ci: Lock down to OIIO 2.5 for icc tests #1799
Merged
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
Background: Intel's icc is essentially discontinued and we think all the renderers using OSL have moved on to the newer icx, but Alex Wells still likes to sometime use icc for its better facilities for some SIMD code generation sleuthing that he does. So we keep it in our CI matrix to esure we don't break it for OSL until it is truly unavoidable or he doesn't feel like he needs it anymore. Some recent changes to OIIO's simd.h, from PR 4187 in that project, seem to tweak what appears to be a compiler bug in icc -- bugs were also noted in OIIO in the course of this PR and some workarounds were put there just for icc to make OIIO's own tests pass. Well, once that went in, OSL's CI tests for the icc case also started failing. It's only texture derivatives that are broken, and in a weird wy that smells like a compiler icc bug with SIMD generation similar to what we discovered on the OIIO side. Like I said, this is a discontinued compiler that no major renderers are still using, so it doesn't seem worth tracking it down any further to find a true fix. So let's just lock that one test case down to OIIO 2.5 (the current release) to avoid the changes that will eventually be in OIIO 2.6/3.0 and hope that we fully retire icc before OIIO 2.5 is no longer supported by OSL. Signed-off-by: Larry Gritz <lg@larrygritz.com>
|
No objections, CI only change -> merging |
lgritz
added a commit
to lgritz/OpenShadingLanguage
that referenced
this pull request
Apr 29, 2024
Background: Intel's icc is essentially discontinued and we think all the renderers using OSL have moved on to the newer icx, but Alex Wells still likes to sometime use icc for its better facilities for some SIMD code generation sleuthing that he does. So we keep it in our CI matrix to esure we don't break it for OSL until it is truly unavoidable or he doesn't feel like he needs it anymore. Some recent changes to OIIO's simd.h, from PR 4187 in that project, seem to tweak what appears to be a compiler bug in icc -- bugs were also noted in OIIO in the course of this PR and some workarounds were put there just for icc to make OIIO's own tests pass. Well, once that went in, OSL's CI tests for the icc case also started failing. It's only texture derivatives that are broken, and in a weird wy that smells like a compiler icc bug with SIMD generation similar to what we discovered on the OIIO side. Like I said, this is a discontinued compiler that no major renderers are still using, so it doesn't seem worth tracking it down any further to find a true fix. So let's just lock that one test case down to OIIO 2.5 (the current release) to avoid the changes that will eventually be in OIIO 2.6/3.0 and hope that we fully retire icc before OIIO 2.5 is no longer supported by OSL. Signed-off-by: Larry Gritz <lg@larrygritz.com>
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.
Background: Intel's icc is essentially discontinued and we think all the renderers using OSL have moved on to the newer icx, but Alex Wells still likes to sometime use icc for its better facilities for some SIMD code generation sleuthing that he does. So we keep it in our CI matrix to esure we don't break it for OSL until it is truly unavoidable or he doesn't feel like he needs it anymore.
Some recent changes to OIIO's simd.h, from PR 4187 in that project, seem to tweak what appears to be a compiler bug in icc -- bugs were also noted in OIIO in the course of this PR and some workarounds were put there just for icc to make OIIO's own tests pass. Well, once that went in, OSL's CI tests for the icc case also started failing. It's only texture derivatives that are broken, and in a weird wy that smells like a compiler icc bug with SIMD generation similar to what we discovered on the OIIO side.
Like I said, this is a discontinued compiler that no major renderers are still using, so it doesn't seem worth tracking it down any further to find a true fix. So let's just lock that one test case down to OIIO 2.5 (the current release) to avoid the changes that will eventually be in OIIO 2.6/3.0 and hope that we fully retire icc before OIIO 2.5 is no longer supported by OSL.