Created attachment 87772 [details]
dmesg

System Environment:
--------------------------
Kernel: (drm-intel-next-queued)ec98c8d1ffdc87c54ec0896db45db5b7bd22eca9
Some additional commit info:
Author: Ville Syrjälä <ville.syrjala@linux.intel.com>
Date:   Fri Oct 11 15:26:26 2013 +0300

    drm/i915: Check 5/6 DDB split only when sprites are enabled

Bug detailed description:
-----------------------------
It fails on PNV/ILK/SNB/IVB/HSW with -fixes,-queued and -nightly kernel.
It's new case.
Following cases also fail:
igt/kms_addfb/bo-too-small-due-to-tiling
igt/kms_addfb/framebuffer-vs-set-tiling
igt/kms_addfb/too-high

output:
Test assertion failure function size_tests, file kms_addfb.c:190:
Failed assertion: drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == -1 && errno == EINVAL
Subtest bo-too-small: FAIL
Test assertion failure function size_tests, file kms_addfb.c:206:
Failed assertion: drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == -1 && errno == EINVAL
Subtest bo-too-small-due-to-tiling: FAIL
Test assertion failure function pitch_tests, file kms_addfb.c:93:
Failed assertion: __gem_set_tiling(fd, gem_bo, I915_TILING_X, 512*4) == -EBUSY
Subtest framebuffer-vs-set-tiling: FAIL
Test assertion failure function size_tests, file kms_addfb.c:180:
Failed assertion: drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == -1 && errno == EINVAL
Subtest too-high: FAIL

Reproduce steps:
----------------------------
1../kms_addfb --run-subtest bo-too-small