[hipblaslt] Fixing build issues for gfx_950_mx_rebase#4465
Merged
Conversation
bnemanich
reviewed
Feb 18, 2026
nakajee
reviewed
Feb 18, 2026
… by separating code paths of MX and Non-MX
NineKa
force-pushed
the
users/hongjche/gfx950_mx_rebase_build
branch
from
ccc586c to
72db936
Compare
NineKa
force-pushed
the
users/hongjche/gfx950_mx_rebase_build
branch
from
72db936 to
3c1f207
Compare
Contributor
Author
|
Build verified for a455dcb with |
nakajee
reviewed
Feb 19, 2026
nakajee
reviewed
Feb 20, 2026
Contributor
Author
|
build verified for 3ef482e |
nakajee
reviewed
Feb 24, 2026
NineKa
force-pushed
the
users/hongjche/gfx950_mx_rebase_build
branch
from
7f724a9 to
f720d64
Compare
nakajee
reviewed
Feb 24, 2026
nakajee
reviewed
Feb 24, 2026
nakajee
reviewed
Feb 24, 2026
nakajee
reviewed
Feb 25, 2026
| else: | ||
| if state["ProblemType"]["Sparse"] and state["MIInputPerThread"] * state["ProblemType"]["MacDataTypeA"].numBytes() > 16: | ||
| autoLRVWA = 1 | ||
| if state["TransposeLDS"] and (not state["DirectToLds"]): |
Contributor
There was a problem hiding this comment.
Better to use "DirectToLdsA" if the check is for A only.
Contributor
There was a problem hiding this comment.
Another thing is why we need to check if it is DirectToLds or not here?
No need to change LocalReadVectorWidth depending on whether it is DirectToLds or not.
nakajee
reviewed
Feb 25, 2026
nakajee
reviewed
Feb 25, 2026
nakajee
reviewed
Feb 25, 2026
| ldsNumBytesA, ldsNumBytesAlignedA, ldsNumBytesB, ldsNumBytesAlignedB, ldsNumBytesMetadata, ldsNumBytesAlignedMetadata, \ | ||
| ldsNumBytesMXSA, ldsNumBytesAlignedMXSA, ldsNumBytesMXSB, ldsNumBytesAlignedMXSB \ | ||
| = calcLdsNumBytes(padA, ldsBlockSizePerPadA, padB, ldsBlockSizePerPadB) | ||
| if (ldsNumBytesAlignedA + ldsNumBytesAlignedB) > state["MaxLDS"]: |
Contributor
There was a problem hiding this comment.
We should consider ldsNumBytesAlignedMXSA and ldsNumBytesAlignedMXSB here
Contributor
There was a problem hiding this comment.
Probably, ldsNumBytesAlignedMetadata as well
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
5 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.
Motivation
Solving various issues blocking hipblaslt from built
Technical Details
Solving various issues blocking hipblaslt from built
Submission Checklist