
+	if (ret == 0 && enable) {
+		struct iolatency_grp *tmp = blkg_to_lat(blkg);
+		struct blk_iolatency *blkiolat = tmp->blkiolat;
+
+		blk_mq_freeze_queue(blkg->q);
+
+		if (enable == 1)
+			atomic_inc(&blkiolat->enabled);
+		else if (enable == -1)
+			atomic_dec(&blkiolat->enabled);
+		else
+			WARN_ON_ONCE(1);
+
+		blk_mq_unfreeze_queue(blkg->q);
+
+		blkg_put(blkg);
+		blk_put_queue(blkg->q);
+	}
 	return ret ?: nbytes;
 }
 
@@ -1024,8 +1056,14 @@ static void iolatency_pd_offline(struct blkg_policy_data *pd)
 {
 	struct iolatency_grp *iolat = pd_to_lat(pd);
 	struct blkcg_gq *blkg = lat_to_blkg(iolat);
+	struct blk_iolatency *blkiolat = iolat->blkiolat;
+	int ret;
 
-	iolatency_set_min_lat_nsec(blkg, 0);
+	ret = iolatency_set_min_lat_nsec(blkg, 0);
+	if (ret == 1)
+		atomic_inc(&blkiolat->enabled);
+	if (ret == -1)
+		atomic_dec(&blkiolat->enabled);
 	iolatency_clear_scaling(blkg);
 }
 
-- 
2.19.1

Powered by blists - more mailing lists