On 10.06.2015 01:53, Ian Romanick wrote: > On 05/29/2015 02:44 AM, Michel Dänzer wrote: >> On 29.05.2015 09:00, [email protected] (Ian Romanick) wrote: >>> Module: Mesa >>> Branch: master >>> Commit: 9b5e92f4ccc6ee1cb9caea947f6efaad2b391cf1 >>> URL: >>> http://cgit.freedesktop.org/mesa/mesa/commit/?id=9b5e92f4ccc6ee1cb9caea947f6efaad2b391cf1 >>> >>> Author: Ian Romanick <[email protected]> >>> Date: Wed Apr 29 16:12:40 2015 -0700 >>> >>> mesa: Allow overriding the version of ES2+ contexts >>> >>> Signed-off-by: Ian Romanick <[email protected]> >>> Reviewed-by: Tapani Pälli <[email protected]> >> >> This commit broke a bunch of GLES related piglit tests for me with >> radeonsi, e.g. egl-create-context-default-major-version-gles: >> >> Unexpected GLES version: 93847553.6 Mesa 10.7.0-devel (git-9b5e92f) >> Expected GLES 1.0 or 1.1. >> PIGLIT: {"result": "fail" } >> >> The numbers before "Mesa" in the version string seem random, and >> sometimes the test happens to pass. > > Does this still occur with Tapani's patch? > > http://lists.freedesktop.org/archives/mesa-dev/2015-June/085808.html
Not sure, but Marek said that Martin's patch http://lists.freedesktop.org/archives/mesa-dev/2015-June/085821.html fixes GLES related issues for him. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer _______________________________________________ mesa-dev mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/mesa-dev
