Merged
Conversation
nixpkgs-ci
bot
added
10.rebuild-linux: 11-100
LunNova
force-pushed
the
lunnova/rocm-7.1
branch
2 times, most recently
from
6629997 to
ca5dbb0
Compare
LunNova
force-pushed
the
lunnova/rocm-7.1
branch
2 times, most recently
from
05759db to
1170705
Compare
LunNova
force-pushed
the
lunnova/rocm-7.1
branch
2 times, most recently
from
8db2357 to
1558c61
Compare
Member
Author
|
Pretty surprised how well this bump is going, successfully built everything already. RCCL link time has gotten worse; it's taking hours and using >64G of RAM :c |
Contributor
https://github.com/GZGavinZhao/rocm-systems/commits/solus-rocm-7.1.x/ ? |
Member
Author
|
! I forgot you already had that ready thanks so much, will swap that in and then burn another overnight big rebuild lol. |
LunNova
force-pushed
the
lunnova/rocm-7.1
branch
2 times, most recently
from
45164cb to
d81ece2
Compare
Member
Author
|
Rebuild after applying the ISA compat patches went well too. |
Contributor
|
Is this expected to fix the The logs generally look like this: To be clear, I do not have this patch installed. |
Member
Author
|
No, as far as I know that's a linux-firmware issue and not related to the ROCm version. #472914 was expected to resolve it (but I guess doesn't work since that should've reached you if you updated unstable recently) You might want to try older kernel and older firmware, and if you find a combination that works you could post that info on the amdgpu issue you linked on freedesktop. I think ROCm/ROCm#5844 is tracking the same thing you're seeing. |
Flakebi
approved these changes
Jan 21, 2026
Member
Flakebi
left a comment
Flakebi
left a comment
There was a problem hiding this comment.
Didn’t build it but the package changes look good to me, thanks!
nixpkgs-ci
bot
added
12.approvals: 1
LunNova
force-pushed
the
lunnova/rocm-7.1
branch
from
d81ece2 to
c4f2ce6
Compare
nixpkgs-ci
bot
added
the
2.status: merge conflict
fabianhjr
approved these changes
Jan 24, 2026
nixpkgs-ci
bot
added
12.approvals: 2
LunNova
force-pushed
the
lunnova/rocm-7.1
branch
from
c4f2ce6 to
ab49426
Compare
LunNova
commented
Jan 25, 2026
Comment on lines
+56
to
+58
| # Vendored upstream PR for fix for segfault when queue allocation fails | ||
| # https://github.com/ROCm/rocm-systems/pull/2850 | ||
| ./queue-failure.patch |
…on fails Found by @06kellyjac, possibly triggered by bug in linux-6.18.4 https://lore.kernel.org/linux-iommu/870872aa-28e9-412a-bac6-8020bf560e4f@amd.com/t/ Resolving this *will not fix the underlying issue*, but does transmute a segfault into an actual error result. ROCm error: out of memory current device: 0, in function stream at /build/source/ggml/src/ggml-hip/../ggml-cuda/common.cuh:1345 hipStreamCreateWithFlags(&streams[device][stream], 0x01)
LunNova
force-pushed
the
lunnova/rocm-7.1
branch
from
ab49426 to
6713c5e
Compare
nixpkgs-ci
bot
removed
the
2.status: merge conflict
LunNova
commented
Jan 25, 2026
Comment on lines
+45
to
+47
| # Fix error: redefinition of 'struct drm_color_ctm_3x4' | ||
| # https://github.com/ROCm/amdsmi/pull/165 | ||
| ./drm-struct-redefinition-fix.patch |
Member
Author
|
Rebased, fixed a conflict, fixed a new issue with amdsmi. Planning to merge later today if this review run goes well. |
Member
Author
|
Member
|
@elis1000 I think the following table might clarify a few solutions: However, 6 months ago, I tried to upgrade ROCm packages I experienced many issues that I had no idea how to solve. Thus I am thankful for the work done by @LunNova , and I will patiently wait for the version 7.2 to becomes stable. |
Contributor
|
@LunNova is it just me or is |
Member
Author
It failed, it's succeeding again, looks like it OOMed on the builder. Not much I can do, when the channel advances a bit more it'll be cached again.
|
Contributor
|
Makes sense; any chance patching in this old draft PR of yours in might fix the sporadic OOMing / build failures? |
Contributor
|
If I'm understanding correctly, a large part of these changes are already applied for ROCm 7.2. |
16 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Closes #462424.
nixpkgs-reviewresultGenerated using
nixpkgs-review.Command:
nixpkgs-review pr 481349Commit:
d81ece2a6fa5d30929ecfe7aba631b1f06520031x86_64-linux✅ 113 packages built:
Things done
passthru.tests.nixpkgs-reviewon this PR. See nixpkgs-review usage../result/bin/.Add a 👍 reaction to pull requests you find important.