Skip to content

[SYCL] Fix bug with not found alloca command. #148

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged

Conversation

romanovvlad
Copy link
Contributor

Fix bug with not found alloca in case of two accessors are created for
the same memory object. The problem is that during searching for
dependencies for the first accessor list of leaf command is modified so
it doesn't contain commands that are considered as dependencies. As a
result search for dependencies for the second accessor fails.
The fix is to update list of leafs after dependencies for all accessors
are found.

Signed-off-by: Vlad Romanov [email protected]

@keryell
Copy link
Contributor

keryell commented May 20, 2019

Good point!

vladimirlaz
vladimirlaz previously approved these changes May 21, 2019
Fix bug with not found alloca in case of two accessors are created for
the same memory object. The problem is that during searching for
dependencies for the first accessor list of leaf command is modified so
it doesn't contain commands that are considered as dependencies. As a
result search for dependencies for the second accessor fails.
The fix is to update list of leafs after dependencies for all accessors
are found.

Signed-off-by: Vlad Romanov <[email protected]>
@romanovvlad romanovvlad force-pushed the private/vromanov/FixNotFoundAlloca branch from 236f8c3 to 25c9f09 Compare May 21, 2019 13:40
@romanovvlad romanovvlad merged commit 2fd74a7 into intel:sycl May 21, 2019
vmaksimo pushed a commit to vmaksimo/llvm that referenced this pull request Mar 3, 2021
  CONFLICT (content): Merge conflict in clang/lib/Driver/Driver.cpp
aelovikov-intel pushed a commit to aelovikov-intel/llvm that referenced this pull request Mar 27, 2023
* [ESIMD] Fix stencil tests timeouts

I analyzed the history of ESIMD stencil tests in CI and found that
stencil tests are sensitive to the time limit. I.e. they are not
hanging but timing out before finishing the job.
Since intel#148 Stencil.cpp was failing several times per day.

This patch reduces back the working set for the two tests allowing
them to finish before in required time limit. Also, added a test
parameter to run with a bigger dimension size.

* Address review comments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants