-
Notifications
You must be signed in to change notification settings - Fork 769
[sycl-post-link] Handle spec constants after device code split #3748
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
bader
merged 3 commits into
intel:sycl
from
AlexeySachkov:private/asachkov/do-spec-const-after-split
May 15, 2021
Merged
[sycl-post-link] Handle spec constants after device code split #3748
bader
merged 3 commits into
intel:sycl
from
AlexeySachkov:private/asachkov/do-spec-const-after-split
May 15, 2021
Conversation
This file contains hidden or 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
Due to changes in `SpecConstatsPass` introduced by intel#3513, device image properties for specialization constants are now computed in a bit different way: in particular they are calculated based on metadata generated by the pass and not by looking at the SPIR-V friendly IR intrinsics left in the module. Such change caused a regression: if device code split was used together with specialization constants, each resulting device image was supplied with list of *all* specialization constants found in the _original_ undivided module. In combination with some DPC++ RT changes it lead to setting non-existing specialization constants. This patch fixed the regression by performing device code split step *before* handling specialization constants.
mlychkov
reviewed
May 13, 2021
llvm/test/tools/sycl-post-link/spec-constants/spec_const_and_split.ll
Outdated
Show resolved
Hide resolved
There is also a typo in commit description: SpecConstatsPass -> SpecConstantsPass . |
Thanks, updated description |
mlychkov
reviewed
May 13, 2021
AlexeySachkov
commented
May 14, 2021
Failures in |
mlychkov
approved these changes
May 14, 2021
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.
Due to changes in
SpecConstantsPass
introduced by #3513,device image properties for specialization constants are now computed in
a bit different way: in particular they are calculated based on metadata
generated by the pass and not by looking at the SPIR-V friendly IR
intrinsics left in the module.
Such change caused a regression: if device code split was used together
with specialization constants, each resulting device image was supplied
with list of all specialization constants found in the original
undivided module. In combination with some DPC++ RT changes it lead to
setting non-existing specialization constants.
This patch fixed the regression by performing device code split step
before handling specialization constants.