Summary: | glsl/glcpp/tests/glcpp-test-cr-lf regression | ||
---|---|---|---|
Product: | Mesa | Reporter: | Vinson Lee <vlee> |
Component: | glsl-compiler | Assignee: | Kenneth Graunke <kenneth> |
Status: | RESOLVED FIXED | QA Contact: | Intel 3D Bugs Mailing List <intel-3d-bugs> |
Severity: | normal | ||
Priority: | medium | CC: | idr, kenneth |
Version: | git | Keywords: | bisected, regression |
Hardware: | x86-64 (AMD64) | ||
OS: | All | ||
Whiteboard: | |||
i915 platform: | i915 features: |
Description
Vinson Lee
2016-03-08 23:26:08 UTC
That is odd. Since Ken added some tests, I would think he would have seen such a failure. Vinson, what were you building this on? (In reply to Ian Romanick from comment #1) > That is odd. Since Ken added some tests, I would think he would have seen > such a failure. > > Vinson, what were you building this on? I can reproduce the make check failure on both Linux and Mac OS X. I didn't run "make check". I ran glcpp-test directly. I didn't know glcpp-test-cr-lf existed. It looks like the new 146-* tests pass normal testing, but fail glcpp-test-cr-lf, with different newline terminators. Presumably it outputs some different number of newlines. I can't tell because apparently running that script directly doesn't work anymore. I'll have to look into it later. My test actually exposed another preprocessor bug. Namely, the character location is wrong for hashes on a single line. # #error foo currently generates: 0:1(3): preprocessor error: #error foo which has the character position as 3. In the cr-lf mode, it generates 4. It ought to be 1, however... Patch on mailing list (passes "make check"; still running it through Piglit): https://lists.freedesktop.org/archives/mesa-dev/2016-March/109456.html commit e032e4ad5a3f73dffb1a8babcfd333954a574ffa Author: Kenneth Graunke <kenneth@whitecape.org> Date: Tue Mar 8 19:03:11 2016 -0800 glcpp: Fix locations when encounting "#<NEWLINE>". |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.