Bug 112302

Summary: [kernel 5.4-rc7][amdgpu] kmemleak reports various leaks
Product: DRI Reporter: erhard_f
Component: DRM/AMDgpuAssignee: Default DRI bug account <dri-devel>
Status: RESOLVED MOVED QA Contact:
Severity: normal    
Priority: not set    
Version: DRI git   
Hardware: x86-64 (AMD64)   
OS: All   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
cat /sys/kernel/debug/kmemleak
none
dmesg (kernel 5.4-rc7)
none
kernel .config (kernel 5.4-rc7) none

Description erhard_f 2019-11-16 11:26:57 UTC
Created attachment 145972 [details]
cat /sys/kernel/debug/kmemleak

[ 1306.389652] kmemleak: 28 new suspected memory leaks (see /sys/kernel/debug/kmemleak)

Running X I get various kmemleak dumps for amdgpu:
[...]
unreferenced object 0xffffa2501f3374c0 (size 208):
  comm "Xorg", pid 694, jiffies 4294885609 (age 1870.427s)
  hex dump (first 32 bytes):
    50 75 33 1f 50 a2 ff ff a0 41 47 c0 ff ff ff ff  Pu3.P....AG.....
    b6 d7 06 a0 06 00 00 00 48 fe aa 8c b5 b1 ff ff  ........H.......
  backtrace:
    [<00000000502b8475>] drm_sched_fence_create+0x1d/0xc0 [gpu_sched]
    [<0000000067f89939>] drm_sched_job_init+0x5b/0xa0 [gpu_sched]
    [<0000000097e803db>] amdgpu_job_submit+0x1e/0x90 [amdgpu]
    [<000000008af66624>] amdgpu_copy_buffer+0x1ab/0x290 [amdgpu]
    [<000000005723dc17>] amdgpu_ttm_copy_mem_to_mem+0x25b/0x540 [amdgpu]
    [<000000009372cf45>] amdgpu_move_blit.constprop.0+0x85/0x1b0 [amdgpu]
    [<00000000b09834ab>] amdgpu_bo_move+0x168/0x2b0 [amdgpu]
    [<000000009f1ae718>] ttm_bo_handle_move_mem+0x126/0x5c0 [ttm]
    [<00000000d0c0e30f>] ttm_bo_validate+0x174/0x1b0 [ttm]
    [<00000000812e63d8>] amdgpu_bo_fault_reserve_notify+0xe2/0x130 [amdgpu]
    [<00000000340ebb51>] ttm_bo_vm_fault+0xa7/0x650 [ttm]
    [<00000000a9630d5e>] __do_fault+0x31/0xf0
    [<000000008cecfe93>] __handle_mm_fault+0xccc/0x1390
    [<00000000984b2f55>] handle_mm_fault+0x15b/0x2f0
    [<0000000053ca6a37>] __do_page_fault+0x1a2/0x420
    [<00000000c546a67a>] page_fault+0x31/0x40
unreferenced object 0xffffa24c4471b040 (size 72):
  comm "sdma0", pid 402, jiffies 4294885609 (age 1870.427s)
  hex dump (first 32 bytes):
    a0 c7 36 59 4c a2 ff ff e0 95 fd c0 ff ff ff ff  ..6YL...........
    a4 2b 08 a0 06 00 00 00 58 e4 2a 61 50 a2 ff ff  .+......X.*aP...
  backtrace:
    [<0000000078aa5208>] amdgpu_fence_emit+0x2b/0x270 [amdgpu]
    [<00000000a223cd76>] amdgpu_ib_schedule+0x311/0x540 [amdgpu]
    [<000000004db3d964>] amdgpu_job_run+0xfc/0x1e0 [amdgpu]
    [<00000000bfb92cc3>] drm_sched_main+0xf7/0x2e0 [gpu_sched]
    [<00000000a857f89f>] kthread+0xf6/0x130
    [<0000000046d81882>] ret_from_fork+0x27/0x50
[...]
Comment 1 erhard_f 2019-11-16 11:27:33 UTC
Created attachment 145973 [details]
dmesg (kernel 5.4-rc7)
Comment 2 erhard_f 2019-11-16 11:28:08 UTC
Created attachment 145974 [details]
kernel .config (kernel 5.4-rc7)
Comment 3 Martin Peres 2019-11-19 10:01:59 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/drm/amd/issues/958.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.