[Relax] add sample_indices in sampling #16675
Merged
Conversation
This file contains 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
MasterJH5574
approved these changes
Mar 5, 2024
MasterJH5574
added a commit
to MasterJH5574/tvm
that referenced
this pull request
Mar 12, 2024
This PR fixes a typo in the samping kernel of top-p/top-k sampling op. Prior to this PR, the kernel has out-of-bound global memory access due to a miss when introducing `sample_indices` in apache#16675. The correctness pass did not reveal this issue by directly running the test or running through pytest. But actually, if we use compute-sanitizer from NVIDIA, it will report the illegal memory access: ``` > compute-sanitizer --tool memcheck --print-limit=5 --launch-timeout 3600 python tests/python/relax/test_frontend_nn_op.py ========= COMPUTE-SANITIZER ========= Invalid __global__ read of size 8 bytes ========= at 0x4e90 in get_index_from_sorted_kernel ========= by thread (7,0,0) in block (0,0,0) ========= Address 0x7fe35ac00238 is out of bounds ========= and is 9 bytes after the nearest allocation at 0x7fe35ac00200 of size 48 bytes ========= Saved host backtrace up to driver entry point at kernel launch time ... ```
MasterJH5574
added a commit
to MasterJH5574/tvm
that referenced
this pull request
Mar 12, 2024
This PR fixes a typo in the samping kernel of top-p/top-k sampling op. Prior to this PR, the kernel has out-of-bound global memory access due to a miss when introducing `sample_indices` in apache#16675. The correctness pass did not reveal this issue by directly running the test or running through pytest. But actually, if we use compute-sanitizer from NVIDIA, it will report the illegal memory access: ``` > compute-sanitizer --tool memcheck --print-limit=5 --launch-timeout 3600 python tests/python/relax/test_frontend_nn_op.py ========= COMPUTE-SANITIZER ========= Invalid __global__ read of size 8 bytes ========= at 0x4e90 in get_index_from_sorted_kernel ========= by thread (7,0,0) in block (0,0,0) ========= Address 0x7fe35ac00238 is out of bounds ========= and is 9 bytes after the nearest allocation at 0x7fe35ac00200 of size 48 bytes ========= Saved host backtrace up to driver entry point at kernel launch time ... ```
tqchen
pushed a commit
that referenced
this pull request
Mar 12, 2024
This PR fixes a typo in the samping kernel of top-p/top-k sampling op. Prior to this PR, the kernel has out-of-bound global memory access due to a miss when introducing `sample_indices` in #16675. The correctness pass did not reveal this issue by directly running the test or running through pytest. But actually, if we use compute-sanitizer from NVIDIA, it will report the illegal memory access: ``` > compute-sanitizer --tool memcheck --print-limit=5 --launch-timeout 3600 python tests/python/relax/test_frontend_nn_op.py ========= COMPUTE-SANITIZER ========= Invalid __global__ read of size 8 bytes ========= at 0x4e90 in get_index_from_sorted_kernel ========= by thread (7,0,0) in block (0,0,0) ========= Address 0x7fe35ac00238 is out of bounds ========= and is 9 bytes after the nearest allocation at 0x7fe35ac00200 of size 48 bytes ========= Saved host backtrace up to driver entry point at kernel launch time ... ```
Lunderberg
pushed a commit
to Lunderberg/tvm
that referenced
this pull request
Mar 12, 2024
thaisacs
pushed a commit
to thaisacs/tvm
that referenced
this pull request
Apr 3, 2024
thaisacs
pushed a commit
to thaisacs/tvm
that referenced
this pull request
Apr 3, 2024
This PR fixes a typo in the samping kernel of top-p/top-k sampling op. Prior to this PR, the kernel has out-of-bound global memory access due to a miss when introducing `sample_indices` in apache#16675. The correctness pass did not reveal this issue by directly running the test or running through pytest. But actually, if we use compute-sanitizer from NVIDIA, it will report the illegal memory access: ``` > compute-sanitizer --tool memcheck --print-limit=5 --launch-timeout 3600 python tests/python/relax/test_frontend_nn_op.py ========= COMPUTE-SANITIZER ========= Invalid __global__ read of size 8 bytes ========= at 0x4e90 in get_index_from_sorted_kernel ========= by thread (7,0,0) in block (0,0,0) ========= Address 0x7fe35ac00238 is out of bounds ========= and is 9 bytes after the nearest allocation at 0x7fe35ac00200 of size 48 bytes ========= Saved host backtrace up to driver entry point at kernel launch time ... ```
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Add sample_indices into multinomial_from_uniform and sample_top_p_top_k_from_sorted_prob