AMDGPU: apply hotswap gfx1250 TDM mask workarounds#3297
Draft
harsh-amd wants to merge 1 commit into
Draft
Conversation
harsh-amd
force-pushed
the
plat-204339-hotswap-mask-workaround
branch
4 times, most recently
from
f8191d6 to
169e3d6
Compare
harsh-amd
added
comgr
harsh-amd
force-pushed
the
plat-204339-hotswap-mask-workaround
branch
from
169e3d6 to
86a7bfa
Compare
harsh-amd
force-pushed
the
plat-204339-hotswap-mask-workaround
branch
from
86a7bfa to
fdafe44
Compare
|
Can we split this into 2 PRs - first one that adds the strict rewrite and the D# workaround sits on top of that ? |
In strict-mode hotswap, apply the PLAT-204339 mask workarounds for both gfx1250 A0 and B0 targets. A0 clears tensor_load_to_lds D# Group 1 wg_mask bits and wraps remaining cluster_load* instructions with M0 save/clear/restore. B0 clears tensor_load_to_lds only for non-cluster waves, using fixed .cluster_dims metadata when available and otherwise checking IB_STS2.CLUSTER_ID dynamically. B0 cluster loads are left unchanged. Required mask patches now fail the rewrite if scratch allocation or emission fails, rather than returning unsafe code. Validation on mi300x: - cmake --build build --target amd_comgr hotswap-rewrite -j$(nproc) - build/bin/llvm-lit -sv build/tools/comgr/test-lit --filter 'hotswap-trampoline-tensor|hotswap-trampoline-cluster|hotswap-inplace' - cmake --build build-asan --target amd_comgr hotswap-rewrite -j$(nproc) - build-asan/bin/llvm-lit -sv build-asan/tools/comgr/test-lit --filter 'hotswap-trampoline-tensor|hotswap-trampoline-cluster|hotswap-inplace'
harsh-amd
force-pushed
the
plat-204339-hotswap-mask-workaround
branch
from
fdafe44 to
f792b49
Compare
harsh-amd
changed the title
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
2 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.
Summary:
Validation on mi300x: