On Mon, Oct 15, 2012 at 7:34 PM, Luís Picciochi Oliveira > Daniel's patch also seems more reasonable to me. > > I ran the intel-gpu-tools tests on this card. Here's some input on the > multiple options we have: > At first impression, the last patch proposed by Daniel works well and > also fixes the reported problems. It also passes more tests on the > mentioned testbench than the "blind backport" I suggested previously. > > Here's a summary of the obtained results: > > With the original debian 3.2 kernel: > $ grep FAIL 3.2.0-3-686-pae-all_tests | wc -l > 13 > > With the "blind" patch I suggested previously: > $ grep FAIL 3.2.0-3-686-pae-patched-all_tests | wc -l > 20 > These tests don't even finish. A kernel BUG is spit to the console and > the system has to be rebooted. > > With Daniel Vetter's patch: > $ grep FAIL 3.2.23-vetterpatch-all_tests | wc -l > 10 > > Although we have less tests failing, some new FAILs appeared with this patch. > Here's the diff between the obtained "FAIL"s: > $ diff fails-debian-unpatched fails-debian-with-vetters-patch > 0a1 >> FAIL: gem_cs_tlb > 4,6c5 > < FAIL: gem_set_tiling_vs_blt > < FAIL: gem_tiled_pread > < FAIL: gem_tiled_pread_pwrite > --- >> FAIL: gem_set_tiling_vs_pwrite > 8d6 > < FAIL: gem_tiled_swapping > 10d7 > < FAIL: gem_tiled_fence_blits > > I'm not sure if those 2 new FAILs are grave enough to hold back this fix.
gem_cs_tlb is harmless, the testcase itself isn't that robust and fails an internal self-check once a while when other things have happened - it should work if you run it alone. The gem_set_tiling_vs_pwrite is more unsettling - does that reliably fail with the little backport and reliably work on a kernel without that backport? Might be we miss another workaround. I'll double-check these tests on my own i915g meanwhile. Yours, Daniel -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch -- To unsubscribe from this list: send the line "unsubscribe stable" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
