Bug 213201 - [KAVERI] memory leak - unreferenced object 0xffff8881700cf988 (size 56)
Summary: [KAVERI] memory leak - unreferenced object 0xffff8881700cf988 (size 56)
Status: NEW
Alias: None
Product: Drivers
Classification: Unclassified
Component: Video(DRI - non Intel) (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: drivers_video-dri
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2021-05-24 15:31 UTC by Erhard F.
Modified: 2021-10-04 23:46 UTC (History)
2 users (show)

See Also:
Kernel Version: 5.13-rc3
Subsystem:
Regression: No
Bisected commit-id:


Attachments
kernel dmesg (5.13-rc3, AMD A10 PRO-7800B) (82.81 KB, text/plain)
2021-05-24 15:31 UTC, Erhard F.
Details
kernel .config (5.13-rc3, AMD A10 PRO-7800B) (108.16 KB, text/plain)
2021-05-24 15:33 UTC, Erhard F.
Details
kernel dmesg (4.4.271, AMD A10 PRO-7800B) (55.26 KB, text/plain)
2021-06-15 19:33 UTC, Erhard F.
Details
kernel .config (4.4.271, AMD A10 PRO-7800B) (84.55 KB, text/plain)
2021-06-15 19:35 UTC, Erhard F.
Details
output of /sys/kernel/debug/kmemleak (kernel 4.4.271) (8.60 KB, text/plain)
2021-06-15 19:36 UTC, Erhard F.
Details
output of /sys/kernel/debug/kmemleak (kernel 5.14-rc2) (2.43 KB, text/plain)
2021-07-19 13:29 UTC, Erhard F.
Details
kernel dmesg (5.14-rc2, AMD A10-8750) (85.02 KB, text/plain)
2021-07-19 13:30 UTC, Erhard F.
Details
kernel .config (5.14-rc2, AMD A10-8750) (109.24 KB, text/plain)
2021-07-19 13:32 UTC, Erhard F.
Details
kernel .config (5.14-rc4, AMD A10-7860K) (108.99 KB, text/plain)
2021-08-06 00:12 UTC, Erhard F.
Details
kernel dmesg (5.14-rc4, AMD A10-7860K) (83.82 KB, text/plain)
2021-08-06 00:13 UTC, Erhard F.
Details
output of /sys/kernel/debug/kmemleak (kernel 5.14-rc4) (2.39 KB, text/plain)
2021-08-06 00:17 UTC, Erhard F.
Details
utput of /sys/kernel/debug/kmemleak (kernel 5.15-rc4, AMD PRO A10-8750B) (2.47 KB, text/plain)
2021-10-04 23:46 UTC, Erhard F.
Details

Description Erhard F. 2021-05-24 15:31:54 UTC
Created attachment 296967 [details]
kernel dmesg (5.13-rc3, AMD A10 PRO-7800B)

Get this shortly after booting kernel 5.13-rc3. Looks like amdgpu is involved:

unreferenced object 0xffff8881700cf988 (size 56):
  comm "systemd-udevd", pid 216, jiffies 4294885043 (age 1739.830s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 0d 01 70 00 00 00 00 00  ..........p.....
    7b 98 01 00 00 c9 ff ff 00 00 00 00 00 00 00 00  {...............
  backtrace:
    [<00000000f3e5fc85>] kmem_cache_alloc+0x10a/0x1e4
    [<00000000fff767c3>] acpi_ps_alloc_op+0x4e/0xdc
    [<00000000445cacf1>] acpi_ps_create_op+0x24e/0x473
    [<000000003452a27a>] acpi_ps_parse_loop+0x25e/0xa39
    [<000000004e3469e0>] acpi_ps_parse_aml+0xf3/0x450
    [<00000000be1f7fd0>] acpi_ps_execute_method+0x278/0x2bb
    [<000000004564e9a8>] acpi_ns_evaluate+0x34f/0x45d
    [<0000000072f73c39>] acpi_evaluate_object+0x201/0x3b5
    [<0000000036379e33>] amdgpu_atcs_call+0x11d/0x19b [amdgpu]
    [<00000000829cc77d>] amdgpu_acpi_init+0xf3/0x7b8 [amdgpu]
    [<00000000671194bb>] amdgpu_driver_load_kms+0x1e5/0x354 [amdgpu]
    [<000000005ad1722a>] amdgpu_pci_probe+0x150/0x1d2 [amdgpu]
    [<00000000e32ddc8a>] pci_device_probe+0xf3/0x17e
    [<00000000f74eb70d>] really_probe+0x215/0x539
    [<00000000af9fda84>] driver_probe_device+0x9d/0xd6
    [<00000000bc527018>] device_driver_attach+0x66/0x8a
unreferenced object 0xffff8881700ceb30 (size 56):
  comm "systemd-udevd", pid 216, jiffies 4294885043 (age 1739.834s)
  hex dump (first 32 bytes):
    88 f9 0c 70 81 88 ff ff 0d 01 2d 00 00 00 00 00  ...p......-.....
    7c 98 01 00 00 c9 ff ff 00 00 00 00 00 00 00 00  |...............
  backtrace:
    [<00000000f3e5fc85>] kmem_cache_alloc+0x10a/0x1e4
    [<00000000fff767c3>] acpi_ps_alloc_op+0x4e/0xdc
    [<00000000445cacf1>] acpi_ps_create_op+0x24e/0x473
    [<000000003452a27a>] acpi_ps_parse_loop+0x25e/0xa39
    [<000000004e3469e0>] acpi_ps_parse_aml+0xf3/0x450
    [<00000000be1f7fd0>] acpi_ps_execute_method+0x278/0x2bb
    [<000000004564e9a8>] acpi_ns_evaluate+0x34f/0x45d
    [<0000000072f73c39>] acpi_evaluate_object+0x201/0x3b5
    [<0000000036379e33>] amdgpu_atcs_call+0x11d/0x19b [amdgpu]
    [<00000000829cc77d>] amdgpu_acpi_init+0xf3/0x7b8 [amdgpu]
    [<00000000671194bb>] amdgpu_driver_load_kms+0x1e5/0x354 [amdgpu]
    [<000000005ad1722a>] amdgpu_pci_probe+0x150/0x1d2 [amdgpu]
    [<00000000e32ddc8a>] pci_device_probe+0xf3/0x17e
    [<00000000f74eb70d>] really_probe+0x215/0x539
    [<00000000af9fda84>] driver_probe_device+0x9d/0xd6
    [<00000000bc527018>] device_driver_attach+0x66/0x8a


 # inxi -bZ
System:    Host: yea Kernel: 5.13.0-rc3-bdver3 x86_64 bits: 64 Console: tty 1 Distro: Gentoo Base System release 2.7 
Machine:   Type: Desktop Mobo: ASRock model: FM2A88M Pro3+ serial: N/A UEFI: American Megatrends v: P2.60 date: 01/11/2016 
CPU:       Info: Quad Core AMD A10 PRO-7800B R7 12 Compute Cores 4C+8G [MCP] speed: 1897 MHz min/max: 1400/3500 MHz 
Graphics:  Device-1: Advanced Micro Devices [AMD/ATI] Kaveri [Radeon R7 Graphics] driver: amdgpu v: kernel 
           Display: server: X.org 1.20.11 driver: amdgpu,ati unloaded: fbdev,modesetting tty: 211x54 
           Message: Advanced graphics data unavailable in console for root. 
Network:   Device-1: Realtek RTL8111/8168/8411 PCI Express Gigabit Ethernet driver: r8169 


 # lspci -s 00:01.0 -vv
00:01.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Kaveri [Radeon R7 Graphics] (prog-if 00 [VGA controller])
	Subsystem: ASRock Incorporation Kaveri [Radeon R7 Graphics]
	Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
	Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
	Latency: 0, Cache Line Size: 64 bytes
	Interrupt: pin A routed to IRQ 56
	IOMMU group: 0
	Region 0: Memory at c0000000 (64-bit, prefetchable) [size=256M]
	Region 2: Memory at d0000000 (64-bit, prefetchable) [size=8M]
	Region 4: I/O ports at f000 [size=256]
	Region 5: Memory at feb00000 (32-bit, non-prefetchable) [size=256K]
	Expansion ROM at 000c0000 [disabled] [size=128K]
	Capabilities: [48] Vendor Specific Information: Len=08 <?>
	Capabilities: [50] Power Management version 3
		Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0-,D1+,D2+,D3hot+,D3cold-)
		Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
	Capabilities: [58] Express (v2) Root Complex Integrated Endpoint, MSI 00
		DevCap:	MaxPayload 256 bytes, PhantFunc 0
			ExtTag+ RBE+ FLReset-
		DevCtl:	CorrErr- NonFatalErr- FatalErr- UnsupReq-
			RlxdOrd+ ExtTag+ PhantFunc- AuxPwr- NoSnoop+
			MaxPayload 256 bytes, MaxReadReq 512 bytes
		DevSta:	CorrErr- NonFatalErr- FatalErr- UnsupReq- AuxPwr- TransPend-
		DevCap2: Completion Timeout: Not Supported, TimeoutDis- NROPrPrP- LTR-
			 10BitTagComp- 10BitTagReq- OBFF Not Supported, ExtFmt+ EETLPPrefix+, MaxEETLPPrefixes 1
			 EmergencyPowerReduction Not Supported, EmergencyPowerReductionInit-
			 FRS-
			 AtomicOpsCap: 32bit- 64bit- 128bitCAS-
		DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis- LTR- OBFF Disabled,
			 AtomicOpsCtl: ReqEn-
	Capabilities: [a0] MSI: Enable+ Count=1/1 Maskable- 64bit+
		Address: 00000000fee02004  Data: 0029
	Capabilities: [100 v1] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?>
	Capabilities: [270 v1] Secondary PCI Express
		LnkCtl3: LnkEquIntrruptEn- PerformEqu-
		LaneErrStat: 0
	Capabilities: [2b0 v1] Address Translation Service (ATS)
		ATSCap:	Invalidate Queue Depth: 00
		ATSCtl:	Enable+, Smallest Translation Unit: 00
	Capabilities: [2c0 v1] Page Request Interface (PRI)
		PRICtl: Enable- Reset-
		PRISta: RF- UPRGI- Stopped+
		Page Request Capacity: 00000020, Page Request Allocation: 00000000
	Capabilities: [2d0 v1] Process Address Space ID (PASID)
		PASIDCap: Exec- Priv-, Max PASID Width: 10
		PASIDCtl: Enable- Exec- Priv-
	Kernel driver in use: amdgpu
	Kernel modules: amdgpu
Comment 1 Erhard F. 2021-05-24 15:33:21 UTC
Created attachment 296969 [details]
kernel .config (5.13-rc3, AMD A10 PRO-7800B)
Comment 2 Alex Deucher 2021-05-24 18:46:19 UTC
The amdgpu acpi handling hasn't really changed in ages.  If this is new for 5.13 can you bisect?
Comment 3 Erhard F. 2021-05-24 19:29:36 UTC
As I bought this machine just recently I can't tell. But I will try to bisect.

I got this Kaveri-based one instead of the Excavator-based one from bug #211875. With bug #211875 it was not fit for it's purpose unfortunately so I sold it again. This Kaveri-based one here fares much better. Only this memory leak but no grave issues otherwise.
Comment 4 Erhard F. 2021-05-24 21:42:45 UTC
Happens on 5.11.x and 5.10.x too. This may take a while...
Comment 5 Erhard F. 2021-06-15 19:33:36 UTC
Created attachment 297375 [details]
kernel dmesg (4.4.271, AMD A10 PRO-7800B)

Bisecting was no success, but I traced the memory leak back to kernel 4.4.x series.

Kernels v4.3.6 and v4.2.8 won't boot on the machine. But probably the leak has been there since amdgpu got first introduced in kernel 4.2 anyhow.

[...]
unreferenced object 0xffff8803ecbef380 (size 56):
  comm "systemd-udevd", pid 222, jiffies 4294880256 (age 1158.457s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 0d 01 70 00 00 00 00 00  ..........p.....
    7b 58 01 00 00 c9 ff ff 00 00 00 00 00 00 00 00  {X..............
  backtrace:
    [<ffffffffa110c0c9>] create_object+0x162/0x280
    [<ffffffffa13d741a>] kmemleak_alloc+0x20/0x3c
    [<ffffffffa10fa921>] slab_post_alloc_hook+0x44/0x65
    [<ffffffffa10fc73a>] kmem_cache_alloc+0xbf/0xe2
    [<ffffffffa1263026>] acpi_os_acquire_object+0x28/0x2a
    [<ffffffffa126306f>] acpi_ps_alloc_op+0x39/0x69
    [<ffffffffa12622ff>] acpi_ps_create_op+0x105/0x234
    [<ffffffffa1261cf2>] acpi_ps_parse_loop+0x1db/0x586
    [<ffffffffa1262b62>] acpi_ps_parse_aml+0x93/0x265
    [<ffffffffa126335c>] acpi_ps_execute_method+0x14e/0x185
    [<ffffffffa125d96c>] acpi_ns_evaluate+0x1bf/0x25b
    [<ffffffffa12604a9>] acpi_evaluate_object+0x13c/0x24c
    [<ffffffffc075864e>] amdgpu_atcs_call+0x83/0xd9 [amdgpu]
    [<ffffffffc0758bdf>] amdgpu_acpi_init+0x62/0x4ef [amdgpu]
    [<ffffffffc06f6028>] amdgpu_driver_load_kms+0x83/0x103 [amdgpu]
    [<ffffffffc04445f1>] drm_dev_register+0x6d/0xae [drm]
unreferenced object 0xffff8803ecbeef00 (size 56):
  comm "systemd-udevd", pid 222, jiffies 4294880256 (age 1158.457s)
  hex dump (first 32 bytes):
    80 f3 be ec 03 88 ff ff 0d 01 2d 00 00 00 00 00  ..........-.....
    7c 58 01 00 00 c9 ff ff 00 00 00 00 00 00 00 00  |X..............
  backtrace:
    [<ffffffffa110c0c9>] create_object+0x162/0x280
    [<ffffffffa13d741a>] kmemleak_alloc+0x20/0x3c
    [<ffffffffa10fa921>] slab_post_alloc_hook+0x44/0x65
    [<ffffffffa10fc73a>] kmem_cache_alloc+0xbf/0xe2
    [<ffffffffa1263026>] acpi_os_acquire_object+0x28/0x2a
    [<ffffffffa126306f>] acpi_ps_alloc_op+0x39/0x69
    [<ffffffffa12622ff>] acpi_ps_create_op+0x105/0x234
    [<ffffffffa1261cf2>] acpi_ps_parse_loop+0x1db/0x586
    [<ffffffffa1262b62>] acpi_ps_parse_aml+0x93/0x265
    [<ffffffffa126335c>] acpi_ps_execute_method+0x14e/0x185
    [<ffffffffa125d96c>] acpi_ns_evaluate+0x1bf/0x25b
    [<ffffffffa12604a9>] acpi_evaluate_object+0x13c/0x24c
    [<ffffffffc075864e>] amdgpu_atcs_call+0x83/0xd9 [amdgpu]
    [<ffffffffc0758bdf>] amdgpu_acpi_init+0x62/0x4ef [amdgpu]
    [<ffffffffc06f6028>] amdgpu_driver_load_kms+0x83/0x103 [amdgpu]
    [<ffffffffc04445f1>] drm_dev_register+0x6d/0xae [drm]
Comment 6 Erhard F. 2021-06-15 19:35:16 UTC
Created attachment 297377 [details]
kernel .config (4.4.271, AMD A10 PRO-7800B)
Comment 7 Erhard F. 2021-06-15 19:36:58 UTC
Created attachment 297379 [details]
output of /sys/kernel/debug/kmemleak (kernel 4.4.271)
Comment 8 Erhard F. 2021-07-19 13:29:36 UTC
Created attachment 297921 [details]
output of /sys/kernel/debug/kmemleak (kernel 5.14-rc2)

Looking slightly different on kernel v5.14-rc2 on another board w. another CPU (ASRock A88M-G/3.1 w. AMD A10-8750)

[...]
unreferenced object 0xffff888142247988 (size 56):
  comm "systemd-udevd", pid 206, jiffies 4294882633 (age 2086.390s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 0d 01 70 00 00 00 00 00  ..........p.....
    13 5e 02 00 00 c9 ff ff 00 00 00 00 00 00 00 00  .^..............
  backtrace:
    [<ffffffffa93fc402>] kmem_cache_alloc+0x10a/0x1e3
    [<ffffffffa9a5d413>] acpi_ps_alloc_op+0x8b/0x1c4
    [<ffffffffa9a5a5ad>] acpi_ps_create_op+0x4b1/0x8ec
    [<ffffffffa9a590b8>] acpi_ps_parse_loop+0x401/0x1062
    [<ffffffffa9a5c360>] acpi_ps_parse_aml+0x1cd/0x6fa
    [<ffffffffa9a5de90>] acpi_ps_execute_method+0x51f/0x57b
    [<ffffffffa9a4da85>] acpi_ns_evaluate+0x64c/0x886
    [<ffffffffa9a55a32>] acpi_evaluate_object+0x335/0x690
    [<ffffffffc115f531>] amdgpu_atcs_call.constprop.0+0x141/0x1bd [amdgpu]
    [<ffffffffc115fa8a>] amdgpu_atcs_pci_probe_handle.isra.0+0x147/0x2a1 [amdgpu]
    [<ffffffffc1160a9a>] amdgpu_acpi_detect+0xd1/0x38e [amdgpu]
    [<ffffffffc1cb30aa>] hid_lookup_collection+0xaa/0x176 [hid]
    [<ffffffffa90017dc>] do_one_initcall+0x159/0x31d
    [<ffffffffa91d0584>] do_init_module+0x1ae/0x573
    [<ffffffffa91d5b2b>] load_module+0x5063/0x5f46
    [<ffffffffa91d6ccc>] __do_sys_finit_module+0xf6/0x145
unreferenced object 0xffff888142246338 (size 56):
  comm "systemd-udevd", pid 206, jiffies 4294882633 (age 2086.390s)
  hex dump (first 32 bytes):
    88 79 24 42 81 88 ff ff 0d 01 2d 00 00 00 00 00  .y$B......-.....
    14 5e 02 00 00 c9 ff ff 00 00 00 00 00 00 00 00  .^..............
  backtrace:
    [<ffffffffa93fc402>] kmem_cache_alloc+0x10a/0x1e3
    [<ffffffffa9a5d413>] acpi_ps_alloc_op+0x8b/0x1c4
    [<ffffffffa9a5a5ad>] acpi_ps_create_op+0x4b1/0x8ec
    [<ffffffffa9a590b8>] acpi_ps_parse_loop+0x401/0x1062
    [<ffffffffa9a5c360>] acpi_ps_parse_aml+0x1cd/0x6fa
    [<ffffffffa9a5de90>] acpi_ps_execute_method+0x51f/0x57b
    [<ffffffffa9a4da85>] acpi_ns_evaluate+0x64c/0x886
    [<ffffffffa9a55a32>] acpi_evaluate_object+0x335/0x690
    [<ffffffffc115f531>] amdgpu_atcs_call.constprop.0+0x141/0x1bd [amdgpu]
    [<ffffffffc115fa8a>] amdgpu_atcs_pci_probe_handle.isra.0+0x147/0x2a1 [amdgpu]
    [<ffffffffc1160a9a>] amdgpu_acpi_detect+0xd1/0x38e [amdgpu]
    [<ffffffffc1cb30aa>] hid_lookup_collection+0xaa/0x176 [hid]
    [<ffffffffa90017dc>] do_one_initcall+0x159/0x31d
    [<ffffffffa91d0584>] do_init_module+0x1ae/0x573
    [<ffffffffa91d5b2b>] load_module+0x5063/0x5f46
    [<ffffffffa91d6ccc>] __do_sys_finit_module+0xf6/0x145


 # inxi -bZ
System:    Host: yea Kernel: 5.14.0-rc2-bdver3 x86_64 bits: 64 Desktop: MATE 1.24.1 
           Distro: Gentoo Base System release 2.7 
Machine:   Type: Desktop Mobo: ASRock model: A88M-G/3.1 serial: N/A 
           UEFI: American Megatrends v: P1.40C date: 11/21/2016 
CPU:       Info: Quad Core AMD A10-8750 Radeon R7 12 Compute Cores 4C+8G [MCP] speed: 1809 MHz 
           min/max: 1400/3500 MHz 
Graphics:  Device-1: AMD Kaveri [Radeon R7 Graphics] driver: amdgpu v: kernel 
           Display: x11 server: X.Org 1.20.11 driver: amdgpu,ati unloaded: fbdev,modesetting 
           resolution: 1920x1080~60Hz 
           OpenGL: renderer: AMD KAVERI (DRM 3.42.0 5.14.0-rc2-bdver3 LLVM 12.0.0) 
           v: 4.6 Mesa 21.1.4 
Network:   Device-1: Realtek RTL8111/8168/8411 PCI Express Gigabit Ethernet driver: r8169
Comment 9 Erhard F. 2021-07-19 13:30:36 UTC
Created attachment 297923 [details]
kernel dmesg (5.14-rc2, AMD A10-8750)
Comment 10 Erhard F. 2021-07-19 13:32:34 UTC
Created attachment 297925 [details]
kernel .config (5.14-rc2, AMD A10-8750)
Comment 11 Erhard F. 2021-08-06 00:12:11 UTC
Created attachment 298215 [details]
kernel .config (5.14-rc4, AMD A10-7860K)
Comment 12 Erhard F. 2021-08-06 00:13:13 UTC
Created attachment 298217 [details]
kernel dmesg (5.14-rc4, AMD A10-7860K)
Comment 13 Erhard F. 2021-08-06 00:17:17 UTC
Created attachment 298219 [details]
output of /sys/kernel/debug/kmemleak (kernel 5.14-rc4)

Same board, another CPU:

unreferenced object 0xffff888102e48bd0 (size 56):
  comm "systemd-udevd", pid 199, jiffies 4294881489 (age 3502.134s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 0d 01 70 00 00 00 00 00  ..........p.....
    7b 5d 02 00 00 c9 ff ff 00 00 00 00 00 00 00 00  {]..............
  backtrace:
    [<ffffffff9e33706e>] kmem_cache_alloc+0x109/0x132
    [<ffffffff9e92d8aa>] acpi_ps_alloc_op+0x8b/0x1c4
    [<ffffffff9e92aa44>] acpi_ps_create_op+0x4b1/0x8ec
    [<ffffffff9e92954f>] acpi_ps_parse_loop+0x401/0x1062
    [<ffffffff9e92c7f7>] acpi_ps_parse_aml+0x1cd/0x6fa
    [<ffffffff9e92e327>] acpi_ps_execute_method+0x51f/0x57b
    [<ffffffff9e91defb>] acpi_ns_evaluate+0x64c/0x886
    [<ffffffff9e925ec9>] acpi_evaluate_object+0x335/0x690
    [<ffffffffc112c200>] amdgpu_atcs_call.constprop.0+0x141/0x1bd [amdgpu]
    [<ffffffffc112c759>] amdgpu_atcs_pci_probe_handle.isra.0+0x147/0x2a1 [amdgpu]
    [<ffffffffc112d769>] amdgpu_acpi_detect+0xd1/0x38e [amdgpu]
    [<ffffffffc1c0c0aa>] 0xffffffffc1c0c0aa
    [<ffffffff9e0006b7>] do_one_initcall+0xe0/0x1fc
    [<ffffffff9e18517d>] do_init_module+0x1c1/0x584
    [<ffffffff9e18a561>] load_module+0x4ea2/0x5cc6
    [<ffffffff9e18b643>] __do_sys_finit_module+0xf6/0x145
unreferenced object 0xffff888102e48480 (size 56):
  comm "systemd-udevd", pid 199, jiffies 4294881489 (age 3502.134s)
  hex dump (first 32 bytes):
    d0 8b e4 02 81 88 ff ff 0d 01 2d 00 00 00 00 00  ..........-.....
    7c 5d 02 00 00 c9 ff ff 00 00 00 00 00 00 00 00  |]..............
  backtrace:
    [<ffffffff9e33706e>] kmem_cache_alloc+0x109/0x132
    [<ffffffff9e92d8aa>] acpi_ps_alloc_op+0x8b/0x1c4
    [<ffffffff9e92aa44>] acpi_ps_create_op+0x4b1/0x8ec
    [<ffffffff9e92954f>] acpi_ps_parse_loop+0x401/0x1062
    [<ffffffff9e92c7f7>] acpi_ps_parse_aml+0x1cd/0x6fa
    [<ffffffff9e92e327>] acpi_ps_execute_method+0x51f/0x57b
    [<ffffffff9e91defb>] acpi_ns_evaluate+0x64c/0x886
    [<ffffffff9e925ec9>] acpi_evaluate_object+0x335/0x690
    [<ffffffffc112c200>] amdgpu_atcs_call.constprop.0+0x141/0x1bd [amdgpu]
    [<ffffffffc112c759>] amdgpu_atcs_pci_probe_handle.isra.0+0x147/0x2a1 [amdgpu]
    [<ffffffffc112d769>] amdgpu_acpi_detect+0xd1/0x38e [amdgpu]
    [<ffffffffc1c0c0aa>] 0xffffffffc1c0c0aa
    [<ffffffff9e0006b7>] do_one_initcall+0xe0/0x1fc
    [<ffffffff9e18517d>] do_init_module+0x1c1/0x584
    [<ffffffff9e18a561>] load_module+0x4ea2/0x5cc6
    [<ffffffff9e18b643>] __do_sys_finit_module+0xf6/0x145
Comment 14 Erhard F. 2021-10-04 23:46:40 UTC
Created attachment 299101 [details]
utput of /sys/kernel/debug/kmemleak (kernel 5.15-rc4, AMD PRO A10-8750B)

Same board, another CPU:

 # cat /sys/kernel/debug/kmemleak
unreferenced object 0xffff8881010b5278 (size 56):
  comm "systemd-udevd", pid 198, jiffies 4294881523 (age 926.930s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 0d 01 70 00 00 00 00 00  ..........p.....
    4b 5f 02 00 00 c9 ff ff 00 00 00 00 00 00 00 00  K_..............
  backtrace:
    [<ffffffff9df51ae1>] acpi_ps_alloc_op+0x8b/0x1c4
    [<ffffffff9df4ec89>] acpi_ps_create_op+0x48c/0x8ca
    [<ffffffff9df4d7b9>] acpi_ps_parse_loop+0x401/0x1062
    [<ffffffff9df50a3e>] acpi_ps_parse_aml+0x1cd/0x6ea
    [<ffffffff9df5255e>] acpi_ps_execute_method+0x51f/0x57b
    [<ffffffff9df42193>] acpi_ns_evaluate+0x64b/0x885
    [<ffffffff9df4a12f>] acpi_evaluate_object+0x333/0x68d
    [<ffffffffc15b70f9>] amdgpu_atcs_call.constprop.0+0x129/0x1f0 [amdgpu]
    [<ffffffffc15b7b79>] amdgpu_atcs_pci_probe_handle.isra.0+0x159/0x2d0 [amdgpu]
    [<ffffffffc15b8950>] amdgpu_acpi_detect+0xb0/0x470 [amdgpu]
    [<ffffffffc243b0a8>] chacha_2block_xor_avx2+0xa8/0x220 [chacha_x86_64]
    [<ffffffff9d001b0b>] do_one_initcall+0xbb/0x3e0
    [<ffffffff9d2a7509>] do_init_module+0x1c9/0x6e0
    [<ffffffff9d2af427>] load_module+0x78c7/0x9bb0
    [<ffffffff9d2b1a40>] __do_sys_finit_module+0x110/0x180
    [<ffffffff9e7f1836>] do_syscall_64+0x66/0x90
unreferenced object 0xffff8881010b5c38 (size 56):
  comm "systemd-udevd", pid 198, jiffies 4294881523 (age 926.930s)
  hex dump (first 32 bytes):
    78 52 0b 01 81 88 ff ff 0d 01 2d 00 00 00 00 00  xR........-.....
    4c 5f 02 00 00 c9 ff ff 00 00 00 00 00 00 00 00  L_..............
  backtrace:
    [<ffffffff9df51ae1>] acpi_ps_alloc_op+0x8b/0x1c4
    [<ffffffff9df4ec89>] acpi_ps_create_op+0x48c/0x8ca
    [<ffffffff9df4d7b9>] acpi_ps_parse_loop+0x401/0x1062
    [<ffffffff9df50a3e>] acpi_ps_parse_aml+0x1cd/0x6ea
    [<ffffffff9df5255e>] acpi_ps_execute_method+0x51f/0x57b
    [<ffffffff9df42193>] acpi_ns_evaluate+0x64b/0x885
    [<ffffffff9df4a12f>] acpi_evaluate_object+0x333/0x68d
    [<ffffffffc15b70f9>] amdgpu_atcs_call.constprop.0+0x129/0x1f0 [amdgpu]
    [<ffffffffc15b7b79>] amdgpu_atcs_pci_probe_handle.isra.0+0x159/0x2d0 [amdgpu]
    [<ffffffffc15b8950>] amdgpu_acpi_detect+0xb0/0x470 [amdgpu]
    [<ffffffffc243b0a8>] chacha_2block_xor_avx2+0xa8/0x220 [chacha_x86_64]
    [<ffffffff9d001b0b>] do_one_initcall+0xbb/0x3e0
    [<ffffffff9d2a7509>] do_init_module+0x1c9/0x6e0
    [<ffffffff9d2af427>] load_module+0x78c7/0x9bb0
    [<ffffffff9d2b1a40>] __do_sys_finit_module+0x110/0x180
    [<ffffffff9e7f1836>] do_syscall_64+0x66/0x90

Note You need to log in before you can comment on or make changes to this bug.