1 ===============
2 GPU Debugging
3 ===============
4
5 GPUVM Debugging
6 ===============
7
8 To aid in debugging GPU virtual memory related
9 number of options module parameters:
10
11 `vm_fault_stop` - If non-0, halt the GPU memor
12
13 `vm_update_mode` - If non-0, use the CPU to up
14 the GPU.
15
16
17 Decoding a GPUVM Page Fault
18 ===========================
19
20 If you see a GPU page fault in the kernel log,
21 out what is going wrong in your application.
22 log may look something like this:
23
24 ::
25
26 [gfxhub0] no-retry page fault (src_id:0 ring:
27 in page starting at address 0x0000800102800
28 VM_L2_PROTECTION_FAULT_STATUS:0x00301030
29 Faulty UTCL2 client ID: TCP (0x8)
30 MORE_FAULTS: 0x0
31 WALKER_ERROR: 0x0
32 PERMISSION_FAULTS: 0x3
33 MAPPING_ERROR: 0x0
34 RW: 0x0
35
36 First you have the memory hub, gfxhub and mmhu
37 hub used for graphics, compute, and sdma on so
38 memory hub used for multi-media and sdma on so
39
40 Next you have the vmid and pasid. If the vmid
41 caused by the kernel driver or firmware. If t
42 a fault in a user application. The pasid is u
43 process id. If the process is active when the
44 information will be printed.
45
46 The GPU virtual address that caused the fault
47
48 The client ID indicates the GPU block that cau
49 Some common client IDs:
50
51 - CB/DB: The color/depth backend of the graphi
52 - CPF: Command Processor Frontend
53 - CPC: Command Processor Compute
54 - CPG: Command Processor Graphics
55 - TCP/SQC/SQG: Shaders
56 - SDMA: SDMA engines
57 - VCN: Video encode/decode engines
58 - JPEG: JPEG engines
59
60 PERMISSION_FAULTS describe what faults were en
61
62 - bit 0: the PTE was not valid
63 - bit 1: the PTE read bit was not set
64 - bit 2: the PTE write bit was not set
65 - bit 3: the PTE execute bit was not set
66
67 Finally, RW, indicates whether the access was
68
69 In the example above, a shader (cliend id = TC
70 an invalid page (PERMISSION_FAULTS = 0x3) at G
71 0x0000800102800000. The user can then inspect
72 descriptor state to determine what caused the
73
74 UMR
75 ===
76
77 `umr <https://gitlab.freedesktop.org/tomstdeni
78 GPU debugging and diagnostics tool. Please se
79 `documentation <https://umr.readthedocs.io/en/
80 about its capabilities.
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.