Cherry-pick hipblaslt changes for rocm 7.0#675
Merged
vamovsik merged 6 commits intoJul 22, 2025
Conversation
This includes 2 changes: - Unrestricted the temp sgprs needed for gsu from being contiguous, avoiding overflow for certain kernels - Account for additional temp sgprs that will be required for code gen, up to physical limits --- 🔁 Imported from [ROCm/hipBLASLt#2184](ROCm/hipBLASLt#2184) 🧑💻 Originally authored by @mahmoodw --------- Co-authored-by: assistant-librarian[bot] <210906412+assistant-librarian[bot]@users.noreply.github.com> Co-authored-by: mahmoodw <wmahmood@amd.com>
There were more SGPR allocations than what is needed for . This PR addresses that while keeping backward compatibility. hipblaslt-test on gfx950: [----------] Global test environment tear-down [==========] 19937 tests from 12 test suites ran. (928859 ms total) [ PASSED ] 19937 tests. hipBLASLt version: 100100 hipBLASLt git version: 36ab695-dirty command line: ./hipblaslt-test hipblaslt-test on gfx942: [----------] Global test environment tear-down [==========] 54591 tests from 11 test suites ran. (3033316 ms total) [ PASSED ] 54591 tests. hipBLASLt version: 100100 hipBLASLt git version: 36ab695 command line: ./hipblaslt-test --- 🔁 Imported from [ROCm/hipBLASLt#2250](ROCm/hipBLASLt#2250) 🧑💻 Originally authored by @aliry95amd Co-authored-by: aliry95amd <ayazdani@amd.com> Co-authored-by: assistant-librarian[bot] <assistant-librarian[bot]@users.noreply.github.com>
This PR fixes a bug that occurred when negative WGM values were used for StreamK kernels.
- removes all major GEMMs from the GridBased folder - adds major GEMMs for Origami libraries in gfx950 - adds fallback kernels for all GEMMs - adds the latest custom kernels for BBS_TN/HHS_TN - some modifications to file names, kernel names, bias datatypes --------- Co-authored-by: aliry95amd <ayazdani@amd.com> Co-authored-by: b-shi <brianshi@amd.com>
Add custom HHS/BBS TN kernels to equality library Cherry-picked from develop. Original PRs: ROCm/hipBLASLt#2193 ROCm/hipBLASLt#2211
AlexBrownAMD
requested review from
aliry95amd,
b-shi,
bstefanuk,
davidd-amd,
eidenyoshida,
msujon-AMD and
yoichiyoshida
This was referenced Jul 15, 2025
b-shi
approved these changes
Jul 15, 2025
aliry95amd
approved these changes
Jul 15, 2025
davidd-amd
approved these changes
Jul 17, 2025
bethune-bryant
approved these changes
Jul 17, 2025
rahulc-gh
added a commit
to ROCm/hipBLASLt
that referenced
this pull request
Jul 22, 2025
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
8 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.
hipblaslt changes required for rocm 7.0: