From: Daniel Gomez <da.gomez@kernel.org>
To: Chuck Lever <chuck.lever@oracle.com>,
Luis Chamberlain <mcgrof@kernel.org>
Cc: kdevops@lists.linux.dev, Daniel Gomez <da.gomez@samsung.com>
Subject: Re: [PATCH 09/10] fstests: tmpfs: expunges: testing
Date: Mon, 16 Jun 2025 12:41:50 +0200 [thread overview]
Message-ID: <ae78957f-06a6-4835-aae4-c32319cfbcec@kernel.org> (raw)
In-Reply-To: <09ef783c-32c5-4558-8c73-8311b6742f29@oracle.com>
On 15/06/2025 18.18, Chuck Lever wrote:
> On 6/15/25 5:58 AM, Daniel Gomez wrote:
> For comparison, here is the expunge list I use for tmpfs:
>
> diff --git a/workflows/fstests/expunges/any/tmpfs/unassigned/all.txt
> b/workflows/fstests/expunges/any/tmpfs/unassigned/all.txt
> new file mode 100644
> index 000000000000..86710a98776b
> --- /dev/null
> +++ b/workflows/fstests/expunges/any/tmpfs/unassigned/all.txt
> @@ -0,0 +1,3 @@
> +generic/449
> +generic/551
> +generic/561
The time spent in generic/449 before it's skipped is quite long, that's why I
added it here.
But rather than augment expunges I think it'd be better to do it directly
in xfstests as done by Hugh [1] in a patch list I got from him some time last
year. (BTW, I think you've probably received the same here [2], right?).
Unfortunately, I got stuck in this review before reaching to that point [3]
and never had the time to follow it up:
[1] https://gitlab.com/da.gomez/xfstests-dev/-/commits/hughd/tmpfs-fixes?ref_type=heads
[2] https://lore.kernel.org/all/5d1e7ea8-be4c-4906-a1d4-835fa46da605@oracle.com/
[3] https://lore.kernel.org/all/20240614061722.1080-1-da.gomez@samsung.com/
This is the actual time spent on generic/449 in my current setup:
[ 1468.203926] run fstests generic/446 at 2025-06-16 08:19:06
[ 1472.522334] run fstests generic/447 at 2025-06-16 08:19:11
[ 1473.208250] run fstests generic/448 at 2025-06-16 08:19:11
[ 1474.527435] run fstests generic/449 at 2025-06-16 08:19:13
[ 6894.771567] run fstests generic/450 at 2025-06-16 09:49:33
[ 6895.431099] run fstests generic/451 at 2025-06-16 09:49:34
>
> So: agreed on 449. Should we consider skipping generic/551 and
> generic/561 too?
>
For generic/551 I'm currently getting this:
[ 7081.496735] run fstests generic/547 at 2025-06-16 09:52:40
[ 7082.015303] run fstests generic/548 at 2025-06-16 09:52:40
[ 7082.689962] run fstests generic/549 at 2025-06-16 09:52:41
[ 7083.406567] run fstests generic/550 at 2025-06-16 09:52:41
[ 7083.887618] run fstests generic/551 at 2025-06-16 09:52:42
[ 7089.441945] aio-dio-write-v invoked oom-killer: gfp_mask=0x140cca(GFP_HIGHUSER_MOVABLE|__GFP_COMP), order=0, oom_score_adj=250
[ 7089.442675] CPU: 2 UID: 0 PID: 1842847 Comm: aio-dio-write-v Not tainted 6.16.0-rc2 #9 PREEMPT(full)
[ 7089.442677] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 2025.02-6 04/08/2025
[ 7089.442678] Call Trace:
[ 7089.442679] <TASK>
[ 7089.442680] dump_stack_lvl+0x56/0x80
[ 7089.442685] dump_header+0x43/0x1b7
[ 7089.442687] oom_kill_process.cold+0x8/0x90
[ 7089.442689] out_of_memory+0x162/0x4f0
[ 7089.442692] __alloc_pages_slowpath.constprop.0+0x9fc/0xda0
[ 7089.442695] __alloc_frozen_pages_noprof+0x359/0x370
[ 7089.442698] alloc_pages_mpol+0x7d/0x1c0
[ 7089.442700] folio_alloc_mpol_noprof+0x14/0xb0
[ 7089.442702] shmem_alloc_folio+0x9c/0xb0
[ 7089.442705] shmem_alloc_and_add_folio+0x174/0x550
[ 7089.442706] ? shmem_allowable_huge_orders+0xc6/0x140
[ 7089.442708] shmem_get_folio_gfp+0x229/0x740
[ 7089.442709] shmem_write_begin+0x51/0x130
[ 7089.442711] generic_perform_write+0xc7/0x2d0
[ 7089.442713] shmem_file_write_iter+0x82/0x90
[ 7089.442715] aio_write+0x110/0x250
[ 7089.442718] ? __alloc_object+0x2f/0x220
[ 7089.442720] ? __link_object+0x142/0x220
[ 7089.442722] ? _raw_spin_unlock_irqrestore+0x23/0x40
[ 7089.442725] ? __create_object+0x5e/0x90
[ 7089.442726] ? io_submit_one+0x7b/0x970
[ 7089.442727] ? kmem_cache_alloc_noprof+0x310/0x3d0
[ 7089.442730] ? _raw_spin_lock_irqsave+0x23/0x50
[ 7089.442732] ? io_submit_one+0x435/0x970
[ 7089.442733] io_submit_one+0x435/0x970
[ 7089.442735] __x64_sys_io_submit+0x81/0x1c0
[ 7089.442736] do_syscall_64+0x50/0x1e0
[ 7089.442739] entry_SYSCALL_64_after_hwframe+0x76/0x7e
[ 7089.442740] RIP: 0033:0x7f15a5122779
[ 7089.442745] Code: Unable to access opcode bytes at 0x7f15a512274f.
[ 7089.442746] RSP: 002b:00007ffdb85b0288 EFLAGS: 00000246 ORIG_RAX: 00000000000000d1
[ 7089.442747] RAX: ffffffffffffffda RBX: 00007f15a50116c8 RCX: 00007f15a5122779
[ 7089.442748] RDX: 000056050750e0b0 RSI: 0000000000000019 RDI: 00007f15a5214000
[ 7089.442749] RBP: 00007f15a5214000 R08: 0000000000000001 R09: 0000000000000003
[ 7089.442750] R10: 00007f15a51fbac0 R11: 0000000000000246 R12: 0000000000000019
[ 7089.442750] R13: 0000000000000000 R14: 000056050750e0b0 R15: 000056050750e0b0
[ 7089.442753] </TASK>
[ 7089.442753] Mem-Info:
[ 7089.459986] active_anon:922060 inactive_anon:12934 isolated_anon:0
active_file:2 inactive_file:55 isolated_file:0
unevictable:0 dirty:11 writeback:0
slab_reclaimable:3359 slab_unreclaimable:25305
mapped:83 shmem:62663 pagetables:2595
sec_pagetables:0 bounce:0
kernel_misc_reclaimable:0
free:20479 free_pcp:2269 free_cma:0
[ 7089.462234] Node 0 active_anon:3688240kB inactive_anon:51736kB active_file:8kB inactive_file:220kB unevictable:0kB isolated(anon):0kB isolated(file):0kB mapped:332kB dirty:44kB writeback:0kB shmem:250652kB
[ 7089.464097] Node 0 DMA free:14368kB boost:0kB min:248kB low:308kB high:368kB reserved_highatomic:0KB free_highatomic:0KB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB w
[ 7089.465762] lowmem_reserve[]: 0 1927 3899 3899 3899
[ 7089.466055] Node 0 DMA32 free:33328kB boost:0kB min:32280kB low:40348kB high:48416kB reserved_highatomic:0KB free_highatomic:0KB active_anon:1930668kB inactive_anon:1896kB active_file:0kB inactive_file:68kB
[ 7089.467823] lowmem_reserve[]: 0 0 1971 1971 1971
[ 7089.468099] Node 0 Normal free:34220kB boost:6144kB min:41196kB low:49956kB high:58716kB reserved_highatomic:0KB free_highatomic:0KB active_anon:1727272kB inactive_anon:80140kB active_file:216kB inactive_fi
[ 7089.470959] lowmem_reserve[]: 0 0 0 0 0
[ 7089.471360] Node 0 DMA: 2*4kB (U) 1*8kB (U) 1*16kB (U) 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 2*1024kB (U) 2*2048kB (M) 2*4096kB (M) = 14368kB
[ 7089.472594] Node 0 DMA32: 2*4kB (UM) 2*8kB (UM) 4*16kB (UE) 5*32kB (UME) 5*64kB (UE) 3*128kB (UE) 5*256kB (UME) 5*512kB (UME) 12*1024kB (UME) 6*2048kB (M) 1*4096kB (M) = 33464kB
[ 7089.474152] Node 0 Normal: 1439*4kB (UME) 152*8kB (UME) 15*16kB (UME) 26*32kB (UME) 55*64kB (UM) 99*128kB (UME) 39*256kB (UME) 0*512kB 0*1024kB 0*2048kB 0*4096kB = 34220kB
[ 7089.475695] Node 0 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=1048576kB
[ 7089.476577] Node 0 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=2048kB
[ 7089.477441] 62699 total pagecache pages
[ 7089.477836] 0 pages in swap cache
[ 7089.478182] Free swap = 0kB
[ 7089.478481] Total swap = 0kB
[ 7089.478825] 1046922 pages RAM
[ 7089.479139] 0 pages HighMem/MovableOnly
[ 7089.479533] 45176 pages reserved
[ 7089.479866] 0 pages cma reserved
[ 7089.480200] 0 pages hwpoisoned
[ 7089.480524] Tasks state (memory values in pages):
[ 7089.481036] [ pid ] uid tgid total_vm rss rss_anon rss_file rss_shmem pgtables_bytes swapents oom_score_adj name
[ 7089.482135] [ 342] 0 342 19052 351 320 29 2 147456 0 -250 systemd-journal
[ 7089.483304] [ 390] 991 390 5859 728 672 56 0 86016 0 0 systemd-resolve
[ 7089.483978] [ 391] 997 391 23058 298 256 42 0 90112 0 0 systemd-timesyn
[ 7089.484657] [ 401] 0 401 8148 573 512 61 0 86016 0 -1000 systemd-udevd
[ 7089.485345] [ 411] 998 411 5242 520 336 184 0 86016 0 0 systemd-network
[ 7089.486005] [ 517] 103 517 1643 179 96 83 0 49152 0 0 rpcbind
[ 7089.486671] [ 614] 996 614 2016 119 64 55 0 61440 0 -900 dbus-daemon
[ 7089.487305] [ 629] 0 629 20227 49 32 17 0 57344 0 0 qemu-ga
[ 7089.487919] [ 630] 0 630 1466 211 160 51 0 53248 0 0 dhclient
[ 7089.488557] [ 633] 0 633 4659 348 288 60 0 73728 0 0 systemd-logind
[ 7089.489243] [ 684] 0 684 1316 58 33 25 0 53248 0 0 blkmapd
[ 7089.489865] [ 773] 63150 773 16062 582 512 70 0 147456 0 0 systemd-journal
[ 7089.490552] [ 783] 0 783 2974 406 320 86 0 65536 0 -1000 sshd
[ 7089.491166] [ 788] 0 788 2042 91 32 59 0 57344 0 0 agetty
[ 7089.491782] [ 789] 0 789 1945 103 32 71 0 61440 0 0 agetty
[ 7089.492474] [ 790] 0 790 4883 579 512 67 0 81920 0 0 sshd-session
[ 7089.493568] [ 795] 1002 795 5507 588 544 44 0 90112 0 100 systemd
[ 7089.494638] [ 799] 1002 799 5587 484 425 59 0 73728 0 100 (sd-pam)
[ 7089.495683] [ 820] 1002 820 5024 695 633 62 0 86016 0 0 sshd-session
[ 7089.496848] [ 1041] 0 1041 4838 545 480 65 0 77824 0 0 sshd-session
[ 7089.497983] [ 1057] 1002 1057 4890 593 521 72 0 73728 0 0 sshd-session
[ 7089.499155] [ 1064] 1002 1064 2262 675 576 99 0 65536 0 0 bash
[ 7089.500220] [ 1119] 1002 1119 672 94 0 94 0 49152 0 0 sh
[ 7089.501316] [ 1120] 1002 1120 4961 338 288 50 0 73728 0 0 sudo
[ 7089.502389] [ 1122] 1002 1122 4961 341 297 44 0 61440 0 0 sudo
[ 7089.503495] [ 1123] 0 1123 2555 153 96 57 0 53248 0 0 su
[ 7089.504540] [ 1125] 0 1125 9117 3472 3418 54 0 110592 0 0 python3
[ 7089.505618] [ 1147] 0 1147 1906 210 160 50 0 57344 0 0 oscheck.sh
[ 7089.506726] [ 1662] 0 1662 2008 1044 992 52 0 61440 0 -500 bash
[ 7089.507603] [ 2357] 1002 2357 4960 375 288 87 0 77824 0 0 sudo
[ 7089.508214] [ 2365] 1002 2365 4960 359 297 62 0 65536 0 0 sudo
[ 7089.508867] [ 2367] 0 2367 1469 104 32 72 0 53248 0 0 dmesg
[ 7089.509482] [1842640] 0 1842640 1779 886 800 86 0 53248 0 250 551
[ 7089.510087] [1842847] 0 1842847 858194 857584 857535 49 0 6922240 0 250 aio-dio-write-v
[ 7089.510794] oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0,global_oom,task_memcg=/system.slice/fstests-generic-551.scope,task=aio-dio-write-v,pid=1842847,uid=0
[ 7089.511761] Out of memory: Killed process 1842847 (aio-dio-write-v) total-vm:3432776kB, anon-rss:3430140kB, file-rss:196kB, shmem-rss:0kB, UID:0 pgtables:6760kB oom_score_adj:250
[ 7090.616260] run fstests generic/552 at 2025-06-16 09:52:49
So, why do you want to skip it?
I think Hugh's patch [1] on generic/551 is unrelated to the above. And maybe my
VM just needs more memory to pass this test?
[1] https://gitlab.com/da.gomez/xfstests-dev/-/commit/cfc6b431403123d70bf25842c80d6693f0c9ceb6
And same question for generic/561, why do you want to skip it? I see the time
spent for skipping the test is actually fine:
[ 7103.693096] run fstests generic/559 at 2025-06-16 09:53:02
[ 7104.356919] run fstests generic/560 at 2025-06-16 09:53:02
[ 7105.154012] run fstests generic/561 at 2025-06-16 09:53:03
[ 7105.614221] run fstests generic/562 at 2025-06-16 09:53:04
[ 7106.056302] run fstests generic/563 at 2025-06-16 09:53:04
cat /data/fstests-install/xfstests/results/debian13-tmpfs-default/6.16.0-rc2/tmpfs_default/generic/561.notrun
duperemove utility required, skipped this test
cat /data/fstests-install/xfstests/results/debian13-tmpfs-default/6.16.0-rc2/tmpfs_default/generic/561.full
umount: /media/scratch: not mounted.
next prev parent reply other threads:[~2025-06-16 10:41 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-06-15 9:58 [PATCH 00/10] Extend kdevops-ci to support more workflows Daniel Gomez
2025-06-15 9:58 ` [PATCH 01/10] .ci: blktests: add 1 section only targets Daniel Gomez
2025-06-15 9:58 ` [PATCH 02/10] defconfigs: add tmpfs support Daniel Gomez
2025-06-15 9:58 ` [PATCH 03/10] .ci: " Daniel Gomez
2025-06-15 9:58 ` [PATCH 04/10] defconfigs: blktests: add blktests_meta support Daniel Gomez
2025-06-15 9:58 ` [PATCH 05/10] gen_nodes: selftests: fix hosts names generation Daniel Gomez
2025-06-15 9:58 ` [PATCH 06/10] gen_hosts: " Daniel Gomez
2025-06-15 9:58 ` [PATCH 07/10] fstests: fix ansible range type error Daniel Gomez
2025-06-15 9:58 ` [PATCH 08/10] selftests: kmod: add timeout makefile parameter cli support Daniel Gomez
2025-06-15 9:58 ` [PATCH 09/10] fstests: tmpfs: expunges: testing Daniel Gomez
2025-06-15 16:18 ` Chuck Lever
2025-06-16 10:41 ` Daniel Gomez [this message]
2025-06-16 14:31 ` Chuck Lever
2025-06-16 20:24 ` Daniel Gomez
2025-06-16 20:36 ` Chuck Lever
2025-06-17 8:34 ` Daniel Gomez
2025-06-17 13:43 ` Chuck Lever
2025-06-21 7:47 ` Daniel Gomez
2025-06-26 19:10 ` Daniel Gomez
2025-06-15 9:58 ` [PATCH 10/10] workflows: add support to change kdevops repo in guests via cli Daniel Gomez
2025-06-26 18:59 ` [PATCH 00/10] Extend kdevops-ci to support more workflows Daniel Gomez
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ae78957f-06a6-4835-aae4-c32319cfbcec@kernel.org \
--to=da.gomez@kernel.org \
--cc=chuck.lever@oracle.com \
--cc=da.gomez@samsung.com \
--cc=kdevops@lists.linux.dev \
--cc=mcgrof@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox