On my IVB system, $ bin/glean -o -v -v -v -t +fragProg1 --quick ---------------------------------------------------------------------- Fragment Program test 1: test a specific set of fragment programs. FAILURE: Program: CMP test Expected color: 0, 0.25, 1, 0 Observed color: 0.501961, 0.25098, 1, 0.501961 fragProg1: FAIL rgba8, db, z24, s8, win+pmap, id 33 52 tests passed, 1 tests failed. git-bisect says: 2999f83bd987fd53e2348ceb887dc2d40096c813 is the first bad commit commit 2999f83bd987fd53e2348ceb887dc2d40096c813 Author: Jason Ekstrand <jason.ekstrand@intel.com> Date: Sat Aug 16 11:38:07 2014 -0700 i965/fs: Derive force_uncompressed from instruction exec_size Signed-off-by: Jason Ekstrand <jason.ekstrand@intel.com> Reviewed-by: Matt Turner <mattst88@gmail.com> :040000 040000 9c30b7bc6ea456e2943cf98e823f77675b0d0585 d1bd4f70e1fc027455e0ef8ee2317786206c5ee9 M src
How about latest master? There were some bugs in the first push that I fixed a couple hours later.
I was testing on 408aa46c. There are a couple commits since then on master, but both are in the freedreno driver. Re-testing at cc355f1c had no change.
I think I found the bug and sent a patch to the list that should fix it. Let me know if it does.
I can reproduce it on (master)16b53005a7df4249fecb6641af0934c32181fdea FAILURE: Program: CMP test Expected color: 0, 0.25, 1, 0 Observed color: 0, 0, 0, 0 fragProg1: FAIL rgba8, db, z24, s8, win+pmap, id 32 0 tests passed, 1 tests failed.
Perhaps fixed now?
Works here.
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.