-
Notifications
You must be signed in to change notification settings - Fork 744
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
[SYCL] Test for fix of linked alloca's deps #1470
Merged
bader
merged 16 commits into
intel:sycl
from
s-kanaev:private/s-kanaev/fix-linked-allocacmd-deps
Aug 3, 2020
Merged
[SYCL] Test for fix of linked alloca's deps #1470
bader
merged 16 commits into
intel:sycl
from
s-kanaev:private/s-kanaev/fix-linked-allocacmd-deps
Aug 3, 2020
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
Signed-off-by: Sergey Kanaev <[email protected]>
romanovvlad
requested changes
Apr 3, 2020
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Please, add a test.
Signed-off-by: Sergey Kanaev <[email protected]>
Done |
Signed-off-by: Sergey Kanaev <[email protected]>
Signed-off-by: Sergey Kanaev <[email protected]>
romanovvlad
requested changes
Apr 9, 2020
Signed-off-by: Sergey Kanaev <[email protected]>
Signed-off-by: Sergey Kanaev <[email protected]>
Signed-off-by: Sergey Kanaev <[email protected]>
s-kanaev
changed the title
[SYCL] Fix linked AllocaCmd dependencies
[SYCL] Test for fix of linked alloca's deps
Jun 17, 2020
Signed-off-by: Sergey Kanaev <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, but I'd like @romanovvlad to take a look
Signed-off-by: Sergey Kanaev <[email protected]>
Signed-off-by: Sergey Kanaev <[email protected]>
romanovvlad
reviewed
Jun 17, 2020
Signed-off-by: Sergey Kanaev <[email protected]>
/summary:run |
…linked-allocacmd-deps
Signed-off-by: Sergey Kanaev <[email protected]>
/summary:run |
@romanovvlad, please, review |
romanovvlad
approved these changes
Aug 3, 2020
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.
The tests provided here are for this case: https://github.com/intel/llvm/blob/sycl/sycl/source/detail/scheduler/graph_builder.cpp#L611.
The fix was introduced with #1471.
Use case fixed by the fix. Imagine us having two device queues (
Q1
andQ2
) and a single bufferB
. We initialize the buffer with host accessor. Also lets have two kernels: kernelK1
executed atQ1
; and kernelK2
which executes atQ2
.K1
writes toB
,K2
reads fromB
. After submittingK1
toQ1
there will also be anAllocaCommand
A1
which allocates buffer on device.K1
depends onA1
via memory objectB
. After submittingK2
toQ2
there will be anotherAllocaCommand
A2
(for queueQ2
and its device).K2
will depend onA2
.A2
, however, should depend on bothA1
andK1
.A2->K1
dependency eliminates data race.