Fix testing with Chrome 65. #792
Merged
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
Chrome 65 was released 2018-3-6. With it, some vertices appear odd in GL rendered lines, but only on the headless GL or xvfb tests. This doesn't affect how the tests are rendered in Chrome 65 on a desktop, so it is likely some sort of Chrome-65-with-mesa issue. This is not a mesa issue, since it was not updated. Since this is just a testing artifact, rather than change the baselines, relax the strictness of the image comparison. Note that if travis allowed pinning to a specific version of Chrome, I'd recommend pinning to version 64 until Chrome fixes their issue.
aashish24
approved these changes
Mar 14, 2018
manthey
added a commit
that referenced
this pull request
May 9, 2018
Chrome 65 and 66 when running in xvfb with openmesa do not compute atan(0, <negative>) correctly. Also, there are occasional int() casts that round down even when they are small integers. These changes make image comparisons more robust by working around the atan issue (this may be the issue detailed in chromium.googlesource.com/angle/angle/commit/da9fb09). I have submitted a new issue to chromium: https://bugs.chromium.org/p/chromium/issues/detail?id=841296 to track this problem. With this commit, the baselines were regenerated with Chrome 64, and the test strictness requirements changed in PR #792 have been reverted. There are subtly differences between Chrome 64 and 66 (font aliasing is slightly different, and I think the other changes are because the shader float precision is subtly different, though I'm not sure of that).
manthey
added a commit
that referenced
this pull request
May 11, 2018
Chrome 65 and 66 when running in xvfb with openmesa do not compute atan(0, <negative>) correctly. Also, there are occasional int() casts that round down even when they are small integers. These changes make image comparisons more robust by working around the atan issue (this may be the issue detailed in chromium.googlesource.com/angle/angle/commit/da9fb09). I have submitted a new issue to chromium: https://bugs.chromium.org/p/chromium/issues/detail?id=841296 to track this problem. With this commit, the baselines were regenerated with Chrome 64, and the test strictness requirements changed in PR #792 have been reverted. There are subtly differences between Chrome 64 and 66 (font aliasing is slightly different, and I think the other changes are because the shader float precision is subtly different, though I'm not sure of that).
manthey
added a commit
that referenced
this pull request
May 11, 2018
Chrome 65 and 66 when running in xvfb with openmesa do not compute atan(0, <negative>) correctly. Also, there are occasional int() casts that round down even when they are small integers. These changes make image comparisons more robust by working around the atan issue (this may be the issue detailed in chromium.googlesource.com/angle/angle/commit/da9fb09). I have submitted a new issue to chromium: https://bugs.chromium.org/p/chromium/issues/detail?id=841296 to track this problem. With this commit, the baselines were regenerated with Chrome 64, and the test strictness requirements changed in PR #792 have been reverted. There are subtly differences between Chrome 64 and 66 (font aliasing is slightly different, and I think the other changes are because the shader float precision is subtly different, though I'm not sure of that).
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.
Chrome 65 was released 2018-3-6. With it, some vertices appear odd in GL rendered lines, but only on the headless GL or xvfb tests. This doesn't affect how the tests are rendered in Chrome 65 on a desktop, so it is likely some sort of Chrome-65-with-mesa issue. This is not a mesa issue, since it was not updated.
Since this is just a testing artifact, rather than change the baselines, relax the strictness of the image comparison. Note that if travis allowed pinning to a specific version of Chrome, I'd recommend pinning to version 64 until Chrome fixes their issue.