From: bugzilla-daemon@bugzilla.kernel.org
To: dri-devel@lists.freedesktop.org
Subject: [Bug 214029] [NAVI] Several memory leaks in amdgpu and ttm
Date: Mon, 30 Aug 2021 13:49:29 +0000 [thread overview]
Message-ID: <bug-214029-2300-sAdPc9KTz3@https.bugzilla.kernel.org/> (raw)
In-Reply-To: <bug-214029-2300@https.bugzilla.kernel.org/>
https://bugzilla.kernel.org/show_bug.cgi?id=214029
Erhard F. (erhard_f@mailbox.org) changed:
What |Removed |Added
----------------------------------------------------------------------------
Attachment #298267|0 |1
is obsolete| |
--- Comment #5 from Erhard F. (erhard_f@mailbox.org) ---
Created attachment 298529
--> https://bugzilla.kernel.org/attachment.cgi?id=298529&action=edit
output of kmemleak (kernel 5.14, AMD FX-8370)
[...]
unreferenced object 0xffff8881dd96f0c0 (size 216):
comm "lxdm-greeter-gt", pid 619, jiffies 4294888177 (age 3729.577s)
hex dump (first 32 bytes):
50 f1 96 dd 81 88 ff ff 20 8b 72 c0 ff ff ff ff P....... .r.....
c0 3a 59 da 08 00 00 00 58 fd a6 08 00 c9 ff ff .:Y.....X.......
backtrace:
[<ffffffffc072514f>] drm_sched_fence_create+0x1f/0x1d0 [gpu_sched]
[<ffffffffc07204de>] drm_sched_job_init+0x10e/0x240 [gpu_sched]
[<ffffffffc159e317>] amdgpu_job_submit+0x27/0x2d0 [amdgpu]
[<ffffffffc118ae6e>] amdgpu_copy_buffer+0x49e/0x700 [amdgpu]
[<ffffffffc118b6ca>] amdgpu_ttm_copy_mem_to_mem+0x5fa/0xf00 [amdgpu]
[<ffffffffc118ce06>] amdgpu_bo_move+0x356/0x2180 [amdgpu]
[<ffffffffc0887897>] ttm_bo_handle_move_mem+0x1c7/0x620 [ttm]
[<ffffffffc088b297>] ttm_bo_validate+0x2c7/0x450 [ttm]
[<ffffffffc1193444>] amdgpu_bo_fault_reserve_notify+0x2a4/0x640 [amdgpu]
[<ffffffffc11a3323>] amdgpu_gem_fault+0x123/0x2d0 [amdgpu]
[<ffffffffb255c853>] __do_fault+0xf3/0x3e0
[<ffffffffb256ea4b>] __handle_mm_fault+0x1bcb/0x2ac0
[<ffffffffb256fa6a>] handle_mm_fault+0x12a/0x490
[<ffffffffb2090909>] do_user_addr_fault+0x259/0xb70
[<ffffffffb37ad665>] exc_page_fault+0x55/0xb0
[<ffffffffb3800acb>] asm_exc_page_fault+0x1b/0x20
unreferenced object 0xffff888120f77100 (size 72):
comm "sdma0", pid 344, jiffies 4294888177 (age 3729.577s)
hex dump (first 32 bytes):
30 f4 ec 3a 81 88 ff ff 20 8a f0 c1 ff ff ff ff 0..:.... .......
a3 49 5d da 08 00 00 00 58 84 a3 0c 82 88 ff ff .I].....X.......
backtrace:
[<ffffffffc1180521>] amdgpu_fence_emit+0x91/0x790 [amdgpu]
[<ffffffffc11ed4cb>] amdgpu_ib_schedule+0x8cb/0x12f0 [amdgpu]
[<ffffffffc159d67e>] amdgpu_job_run+0x35e/0x790 [amdgpu]
[<ffffffffc072392e>] drm_sched_main+0x64e/0xc60 [gpu_sched]
[<ffffffffb212fdf2>] kthread+0x342/0x410
[<ffffffffb20030d2>] ret_from_fork+0x22/0x30
unreferenced object 0xffff888246342640 (size 216):
comm "mate-session-ch", pid 718, jiffies 4294890392 (age 3722.200s)
hex dump (first 32 bytes):
d0 26 34 46 82 88 ff ff 20 8b 72 c0 ff ff ff ff .&4F.... .r.....
03 71 65 92 0a 00 00 00 58 fd a6 08 00 c9 ff ff .qe.....X.......
backtrace:
[<ffffffffc072514f>] drm_sched_fence_create+0x1f/0x1d0 [gpu_sched]
[<ffffffffc07204de>] drm_sched_job_init+0x10e/0x240 [gpu_sched]
[<ffffffffc159e317>] amdgpu_job_submit+0x27/0x2d0 [amdgpu]
[<ffffffffc118ae6e>] amdgpu_copy_buffer+0x49e/0x700 [amdgpu]
[<ffffffffc118b6ca>] amdgpu_ttm_copy_mem_to_mem+0x5fa/0xf00 [amdgpu]
[<ffffffffc118ce06>] amdgpu_bo_move+0x356/0x2180 [amdgpu]
[<ffffffffc0887897>] ttm_bo_handle_move_mem+0x1c7/0x620 [ttm]
[<ffffffffc088b297>] ttm_bo_validate+0x2c7/0x450 [ttm]
[<ffffffffc1193444>] amdgpu_bo_fault_reserve_notify+0x2a4/0x640 [amdgpu]
[<ffffffffc11a3323>] amdgpu_gem_fault+0x123/0x2d0 [amdgpu]
[<ffffffffb255c853>] __do_fault+0xf3/0x3e0
[<ffffffffb256ea4b>] __handle_mm_fault+0x1bcb/0x2ac0
[<ffffffffb256fa6a>] handle_mm_fault+0x12a/0x490
[<ffffffffb2090909>] do_user_addr_fault+0x259/0xb70
[<ffffffffb37ad665>] exc_page_fault+0x55/0xb0
[<ffffffffb3800acb>] asm_exc_page_fault+0x1b/0x20
[...]
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are watching the assignee of the bug.
next prev parent reply other threads:[~2021-08-30 13:49 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-10 18:34 [Bug 214029] New: [NAVI] Several memory leaks in amdgpu and ttm bugzilla-daemon
2021-08-10 18:35 ` [Bug 214029] " bugzilla-daemon
2021-08-10 18:36 ` bugzilla-daemon
2021-08-30 13:45 ` bugzilla-daemon
2021-08-30 13:45 ` bugzilla-daemon
2021-08-30 13:49 ` bugzilla-daemon [this message]
2021-08-31 9:39 ` bugzilla-daemon
2021-08-31 10:20 ` bugzilla-daemon
2021-09-13 19:57 ` bugzilla-daemon
2021-09-13 19:58 ` bugzilla-daemon
2021-09-13 20:00 ` bugzilla-daemon
2021-09-20 16:29 ` bugzilla-daemon
2021-09-20 16:30 ` bugzilla-daemon
2021-09-20 16:35 ` bugzilla-daemon
2021-09-20 16:35 ` [Bug 214029] [bisected] " bugzilla-daemon
2021-09-22 22:05 ` bugzilla-daemon
2021-09-22 22:05 ` [Bug 214029] [bisected] [amdgpu] " bugzilla-daemon
2021-09-28 15:21 ` bugzilla-daemon
2021-09-28 15:25 ` bugzilla-daemon
2021-10-15 22:53 ` bugzilla-daemon
2021-10-15 22:55 ` bugzilla-daemon
2021-10-20 17:46 ` bugzilla-daemon
2021-10-21 7:37 ` bugzilla-daemon
2021-10-21 8:44 ` bugzilla-daemon
2021-10-21 12:39 ` bugzilla-daemon
2021-11-03 9:45 ` bugzilla-daemon
2021-11-03 15:07 ` bugzilla-daemon
2021-11-03 17:54 ` bugzilla-daemon
2021-11-04 7:16 ` bugzilla-daemon
2021-11-04 7:44 ` bugzilla-daemon
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=bug-214029-2300-sAdPc9KTz3@https.bugzilla.kernel.org/ \
--to=bugzilla-daemon@bugzilla.kernel.org \
--cc=dri-devel@lists.freedesktop.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