* [PATCH] md/raid10: fix deadlock with check operation and nowait requests
@ 2026-02-10 5:09 Josh Hunt
2026-02-10 10:17 ` kernel test robot
` (2 more replies)
0 siblings, 3 replies; 5+ messages in thread
From: Josh Hunt @ 2026-02-10 5:09 UTC (permalink / raw)
To: song, yukuai, linan122, linux-raid; +Cc: ncroxon, Josh Hunt, stable
When an array check is running it will raise the barrier at which point
normal requests will become blocked and increment the nr_pending value to
signal there is work pending inside of wait_barrier(). NOWAIT requests
do not block and so will return immediately with an error, and additionally
do not increment nr_pending in wait_barrier(). Upstream change
43806c3d5b9b ("raid10: cleanup memleak at raid10_make_request") added a
call to raid_end_bio_io() to fix a memory leak when NOWAIT requests hit
this condition. raid_end_bio_io() eventually calls allow_barrier() and
it will unconditionally do an atomic_dec_and_test(&conf->nr_pending) even
though the corresponding increment on nr_pending didn't happen in the
NOWAIT case.
This can be easily seen by starting a check operation while an application
is doing nowait IO on the same array. This results in a deadlocked state
due to nr_pending value underflowing and so the md resync thread gets
stuck waiting for nr_pending to == 0.
Output of r10conf state of the array when we hit this condition:
crash> struct r10conf.barrier,nr_pending,nr_waiting,nr_queued <addr of r10conf>
barrier = 1,
nr_pending = {
counter = -41
},
nr_waiting = 15,
nr_queued = 0,
Example of md_sync thread stuck waiting on raise_barrier() and other
requests stuck in wait_barrier():
md1_resync
[<0>] raise_barrier+0xce/0x1c0
[<0>] raid10_sync_request+0x1ca/0x1ed0
[<0>] md_do_sync+0x779/0x1110
[<0>] md_thread+0x90/0x160
[<0>] kthread+0xbe/0xf0
[<0>] ret_from_fork+0x34/0x50
[<0>] ret_from_fork_asm+0x1a/0x30
kworker/u1040:2+flush-253:4
[<0>] wait_barrier+0x1de/0x220
[<0>] regular_request_wait+0x30/0x180
[<0>] raid10_make_request+0x261/0x1000
[<0>] md_handle_request+0x13b/0x230
[<0>] __submit_bio+0x107/0x1f0
[<0>] submit_bio_noacct_nocheck+0x16f/0x390
[<0>] ext4_io_submit+0x24/0x40
[<0>] ext4_do_writepages+0x254/0xc80
[<0>] ext4_writepages+0x84/0x120
[<0>] do_writepages+0x7a/0x260
[<0>] __writeback_single_inode+0x3d/0x300
[<0>] writeback_sb_inodes+0x1dd/0x470
[<0>] __writeback_inodes_wb+0x4c/0xe0
[<0>] wb_writeback+0x18b/0x2d0
[<0>] wb_workfn+0x2a1/0x400
[<0>] process_one_work+0x149/0x330
[<0>] worker_thread+0x2d2/0x410
[<0>] kthread+0xbe/0xf0
[<0>] ret_from_fork+0x34/0x50
[<0>] ret_from_fork_asm+0x1a/0x30
Fixes: 43806c3d5b9b ("raid10: cleanup memleak at raid10_make_request")
Cc: stable@vger.kernel.org
Signed-off-by: Josh Hunt <johunt@akamai.com>
---
drivers/md/raid10.c | 36 +++++++++++++++++++++++++-----------
1 file changed, 25 insertions(+), 11 deletions(-)
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 9debb20cf129..184b5b3906d1 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -68,6 +68,7 @@
*/
static void allow_barrier(struct r10conf *conf);
+static void allow_barrier_nowait(struct r10conf *conf);
static void lower_barrier(struct r10conf *conf);
static int _enough(struct r10conf *conf, int previous, int ignore);
static int enough(struct r10conf *conf, int ignore);
@@ -317,7 +318,7 @@ static void reschedule_retry(struct r10bio *r10_bio)
* operation and are ready to return a success/failure code to the buffer
* cache layer.
*/
-static void raid_end_bio_io(struct r10bio *r10_bio)
+static void raid_end_bio_io(struct r10bio *r10_bio, bool adjust_pending)
{
struct bio *bio = r10_bio->master_bio;
struct r10conf *conf = r10_bio->mddev->private;
@@ -332,7 +333,10 @@ static void raid_end_bio_io(struct r10bio *r10_bio)
* Wake up any possible resync thread that waits for the device
* to go idle.
*/
- allow_barrier(conf);
+ if (adjust_pending)
+ allow_barrier(conf);
+ else
+ allow_barrier_nowait(conf);
free_r10bio(r10_bio);
}
@@ -414,7 +418,7 @@ static void raid10_end_read_request(struct bio *bio)
uptodate = 1;
}
if (uptodate) {
- raid_end_bio_io(r10_bio);
+ raid_end_bio_io(r10_bio, true);
rdev_dec_pending(rdev, conf->mddev);
} else {
/*
@@ -446,7 +450,7 @@ static void one_write_done(struct r10bio *r10_bio)
if (test_bit(R10BIO_MadeGood, &r10_bio->state))
reschedule_retry(r10_bio);
else
- raid_end_bio_io(r10_bio);
+ raid_end_bio_io(r10_bio, true);
}
}
}
@@ -1030,13 +1034,23 @@ static bool wait_barrier(struct r10conf *conf, bool nowait)
return ret;
}
-static void allow_barrier(struct r10conf *conf)
+static void __allow_barrier(struct r10conf *conf, bool adjust_pending)
{
- if ((atomic_dec_and_test(&conf->nr_pending)) ||
+ if ((adjust_pending && atomic_dec_and_test(&conf->nr_pending)) ||
(conf->array_freeze_pending))
wake_up_barrier(conf);
}
+static void allow_barrier(struct r10conf *conf)
+{
+ __allow_barrier(conf, true);
+}
+
+static void allow_barrier_nowait(struct r10conf *conf)
+{
+ __allow_barrier(conf, false);
+}
+
static void freeze_array(struct r10conf *conf, int extra)
{
/* stop syncio and normal IO and wait for everything to
@@ -1184,7 +1198,7 @@ static void raid10_read_request(struct mddev *mddev, struct bio *bio,
}
if (!regular_request_wait(mddev, conf, bio, r10_bio->sectors)) {
- raid_end_bio_io(r10_bio);
+ raid_end_bio_io(r10_bio, false);
return;
}
@@ -1195,7 +1209,7 @@ static void raid10_read_request(struct mddev *mddev, struct bio *bio,
mdname(mddev), b,
(unsigned long long)r10_bio->sector);
}
- raid_end_bio_io(r10_bio);
+ raid_end_bio_io(r10_bio, true);
return;
}
if (err_rdev)
@@ -1372,7 +1386,7 @@ static void raid10_write_request(struct mddev *mddev, struct bio *bio,
sectors = r10_bio->sectors;
if (!regular_request_wait(mddev, conf, bio, sectors)) {
- raid_end_bio_io(r10_bio);
+ raid_end_bio_io(r10_bio, false);
return;
}
@@ -2952,7 +2966,7 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio)
if (test_bit(R10BIO_WriteError,
&r10_bio->state))
close_write(r10_bio);
- raid_end_bio_io(r10_bio);
+ raid_end_bio_io(r10_bio, true);
}
}
}
@@ -2987,7 +3001,7 @@ static void raid10d(struct md_thread *thread)
if (test_bit(R10BIO_WriteError,
&r10_bio->state))
close_write(r10_bio);
- raid_end_bio_io(r10_bio);
+ raid_end_bio_io(r10_bio, true);
}
}
--
2.34.1
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] md/raid10: fix deadlock with check operation and nowait requests
2026-02-10 5:09 [PATCH] md/raid10: fix deadlock with check operation and nowait requests Josh Hunt
@ 2026-02-10 10:17 ` kernel test robot
2026-02-10 10:18 ` kernel test robot
2026-02-10 11:14 ` kernel test robot
2 siblings, 0 replies; 5+ messages in thread
From: kernel test robot @ 2026-02-10 10:17 UTC (permalink / raw)
To: Josh Hunt, song, yukuai, linan122, linux-raid
Cc: oe-kbuild-all, ncroxon, Josh Hunt, stable
Hi Josh,
kernel test robot noticed the following build errors:
[auto build test ERROR on linus/master]
[also build test ERROR on v6.19 next-20260209]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Josh-Hunt/md-raid10-fix-deadlock-with-check-operation-and-nowait-requests/20260210-135305
base: linus/master
patch link: https://lore.kernel.org/r/20260210050942.3731656-1-johunt%40akamai.com
patch subject: [PATCH] md/raid10: fix deadlock with check operation and nowait requests
config: nios2-allmodconfig (https://download.01.org/0day-ci/archive/20260210/202602101844.0pRyZv4D-lkp@intel.com/config)
compiler: nios2-linux-gcc (GCC) 11.5.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20260210/202602101844.0pRyZv4D-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202602101844.0pRyZv4D-lkp@intel.com/
All errors (new ones prefixed by >>):
drivers/md/raid10.c: In function 'raid10_read_request':
>> drivers/md/raid10.c:1257:9: error: too few arguments to function 'raid_end_bio_io'
1257 | raid_end_bio_io(r10_bio);
| ^~~~~~~~~~~~~~~
drivers/md/raid10.c:321:13: note: declared here
321 | static void raid_end_bio_io(struct r10bio *r10_bio, bool adjust_pending)
| ^~~~~~~~~~~~~~~
drivers/md/raid10.c: In function 'raid10_write_request':
drivers/md/raid10.c:1540:9: error: too few arguments to function 'raid_end_bio_io'
1540 | raid_end_bio_io(r10_bio);
| ^~~~~~~~~~~~~~~
drivers/md/raid10.c:321:13: note: declared here
321 | static void raid_end_bio_io(struct r10bio *r10_bio, bool adjust_pending)
| ^~~~~~~~~~~~~~~
vim +/raid_end_bio_io +1257 drivers/md/raid10.c
caea3c47ad5152 Guoqing Jiang 2018-12-07 1161
bb5f1ed70bc3bb Robert LeBlanc 2016-12-05 1162 static void raid10_read_request(struct mddev *mddev, struct bio *bio,
820455238366a7 Yu Kuai 2023-06-22 1163 struct r10bio *r10_bio, bool io_accounting)
^1da177e4c3f41 Linus Torvalds 2005-04-16 1164 {
e879a8793f915a NeilBrown 2011-10-11 1165 struct r10conf *conf = mddev->private;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1166 struct bio *read_bio;
d4432c23be957f NeilBrown 2011-07-28 1167 int max_sectors;
bb5f1ed70bc3bb Robert LeBlanc 2016-12-05 1168 struct md_rdev *rdev;
545250f2480911 NeilBrown 2017-04-05 1169 char b[BDEVNAME_SIZE];
545250f2480911 NeilBrown 2017-04-05 1170 int slot = r10_bio->read_slot;
545250f2480911 NeilBrown 2017-04-05 1171 struct md_rdev *err_rdev = NULL;
545250f2480911 NeilBrown 2017-04-05 1172 gfp_t gfp = GFP_NOIO;
9b622e2bbcf049 Tomasz Majchrzak 2016-07-28 1173
93decc563637c4 Kevin Vigor 2020-11-06 1174 if (slot >= 0 && r10_bio->devs[slot].rdev) {
545250f2480911 NeilBrown 2017-04-05 1175 /*
545250f2480911 NeilBrown 2017-04-05 1176 * This is an error retry, but we cannot
545250f2480911 NeilBrown 2017-04-05 1177 * safely dereference the rdev in the r10_bio,
545250f2480911 NeilBrown 2017-04-05 1178 * we must use the one in conf.
545250f2480911 NeilBrown 2017-04-05 1179 * If it has already been disconnected (unlikely)
545250f2480911 NeilBrown 2017-04-05 1180 * we lose the device name in error messages.
545250f2480911 NeilBrown 2017-04-05 1181 */
545250f2480911 NeilBrown 2017-04-05 1182 int disk;
545250f2480911 NeilBrown 2017-04-05 1183 /*
545250f2480911 NeilBrown 2017-04-05 1184 * As we are blocking raid10, it is a little safer to
545250f2480911 NeilBrown 2017-04-05 1185 * use __GFP_HIGH.
545250f2480911 NeilBrown 2017-04-05 1186 */
545250f2480911 NeilBrown 2017-04-05 1187 gfp = GFP_NOIO | __GFP_HIGH;
545250f2480911 NeilBrown 2017-04-05 1188
545250f2480911 NeilBrown 2017-04-05 1189 disk = r10_bio->devs[slot].devnum;
a448af25becf4b Yu Kuai 2023-11-25 1190 err_rdev = conf->mirrors[disk].rdev;
545250f2480911 NeilBrown 2017-04-05 1191 if (err_rdev)
900d156bac2bc4 Christoph Hellwig 2022-07-13 1192 snprintf(b, sizeof(b), "%pg", err_rdev->bdev);
545250f2480911 NeilBrown 2017-04-05 1193 else {
545250f2480911 NeilBrown 2017-04-05 1194 strcpy(b, "???");
545250f2480911 NeilBrown 2017-04-05 1195 /* This never gets dereferenced */
545250f2480911 NeilBrown 2017-04-05 1196 err_rdev = r10_bio->devs[slot].rdev;
545250f2480911 NeilBrown 2017-04-05 1197 }
545250f2480911 NeilBrown 2017-04-05 1198 }
856e08e23762df NeilBrown 2011-07-28 1199
43806c3d5b9bb7 Nigel Croxon 2025-07-03 1200 if (!regular_request_wait(mddev, conf, bio, r10_bio->sectors)) {
4e9814d1943b0e Josh Hunt 2026-02-10 1201 raid_end_bio_io(r10_bio, false);
c9aa889b035fca Vishal Verma 2021-12-21 1202 return;
43806c3d5b9bb7 Nigel Croxon 2025-07-03 1203 }
43806c3d5b9bb7 Nigel Croxon 2025-07-03 1204
96c3fd1f380237 NeilBrown 2011-12-23 1205 rdev = read_balance(conf, r10_bio, &max_sectors);
96c3fd1f380237 NeilBrown 2011-12-23 1206 if (!rdev) {
545250f2480911 NeilBrown 2017-04-05 1207 if (err_rdev) {
545250f2480911 NeilBrown 2017-04-05 1208 pr_crit_ratelimited("md/raid10:%s: %s: unrecoverable I/O read error for block %llu\n",
545250f2480911 NeilBrown 2017-04-05 1209 mdname(mddev), b,
545250f2480911 NeilBrown 2017-04-05 1210 (unsigned long long)r10_bio->sector);
545250f2480911 NeilBrown 2017-04-05 1211 }
4e9814d1943b0e Josh Hunt 2026-02-10 1212 raid_end_bio_io(r10_bio, true);
5a7bbad27a4103 Christoph Hellwig 2011-09-12 1213 return;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1214 }
545250f2480911 NeilBrown 2017-04-05 1215 if (err_rdev)
913cce5a1e588e Christoph Hellwig 2022-05-12 1216 pr_err_ratelimited("md/raid10:%s: %pg: redirecting sector %llu to another mirror\n",
545250f2480911 NeilBrown 2017-04-05 1217 mdname(mddev),
913cce5a1e588e Christoph Hellwig 2022-05-12 1218 rdev->bdev,
545250f2480911 NeilBrown 2017-04-05 1219 (unsigned long long)r10_bio->sector);
fc9977dd069e4f NeilBrown 2017-04-05 1220 if (max_sectors < bio_sectors(bio)) {
e820d55cb99dd9 Guoqing Jiang 2018-12-19 1221 allow_barrier(conf);
6fc07785d9b892 Yu Kuai 2025-09-10 1222 bio = bio_submit_split_bioset(bio, max_sectors,
6fc07785d9b892 Yu Kuai 2025-09-10 1223 &conf->bio_split);
c9aa889b035fca Vishal Verma 2021-12-21 1224 wait_barrier(conf, false);
6fc07785d9b892 Yu Kuai 2025-09-10 1225 if (!bio) {
6fc07785d9b892 Yu Kuai 2025-09-10 1226 set_bit(R10BIO_Returned, &r10_bio->state);
4cf58d95290973 John Garry 2024-11-11 1227 goto err_handle;
4cf58d95290973 John Garry 2024-11-11 1228 }
22f166218f7313 Yu Kuai 2025-09-10 1229
fc9977dd069e4f NeilBrown 2017-04-05 1230 r10_bio->master_bio = bio;
fc9977dd069e4f NeilBrown 2017-04-05 1231 r10_bio->sectors = max_sectors;
fc9977dd069e4f NeilBrown 2017-04-05 1232 }
96c3fd1f380237 NeilBrown 2011-12-23 1233 slot = r10_bio->read_slot;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1234
820455238366a7 Yu Kuai 2023-06-22 1235 if (io_accounting) {
820455238366a7 Yu Kuai 2023-06-22 1236 md_account_bio(mddev, &bio);
820455238366a7 Yu Kuai 2023-06-22 1237 r10_bio->master_bio = bio;
820455238366a7 Yu Kuai 2023-06-22 1238 }
abfc426d1b2fb2 Christoph Hellwig 2022-02-02 1239 read_bio = bio_alloc_clone(rdev->bdev, bio, gfp, &mddev->bio_set);
5fa31c49928139 Zheng Qixing 2025-07-02 1240 read_bio->bi_opf &= ~REQ_NOWAIT;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1241
^1da177e4c3f41 Linus Torvalds 2005-04-16 1242 r10_bio->devs[slot].bio = read_bio;
abbf098e6e1e23 NeilBrown 2011-12-23 1243 r10_bio->devs[slot].rdev = rdev;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1244
4f024f3797c43c Kent Overstreet 2013-10-11 1245 read_bio->bi_iter.bi_sector = r10_bio->devs[slot].addr +
f8c9e74ff0832f NeilBrown 2012-05-21 1246 choose_data_offset(r10_bio, rdev);
^1da177e4c3f41 Linus Torvalds 2005-04-16 1247 read_bio->bi_end_io = raid10_end_read_request;
8d3ca83dcf9ca3 NeilBrown 2016-11-18 1248 if (test_bit(FailFast, &rdev->flags) &&
8d3ca83dcf9ca3 NeilBrown 2016-11-18 1249 test_bit(R10BIO_FailFast, &r10_bio->state))
8d3ca83dcf9ca3 NeilBrown 2016-11-18 1250 read_bio->bi_opf |= MD_FAILFAST;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1251 read_bio->bi_private = r10_bio;
c396b90e502691 Christoph Hellwig 2024-03-03 1252 mddev_trace_remap(mddev, read_bio, r10_bio->sector);
ed00aabd5eb9fb Christoph Hellwig 2020-07-01 1253 submit_bio_noacct(read_bio);
5a7bbad27a4103 Christoph Hellwig 2011-09-12 1254 return;
4cf58d95290973 John Garry 2024-11-11 1255 err_handle:
4cf58d95290973 John Garry 2024-11-11 1256 atomic_dec(&rdev->nr_pending);
4cf58d95290973 John Garry 2024-11-11 @1257 raid_end_bio_io(r10_bio);
^1da177e4c3f41 Linus Torvalds 2005-04-16 1258 }
^1da177e4c3f41 Linus Torvalds 2005-04-16 1259
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] md/raid10: fix deadlock with check operation and nowait requests
2026-02-10 5:09 [PATCH] md/raid10: fix deadlock with check operation and nowait requests Josh Hunt
2026-02-10 10:17 ` kernel test robot
@ 2026-02-10 10:18 ` kernel test robot
2026-02-10 11:14 ` kernel test robot
2 siblings, 0 replies; 5+ messages in thread
From: kernel test robot @ 2026-02-10 10:18 UTC (permalink / raw)
To: Josh Hunt, song, yukuai, linan122, linux-raid
Cc: oe-kbuild-all, ncroxon, Josh Hunt, stable
Hi Josh,
kernel test robot noticed the following build errors:
[auto build test ERROR on linus/master]
[also build test ERROR on v6.19 next-20260209]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Josh-Hunt/md-raid10-fix-deadlock-with-check-operation-and-nowait-requests/20260210-135305
base: linus/master
patch link: https://lore.kernel.org/r/20260210050942.3731656-1-johunt%40akamai.com
patch subject: [PATCH] md/raid10: fix deadlock with check operation and nowait requests
config: m68k-defconfig (https://download.01.org/0day-ci/archive/20260210/202602101850.DC3BeMD5-lkp@intel.com/config)
compiler: m68k-linux-gcc (GCC) 15.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20260210/202602101850.DC3BeMD5-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202602101850.DC3BeMD5-lkp@intel.com/
All errors (new ones prefixed by >>):
drivers/md/raid10.c: In function 'raid10_read_request':
>> drivers/md/raid10.c:1257:9: error: too few arguments to function 'raid_end_bio_io'; expected 2, have 1
1257 | raid_end_bio_io(r10_bio);
| ^~~~~~~~~~~~~~~
drivers/md/raid10.c:321:13: note: declared here
321 | static void raid_end_bio_io(struct r10bio *r10_bio, bool adjust_pending)
| ^~~~~~~~~~~~~~~
drivers/md/raid10.c: In function 'raid10_write_request':
drivers/md/raid10.c:1540:9: error: too few arguments to function 'raid_end_bio_io'; expected 2, have 1
1540 | raid_end_bio_io(r10_bio);
| ^~~~~~~~~~~~~~~
drivers/md/raid10.c:321:13: note: declared here
321 | static void raid_end_bio_io(struct r10bio *r10_bio, bool adjust_pending)
| ^~~~~~~~~~~~~~~
vim +/raid_end_bio_io +1257 drivers/md/raid10.c
caea3c47ad5152 Guoqing Jiang 2018-12-07 1161
bb5f1ed70bc3bb Robert LeBlanc 2016-12-05 1162 static void raid10_read_request(struct mddev *mddev, struct bio *bio,
820455238366a7 Yu Kuai 2023-06-22 1163 struct r10bio *r10_bio, bool io_accounting)
^1da177e4c3f41 Linus Torvalds 2005-04-16 1164 {
e879a8793f915a NeilBrown 2011-10-11 1165 struct r10conf *conf = mddev->private;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1166 struct bio *read_bio;
d4432c23be957f NeilBrown 2011-07-28 1167 int max_sectors;
bb5f1ed70bc3bb Robert LeBlanc 2016-12-05 1168 struct md_rdev *rdev;
545250f2480911 NeilBrown 2017-04-05 1169 char b[BDEVNAME_SIZE];
545250f2480911 NeilBrown 2017-04-05 1170 int slot = r10_bio->read_slot;
545250f2480911 NeilBrown 2017-04-05 1171 struct md_rdev *err_rdev = NULL;
545250f2480911 NeilBrown 2017-04-05 1172 gfp_t gfp = GFP_NOIO;
9b622e2bbcf049 Tomasz Majchrzak 2016-07-28 1173
93decc563637c4 Kevin Vigor 2020-11-06 1174 if (slot >= 0 && r10_bio->devs[slot].rdev) {
545250f2480911 NeilBrown 2017-04-05 1175 /*
545250f2480911 NeilBrown 2017-04-05 1176 * This is an error retry, but we cannot
545250f2480911 NeilBrown 2017-04-05 1177 * safely dereference the rdev in the r10_bio,
545250f2480911 NeilBrown 2017-04-05 1178 * we must use the one in conf.
545250f2480911 NeilBrown 2017-04-05 1179 * If it has already been disconnected (unlikely)
545250f2480911 NeilBrown 2017-04-05 1180 * we lose the device name in error messages.
545250f2480911 NeilBrown 2017-04-05 1181 */
545250f2480911 NeilBrown 2017-04-05 1182 int disk;
545250f2480911 NeilBrown 2017-04-05 1183 /*
545250f2480911 NeilBrown 2017-04-05 1184 * As we are blocking raid10, it is a little safer to
545250f2480911 NeilBrown 2017-04-05 1185 * use __GFP_HIGH.
545250f2480911 NeilBrown 2017-04-05 1186 */
545250f2480911 NeilBrown 2017-04-05 1187 gfp = GFP_NOIO | __GFP_HIGH;
545250f2480911 NeilBrown 2017-04-05 1188
545250f2480911 NeilBrown 2017-04-05 1189 disk = r10_bio->devs[slot].devnum;
a448af25becf4b Yu Kuai 2023-11-25 1190 err_rdev = conf->mirrors[disk].rdev;
545250f2480911 NeilBrown 2017-04-05 1191 if (err_rdev)
900d156bac2bc4 Christoph Hellwig 2022-07-13 1192 snprintf(b, sizeof(b), "%pg", err_rdev->bdev);
545250f2480911 NeilBrown 2017-04-05 1193 else {
545250f2480911 NeilBrown 2017-04-05 1194 strcpy(b, "???");
545250f2480911 NeilBrown 2017-04-05 1195 /* This never gets dereferenced */
545250f2480911 NeilBrown 2017-04-05 1196 err_rdev = r10_bio->devs[slot].rdev;
545250f2480911 NeilBrown 2017-04-05 1197 }
545250f2480911 NeilBrown 2017-04-05 1198 }
856e08e23762df NeilBrown 2011-07-28 1199
43806c3d5b9bb7 Nigel Croxon 2025-07-03 1200 if (!regular_request_wait(mddev, conf, bio, r10_bio->sectors)) {
4e9814d1943b0e Josh Hunt 2026-02-10 1201 raid_end_bio_io(r10_bio, false);
c9aa889b035fca Vishal Verma 2021-12-21 1202 return;
43806c3d5b9bb7 Nigel Croxon 2025-07-03 1203 }
43806c3d5b9bb7 Nigel Croxon 2025-07-03 1204
96c3fd1f380237 NeilBrown 2011-12-23 1205 rdev = read_balance(conf, r10_bio, &max_sectors);
96c3fd1f380237 NeilBrown 2011-12-23 1206 if (!rdev) {
545250f2480911 NeilBrown 2017-04-05 1207 if (err_rdev) {
545250f2480911 NeilBrown 2017-04-05 1208 pr_crit_ratelimited("md/raid10:%s: %s: unrecoverable I/O read error for block %llu\n",
545250f2480911 NeilBrown 2017-04-05 1209 mdname(mddev), b,
545250f2480911 NeilBrown 2017-04-05 1210 (unsigned long long)r10_bio->sector);
545250f2480911 NeilBrown 2017-04-05 1211 }
4e9814d1943b0e Josh Hunt 2026-02-10 1212 raid_end_bio_io(r10_bio, true);
5a7bbad27a4103 Christoph Hellwig 2011-09-12 1213 return;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1214 }
545250f2480911 NeilBrown 2017-04-05 1215 if (err_rdev)
913cce5a1e588e Christoph Hellwig 2022-05-12 1216 pr_err_ratelimited("md/raid10:%s: %pg: redirecting sector %llu to another mirror\n",
545250f2480911 NeilBrown 2017-04-05 1217 mdname(mddev),
913cce5a1e588e Christoph Hellwig 2022-05-12 1218 rdev->bdev,
545250f2480911 NeilBrown 2017-04-05 1219 (unsigned long long)r10_bio->sector);
fc9977dd069e4f NeilBrown 2017-04-05 1220 if (max_sectors < bio_sectors(bio)) {
e820d55cb99dd9 Guoqing Jiang 2018-12-19 1221 allow_barrier(conf);
6fc07785d9b892 Yu Kuai 2025-09-10 1222 bio = bio_submit_split_bioset(bio, max_sectors,
6fc07785d9b892 Yu Kuai 2025-09-10 1223 &conf->bio_split);
c9aa889b035fca Vishal Verma 2021-12-21 1224 wait_barrier(conf, false);
6fc07785d9b892 Yu Kuai 2025-09-10 1225 if (!bio) {
6fc07785d9b892 Yu Kuai 2025-09-10 1226 set_bit(R10BIO_Returned, &r10_bio->state);
4cf58d95290973 John Garry 2024-11-11 1227 goto err_handle;
4cf58d95290973 John Garry 2024-11-11 1228 }
22f166218f7313 Yu Kuai 2025-09-10 1229
fc9977dd069e4f NeilBrown 2017-04-05 1230 r10_bio->master_bio = bio;
fc9977dd069e4f NeilBrown 2017-04-05 1231 r10_bio->sectors = max_sectors;
fc9977dd069e4f NeilBrown 2017-04-05 1232 }
96c3fd1f380237 NeilBrown 2011-12-23 1233 slot = r10_bio->read_slot;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1234
820455238366a7 Yu Kuai 2023-06-22 1235 if (io_accounting) {
820455238366a7 Yu Kuai 2023-06-22 1236 md_account_bio(mddev, &bio);
820455238366a7 Yu Kuai 2023-06-22 1237 r10_bio->master_bio = bio;
820455238366a7 Yu Kuai 2023-06-22 1238 }
abfc426d1b2fb2 Christoph Hellwig 2022-02-02 1239 read_bio = bio_alloc_clone(rdev->bdev, bio, gfp, &mddev->bio_set);
5fa31c49928139 Zheng Qixing 2025-07-02 1240 read_bio->bi_opf &= ~REQ_NOWAIT;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1241
^1da177e4c3f41 Linus Torvalds 2005-04-16 1242 r10_bio->devs[slot].bio = read_bio;
abbf098e6e1e23 NeilBrown 2011-12-23 1243 r10_bio->devs[slot].rdev = rdev;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1244
4f024f3797c43c Kent Overstreet 2013-10-11 1245 read_bio->bi_iter.bi_sector = r10_bio->devs[slot].addr +
f8c9e74ff0832f NeilBrown 2012-05-21 1246 choose_data_offset(r10_bio, rdev);
^1da177e4c3f41 Linus Torvalds 2005-04-16 1247 read_bio->bi_end_io = raid10_end_read_request;
8d3ca83dcf9ca3 NeilBrown 2016-11-18 1248 if (test_bit(FailFast, &rdev->flags) &&
8d3ca83dcf9ca3 NeilBrown 2016-11-18 1249 test_bit(R10BIO_FailFast, &r10_bio->state))
8d3ca83dcf9ca3 NeilBrown 2016-11-18 1250 read_bio->bi_opf |= MD_FAILFAST;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1251 read_bio->bi_private = r10_bio;
c396b90e502691 Christoph Hellwig 2024-03-03 1252 mddev_trace_remap(mddev, read_bio, r10_bio->sector);
ed00aabd5eb9fb Christoph Hellwig 2020-07-01 1253 submit_bio_noacct(read_bio);
5a7bbad27a4103 Christoph Hellwig 2011-09-12 1254 return;
4cf58d95290973 John Garry 2024-11-11 1255 err_handle:
4cf58d95290973 John Garry 2024-11-11 1256 atomic_dec(&rdev->nr_pending);
4cf58d95290973 John Garry 2024-11-11 @1257 raid_end_bio_io(r10_bio);
^1da177e4c3f41 Linus Torvalds 2005-04-16 1258 }
^1da177e4c3f41 Linus Torvalds 2005-04-16 1259
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] md/raid10: fix deadlock with check operation and nowait requests
2026-02-10 5:09 [PATCH] md/raid10: fix deadlock with check operation and nowait requests Josh Hunt
2026-02-10 10:17 ` kernel test robot
2026-02-10 10:18 ` kernel test robot
@ 2026-02-10 11:14 ` kernel test robot
2026-02-10 11:56 ` Josh Hunt
2 siblings, 1 reply; 5+ messages in thread
From: kernel test robot @ 2026-02-10 11:14 UTC (permalink / raw)
To: Josh Hunt, song, yukuai, linan122, linux-raid
Cc: oe-kbuild-all, ncroxon, Josh Hunt, stable
Hi Josh,
kernel test robot noticed the following build errors:
[auto build test ERROR on linus/master]
[also build test ERROR on v6.19 next-20260209]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Josh-Hunt/md-raid10-fix-deadlock-with-check-operation-and-nowait-requests/20260210-135305
base: linus/master
patch link: https://lore.kernel.org/r/20260210050942.3731656-1-johunt%40akamai.com
patch subject: [PATCH] md/raid10: fix deadlock with check operation and nowait requests
config: x86_64-rhel-9.4 (https://download.01.org/0day-ci/archive/20260210/202602101220.J4BofeDD-lkp@intel.com/config)
compiler: gcc-14 (Debian 14.2.0-19) 14.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20260210/202602101220.J4BofeDD-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202602101220.J4BofeDD-lkp@intel.com/
All errors (new ones prefixed by >>):
drivers/md/raid10.c: In function 'raid10_read_request':
>> drivers/md/raid10.c:1257:9: error: too few arguments to function 'raid_end_bio_io'
1257 | raid_end_bio_io(r10_bio);
| ^~~~~~~~~~~~~~~
drivers/md/raid10.c:321:13: note: declared here
321 | static void raid_end_bio_io(struct r10bio *r10_bio, bool adjust_pending)
| ^~~~~~~~~~~~~~~
drivers/md/raid10.c: In function 'raid10_write_request':
drivers/md/raid10.c:1540:9: error: too few arguments to function 'raid_end_bio_io'
1540 | raid_end_bio_io(r10_bio);
| ^~~~~~~~~~~~~~~
drivers/md/raid10.c:321:13: note: declared here
321 | static void raid_end_bio_io(struct r10bio *r10_bio, bool adjust_pending)
| ^~~~~~~~~~~~~~~
vim +/raid_end_bio_io +1257 drivers/md/raid10.c
caea3c47ad5152 Guoqing Jiang 2018-12-07 1161
bb5f1ed70bc3bb Robert LeBlanc 2016-12-05 1162 static void raid10_read_request(struct mddev *mddev, struct bio *bio,
820455238366a7 Yu Kuai 2023-06-22 1163 struct r10bio *r10_bio, bool io_accounting)
^1da177e4c3f41 Linus Torvalds 2005-04-16 1164 {
e879a8793f915a NeilBrown 2011-10-11 1165 struct r10conf *conf = mddev->private;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1166 struct bio *read_bio;
d4432c23be957f NeilBrown 2011-07-28 1167 int max_sectors;
bb5f1ed70bc3bb Robert LeBlanc 2016-12-05 1168 struct md_rdev *rdev;
545250f2480911 NeilBrown 2017-04-05 1169 char b[BDEVNAME_SIZE];
545250f2480911 NeilBrown 2017-04-05 1170 int slot = r10_bio->read_slot;
545250f2480911 NeilBrown 2017-04-05 1171 struct md_rdev *err_rdev = NULL;
545250f2480911 NeilBrown 2017-04-05 1172 gfp_t gfp = GFP_NOIO;
9b622e2bbcf049 Tomasz Majchrzak 2016-07-28 1173
93decc563637c4 Kevin Vigor 2020-11-06 1174 if (slot >= 0 && r10_bio->devs[slot].rdev) {
545250f2480911 NeilBrown 2017-04-05 1175 /*
545250f2480911 NeilBrown 2017-04-05 1176 * This is an error retry, but we cannot
545250f2480911 NeilBrown 2017-04-05 1177 * safely dereference the rdev in the r10_bio,
545250f2480911 NeilBrown 2017-04-05 1178 * we must use the one in conf.
545250f2480911 NeilBrown 2017-04-05 1179 * If it has already been disconnected (unlikely)
545250f2480911 NeilBrown 2017-04-05 1180 * we lose the device name in error messages.
545250f2480911 NeilBrown 2017-04-05 1181 */
545250f2480911 NeilBrown 2017-04-05 1182 int disk;
545250f2480911 NeilBrown 2017-04-05 1183 /*
545250f2480911 NeilBrown 2017-04-05 1184 * As we are blocking raid10, it is a little safer to
545250f2480911 NeilBrown 2017-04-05 1185 * use __GFP_HIGH.
545250f2480911 NeilBrown 2017-04-05 1186 */
545250f2480911 NeilBrown 2017-04-05 1187 gfp = GFP_NOIO | __GFP_HIGH;
545250f2480911 NeilBrown 2017-04-05 1188
545250f2480911 NeilBrown 2017-04-05 1189 disk = r10_bio->devs[slot].devnum;
a448af25becf4b Yu Kuai 2023-11-25 1190 err_rdev = conf->mirrors[disk].rdev;
545250f2480911 NeilBrown 2017-04-05 1191 if (err_rdev)
900d156bac2bc4 Christoph Hellwig 2022-07-13 1192 snprintf(b, sizeof(b), "%pg", err_rdev->bdev);
545250f2480911 NeilBrown 2017-04-05 1193 else {
545250f2480911 NeilBrown 2017-04-05 1194 strcpy(b, "???");
545250f2480911 NeilBrown 2017-04-05 1195 /* This never gets dereferenced */
545250f2480911 NeilBrown 2017-04-05 1196 err_rdev = r10_bio->devs[slot].rdev;
545250f2480911 NeilBrown 2017-04-05 1197 }
545250f2480911 NeilBrown 2017-04-05 1198 }
856e08e23762df NeilBrown 2011-07-28 1199
43806c3d5b9bb7 Nigel Croxon 2025-07-03 1200 if (!regular_request_wait(mddev, conf, bio, r10_bio->sectors)) {
4e9814d1943b0e Josh Hunt 2026-02-10 1201 raid_end_bio_io(r10_bio, false);
c9aa889b035fca Vishal Verma 2021-12-21 1202 return;
43806c3d5b9bb7 Nigel Croxon 2025-07-03 1203 }
43806c3d5b9bb7 Nigel Croxon 2025-07-03 1204
96c3fd1f380237 NeilBrown 2011-12-23 1205 rdev = read_balance(conf, r10_bio, &max_sectors);
96c3fd1f380237 NeilBrown 2011-12-23 1206 if (!rdev) {
545250f2480911 NeilBrown 2017-04-05 1207 if (err_rdev) {
545250f2480911 NeilBrown 2017-04-05 1208 pr_crit_ratelimited("md/raid10:%s: %s: unrecoverable I/O read error for block %llu\n",
545250f2480911 NeilBrown 2017-04-05 1209 mdname(mddev), b,
545250f2480911 NeilBrown 2017-04-05 1210 (unsigned long long)r10_bio->sector);
545250f2480911 NeilBrown 2017-04-05 1211 }
4e9814d1943b0e Josh Hunt 2026-02-10 1212 raid_end_bio_io(r10_bio, true);
5a7bbad27a4103 Christoph Hellwig 2011-09-12 1213 return;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1214 }
545250f2480911 NeilBrown 2017-04-05 1215 if (err_rdev)
913cce5a1e588e Christoph Hellwig 2022-05-12 1216 pr_err_ratelimited("md/raid10:%s: %pg: redirecting sector %llu to another mirror\n",
545250f2480911 NeilBrown 2017-04-05 1217 mdname(mddev),
913cce5a1e588e Christoph Hellwig 2022-05-12 1218 rdev->bdev,
545250f2480911 NeilBrown 2017-04-05 1219 (unsigned long long)r10_bio->sector);
fc9977dd069e4f NeilBrown 2017-04-05 1220 if (max_sectors < bio_sectors(bio)) {
e820d55cb99dd9 Guoqing Jiang 2018-12-19 1221 allow_barrier(conf);
6fc07785d9b892 Yu Kuai 2025-09-10 1222 bio = bio_submit_split_bioset(bio, max_sectors,
6fc07785d9b892 Yu Kuai 2025-09-10 1223 &conf->bio_split);
c9aa889b035fca Vishal Verma 2021-12-21 1224 wait_barrier(conf, false);
6fc07785d9b892 Yu Kuai 2025-09-10 1225 if (!bio) {
6fc07785d9b892 Yu Kuai 2025-09-10 1226 set_bit(R10BIO_Returned, &r10_bio->state);
4cf58d95290973 John Garry 2024-11-11 1227 goto err_handle;
4cf58d95290973 John Garry 2024-11-11 1228 }
22f166218f7313 Yu Kuai 2025-09-10 1229
fc9977dd069e4f NeilBrown 2017-04-05 1230 r10_bio->master_bio = bio;
fc9977dd069e4f NeilBrown 2017-04-05 1231 r10_bio->sectors = max_sectors;
fc9977dd069e4f NeilBrown 2017-04-05 1232 }
96c3fd1f380237 NeilBrown 2011-12-23 1233 slot = r10_bio->read_slot;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1234
820455238366a7 Yu Kuai 2023-06-22 1235 if (io_accounting) {
820455238366a7 Yu Kuai 2023-06-22 1236 md_account_bio(mddev, &bio);
820455238366a7 Yu Kuai 2023-06-22 1237 r10_bio->master_bio = bio;
820455238366a7 Yu Kuai 2023-06-22 1238 }
abfc426d1b2fb2 Christoph Hellwig 2022-02-02 1239 read_bio = bio_alloc_clone(rdev->bdev, bio, gfp, &mddev->bio_set);
5fa31c49928139 Zheng Qixing 2025-07-02 1240 read_bio->bi_opf &= ~REQ_NOWAIT;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1241
^1da177e4c3f41 Linus Torvalds 2005-04-16 1242 r10_bio->devs[slot].bio = read_bio;
abbf098e6e1e23 NeilBrown 2011-12-23 1243 r10_bio->devs[slot].rdev = rdev;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1244
4f024f3797c43c Kent Overstreet 2013-10-11 1245 read_bio->bi_iter.bi_sector = r10_bio->devs[slot].addr +
f8c9e74ff0832f NeilBrown 2012-05-21 1246 choose_data_offset(r10_bio, rdev);
^1da177e4c3f41 Linus Torvalds 2005-04-16 1247 read_bio->bi_end_io = raid10_end_read_request;
8d3ca83dcf9ca3 NeilBrown 2016-11-18 1248 if (test_bit(FailFast, &rdev->flags) &&
8d3ca83dcf9ca3 NeilBrown 2016-11-18 1249 test_bit(R10BIO_FailFast, &r10_bio->state))
8d3ca83dcf9ca3 NeilBrown 2016-11-18 1250 read_bio->bi_opf |= MD_FAILFAST;
^1da177e4c3f41 Linus Torvalds 2005-04-16 1251 read_bio->bi_private = r10_bio;
c396b90e502691 Christoph Hellwig 2024-03-03 1252 mddev_trace_remap(mddev, read_bio, r10_bio->sector);
ed00aabd5eb9fb Christoph Hellwig 2020-07-01 1253 submit_bio_noacct(read_bio);
5a7bbad27a4103 Christoph Hellwig 2011-09-12 1254 return;
4cf58d95290973 John Garry 2024-11-11 1255 err_handle:
4cf58d95290973 John Garry 2024-11-11 1256 atomic_dec(&rdev->nr_pending);
4cf58d95290973 John Garry 2024-11-11 @1257 raid_end_bio_io(r10_bio);
^1da177e4c3f41 Linus Torvalds 2005-04-16 1258 }
^1da177e4c3f41 Linus Torvalds 2005-04-16 1259
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] md/raid10: fix deadlock with check operation and nowait requests
2026-02-10 11:14 ` kernel test robot
@ 2026-02-10 11:56 ` Josh Hunt
0 siblings, 0 replies; 5+ messages in thread
From: Josh Hunt @ 2026-02-10 11:56 UTC (permalink / raw)
To: kernel test robot, song, yukuai, linan122, linux-raid
Cc: oe-kbuild-all, ncroxon, stable
On 2/10/26 3:14 AM, kernel test robot wrote:
> !-------------------------------------------------------------------|
> This Message Is From an External Sender
> This message came from outside your organization.
> |-------------------------------------------------------------------!
>
> Hi Josh,
>
> kernel test robot noticed the following build errors:
>
> [auto build test ERROR on linus/master]
> [also build test ERROR on v6.19 next-20260209]
> [If your patch is applied to the wrong git tree, kindly drop us a note.
> And when submitting patch, we suggest to use '--base' as documented in
> https://urldefense.com/v3/__https://git-scm.com/docs/git-format-patch*_base_tree_information__;Iw!!GjvTz_vk!QT9JklLw7YgSInuzDS_0EDSnJkZTTG057ef2CiWO2UwIN0aUk9RGNuVtQ6XCy8sTbfadURM$ ]
>
> url: https://urldefense.com/v3/__https://github.com/intel-lab-lkp/linux/commits/Josh-Hunt/md-raid10-fix-deadlock-with-check-operation-and-nowait-requests/20260210-135305__;!!GjvTz_vk!QT9JklLw7YgSInuzDS_0EDSnJkZTTG057ef2CiWO2UwIN0aUk9RGNuVtQ6XCy8sTilUfWlw$
> base: linus/master
> patch link: https://urldefense.com/v3/__https://lore.kernel.org/r/20260210050942.3731656-1-johunt*40akamai.com__;JQ!!GjvTz_vk!QT9JklLw7YgSInuzDS_0EDSnJkZTTG057ef2CiWO2UwIN0aUk9RGNuVtQ6XCy8sT0QqCm3Y$
> patch subject: [PATCH] md/raid10: fix deadlock with check operation and nowait requests
> config: x86_64-rhel-9.4 (https://urldefense.com/v3/__https://download.01.org/0day-ci/archive/20260210/202602101220.J4BofeDD-lkp@intel.com/config__;!!GjvTz_vk!QT9JklLw7YgSInuzDS_0EDSnJkZTTG057ef2CiWO2UwIN0aUk9RGNuVtQ6XCy8sTZp6wg4c$ )
> compiler: gcc-14 (Debian 14.2.0-19) 14.2.0
> reproduce (this is a W=1 build): (https://urldefense.com/v3/__https://download.01.org/0day-ci/archive/20260210/202602101220.J4BofeDD-lkp@intel.com/reproduce__;!!GjvTz_vk!QT9JklLw7YgSInuzDS_0EDSnJkZTTG057ef2CiWO2UwIN0aUk9RGNuVtQ6XCy8sTFU3cM60$ )
>
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@intel.com>
> | Closes: https://urldefense.com/v3/__https://lore.kernel.org/oe-kbuild-all/202602101220.J4BofeDD-lkp@intel.com/__;!!GjvTz_vk!QT9JklLw7YgSInuzDS_0EDSnJkZTTG057ef2CiWO2UwIN0aUk9RGNuVtQ6XCy8sTUPVdYDI$
>
> All errors (new ones prefixed by >>):
>
> drivers/md/raid10.c: In function 'raid10_read_request':
>>> drivers/md/raid10.c:1257:9: error: too few arguments to function 'raid_end_bio_io'
> 1257 | raid_end_bio_io(r10_bio);
> | ^~~~~~~~~~~~~~~
> drivers/md/raid10.c:321:13: note: declared here
> 321 | static void raid_end_bio_io(struct r10bio *r10_bio, bool adjust_pending)
> | ^~~~~~~~~~~~~~~
> drivers/md/raid10.c: In function 'raid10_write_request':
> drivers/md/raid10.c:1540:9: error: too few arguments to function 'raid_end_bio_io'
> 1540 | raid_end_bio_io(r10_bio);
> | ^~~~~~~~~~~~~~~
> drivers/md/raid10.c:321:13: note: declared here
> 321 | static void raid_end_bio_io(struct r10bio *r10_bio, bool adjust_pending)
> | ^~~~~~~~~~~~~~~
>
>
> vim +/raid_end_bio_io +1257 drivers/md/raid10.c
>
> caea3c47ad5152 Guoqing Jiang 2018-12-07 1161
> bb5f1ed70bc3bb Robert LeBlanc 2016-12-05 1162 static void raid10_read_request(struct mddev *mddev, struct bio *bio,
> 820455238366a7 Yu Kuai 2023-06-22 1163 struct r10bio *r10_bio, bool io_accounting)
> ^1da177e4c3f41 Linus Torvalds 2005-04-16 1164 {
> e879a8793f915a NeilBrown 2011-10-11 1165 struct r10conf *conf = mddev->private;
> ^1da177e4c3f41 Linus Torvalds 2005-04-16 1166 struct bio *read_bio;
> d4432c23be957f NeilBrown 2011-07-28 1167 int max_sectors;
> bb5f1ed70bc3bb Robert LeBlanc 2016-12-05 1168 struct md_rdev *rdev;
> 545250f2480911 NeilBrown 2017-04-05 1169 char b[BDEVNAME_SIZE];
> 545250f2480911 NeilBrown 2017-04-05 1170 int slot = r10_bio->read_slot;
> 545250f2480911 NeilBrown 2017-04-05 1171 struct md_rdev *err_rdev = NULL;
> 545250f2480911 NeilBrown 2017-04-05 1172 gfp_t gfp = GFP_NOIO;
> 9b622e2bbcf049 Tomasz Majchrzak 2016-07-28 1173
> 93decc563637c4 Kevin Vigor 2020-11-06 1174 if (slot >= 0 && r10_bio->devs[slot].rdev) {
> 545250f2480911 NeilBrown 2017-04-05 1175 /*
> 545250f2480911 NeilBrown 2017-04-05 1176 * This is an error retry, but we cannot
> 545250f2480911 NeilBrown 2017-04-05 1177 * safely dereference the rdev in the r10_bio,
> 545250f2480911 NeilBrown 2017-04-05 1178 * we must use the one in conf.
> 545250f2480911 NeilBrown 2017-04-05 1179 * If it has already been disconnected (unlikely)
> 545250f2480911 NeilBrown 2017-04-05 1180 * we lose the device name in error messages.
> 545250f2480911 NeilBrown 2017-04-05 1181 */
> 545250f2480911 NeilBrown 2017-04-05 1182 int disk;
> 545250f2480911 NeilBrown 2017-04-05 1183 /*
> 545250f2480911 NeilBrown 2017-04-05 1184 * As we are blocking raid10, it is a little safer to
> 545250f2480911 NeilBrown 2017-04-05 1185 * use __GFP_HIGH.
> 545250f2480911 NeilBrown 2017-04-05 1186 */
> 545250f2480911 NeilBrown 2017-04-05 1187 gfp = GFP_NOIO | __GFP_HIGH;
> 545250f2480911 NeilBrown 2017-04-05 1188
> 545250f2480911 NeilBrown 2017-04-05 1189 disk = r10_bio->devs[slot].devnum;
> a448af25becf4b Yu Kuai 2023-11-25 1190 err_rdev = conf->mirrors[disk].rdev;
> 545250f2480911 NeilBrown 2017-04-05 1191 if (err_rdev)
> 900d156bac2bc4 Christoph Hellwig 2022-07-13 1192 snprintf(b, sizeof(b), "%pg", err_rdev->bdev);
> 545250f2480911 NeilBrown 2017-04-05 1193 else {
> 545250f2480911 NeilBrown 2017-04-05 1194 strcpy(b, "???");
> 545250f2480911 NeilBrown 2017-04-05 1195 /* This never gets dereferenced */
> 545250f2480911 NeilBrown 2017-04-05 1196 err_rdev = r10_bio->devs[slot].rdev;
> 545250f2480911 NeilBrown 2017-04-05 1197 }
> 545250f2480911 NeilBrown 2017-04-05 1198 }
> 856e08e23762df NeilBrown 2011-07-28 1199
> 43806c3d5b9bb7 Nigel Croxon 2025-07-03 1200 if (!regular_request_wait(mddev, conf, bio, r10_bio->sectors)) {
> 4e9814d1943b0e Josh Hunt 2026-02-10 1201 raid_end_bio_io(r10_bio, false);
> c9aa889b035fca Vishal Verma 2021-12-21 1202 return;
> 43806c3d5b9bb7 Nigel Croxon 2025-07-03 1203 }
> 43806c3d5b9bb7 Nigel Croxon 2025-07-03 1204
> 96c3fd1f380237 NeilBrown 2011-12-23 1205 rdev = read_balance(conf, r10_bio, &max_sectors);
> 96c3fd1f380237 NeilBrown 2011-12-23 1206 if (!rdev) {
> 545250f2480911 NeilBrown 2017-04-05 1207 if (err_rdev) {
> 545250f2480911 NeilBrown 2017-04-05 1208 pr_crit_ratelimited("md/raid10:%s: %s: unrecoverable I/O read error for block %llu\n",
> 545250f2480911 NeilBrown 2017-04-05 1209 mdname(mddev), b,
> 545250f2480911 NeilBrown 2017-04-05 1210 (unsigned long long)r10_bio->sector);
> 545250f2480911 NeilBrown 2017-04-05 1211 }
> 4e9814d1943b0e Josh Hunt 2026-02-10 1212 raid_end_bio_io(r10_bio, true);
> 5a7bbad27a4103 Christoph Hellwig 2011-09-12 1213 return;
> ^1da177e4c3f41 Linus Torvalds 2005-04-16 1214 }
> 545250f2480911 NeilBrown 2017-04-05 1215 if (err_rdev)
> 913cce5a1e588e Christoph Hellwig 2022-05-12 1216 pr_err_ratelimited("md/raid10:%s: %pg: redirecting sector %llu to another mirror\n",
> 545250f2480911 NeilBrown 2017-04-05 1217 mdname(mddev),
> 913cce5a1e588e Christoph Hellwig 2022-05-12 1218 rdev->bdev,
> 545250f2480911 NeilBrown 2017-04-05 1219 (unsigned long long)r10_bio->sector);
> fc9977dd069e4f NeilBrown 2017-04-05 1220 if (max_sectors < bio_sectors(bio)) {
> e820d55cb99dd9 Guoqing Jiang 2018-12-19 1221 allow_barrier(conf);
> 6fc07785d9b892 Yu Kuai 2025-09-10 1222 bio = bio_submit_split_bioset(bio, max_sectors,
> 6fc07785d9b892 Yu Kuai 2025-09-10 1223 &conf->bio_split);
> c9aa889b035fca Vishal Verma 2021-12-21 1224 wait_barrier(conf, false);
> 6fc07785d9b892 Yu Kuai 2025-09-10 1225 if (!bio) {
> 6fc07785d9b892 Yu Kuai 2025-09-10 1226 set_bit(R10BIO_Returned, &r10_bio->state);
> 4cf58d95290973 John Garry 2024-11-11 1227 goto err_handle;
> 4cf58d95290973 John Garry 2024-11-11 1228 }
> 22f166218f7313 Yu Kuai 2025-09-10 1229
> fc9977dd069e4f NeilBrown 2017-04-05 1230 r10_bio->master_bio = bio;
> fc9977dd069e4f NeilBrown 2017-04-05 1231 r10_bio->sectors = max_sectors;
> fc9977dd069e4f NeilBrown 2017-04-05 1232 }
> 96c3fd1f380237 NeilBrown 2011-12-23 1233 slot = r10_bio->read_slot;
> ^1da177e4c3f41 Linus Torvalds 2005-04-16 1234
> 820455238366a7 Yu Kuai 2023-06-22 1235 if (io_accounting) {
> 820455238366a7 Yu Kuai 2023-06-22 1236 md_account_bio(mddev, &bio);
> 820455238366a7 Yu Kuai 2023-06-22 1237 r10_bio->master_bio = bio;
> 820455238366a7 Yu Kuai 2023-06-22 1238 }
> abfc426d1b2fb2 Christoph Hellwig 2022-02-02 1239 read_bio = bio_alloc_clone(rdev->bdev, bio, gfp, &mddev->bio_set);
> 5fa31c49928139 Zheng Qixing 2025-07-02 1240 read_bio->bi_opf &= ~REQ_NOWAIT;
> ^1da177e4c3f41 Linus Torvalds 2005-04-16 1241
> ^1da177e4c3f41 Linus Torvalds 2005-04-16 1242 r10_bio->devs[slot].bio = read_bio;
> abbf098e6e1e23 NeilBrown 2011-12-23 1243 r10_bio->devs[slot].rdev = rdev;
> ^1da177e4c3f41 Linus Torvalds 2005-04-16 1244
> 4f024f3797c43c Kent Overstreet 2013-10-11 1245 read_bio->bi_iter.bi_sector = r10_bio->devs[slot].addr +
> f8c9e74ff0832f NeilBrown 2012-05-21 1246 choose_data_offset(r10_bio, rdev);
> ^1da177e4c3f41 Linus Torvalds 2005-04-16 1247 read_bio->bi_end_io = raid10_end_read_request;
> 8d3ca83dcf9ca3 NeilBrown 2016-11-18 1248 if (test_bit(FailFast, &rdev->flags) &&
> 8d3ca83dcf9ca3 NeilBrown 2016-11-18 1249 test_bit(R10BIO_FailFast, &r10_bio->state))
> 8d3ca83dcf9ca3 NeilBrown 2016-11-18 1250 read_bio->bi_opf |= MD_FAILFAST;
> ^1da177e4c3f41 Linus Torvalds 2005-04-16 1251 read_bio->bi_private = r10_bio;
> c396b90e502691 Christoph Hellwig 2024-03-03 1252 mddev_trace_remap(mddev, read_bio, r10_bio->sector);
> ed00aabd5eb9fb Christoph Hellwig 2020-07-01 1253 submit_bio_noacct(read_bio);
> 5a7bbad27a4103 Christoph Hellwig 2011-09-12 1254 return;
> 4cf58d95290973 John Garry 2024-11-11 1255 err_handle:
> 4cf58d95290973 John Garry 2024-11-11 1256 atomic_dec(&rdev->nr_pending);
> 4cf58d95290973 John Garry 2024-11-11 @1257 raid_end_bio_io(r10_bio);
> ^1da177e4c3f41 Linus Torvalds 2005-04-16 1258 }
> ^1da177e4c3f41 Linus Torvalds 2005-04-16 1259
>
Apologies. I cherry-picked this from my 6.12.y branch which is where we
hit the issue, but looks like I forgot to build test it. Will send a v2.
Josh
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2026-02-10 12:47 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-02-10 5:09 [PATCH] md/raid10: fix deadlock with check operation and nowait requests Josh Hunt
2026-02-10 10:17 ` kernel test robot
2026-02-10 10:18 ` kernel test robot
2026-02-10 11:14 ` kernel test robot
2026-02-10 11:56 ` Josh Hunt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox