fix: Do not mutate shared _gemm_output_3d in CpuGemmConv2d::run()#1275
Open
fix: Do not mutate shared _gemm_output_3d in CpuGemmConv2d::run()#1275
Conversation
CpuGemmConv2d::run() was mutating the shared member _gemm_output_3d by extending its padding before soft_init()/import_memory(). When the same operator instance is reused across runs, this can cause later extend_padding() calls to fail. It is also unsafe when the operator is used from multiple threads. Use a local TensorInfo copy in run() for padding extension and soft_init()/import_memory(), leaving _gemm_output_3d unchanged. Added a new test: RepeatedRunDoesNotReuseImportedGemm3dTensorInfo. Change-Id: I3e4e2d25cabf85724ecf126b1c93df6733ee7d48 Signed-off-by: Pablo Marquez Tello <pablo.tello@arm.com>
morgolock
force-pushed
the
pr/fix_shared_tinfo_cpugemm
branch
from
f823e19 to
a1d5f71
Compare
gunes-arm
reviewed
Apr 2, 2026
| * - The second run does not throw | ||
| * - Both runs compute the same output | ||
| */ | ||
| TEST_CASE(RepeatedRunDoesNotReuseImportedGemm3dTensorInfo, framework::DatasetMode::ALL) |
Contributor
There was a problem hiding this comment.
Is this really failing w/o this fix? I've been testing it but couldn't make it fail. I think it's happening because TensorAllocator's destructor is marking the TensorInfo as resizable again and when run() finishes the allocator that's been soft initialized is destructed and it marks the information object resizable again. This makes this problem only visible in situations where the usage of the same object is multi-threaded.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
CpuGemmConv2d::run() was mutating the shared member _gemm_output_3d by extending its padding before soft_init()/import_memory().
When the same operator instance is reused across runs, this can cause later extend_padding() calls to fail. It is also unsafe when the operator is used from multiple threads.
Use a local TensorInfo copy in run() for padding extension and soft_init()/import_memory(), leaving _gemm_output_3d unchanged.
Added a new test: RepeatedRunDoesNotReuseImportedGemm3dTensorInfo.
Change-Id: I3e4e2d25cabf85724ecf126b1c93df6733ee7d48