[OpenMP] Remove optimization skipping reduction struct initialization #65697
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
jdoerfert
requested review from
alexey-bataev,
jdoerfert,
shiltian and
jhuber6
shiltian
changed the title
shiltian
reviewed
Sep 8, 2023
shiltian
reviewed
Sep 8, 2023
| @@ -2512,11 +2512,6 @@ void *__kmp_task_reduction_init(int gtid, int num, T *data) { | |||
| KMP_ASSERT(tg != NULL); | |||
| KMP_ASSERT(data != NULL); | |||
| KMP_ASSERT(num > 0); | |||
| if (nth == 1) { | |||
There was a problem hiding this comment.
Instead of completely removing the branch, it might be better to check if HHT is enabled. If no, then we can safely return here.
There was a problem hiding this comment.
I'm not sure how important single threaded reduction costs really are, but sure.
shiltian
approved these changes
Sep 8, 2023
hyviquel
reviewed
Sep 8, 2023
This commit ensures the initialization of the reduction struct in task loops, even when the number of threads in a team is 1, provided that hidden helper threads are active. Previously, an optimization skipped this initialization in cases where there was only a single thread in the team, which led to a bug. The bug behavior resulted in a segmentation fault when hidden helper threads were active. This occurred because the reduction struct was not initialized when the task group was created by a team with only one thread, causing the hidden helper thread to encounter issues when accessing the reduction data. In addition to this fix, a LIT test for issue llvm#57522 is included to validate the resolution and prevent any regressions. Fixes: llvm#57522
rodrigo-ceccato
force-pushed
the
target_in_reduction_fix
branch
from
35dd2e8
to
225afe4
Compare
|
Squashed and updated the commit message. |
ZijunZhaoCCK
pushed a commit
to ZijunZhaoCCK/llvm-project
that referenced
this pull request
Sep 19, 2023
…llvm#65697) This commit removes an optimization that skips the initialization of the reduction struct if the number of threads in a team is 1. This optimization caused a bug with Hidden Helper Threads. When the task group is initially initialized by the master thread but a Hidden Helper Thread executes a target nowait region, it requires the reduction struct initialization to properly accumulate the data. This commit also adds a LIT test for issue llvm#57522 to ensure that the issue is properly addressed and that the optimization removal does not introduce any regressions. Fixes: llvm#57522
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.
This commit removes an optimization that skips the initialization of the
reduction struct if the number of threads in a team is 1. This optimization
caused a bug with Hidden Helper Threads. When the task group is initially
initialized by the master thread but a Hidden Helper Thread executes a target
nowait region, it requires the reduction struct initialization to properly
accumulate the data.
This commit also adds a LIT test for issue #57522 to ensure that the issue is
properly addressed and that the optimization removal does not introduce any
regressions.
Fixes: #57522