Re: [LKP] [lkp-robot] [mm] 9092c71bb7: blogbench.write_score -12.3% regression
by Huang, Ying
Josef Bacik <josef(a)toxicpanda.com> writes:
> On Thu, Aug 02, 2018 at 01:55:23PM +0800, Huang, Ying wrote:
>> "Huang, Ying" <ying.huang(a)intel.com> writes:
>>
>> > Hi, Chris,
>> >
>> > Chris Mason <clm(a)fb.com> writes:
>> >
>> >> On 19 Jun 2018, at 23:51, Huang, Ying wrote:
>> >>>>> "Huang, Ying" <ying.huang(a)intel.com> writes:
>> >>>>>
>> >>>>>> Hi, Josef,
>> >>>>>>
>> >>>>>> Do you have time to take a look at the regression?
>> >>>>>>
>> >>>>>> kernel test robot <xiaolong.ye(a)intel.com> writes:
>> >>>>>>
>> >>>>>>> Greeting,
>> >>>>>>>
>> >>>>>>> FYI, we noticed a -12.3% regression of blogbench.write_score and
>> >>>>>>> a +9.6% improvement
>> >>>>>>> of blogbench.read_score due to commit:
>> >>>>>>>
>> >>>>>>>
>> >>>>>>> commit: 9092c71bb724dba2ecba849eae69e5c9d39bd3d2 ("mm: use
>> >>>>>>> sc->priority for slab shrink targets")
>> >>>>>>> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git
>> >>>>>>> master
>> >>>>>>>
>> >>>>>>> in testcase: blogbench
>> >>>>>>> on test machine: 16 threads Intel(R) Xeon(R) CPU D-1541 @
>> >>>>>>> 2.10GHz with 8G memory
>> >>>>>>> with following parameters:
>> >>>>>>>
>> >>>>>>> disk: 1SSD
>> >>>>>>> fs: btrfs
>> >>>>>>> cpufreq_governor: performance
>> >>>>>>>
>> >>>>>>> test-description: Blogbench is a portable filesystem benchmark
>> >>>>>>> that tries to reproduce the load of a real-world busy file
>> >>>>>>> server.
>> >>>>>>> test-url:
>> >>
>> >> I'm surprised, this patch is a big win in production here at FB. I'll
>> >> have to reproduce these results to better understand what is going on.
>> >> My first guess is that since we have fewer inodes in slab, we're
>> >> reading more inodes from disk in order to do the writes.
>> >>
>> >> But that should also make our read scores lower.
>> >
>> > Any update on this?
>>
>> Ping.
>>
>
> I can't reproduce this, and what's more it appears that blogbench doesn't use
> much memory at all. I have the slab shrinking tracepoints on and we never go
> into this code at all, so I'm pretty sure these results are bogus. How are you
> running blogbench? I'm doing blogbench -d /whatever, if I need to be doing
> something else let me know. But from what I can tell this thing uses less than
> 100m of memory, and on an 8gig of ram box we're never going to trip over this
> code. Thanks,
Thanks for looking at this. In my testing, blogbench will eat up system
memory. Please check the vmstat result attached. The SSD disk size is
about 745GB.
Best Regards,
Huang, Ying
3 years
c82c6d59af ("exec: do unshare_files after de_thread"): BUG: unable to handle kernel paging request at 6b6b6b6b
by kernel test robot
Greetings,
0day kernel testing robot got the below dmesg and the first bad commit is
https://github.com/0day-ci/linux/commits/Jeff-Layton/exec-separate-thread...
commit c82c6d59af8195d10a4956c825cb5a024cb8c169
Author: Jeff Layton <jlayton(a)kernel.org>
AuthorDate: Mon Aug 27 13:47:22 2018 -0400
Commit: 0day robot <lkp(a)intel.com>
CommitDate: Tue Aug 28 02:11:45 2018 +0800
exec: do unshare_files after de_thread
POSIX mandates that open fds and their associated file locks should be
preserved across an execve. This works, unless the process is
multithreaded at the time that execve is called.
In that case, we'll end up unsharing the files_struct but the locks will
still have their fl_owner set to the address of the old one. Eventually,
when the other threads die and the last reference to the old
files_struct is put, any POSIX locks get torn down since it looks like
a close occurred on them.
The result is that all of your open files will be intact with none of
the locks you held before execve. The simple answer to this is "use OFD
locks", but this is a nasty surprise and it violates the spec.
Fix this by doing unshare_files later during exec, after we've already
killed off the other threads in the process. This helps ensure that we
only unshare the files_struct during exec when it is truly shared with
other processes.
Note that because the unshare_files call is now done just after
de_thread, we need a mechanism to pass the displaced files_struct back
up to __do_execve_file. This is done via a new displaced_files field
inside the linux_binprm.
Cc: Eric W. Biederman <ebiederm(a)xmission.com>
Reported-by: Daniel P. Berrangé <berrange(a)redhat.com>
Signed-off-by: Jeff Layton <jlayton(a)kernel.org>
83a164d8c5 exec: delay clone(CLONE_FILES) if task associated with current files_struct is exec'ing
c82c6d59af exec: do unshare_files after de_thread
+------------------------------------------+------------+------------+
| | 83a164d8c5 | c82c6d59af |
+------------------------------------------+------------+------------+
| boot_successes | 34 | 0 |
| boot_failures | 1 | 11 |
| Mem-Info | 1 | |
| BUG:unable_to_handle_kernel | 0 | 11 |
| Oops:#[##] | 0 | 11 |
| EIP:put_files_struct | 0 | 11 |
| Kernel_panic-not_syncing:Fatal_exception | 0 | 11 |
+------------------------------------------+------------+------------+
[ 15.899443] Freeing unused kernel image memory: 1112K
[ 15.906651] Write protecting the kernel text: 16084k
[ 15.910352] Write protecting the kernel read-only data: 7288k
[ 15.913400] rodata_test: all tests were successful
[ 15.916180] Run /init as init process
[ 15.933048] BUG: unable to handle kernel paging request at 6b6b6b6b
[ 15.935746] *pde = 00000000
[ 15.938288] Oops: 0002 [#1] PREEMPT PTI
[ 15.940994] CPU: 0 PID: 1 Comm: init Tainted: G S 4.19.0-rc1-00003-gc82c6d5 #1
[ 15.944301] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014
[ 15.960693] EIP: put_files_struct+0x9/0x9d
[ 15.963422] Code: c3 b8 01 00 00 00 e8 1d 42 f5 ff 8b 9b 08 03 00 00 85 db 74 02 ff 03 e8 48 f8 ff ff 89 d8 5b 5d c3 55 89 e5 57 56 53 83 ec 08 <ff> 08 0f 85 85 00 00 00 8b 70 10 89 c3 31 ff 8d 04 fd 00 00 00 00
[ 15.970956] EAX: 6b6b6b6b EBX: 00000000 ECX: 00000006 EDX: 00000001
[ 15.987147] ESI: cccb2008 EDI: 00000001 EBP: cf455f50 ESP: cf455f3c
[ 15.989962] DS: 007b ES: 007b FS: 0000 GS: 00e0 SS: 0068 EFLAGS: 00210296
[ 15.993455] CR0: 80050033 CR2: 6b6b6b6b CR3: 0ccd8000 CR4: 000406d0
[ 15.997017] Call Trace:
[ 16.012680] __do_execve_file+0x4c3/0x5eb
[ 16.015287] do_execve+0x16/0x18
[ 16.018012] run_init_process+0x2c/0x31
[ 16.021021] ? rest_init+0xaa/0xaa
[ 16.023640] kernel_init+0x5e/0xdf
[ 16.039550] ret_from_fork+0x19/0x30
[ 16.042439] EIP: 0x47e59d80
[ 16.044798] Code: Bad RIP value.
[ 16.046921] EAX: 00000000 EBX: 00000000 ECX: 00000000 EDX: 00000000
[ 16.049419] ESI: 00000000 EDI: 00000000 EBP: 00000000 ESP: bff94040
[ 16.065177] DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 007b EFLAGS: 00000200
[ 16.068262] CR2: 000000006b6b6b6b
[ 16.070699] ---[ end trace 054462f0d032c488 ]---
[ 16.073039] EIP: put_files_struct+0x9/0x9d
# HH:MM RESULT GOOD BAD GOOD_BUT_DIRTY DIRTY_NOT_BAD
git bisect start e8dbb3d2b0c3d4a6c99e4cec461a1ca35c345be8 5b394b2ddf0347bef56e50c69a58773c94343ff3 --
git bisect bad 410931a20c8df65ca39586b6c5bf5b19adbb9684 # 16:56 B 0 11 36 11 Merge 'linux-review/Dexuan-Cui/hv_netvsc-Fix-a-deadlock-by-getting-rtnl_lock-earlier-in-netvsc_probe/20180823-105718' into devel-hourly-2018082814
git bisect bad daa847f66bac12fae037648808a73a3a31ce03f2 # 17:15 B 0 11 37 12 Merge 'linux-review/UPDATE-20180828-102007/Rob-Herring/ARM-omap-Convert-to-using-pOFn-instead-of-device_node-name/20180828-100534' into devel-hourly-2018082814
git bisect good 0655e86575fa37071b4f5312de3b609c8e5fa882 # 17:36 G 11 0 4 6 Merge 'linux-review/Colin-King/ALSA-hdspm-fix-spelling-mistake-Initializeing-Initializing/20180827-084806' into devel-hourly-2018082814
git bisect bad b37e25837b055727e0680b9ca7a59fb4b6809ffd # 17:53 B 0 10 34 10 Merge 'linux-review/UPDATE-20180828-125222/Rob-Herring/ARM-omap-Convert-to-using-pOFn-instead-of-device_node-name/20180828-100534' into devel-hourly-2018082814
git bisect good 9657b13b107d24727a600467fdff5beffd5199e9 # 18:00 G 10 0 3 3 Merge 'linux-review/Ding-Xiang/Bluetooth-bt3c_cs-Fix-obsolete-function/20180827-105841' into devel-hourly-2018082814
git bisect bad cbd532a9942da135edfce12188e98b00827916cc # 18:07 B 0 1 26 11 Merge 'linux-review/Matthias-Kaehlcke/iio-adc-qcom-spmi-adc5-Verify-channel-numbers-from-DT/20180828-051624' into devel-hourly-2018082814
git bisect bad e1b04817326da403c24b424099c3bd0d683f681c # 18:42 B 0 2 27 11 Merge 'linux-review/Vincent-Whitchurch/kmemleak-Add-option-to-print-warnings-to-dmesg/20180827-165045' into devel-hourly-2018082814
git bisect good 3762ab54ed21ac3cf3dcee5dd8d5b786a6bf357e # 18:57 G 11 0 6 8 Merge 'linux-review/UPDATE-20180828-104326/Rob-Herring/ARM-omap-Convert-to-using-pOFn-instead-of-device_node-name/20180828-100534' into devel-hourly-2018082814
git bisect good 77275292632d915316abc78e4f41d90182f94355 # 19:19 G 11 0 4 4 Merge 'linux-review/Jiri-Pirko/net-sched-fix-extack-error-message-when-chain-is-failed-to-be-created/20180828-031100' into devel-hourly-2018082814
git bisect bad 4fdc76e645ab23e73b5c617af647124edc810743 # 19:29 B 0 3 28 11 Merge 'linux-review/Jeff-Layton/exec-separate-thread_count-for-files_struct/20180828-021143' into devel-hourly-2018082814
git bisect good 83a164d8c5c69f8947147001c18016291f27978f # 19:52 G 11 0 0 7 exec: delay clone(CLONE_FILES) if task associated with current files_struct is exec'ing
git bisect bad c82c6d59af8195d10a4956c825cb5a024cb8c169 # 20:12 B 0 11 25 0 exec: do unshare_files after de_thread
# first bad commit: [c82c6d59af8195d10a4956c825cb5a024cb8c169] exec: do unshare_files after de_thread
git bisect good 83a164d8c5c69f8947147001c18016291f27978f # 20:17 G 31 0 9 16 exec: delay clone(CLONE_FILES) if task associated with current files_struct is exec'ing
# extra tests with debug options
git bisect bad c82c6d59af8195d10a4956c825cb5a024cb8c169 # 20:25 B 0 11 25 0 exec: do unshare_files after de_thread
# extra tests on HEAD of linux-devel/devel-hourly-2018082814
git bisect bad e8dbb3d2b0c3d4a6c99e4cec461a1ca35c345be8 # 20:25 B 0 21 38 0 0day head guard for 'devel-hourly-2018082814'
# extra tests on tree/branch linux-review/Jeff-Layton/exec-separate-thread_count-for-files_struct/20180828-021143
git bisect bad c82c6d59af8195d10a4956c825cb5a024cb8c169 # 20:28 B 0 11 25 0 exec: do unshare_files after de_thread
# extra tests with first bad commit reverted
git bisect good 533258b19d3044ed40b456a8e15a53f67fe2baaa # 20:43 G 11 0 1 1 Revert "exec: do unshare_files after de_thread"
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/lkp Intel Corporation
3 years
[lkp-robot] [confidence: ] 6ce3dd6eec [ 10.247585] WARNING: CPU: 0 PID: 210 at block/blk-mq.c:647 blk_mq_start_request
by kernel test robot
Greetings,
0day kernel testing robot got the below dmesg and the first bad commit is
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
commit 6ce3dd6eec114930cf2035a8bcb1e80477ed79a8
Author: Ming Lei <ming.lei(a)redhat.com>
AuthorDate: Tue Jul 10 09:03:31 2018 +0800
Commit: Jens Axboe <axboe(a)kernel.dk>
CommitDate: Tue Jul 17 16:04:00 2018 -0600
blk-mq: issue directly if hw queue isn't busy in case of 'none'
In case of 'none' io scheduler, when hw queue isn't busy, it isn't
necessary to enqueue request to sw queue and dequeue it from
sw queue because request may be submitted to hw queue asap without
extra cost, meantime there shouldn't be much request in sw queue,
and we don't need to worry about effect on IO merge.
There are still some single hw queue SCSI HBAs(HPSA, megaraid_sas, ...)
which may connect high performance devices, so 'none' is often required
for obtaining good performance.
This patch improves IOPS and decreases CPU unilization on megaraid_sas,
per Kashyap's test.
Cc: Kashyap Desai <kashyap.desai(a)broadcom.com>
Cc: Laurence Oberman <loberman(a)redhat.com>
Cc: Omar Sandoval <osandov(a)fb.com>
Cc: Christoph Hellwig <hch(a)lst.de>
Cc: Bart Van Assche <bart.vanassche(a)wdc.com>
Cc: Hannes Reinecke <hare(a)suse.de>
Reported-by: Kashyap Desai <kashyap.desai(a)broadcom.com>
Tested-by: Kashyap Desai <kashyap.desai(a)broadcom.com>
Signed-off-by: Ming Lei <ming.lei(a)redhat.com>
Signed-off-by: Jens Axboe <axboe(a)kernel.dk>
71e9690b59 blk-iolatency: truncate our current time
6ce3dd6eec blk-mq: issue directly if hw queue isn't busy in case of 'none'
89cf553533 Add linux-next specific files for 20180720
+-------------------------------------------------+------------+------------+---------------+
| | 71e9690b59 | 6ce3dd6eec | next-20180720 |
+-------------------------------------------------+------------+------------+---------------+
| boot_successes | 32 | 4 | 18 |
| boot_failures | 13 | 12 | 12 |
| BUG:workqueue_lockup-pool | 13 | 2 | 3 |
| WARNING:at_block/blk-mq.c:#blk_mq_start_request | 0 | 12 | 11 |
| RIP:blk_mq_start_request | 0 | 12 | 11 |
+-------------------------------------------------+------------+------------+---------------+
[ 10.225908] UDF-fs: error (device nbd11): udf_read_tagged: read failed, block=512, location=512
[ 10.230004] UDF-fs: error (device nbd6): udf_read_tagged: read failed, block=512, location=512
[ 10.234011] UDF-fs: error (device nbd15): udf_read_tagged: read failed, block=512, location=512
[ 10.238978] UDF-fs: error (device nbd11): udf_read_tagged: read failed, block=256, location=256
[ 10.243423] UDF-fs: error (device nbd6): udf_read_tagged: read failed, block=256, location=256
[ 10.247585] WARNING: CPU: 0 PID: 210 at block/blk-mq.c:647 blk_mq_start_request+0x57/0x84
[ 10.255769] CPU: 0 PID: 210 Comm: kworker/0:1H Not tainted 4.18.0-rc4-00068-g6ce3dd6 #1
[ 10.258705] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014
[ 10.261751] Workqueue: kblockd blk_mq_run_work_fn
[ 10.263469] RIP: 0010:blk_mq_start_request+0x57/0x84
[ 10.265231] Code: 73 1e e8 32 9c b9 ff 81 4b 18 00 00 02 00 48 89 83 b0 00 00 00 48 89 de 48 89 ef e8 4a e5 00 00 8b 83 d4 00 00 00 85 c0 74 02 <0f> 0b 48 89 df e8 65 dd ff ff c7 83 d4 00 00 00 01 00 00 00 83 bd
[ 10.271321] RSP: 0018:ffffc900002a7cb0 EFLAGS: 00010202
[ 10.273134] RAX: 0000000000000001 RBX: ffff88001c61ccc0 RCX: 00000000000000ff
[ 10.275339] RDX: 0000000000000000 RSI: ffff88001c5ae300 RDI: ffff88001c61ccc0
[ 10.277534] RBP: ffff88001c5f7008 R08: 0000000437ddd47e R09: 0000000000000000
[ 10.279729] R10: 0000000000000000 R11: ffff88001c61cd00 R12: 000000000000000a
[ 10.287969] R13: 0000000000000000 R14: ffff88001c5f7000 R15: ffff88001c308a00
[ 10.290141] FS: 0000000000000000(0000) GS:ffffffff82e98000(0000) knlGS:0000000000000000
[ 10.293062] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 10.294955] CR2: 00007f661eaba000 CR3: 0000000002e75002 CR4: 00000000000606b0
[ 10.297108] Call Trace:
[ 10.298313] nbd_queue_rq+0x105/0x3c5
[ 10.310816] ? kvm_sched_clock_read+0x5/0xd
[ 10.312427] ? __lock_acquire+0x217/0x7ae
[ 10.314125] blk_mq_dispatch_rq_list+0x2db/0x4ec
[ 10.315808] ? blk_mq_flush_busy_ctxs+0x5d/0x145
[ 10.317460] blk_mq_sched_dispatch_requests+0x127/0x165
[ 10.319212] __blk_mq_run_hw_queue+0x9d/0xc4
[ 10.320769] process_one_work+0x205/0x36f
[ 10.322277] ? process_one_work+0x1a4/0x36f
[ 10.327153] worker_thread+0x1ec/0x2bb
[ 10.329387] ? process_scheduled_works+0x27/0x27
[ 10.331067] kthread+0x117/0x11f
[ 10.332477] ? kthread_flush_work_fn+0x9/0x9
[ 10.348547] ret_from_fork+0x35/0x40
[ 10.349980] ---[ end trace 29f1a54dd1591b2f ]---
[ 10.352707] NILFS (nbd9): unable to read superblock
# HH:MM RESULT GOOD BAD GOOD_BUT_DIRTY DIRTY_NOT_BAD
git bisect start 01c75271c97333b09c56dc9faf56efab9bddd395 7daf201d7fe8334e2d2364d4e8ed3394ec9af819 --
git bisect bad 09ab3047b1a130d322a0d250fef359b1b910b2b5 # 01:45 B 0 1 15 0 Merge 'pcmoore/next' into devel-catchup-201807190533
git bisect bad d2f660d44ec337305d2015a59542b666ea098a80 # 02:09 B 2 3 2 2 Merge 'linux-review/Katsuhiro-Suzuki/media-dvb-frontends-add-Socionext-SC1501A-ISDB-S-T-demodulator-driver/20180719-045753' into devel-catchup-201807190533
git bisect good bb9f3d6e0cd4247dc8261f6045f27eef059c258c # 02:35 G 11 0 11 11 Merge 'linux-review/Ernesto-A-Fern-ndez/hfsplus-fix-decomposition-of-Hangul-characters/20180719-051928' into devel-catchup-201807190533
git bisect good 08d5773f0d3f83b5f402025cb74f1ad4f9be2c3a # 03:04 G 11 0 4 4 Merge 'vfio/for-linus' into devel-catchup-201807190533
git bisect good 1452167e29e13d331b052fb5e86af82aeb7fa30e # 03:39 G 11 0 4 4 Merge 'linux-review/Srinath-Mannam/mmc-host-iproc-Add-ACPI-support-to-IPROC-SDHCI/20180719-043256' into devel-catchup-201807190533
git bisect bad 4fa583c08a968c296b131f5bf4221b1f8a78b18c # 03:56 B 0 10 25 0 Merge 'linux-review/RAGHU-Halharvi/pktcdvd-checkpatch-remove-static-initialise-null/20180719-040038' into devel-catchup-201807190533
git bisect good b351f0c76c3eb94c9ccfb68d0b23899a35e47f27 # 04:17 G 12 0 6 6 Documentation: add a doc for blk-iolatency
git bisect good 884b031b288bae15397dd07b084a41ffb44f99e4 # 04:31 G 11 0 4 4 lightnvm: pblk: mark expected switch fall-through
git bisect bad bdca3c87fb7ad1cc61d231d37eb0d8f90d001e0c # 04:59 B 0 2 17 0 block: Track DISCARD statistics and output them in stat and diskstat
git bisect bad 6ce3dd6eec114930cf2035a8bcb1e80477ed79a8 # 05:12 B 0 6 21 0 blk-mq: issue directly if hw queue isn't busy in case of 'none'
git bisect good f6352103d2e0ad2d2066725eb19bfdfb8763239b # 05:28 G 11 0 4 4 lightnvm: pblk: assume that chunks are closed on 1.2 devices
git bisect good 71e9690b59e7349156025a514c29c29ef55b0175 # 05:48 G 11 0 5 5 blk-iolatency: truncate our current time
# first bad commit: [6ce3dd6eec114930cf2035a8bcb1e80477ed79a8] blk-mq: issue directly if hw queue isn't busy in case of 'none'
git bisect good 71e9690b59e7349156025a514c29c29ef55b0175 # 05:54 G 36 0 8 13 blk-iolatency: truncate our current time
# extra tests with debug options
git bisect bad 6ce3dd6eec114930cf2035a8bcb1e80477ed79a8 # 06:32 B 0 1 16 0 blk-mq: issue directly if hw queue isn't busy in case of 'none'
# extra tests on HEAD of linux-devel/devel-catchup-201807190533
git bisect bad 01c75271c97333b09c56dc9faf56efab9bddd395 # 06:33 B 0 319 383 46 0day head guard for 'devel-catchup-201807190533'
# extra tests on tree/branch linux-next/master
git bisect bad 89cf553533084a35b44f533d59198497d3319d69 # 07:02 B 1 1 1 1 Add linux-next specific files for 20180720
# extra tests with first bad commit reverted
git bisect good e1aef67c37bb7d5591d822552d4aff6c4b5ab0de # 07:29 G 12 0 2 2 Revert "blk-mq: issue directly if hw queue isn't busy in case of 'none'"
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/lkp Intel Corporation
3 years
[fs, elf] 4ed2863951: 32bit.hugetlb_minimal_copy=no_xb_linkhuge.killed_by_signal
by Rong Chen
FYI, we noticed the following commit (built with gcc-7):
commit 4ed28639519c7bad5f518e70b3284c6e0763e650 ("fs, elf: drop
MAP_FIXED usage from elf_map")
git://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git
tags/next-20180412~2^2~67
in testcase: libhugetlbfs-test
with following parameters:
pagesize: 2MB
pagenum: 32
on test machine: 88 threads Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz
with 64G memory
caused below changes (please refer to attached dmesg/kmsg for entire
log/backtrace):
HUGETLB_MINIMAL_COPY=no xB.linkhuge (2M: 32): <-- Killed by signal
HUGETLB_MINIMAL_COPY=no xB.linkhuge (2M: 64): PASS
HUGETLB_MINIMAL_COPY=no xBDT.linkhuge (2M: 32): PASS
HUGETLB_MINIMAL_COPY=no xBDT.linkhuge (2M: 64): PASS
HUGETLB_ELFMAP=no xB.linkhuge (2M: 32):
<-- Killed by signal
HUGETLB_ELFMAP=no xB.linkhuge (2M: 64): PASS
HUGETLB_ELFMAP=no xBDT.linkhuge (2M: 32): PASS
HUGETLB_ELFMAP=no xBDT.linkhuge (2M: 64): PASS
HUGETLB_SHARE=1 xB.linkshare (2M: 32):
<-- Killed by signal
HUGETLB_SHARE=1 xB.linkshare (2M: 64): PASS
HUGETLB_SHARE=1 xBDT.linkshare (2M: 32): PASS
HUGETLB_SHARE=1 xBDT.linkshare (2M: 64): PASS
HUGETLB_SHARE=1 xB.linkshare (2M: 32):
<-- Killed by signal
HUGETLB_SHARE=1 xB.linkshare (2M: 64): PASS
HUGETLB_SHARE=1 xBDT.linkshare (2M: 32): PASS
HUGETLB_SHARE=1 xBDT.linkshare (2M: 64): PASS
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached
in this email
Thanks,
Rong, Chen
3 years
13bc396977 [ 2.513606] kernel BUG at mm/slab.c:2159!
by kernel test robot
Greetings,
0day kernel testing robot got the below dmesg and the first bad commit is
https://git.kernel.org/pub/scm/linux/kernel/git/mel/linux.git mm-pagealloc-irqpvec-v1r11
commit 13bc396977b94cf8398d115f11cafd91e2101a2a
Author: Mel Gorman <mgorman(a)techsingularity.net>
AuthorDate: Wed Oct 25 17:17:30 2017 +0100
Commit: Mel Gorman <mgorman(a)techsingularity.net>
CommitDate: Wed Aug 22 14:14:37 2018 +0100
mm, pagealloc: Use a pagevec as an order-0 buffer for allocations from irq context
The per-allocator being preempt-safe poses a problem for allocations from
interrupt context -- hard irq, soft irq and those with BH disabled. These
must go through the main buddy allocator with a single lock being contended.
It's non-trivial to just allow softirq allocations to use the per-cpu
allocator as it forces us to check for irqs_disabled which is extremely
expensive in itself. Similarly, we could split per-cpu allocator from per-cpu
stats and then split the per-cpu allocator in two for IRQ and !IRQ contexts
but that is a lot of complexity that may not be necessary.
This patch uses a per-zone pagevec-like structure which gives a fixed-size
buffer of order-0 pages. The bulk of the complexity is dealing with refilling
and draining the pagevecs but at the most basic level, the fast path for
an allocation and free of a page is an irq-safe lock, update a counter,
fetch a pointer, irq-restore unlock. This should be a far shorter path
than the full buddy allocator while still avoiding disabling irqs for
allocations from other contexts.
No data is available on this one as it requires a high-speed network but
it does not crash.
Note however that this could form the basis of a bulk page allocator by
having a caller allocate the page and allocate it as pagevec_large and
then slightly modify irq_alloc_page and irq_free_page. At minimum that
would give batching of a static size. If the caller wanted to specify
batch sizes then it could be added to the pagevec_large structure.
Signed-off-by: Mel Gorman <mgorman(a)techsingularity.net>
6b7b374479 mm, page_alloc: Split out helper that takes a single page from the buddy list
13bc396977 mm, pagealloc: Use a pagevec as an order-0 buffer for allocations from irq context
+-------------------------------------------------------+------------+------------+
| | 6b7b374479 | 13bc396977 |
+-------------------------------------------------------+------------+------------+
| boot_successes | 37 | 0 |
| boot_failures | 0 | 40 |
| kernel_BUG_at_mm/slab.c | 0 | 40 |
| invalid_opcode:#[##] | 0 | 40 |
| EIP:check_irq_off | 0 | 40 |
| Kernel_panic-not_syncing:Fatal_exception_in_interrupt | 0 | 40 |
+-------------------------------------------------------+------------+------------+
[ 2.507243] No PC Engines board detected
[ 2.508934] hidraw: raw HID events driver (C) Jiri Kosina
[ 2.510704] hdaps: supported laptop not found!
[ 2.511369] hdaps: driver init failed (ret=-19)!
[ 2.513068] ------------[ cut here ]------------
[ 2.513606] kernel BUG at mm/slab.c:2159!
[ 2.514241] invalid opcode: 0000 [#1] SMP PTI
[ 2.514712] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.18.0-10891-g13bc396 #1
[ 2.515357] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014
[ 2.515357] EIP: check_irq_off+0x18/0x20
[ 2.515357] Code: f6 c4 02 74 05 5d c3 8d 76 00 0f 0b 8d b6 00 00 00 00 55 89 e5 e8 c4 6e e4 ff 9c 58 8d 74 26 00 f6 c4 02 75 05 5d c3 8d 76 00 <0f> 0b 8d b6 00 00 00 00 55 89 e5 57 56 53 83 ec 08 e8 9e 6e e4 ff
[ 2.515357] EAX: 00200246 EBX: c1800540 ECX: c217baa4 EDX: 00000004
[ 2.515357] ESI: 00000003 EDI: c235c4a0 EBP: c186de6c ESP: c186de6c
[ 2.515357] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 EFLAGS: 00210202
[ 2.515357] CR0: 80050033 CR2: 00000000 CR3: 02394000 CR4: 00040690
[ 2.515357] Call Trace:
[ 2.515357] ? cache_alloc_refill+0x6e3/0x9d0
[ 2.515357] ? kmem_cache_alloc_trace+0x250/0x390
[ 2.515357] ? llc_sap_open+0x19/0x100
[ 2.515357] ? llc_sap_open+0x74/0x100
[ 2.515357] ? eth_offload_init+0x16/0x16
[ 2.515357] ? register_snap_client+0xc0/0xc0
[ 2.515357] ? llc_sap_open+0x74/0x100
[ 2.515357] ? eth_offload_init+0x16/0x16
[ 2.515357] ? snap_init+0x17/0x36
[ 2.515357] ? do_one_initcall+0xab/0x3af
[ 2.515357] ? find_held_lock+0x32/0xb0
[ 2.515357] ? do_early_param+0x78/0x78
[ 2.515357] ? rcu_read_lock_sched_held+0x4f/0xa0
[ 2.515357] ? trace_initcall_level+0x91/0x10a
[ 2.515357] ? kernel_init_freeable+0x118/0x196
[ 2.515357] ? rest_init+0xb0/0xb0
[ 2.515357] ? kernel_init+0x10/0x110
[ 2.515357] ? rest_init+0xb0/0xb0
[ 2.515357] ? schedule_tail_wrapper+0x6/0x8
[ 2.515357] ? ret_from_fork+0x19/0x24
[ 2.515357] Modules linked in:
[ 2.533144] ---[ end trace 26926f8dd27c8ba0 ]---
[ 2.533879] EIP: check_irq_off+0x18/0x20
# HH:MM RESULT GOOD BAD GOOD_BUT_DIRTY DIRTY_NOT_BAD
git bisect start b83d0d6729c55d98296fb3ec6c1ad349afe77c12 94710cac0ef4ee177a63b5227664b38c95bbf703 --
git bisect bad b53273923f33b17a3ce1d67e455886d3a44d6470 # 11:12 B 0 11 26 0 Merge 'linux-review/Dmitry-Osipenko/IOMMU-Tegra-GART-driver-clean-up-and-optimization/20180819-003428' into devel-hourly-2018082223
git bisect bad c8fbcee7ac350fcfff8fdea7d113450b5d434fec # 11:39 B 0 11 26 0 Merge 'linux-review/Li-RongQing/mm-introduce-kvvirt_to_page-helper/20180817-000254' into devel-hourly-2018082223
git bisect bad f9048fcd7ef13a5774a0a03ea05479bfae80fc39 # 12:13 B 0 11 30 4 Merge 'linux-review/Wolfram-Sang/macintosh-therm_windtunnel-drop-using-attach_adapter/20180822-074654' into devel-hourly-2018082223
git bisect good d0944f7f2485328e3d5346ff9e333dae9ed9ebf2 # 12:41 G 10 0 1 1 Merge 'jpirko-mlxsw/petrm_qos' into devel-hourly-2018082223
git bisect good 78a4a0c7af57b3e713a81c343984593baa9baf65 # 12:58 G 11 0 3 3 Merge 'linux-review/Lad-Prabhakar/net-dsa-add-support-for-ksz9897-ethernet-switch/20180816-083425' into devel-hourly-2018082223
git bisect good 77f80257c38d77bbd6ba17e66f981df0974ea0b9 # 13:30 G 10 0 1 1 Merge 'linux-review/Bhaskar-Singh/staging-rtl8188eu-Fix-spelling-mistake/20180821-103113' into devel-hourly-2018082223
git bisect good 45b317f71f9261e17a720996d63f2d765f8af519 # 13:53 G 11 0 2 2 Merge 'linux-review/Toshiaki-Makita/veth-Free-queues-on-link-delete/20180816-082842' into devel-hourly-2018082223
git bisect bad 75f530fff2b5b8f00494d084a7fe4a74158e6bbb # 14:34 B 0 2 19 2 Merge 'linux-review/Chao-Yu/staging-erofs-fix-some-issues-and-clean-up-codes/20180822-135201' into devel-hourly-2018082223
git bisect good f3d135217b4c6f0c9a8ca7592f08e82e0c86e162 # 14:58 G 11 0 1 1 Merge 'linux-review/Ian-W-MORRISON/Use-default-cfg-file-name-for-RTL8723BS-devices-with-id-of-OBDA8723/20180814-152511' into devel-hourly-2018082223
git bisect bad 2c46f90c8dc1925b95e9ff1fb8dc3582cd3c010e # 15:28 B 0 11 26 0 Merge 'mel/mm-pagealloc-irqpvec-v1r11' into devel-hourly-2018082223
git bisect good c6eca9dabd29dc42f3c1bc5cc9659ed5d0ecec6f # 15:56 G 11 0 0 0 mm: define pagevec_large
git bisect bad 13bc396977b94cf8398d115f11cafd91e2101a2a # 16:22 B 0 8 23 0 mm, pagealloc: Use a pagevec as an order-0 buffer for allocations from irq context
git bisect good 6b7b3744791c43d715459d87112bfc7ca4606166 # 16:45 G 11 0 1 1 mm, page_alloc: Split out helper that takes a single page from the buddy list
# first bad commit: [13bc396977b94cf8398d115f11cafd91e2101a2a] mm, pagealloc: Use a pagevec as an order-0 buffer for allocations from irq context
git bisect good 6b7b3744791c43d715459d87112bfc7ca4606166 # 16:50 G 32 0 1 2 mm, page_alloc: Split out helper that takes a single page from the buddy list
# extra tests with debug options
git bisect bad 13bc396977b94cf8398d115f11cafd91e2101a2a # 17:16 B 0 11 26 0 mm, pagealloc: Use a pagevec as an order-0 buffer for allocations from irq context
# extra tests on HEAD of linux-devel/devel-hourly-2018082223
git bisect bad b83d0d6729c55d98296fb3ec6c1ad349afe77c12 # 17:16 B 0 43 61 0 0day head guard for 'devel-hourly-2018082223'
# extra tests on tree/branch mel/mm-pagealloc-irqpvec-v1r11
git bisect bad 13bc396977b94cf8398d115f11cafd91e2101a2a # 17:19 B 0 40 55 0 mm, pagealloc: Use a pagevec as an order-0 buffer for allocations from irq context
# extra tests with first bad commit reverted
git bisect good c3c843f60588cf9732121170c1b3d70ffc72437e # 17:51 G 11 0 2 2 Revert "mm, pagealloc: Use a pagevec as an order-0 buffer for allocations from irq context"
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/lkp Intel Corporation
3 years
cb778823d6 ("mm: Streamline deferred_init_pages and .."): kernel BUG at include/linux/mm.h:970!
by kernel test robot
Greetings,
0day kernel testing robot got the below dmesg and the first bad commit is
https://github.com/0day-ci/linux/commits/Alexander-Duyck/mm-Streamline-de...
commit cb778823d69f58446a2b28a7ec97ac8d684f153b
Author: Alexander Duyck <alexander.h.duyck(a)intel.com>
AuthorDate: Wed Aug 22 19:29:02 2018 -0700
Commit: 0day robot <lkp(a)intel.com>
CommitDate: Thu Aug 23 18:22:25 2018 +0800
mm: Streamline deferred_init_pages and deferred_free_pages
>From what I could tell the deferred_init_pages and deferred_free_pages were
running less than optimally as there were a number of checks that seemed
like they only needed to be run once per page block instead of being run
per page.
For example there is the pfn_valid check which either needs to be run for
every page if the architecture supports holes, or once per page block if it
doesn't. We can get around needing to perform these checks by just using
pfn_valid_within within the loop and running pfn_valid only on the first
access to any given page.
Also in the case of either a node ID mismatch or the pfn_valid check
failing on an architecture that doesn't support holes it doesn't make sense
to initialize pages for an invalid page block. So to skip over those pages
I have opted to OR in the page block mask to allow us to skip to the end of
a given page block.
With this patch I am seeing a modest improvement in boot time as shown
below on a system with 64GB on each node:
-- before --
[ 2.945572] node 0 initialised, 15432905 pages in 636ms
[ 2.968575] node 1 initialised, 15957078 pages in 659ms
-- after --
[ 2.770127] node 0 initialised, 15432905 pages in 457ms
[ 2.785129] node 1 initialised, 15957078 pages in 472ms
Signed-off-by: Alexander Duyck <alexander.h.duyck(a)intel.com>
815f0ddb34 include/linux/compiler*.h: make compiler-*.h mutually exclusive
cb778823d6 mm: Streamline deferred_init_pages and deferred_free_pages
+------------------------------------------+------------+------------+
| | 815f0ddb34 | cb778823d6 |
+------------------------------------------+------------+------------+
| boot_successes | 39 | 2 |
| boot_failures | 0 | 17 |
| kernel_BUG_at_include/linux/mm.h | 0 | 17 |
| invalid_opcode:#[##] | 0 | 17 |
| RIP:page_to_nid | 0 | 17 |
| Kernel_panic-not_syncing:Fatal_exception | 0 | 17 |
+------------------------------------------+------------+------------+
[ 0.256576] smpboot: Total of 2 processors activated (10774.02 BogoMIPS)
[ 0.257374] page:ffffea0000400040 is uninitialized and poisoned
[ 0.257374] raw: ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff
[ 0.257374] page dumped because: VM_BUG_ON_PAGE(PagePoisoned(p))
[ 0.257374] ------------[ cut here ]------------
[ 0.257374] kernel BUG at include/linux/mm.h:970!
[ 0.257374] invalid opcode: 0000 [#1] PREEMPT SMP PTI
[ 0.257374] CPU: 1 PID: 20 Comm: pgdatinit0 Not tainted 4.18.0-11683-gcb77882 #1
[ 0.257374] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014
[ 0.257374] RIP: 0010:page_to_nid+0x10/0x12
[ 0.257374] Code: 60 04 00 00 00 ba 08 00 00 00 0f 45 c2 eb 09 31 c0 eb 05 b8 04 00 00 00 5d c3 55 48 c7 c6 f7 eb 3b 82 48 89 e5 e8 5d af 01 00 <0f> 0b 8b 4e 30 31 c0 81 e1 80 00 00 f0 81 f9 00 00 00 f0 75 46 3b
[ 0.257374] RSP: 0000:ffffc90000173da8 EFLAGS: 00010056
[ 0.257374] RAX: 0000000000000034 RBX: 0000000000000000 RCX: 00000000c37f7f5b
[ 0.257374] RDX: 0000000000000000 RSI: 00000000c09d5bd1 RDI: 0000000000000046
[ 0.257374] RBP: ffffc90000173da8 R08: 0000000000000001 R09: 0000000000000000
[ 0.257374] R10: ffffc90000173de8 R11: ffffffff837542c7 R12: 0000000000000001
[ 0.257374] R13: 0000000000000001 R14: ffffea0000400040 R15: 0000000000010001
[ 0.257374] FS: 0000000000000000(0000) GS:ffff88001dc00000(0000) knlGS:0000000000000000
[ 0.257374] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 0.257374] CR2: 00000000ffffffff CR3: 0000000002612000 CR4: 00000000000006a0
[ 0.257374] Call Trace:
[ 0.257374] page_zone+0x12/0x30
[ 0.257374] set_pfnblock_flags_mask+0x40/0xc4
[ 0.257374] set_pageblock_migratetype+0x3c/0x3e
[ 0.257374] deferred_init_pages+0x97/0x12a
[ 0.257374] deferred_init_memmap+0x176/0x28b
[ 0.257374] ? deferred_free_pages+0x106/0x106
[ 0.257374] kthread+0xf4/0xfc
[ 0.257374] ? __list_del_entry+0x22/0x22
[ 0.257374] ret_from_fork+0x1f/0x30
[ 0.257374] ---[ end trace 9fa3051602685002 ]---
[ 0.257374] RIP: 0010:page_to_nid+0x10/0x12
# HH:MM RESULT GOOD BAD GOOD_BUT_DIRTY DIRTY_NOT_BAD
git bisect start 28956ec1ed6485fde45266eb039c74225eef802c 94710cac0ef4ee177a63b5227664b38c95bbf703 --
git bisect bad 8be39bcbaadb0b1e275b3efcfffea6bcc28c43b0 # 03:52 B 0 11 25 0 Merge 'linux-review/Suraj-Jitindar-Singh/KVM-PPC-Book3S-HV-Set-fault_dsisr-on-H_INST_STORAGE-interrupts/20180823-153113' into devel-spot-201808232206
git bisect good 750bb1e379459e52fd637c88b6c92c0830fd70a1 # 04:34 G 11 0 0 0 Merge 'bpf/master' into devel-spot-201808232206
git bisect bad b6e9c567c0f6a0c26541f4c8c3804efdc3a31b03 # 04:59 B 0 11 27 2 Merge 'linux-review/Weikang-Shi/fs-fix-local-var-type/20180823-180758' into devel-spot-201808232206
git bisect good 241efec1f31adad3dd309de4ce0077fcb4d6868f # 05:26 G 11 0 0 0 Merge 'driver-core/debugfs_cleanup' into devel-spot-201808232206
git bisect good 49dacbd919b87956e327906dc1cf146d7ebc41b7 # 05:46 G 10 0 0 0 Merge 'linux-review/Huijin-Park/mtd-cast-to-u64-to-avoid-unexpected-error/20180823-182723' into devel-spot-201808232206
git bisect bad 2bfdf212570d00a0086691af3703cb3d7004030c # 06:12 B 0 11 28 3 Merge 'linux-review/Alexander-Duyck/mm-Streamline-deferred_init_pages-and-deferred_free_pages/20180823-182222' into devel-spot-201808232206
git bisect bad cb778823d69f58446a2b28a7ec97ac8d684f153b # 06:39 B 0 3 17 0 mm: Streamline deferred_init_pages and deferred_free_pages
# first bad commit: [cb778823d69f58446a2b28a7ec97ac8d684f153b] mm: Streamline deferred_init_pages and deferred_free_pages
git bisect good 815f0ddb346c196018d4d8f8f55c12b83da1de3f # 07:08 G 33 0 0 0 include/linux/compiler*.h: make compiler-*.h mutually exclusive
# extra tests with debug options
git bisect bad cb778823d69f58446a2b28a7ec97ac8d684f153b # 07:26 B 0 1 15 0 mm: Streamline deferred_init_pages and deferred_free_pages
# extra tests on HEAD of linux-devel/devel-spot-201808232206
git bisect bad 28956ec1ed6485fde45266eb039c74225eef802c # 07:33 B 0 343 360 0 0day head guard for 'devel-spot-201808232206'
# extra tests on tree/branch linux-review/Alexander-Duyck/mm-Streamline-deferred_init_pages-and-deferred_free_pages/20180823-182222
git bisect bad cb778823d69f58446a2b28a7ec97ac8d684f153b # 07:37 B 0 17 32 0 mm: Streamline deferred_init_pages and deferred_free_pages
# extra tests with first bad commit reverted
git bisect good 54f39047afafded0b2db7053045bbe695226b9c2 # 07:58 G 11 0 0 0 Revert "mm: Streamline deferred_init_pages and deferred_free_pages"
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/lkp Intel Corporation
3 years
[lkp-robot] [mm, huge page] ef50cf1cfd: [No primary change]
by kernel test robot
Greeting,
There is no primary kpi change in this test, below is the data collected through multiple monitors running background just for your information.
commit: ef50cf1cfddbda7f89e552e99214314dec30658c ("mm, huge page: copy target sub-page last when copy huge page")
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
in testcase: vm-scalability
on test machine: 88 threads Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz with 128G memory
with following parameters:
runtime: 300s
size: 8T
test: anon-cow-seq-hugetlb
cpufreq_governor: performance
test-description: The motivation behind this suite is to exercise functions and regions of the mm/ of the Linux kernel which are of interest to us.
test-url: https://git.kernel.org/cgit/linux/kernel/git/wfg/vm-scalability.git/
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
vm-scalability.time.user_time
9500 +-+------------------------------------------------------------------+
| + .+ |
9000 +-+ .+ +. + : +.+.+ + : |
| .+.+ + + + : + : + : |
| .+ + + : + : .+. .+. .+. .|
8500 +-+.+.+ + +.+.+.+.+.+ + +.+ +.+.+ |
| |
8000 +-+ |
| |
7500 +-+ |
| O |
O O O O |
7000 +-+ O O O |
| O O O O O O O O O O O O O |
6500 +-+------------------------------------------------------------------+
vm-scalability.time.percent_of_cpu_this_job_got
6050 +-+------------------------------------------------------------------+
6000 +-+ .+ + .+ +.+ |
| +. .+.+ + + + .+.+ + .. + .+. .|
5950 +-+. + + +.+.+ + + +. .+. .+.+ +. .+.+.+. .+ |
5900 +-+ + + +.+ + + |
5850 +-+ |
5800 +-+ |
| |
5750 +-+ O |
5700 O-+ O O |
5650 +-+ |
5600 +-O O O O O O O O O O O |
| O O O O |
5550 +-+ O O |
5500 +-+------------------------------------------------------------------+
vm-scalability.time.minor_page_faults
5e+06 +-+---------------------------------------------------------------+
4.9e+06 +-+ O O |
O O O O O O O O O |
4.8e+06 +-+ O O O O O O O O O |
4.7e+06 +-+ O |
| |
4.6e+06 +-+ |
4.5e+06 +-+ |
4.4e+06 +-+ |
| |
4.3e+06 +-+ |
4.2e+06 +-+ |
| .+.+.++.+.+.+. .+. .+.+. .+. .+.+. .+ .+.+. .+. .|
4.1e+06 +-+.+.+.+.+ +.+ + + + + + +.+.+ + |
4e+06 +-+---------------------------------------------------------------+
vm-scalability.time.voluntary_context_switches
1.7e+06 +-+-----------O--O------------------------------------------------+
| O O |
1.6e+06 +-+ O O O O O O |
O O O O O O O O |
| O |
1.5e+06 +-+ O O |
| |
1.4e+06 +-+ |
| |
1.3e+06 +-+ |
| + +. |
| .+ .+ + + + +. .+ |
1.2e+06 +-+. .+.+ +. .+ + .+. .+.+. .+ + + :.+.+.+. .+.+.|
| +.+.+.+ + + + + + +.+ |
1.1e+06 +-+---------------------------------------------------------------+
vm-scalability.throughput
3.4e+07 +-+--------------------------------------------------------------+
3.35e+07 O-O O O O O O O O O O OO O O |
| O O O O O O |
3.3e+07 +-+ |
3.25e+07 +-+ |
3.2e+07 +-+ |
3.15e+07 +-+ |
| |
3.1e+07 +-+ |
3.05e+07 +-+ |
3e+07 +-+ |
2.95e+07 +-+ |
|. .+. .+. .+. .+. .+. .+.+.+.+. .+.+ .+. |
2.9e+07 +-+.+.+ ++.+ + + +.+.+.+.++.+ + + +.+.+ +.|
2.85e+07 +-+--------------------------------------------------------------+
vm-scalability.median
390000 +-+----------------------------------------------------------------+
385000 +-+ O |
O O O O O O |
380000 +-+ O O O O O O O O O O OO O |
375000 +-+ O |
370000 +-+ |
365000 +-+ |
| |
360000 +-+ |
355000 +-+ |
350000 +-+ |
345000 +-+ |
|.+.+.+.+.+.+. .+.+.+.+.+.+.+. .+.++.+.+.+.+.+.+.+.+.+.+.+.+.+.+.+.|
340000 +-+ + + |
335000 +-+----------------------------------------------------------------+
vm-scalability.workload
7.2e+09 +-+-----------O--O------------------------------------------------+
O O O O O O O O O O O O O O O O O |
7.1e+09 +-+ O O |
7e+09 +-+ |
| |
6.9e+09 +-+ |
6.8e+09 +-+ |
| |
6.7e+09 +-+ |
6.6e+09 +-+ |
| |
6.5e+09 +-+ |
6.4e+09 +-+ |
| .+.+. .+. .+. .+.+. .+ .+. .|
6.3e+09 +-+---------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.
Thanks,
Rong, Chen
3 years
[lkp-robot] [x86] 19efe000d3: PANIC:double_fault
by kernel test robot
FYI, we noticed the following commit (built with gcc-5):
commit: 19efe000d3258032d9a1dfb25313a092f9454da0 ("x86: Remap the IRQ stack so it has guard pages")
https://git.kernel.org/cgit/linux/kernel/git/luto/linux.git x86/guard_pages
in testcase: trinity
with following parameters:
runtime: 300s
test-description: Trinity is a linux system call fuzz tester.
test-url: http://codemonkey.org.uk/projects/trinity/
on test machine: qemu-system-x86_64 -enable-kvm -cpu IvyBridge -m 420M
caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace):
+------------------------------------------------------------------+------------+------------+
| | 0d997f71d5 | 19efe000d3 |
+------------------------------------------------------------------+------------+------------+
| boot_successes | 6 | 0 |
| boot_failures | 5 | 32 |
| invoked_oom-killer:gfp_mask=0x | 5 | |
| Mem-Info | 5 | |
| Out_of_memory:Kill_process | 3 | |
| Kernel_panic-not_syncing:Out_of_memory_and_no_killable_processes | 3 | |
| RIP:__put_user_4 | 1 | |
| PANIC:double_fault | 0 | 32 |
| RIP:trace_hardirqs_off_thunk | 0 | 32 |
| Kernel_panic-not_syncing:Machine_halted | 0 | 32 |
| WARNING:kernel_stack | 0 | 32 |
+------------------------------------------------------------------+------------+------------+
[ 0.004000] memory used by lock dependency info: 7871 kB
[ 0.004000] per task-struct memory footprint: 2688 bytes
[ 0.004000] ACPI: Core revision 20180531
[ 0.004000] clocksource: hpet: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 19112604467 ns
[ 0.004000] hpet clockevent registered
[ 0.004000] PANIC: double fault, error_code: 0x0
[ 0.004000] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.18.0-rc4-00074-g19efe000 #2
[ 0.004000] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014
[ 0.004000] RIP: 0010:trace_hardirqs_off_thunk+0xb/0x1c
[ 0.004000] Code: 5f 5d c3 55 48 89 e5 57 56 52 51 50 41 50 41 51 41 52 41 53 48 8b 7d 08 e8 f5 4e 2b 00 eb 34 55 48 89 e5 57 56 52 51 50 41 50 <41> 51 41 52 41 53 48 8b 7d 08 e8 eb 52 2b 00 eb 18 55 48 89 e5 57
[ 0.004000] RSP: 0000:ffffc90000000000 EFLAGS: 00010087
[ 0.004000] RAX: 0000000082800a97 RBX: 0000000000000001 RCX: ffffffff82800a97
[ 0.004000] RDX: 0000000000000000 RSI: ffffffff82800f68 RDI: ffffffff83678c68
[ 0.004000] RBP: ffffc90000000030 R08: 0000000000000000 R09: 0000000000000000
[ 0.004000] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
[ 0.004000] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
[ 0.004000] FS: 0000000000000000(0000) GS:ffff880013400000(0000) knlGS:0000000000000000
[ 0.004000] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 0.004000] CR2: ffffc8fffffffff8 CR3: 000000000366e000 CR4: 00000000000406b0
[ 0.004000] Call Trace:
[ 0.004000] Kernel panic - not syncing: Machine halted.
[ 0.004000] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.18.0-rc4-00074-g19efe000 #2
[ 0.004000] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014
[ 0.004000] Call Trace:
[ 0.004000] <#DF>
[ 0.004000] dump_stack+0x9b/0xe7
[ 0.004000] panic+0x1ad/0x325
[ 0.004000] ? refcount_error_report+0x267/0x267
[ 0.004000] df_debug+0x32/0x32
[ 0.004000] do_double_fault+0x1b3/0x1c7
[ 0.004000] double_fault+0x23/0x30
[ 0.004000] RIP: 0010:trace_hardirqs_off_thunk+0xb/0x1c
[ 0.004000] Code: 5f 5d c3 55 48 89 e5 57 56 52 51 50 41 50 41 51 41 52 41 53 48 8b 7d 08 e8 f5 4e 2b 00 eb 34 55 48 89 e5 57 56 52 51 50 41 50 <41> 51 41 52 41 53 48 8b 7d 08 e8 eb 52 2b 00 eb 18 55 48 89 e5 57
[ 0.004000] RSP: 0000:ffffc90000000000 EFLAGS: 00010087
[ 0.004000] RAX: 0000000082800a97 RBX: 0000000000000001 RCX: ffffffff82800a97
[ 0.004000] RDX: 0000000000000000 RSI: ffffffff82800f68 RDI: ffffffff83678c68
[ 0.004000] RBP: ffffc90000000030 R08: 0000000000000000 R09: 0000000000000000
[ 0.004000] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
[ 0.004000] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
[ 0.004000] ? native_iret+0x7/0x7
[ 0.004000] ? async_page_fault+0x8/0x30
[ 0.004000] WARNING: kernel stack regs at (____ptrval____) in swapper/0:0 has bad 'bp' value (____ptrval____)
[ 0.004000] unwind stack type:0 next_sp:(____ptrval____) mask:0x20 graph_idx:0
[ 0.004000] (____ptrval____): fffffe0000008e00 (0xfffffe0000008e00)
[ 0.004000] (____ptrval____): ffffffff81045a52 (show_trace_log_lvl+0x1ec/0x2c8)
[ 0.004000] (____ptrval____): ffffffff82800f68 (async_page_fault+0x8/0x30)
[ 0.004000] (____ptrval____): fffffe0000008fd8 (0xfffffe0000008fd8)
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email
Thanks,
lkp
3 years
[lkp-robot] [vfs] e950564b97: WARNING:at_lib/list_debug.c:#__list_add_valid
by kernel test robot
FYI, we noticed the following commit (built with gcc-7):
commit: e950564b97fd0f541b02eb207685d0746f5ecf29 ("vfs: don't evict uninitialized inode")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
in testcase: speccpu
with following parameters:
nr_task: 100%
cpu_feature: avx
run_type: speed
test: int:462.libquantum
cpufreq_governor: performance
on test machine: 192 threads Skylake-4S with 768G memory
caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace):
+----------------+------------+------------+
| | a6cbedfa87 | e950564b97 |
+----------------+------------+------------+
| boot_successes | 4 | 8 |
+----------------+------------+------------+
kern :warn : [ 29.190916] WARNING: CPU: 102 PID: 1832 at lib/list_debug.c:31 __list_add_valid+0x67/0x70
kern :warn : [ 29.202467] Modules linked in: overlay btrfs xor zstd_decompress zstd_compress xxhash raid6_pq dm_mod sg sd_mod snd_hda_codec_hdmi sr_mod cdrom mgag200 intel_rapl snd_hda_intel ttm x86_pkg_temp_thermal intel_powerclamp snd_hda_codec coretemp kvm_intel drm_kms_helper snd_hda_core snd_hwdep kvm snd_pcm irqbypass crct10dif_pclmul snd_timer syscopyarea crc32_pclmul crc32c_intel sysfillrect sysimgblt snd pcbc ghash_clmulni_intel fb_sys_fops aesni_intel crypto_simd ahci cryptd libahci pcspkr glue_helper soundcore ipmi_si drm libata wmi ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad
kern :warn : [ 29.265210] CPU: 102 PID: 1832 Comm: lkp-modules Not tainted 4.18.0-rc1-00010-ge950564 #1
kern :warn : [ 29.275443] RIP: 0010:__list_add_valid+0x67/0x70
kern :warn : [ 29.281693] Code: c1 4c 89 c6 48 c7 c7 98 5f 11 82 e8 d3 a4 bf ff 0f 0b 31 c0 c3 48 89 f2 4c 89 c1 48 89 fe 48 c7 c7 e8 5f 11 82 e8 b9 a4 bf ff <0f> 0b 31 c0 c3 0f 1f 40 00 48 8b 07 48 b9 00 01 00 00 00 00 ad de
kern :warn : [ 29.305071] RSP: 0018:ffffc9001d573ab0 EFLAGS: 00010286
kern :warn : [ 29.312069] RAX: 0000000000000000 RBX: ffff882f145d0830 RCX: ffffffff824635c8
kern :warn : [ 29.321234] RDX: 0000000000000001 RSI: 0000000000000092 RDI: 0000000000000246
kern :warn : [ 29.330428] RBP: ffff882f79056800 R08: 0000000000000bc8 R09: 0000000000aaaaaa
kern :warn : [ 29.339580] R10: ffffc9001d573b70 R11: 00000000ffffffff R12: ffff882f145d0950
kern :warn : [ 29.348714] R13: ffff882f145d0950 R14: ffff882f79056d88 R15: ffffffffa0354a40
kern :warn : [ 29.357757] FS: 00007faaf6315700(0000) GS:ffff882f80d80000(0000) knlGS:0000000000000000
kern :warn : [ 29.367980] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
kern :warn : [ 29.375592] CR2: 0000000001a7ae40 CR3: 0000002f5377c002 CR4: 00000000007606e0
kern :warn : [ 29.384794] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
kern :warn : [ 29.394009] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
kern :warn : [ 29.403227] PKRU: 55555554
kern :warn : [ 29.407435] Call Trace:
kern :warn : [ 29.411442] inode_sb_list_add+0x47/0x90
kern :warn : [ 29.417107] ? ovl_inode_test+0x20/0x20 [overlay]
kern :warn : [ 29.423629] inode_insert5+0x13d/0x1f0
kern :warn : [ 29.429116] ovl_get_inode+0xbe/0x3a0 [overlay]
kern :warn : [ 29.435492] ovl_instantiate+0x73/0x160 [overlay]
kern :warn : [ 29.442075] ovl_create_or_link+0x1ac/0x5d0 [overlay]
kern :warn : [ 29.449064] ? selinux_inode_alloc_security+0x32/0x80
kern :warn : [ 29.456067] ? security_inode_alloc+0x2a/0x40
kern :warn : [ 29.462307] ? inode_init_always+0x123/0x1e0
kern :warn : [ 29.468452] ? inode_sb_list_add+0x47/0x90
kern :warn : [ 29.474332] ? ovl_fill_inode+0xc5/0x120 [overlay]
kern :warn : [ 29.481079] ovl_create_object+0xb3/0xf0 [overlay]
kern :warn : [ 29.487809] path_openat+0x13c4/0x16c0
kern :warn : [ 29.493394] do_filp_open+0x9b/0x110
kern :warn : [ 29.498723] ? do_sys_open+0x1bd/0x250
kern :warn : [ 29.504249] do_sys_open+0x1bd/0x250
kern :warn : [ 29.509571] do_syscall_64+0x5b/0x180
kern :warn : [ 29.514968] entry_SYSCALL_64_after_hwframe+0x44/0xa9
kern :warn : [ 29.521915] RIP: 0033:0x7faaf5e2a050
kern :warn : [ 29.527170] Code: 48 8b 15 3b 9e 2c 00 f7 d8 64 89 02 48 83 c8 ff c3 66 0f 1f 84 00 00 00 00 00 83 3d 8d f7 2c 00 00 75 10 b8 02 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 31 c3 48 83 ec 08 e8 1e 92 01 00 48 89 04 24
kern :warn : [ 29.550863] RSP: 002b:00007fffeb16c7a8 EFLAGS: 00000246 ORIG_RAX: 0000000000000002
kern :warn : [ 29.560580] RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007faaf5e2a050
kern :warn : [ 29.569778] RDX: 00000000000001b6 RSI: 0000000000000241 RDI: 0000000001a79490
kern :warn : [ 29.578926] RBP: 0000000001a7b3b0 R08: 0000000000000000 R09: 0000000001a7b3b0
kern :warn : [ 29.588053] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000001a7ae20
kern :warn : [ 29.597148] R13: 0000000001a79490 R14: 0000000001a79550 R15: 0000000001a7ae00
kern :warn : [ 29.606208] ---[ end trace 9cc284a0fe57e045 ]---
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
Thanks,
rongc
3 years
[lkp-robot] [tracing] c3bc8fd637: WARNING:at_kernel/rcu/tree.c:#rcu_irq_enter
by kernel test robot
FYI, we noticed the following commit (built with gcc-6):
commit: c3bc8fd637a9623f5c507bd18f9677effbddf584 ("tracing: Centralize preemptirq tracepoints and unify their usage")
https://git.kernel.org/cgit/linux/kernel/git/rostedt/linux-trace.git ftrace/core
in testcase: trinity
with following parameters:
runtime: 300s
test-description: Trinity is a linux system call fuzz tester.
test-url: http://codemonkey.org.uk/projects/trinity/
on test machine: qemu-system-x86_64 -enable-kvm -cpu host -smp 2 -m 4G
caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace):
+------------------------------------------------------------------+------------+------------+
| | e6753f23d9 | c3bc8fd637 |
+------------------------------------------------------------------+------------+------------+
| boot_successes | 19 | 4 |
| boot_failures | 2 | 15 |
| invoked_oom-killer:gfp_mask=0x | 2 | 3 |
| Mem-Info | 2 | 3 |
| Out_of_memory:Kill_process | 1 | 3 |
| RIP:copy_user_generic_string | 1 | 1 |
| RIP:__clear_user | 1 | |
| Kernel_panic-not_syncing:Out_of_memory_and_no_killable_processes | 1 | 1 |
| WARNING:at_kernel/rcu/tree.c:#rcu_irq_enter | 0 | 12 |
| RIP:rcu_irq_enter | 0 | 12 |
| RIP:arch_local_irq_restore | 0 | 12 |
| WARNING:at_kernel/events/core.c:#perf_event_task_tick | 0 | 12 |
| RIP:perf_event_task_tick | 0 | 12 |
| WARNING:at_kernel/rcu/tree.c:#rcu_irq_exit | 0 | 11 |
| RIP:rcu_irq_exit | 0 | 11 |
| RIP:check_poison_obj | 0 | 1 |
| RIP:get_page_from_freelist | 0 | 1 |
| RIP:locks_remove_posix | 0 | 1 |
+------------------------------------------------------------------+------------+------------+
[ 11.209512] WARNING: CPU: 0 PID: 24 at kernel/rcu/tree.c:999 rcu_irq_enter+0x55/0x77
[ 11.209512] CPU: 0 PID: 24 Comm: kworker/0:1 Tainted: G T 4.18.0-rc6-00023-gc3bc8fd #1
[ 11.209512] Workqueue: events test_mutex_work
[ 11.209512] RIP: 0010:rcu_irq_enter+0x55/0x77
[ 11.209512] Code: 00 00 00 75 27 83 b8 d8 0a 00 00 00 74 1e 80 3d 48 e3 f0 01 00 75 15 48 c7 c7 1a 1a 8b 82 c6 05 38 e3 f0 01 01 e8 fc f5 f9 ff <0f> 0b 48 83 7b 08 00 75 13 65 48 8b 04 25 c0 4e 01 00 c7 80 40 19
[ 11.209512] RSP: 0000:ffff88013f403fc8 EFLAGS: 00010082
[ 11.209512] RAX: 0000000000000000 RBX: ffff88013f419c60 RCX: ffff88013f403e44
[ 11.209512] RDX: 0000000000000003 RSI: 0000000000000003 RDI: 0000000000000046
[ 11.209512] RBP: ffff88013f403fe8 R08: 0000000000000001 R09: 0000000000000001
[ 11.209512] R10: 0000000000000000 R11: ffffffff82c98b38 R12: 0000000000000000
[ 11.209512] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
[ 11.209512] FS: 0000000000000000(0000) GS:ffff88013f400000(0000) knlGS:0000000000000000
[ 11.209512] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 11.209512] CR2: 0000000000000000 CR3: 0000000002c57001 CR4: 00000000001606f0
[ 11.209512] Call Trace:
[ 11.209512] <IRQ>
[ 11.209512] irq_enter+0xa/0x60
[ 11.209512] scheduler_ipi+0x4b/0x8a
[ 11.209512] reschedule_interrupt+0xf/0x20
[ 11.209512] </IRQ>
[ 11.209512] RIP: 0010:arch_local_irq_restore+0x2/0x8
[ 11.209512] Code: 15 48 c7 c6 19 8c 8c 82 48 c7 c7 ca 58 8b 82 e8 da 0b fc ff 0f 0b 48 c7 43 58 00 00 00 00 5b c3 9c 58 0f 1f 44 00 00 c3 57 9d <0f> 1f 44 00 00 c3 e8 eb ff ff ff 48 89 c2 fa 66 0f 1f 44 00 00 48
[ 11.209512] RSP: 0000:ffffc900000cbe60 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff02
[ 11.209512] RAX: ffff88013ead8100 RBX: ffff88013ea0aa80 RCX: 0000000000000000
[ 11.209512] RDX: 0000000000000001 RSI: ffffffff83e00190 RDI: 0000000000000246
[ 11.209512] RBP: ffffc900000cbed0 R08: ffffffff83e00190 R09: ffffffff8323b3ec
[ 11.209512] R10: ffffc900000cbdf8 R11: 0000000000000002 R12: ffff88013f4217c0
[ 11.209512] R13: ffff88013f424600 R14: 0000000000000000 R15: 0000000000000000
[ 11.209512] process_one_work+0x1ff/0x4a9
[ 11.209512] ? process_one_work+0x199/0x4a9
[ 11.209512] ? process_scheduled_works+0x2c/0x2c
[ 11.209512] worker_thread+0x1b2/0x261
[ 11.209512] ? process_scheduled_works+0x2c/0x2c
[ 11.209512] kthread+0xf5/0xfa
[ 11.209512] ? queuing_blocked+0x32/0x32
[ 11.209512] ret_from_fork+0x3a/0x50
[ 11.209512] irq event stamp: 71539
[ 11.209512] hardirqs last enabled at (71539): [<0000000000000000>] (null)
[ 11.209512] hardirqs last disabled at (71538): [<0000000000000000>] (null)
[ 11.209512] softirqs last enabled at (71496): [<ffffffff8220030d>] __do_softirq+0x30d/0x38f
[ 11.209512] softirqs last disabled at (71489): [<ffffffff810daae3>] irq_exit+0x63/0xd4
[ 11.209512] _warn_unseeded_randomness: 1 callbacks suppressed
[ 11.209512] random: get_random_bytes called from init_oops_id+0x22/0x31 with crng_init=0
[ 11.209512] ---[ end trace ad4d30e5b1ea3d93 ]---
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email
Thanks,
lkp
3 years