Fix some possible thread-id overflow calculations #17473
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
davidwendt
added
bug
|
Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
|
/ok to test |
davidwendt
commented
Dec 2, 2024
| @@ -413,7 +413,7 @@ CUDF_KERNEL void compute_segment_sizes(device_span<column_device_view const> col | |||
| size_type max_branch_depth) | |||
| { | |||
| extern __shared__ row_span thread_branch_stacks[]; | |||
| int const tid = threadIdx.x + blockIdx.x * blockDim.x; | |||
| auto const tid = static_cast<size_type>(cudf::detail::grid_1d::global_thread_id()); | |||
There was a problem hiding this comment.
Merely for clarity. Prefer the type be size_type rather than int.
davidwendt
commented
Dec 2, 2024
| thread_index_type const stride = blockDim.x * gridDim.x; | ||
| auto const block_size = static_cast<thread_index_type>(blockDim.x); | ||
| thread_index_type const start = threadIdx.x + blockIdx.x * block_size; | ||
| thread_index_type const stride = block_size * gridDim.x; |
There was a problem hiding this comment.
This is explicit up-casting for multiplication.
davidwendt
commented
Dec 2, 2024
| @@ -138,7 +138,7 @@ CUDF_KERNEL void compute_row_partition_numbers(row_hasher_t the_hasher, | |||
| auto const stride = cudf::detail::grid_1d::grid_stride(); | |||
|
|
|||
| // Initialize local histogram | |||
| size_type partition_number = threadIdx.x; | |||
| thread_index_type partition_number = threadIdx.x; | |||
| while (partition_number < num_partitions) { | |||
| shared_partition_sizes[partition_number] = 0; | |||
| partition_number += blockDim.x; | |||
There was a problem hiding this comment.
If num_partitions is close to max<size_type> then partition_number += blockDim.x could overlfow size_type.
|
/ok to test |
|
/ok to test |
|
/ok to test |
|
/ok to test |
|
/ok to test |
davidwendt
added
3 - Ready for Review
vuule
approved these changes
Dec 9, 2024
|
/merge |
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.
Description
Fixes some possible thread-id calculations or usages that may possibly overflow
int32type orsize_type.Reference #10368
Checklist