Current issue state, recent activity, and per-issue timelines from the indexed issue data.
| Date | Opened | Closed | Comments | Events | Open Backlog |
|---|---|---|---|---|---|
| 2026-08-24 | 0 | 0 | 0 | 0 | 0 |
| 2026-08-23 | 0 | 0 | 1 | 1 | 1 |
| 2026-08-22 | 0 | 0 | 3 | 5 | 7 |
| 2026-08-21 | 1 | 1 | 0 | 0 | 0 |
| 2026-08-20 | 0 | 0 | 0 | 0 | 0 |
| 2026-08-19 | 1 | 0 | 0 | 0 | 0 |
| 2026-08-18 | 1 | 0 | 0 | 0 | 0 |
| 2026-08-17 | 0 | 0 | 0 | 0 | 0 |
| 2026-08-16 | 0 | 0 | 0 | 0 | 0 |
| 2026-08-15 | 0 | 0 | 0 | 0 | 0 |
| 2026-08-14 | 0 | 0 | 0 | 0 | 0 |
| 2026-08-13 | 0 | 0 | 0 | 0 | 0 |
| 2026-08-12 | 0 | 0 | 0 | 0 | 0 |
| 2026-08-11 | 0 | 0 | 0 | 0 | 0 |
Opened: 3
Closed: 1
Comments: 4
Events: 6
| Issue | Author | State | Labels | Comments | Reactions | Updated |
|---|---|---|---|---|---|---|
#6630 [Issue]: gfx1201 R9700 ROCm 7.14 llama.cpp generation hang, MES queue failure and PSP reset -62; Vulkan passes Opened 5 days ago | angelhalo | open | No labels | 2 | 1 | 1 day ago |
#6406 [Issue]: CPU keeps spinning at 100% as soon as first PyTorch function on 'cuda' device is called Opened 1 month ago | bitgamma | open | status: assessed | 6 | 0 | 2 days ago |
#6186 [Issue]: gfx1151 Constant PERMISSION_FAULTS (0x3) on Strix Halo (Ryzen AI MAX 395) during ROCm workloads Opened 4 months ago | valeriy-idatica | open | status: triage | 6 | 0 | 2 days ago |
#6627 [Issue]: Adding `-g` makes my HIP program crash with "illegal memory access" (works fine without `-g`) — is this a compiler bug? Opened 6 days ago | Yujif1Aero | open | status: triage | 4 | 0 | 2 days ago |
#5107 [Issue]: 100% GPU usage for multiple models with pipeline parallel Opened 1 year ago | DKingAlpha | open | Under Investigation | 25 | 0 | 3 days ago |
#6413 [Issue]: gfx1201 (Radeon AI PRO R9700) GPU page fault (PERMISSION_FAULTS) + GPU reset during sustained PyTorch/ROCm training, reproducible 2x same session Opened 1 month ago | FR-Mister-T | open - reopened | status: triage | 16 | 1 | 3 days ago |
#5825 [Issue]: ROCm OpenCL leaves GPU compute state wedged on gfx1201; subsequent Vulkan apps fail with VK_ERROR_DEVICE_LOST Opened 8 months ago | xplatinum | closed - completed | status: triage | 9 | 3 | 3 days ago |
#6596 TK Opened 15 days ago | Octomal | closed - completed | No labels | 0 | 6 | 3 days ago |
#6637 Repository deprecation notice: transition to ROCm/TheRock Opened 3 days ago | amitkumar-amd | open | No labels | 0 | 0 | 3 days ago |
