Bug 90190

Created attachment 115363 [details]
dmesg

==System Environment==
--------------------------
Regression: YES

Non-working platforms: IVB

==kernel==
--------------------------
drm-intel-nightly/92bb36c80e561f82b1f4b63cc269a71833137841
commit 92bb36c80e561f82b1f4b63cc269a71833137841
Author: Jani Nikula <jani.nikula@intel.com>
Date:   Fri Apr 24 00:26:53 2015 +0300

    drm-intel-nightly: 2015y-04m-23d-21h-26m-24s UTC integration manifest

==Bug detailed description==
-----------------------------
Fail to start X on IVB with drm-intel-nightly and drm-intel-next-queued kernel.It doesn't happen on drm-intel-fixes kernel.

Bisect shows:0875546c5318c85c13d07014af5350e9000bc9e9 is the first bad commit.
commit 0875546c5318c85c13d07014af5350e9000bc9e9
Author:     Daniel Vetter <daniel.vetter@ffwll.ch>
AuthorDate: Mon Apr 20 09:04:05 2015 -0700
Commit:     Daniel Vetter <daniel.vetter@ffwll.ch>
CommitDate: Thu Apr 23 21:06:39 2015 +0200

    drm/i915: Fix up the vma aliasing ppgtt binding

    Currently we have the problem that the decision whether ptes need to
    be (re)written is splattered all over the codebase. Move all that into
    i915_vma_bind. This needs a few changes:
    - Just reuse the PIN_* flags for i915_vma_bind and do the conversion
      to vma->bound in there to avoid duplicating the conversion code all
      over.
    - We need to make binding for EXECBUF (i.e. pick aliasing ppgtt if
      around) explicit, add PIN_USER for that.
    - Two callers want to update ptes, give them a PIN_UPDATE for that.

    Of course we still want to avoid double-binding, but that should be
    taken care of:
    - A ppgtt vma will only ever see PIN_USER, so no issue with
      double-binding.
    - A ggtt vma with aliasing ppgtt needs both types of binding, and we
      track that properly now.
    - A ggtt vma without aliasing ppgtt could be bound twice. In the
      lower-level ->bind_vma functions hence unconditionally set
      GLOBAL_BIND when writing the ggtt ptes.

    There's still a bit room for cleanup, but that's for follow-up
    patches.

    v2: Fixup fumbles.

    v3: s/PIN_EXECBUF/PIN_USER/ for clearer meaning, suggested by Chris.

    Cc: Chris Wilson <chris@chris-wilson.co.uk>
    Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
    Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

output:
[root@x-ivb9 ~]# xinit&
[1] 3724
[root@x-ivb9 ~]#

X.Org X Server 1.17.99
Release Date: 2015-02-02
X Protocol Version 11, Revision 0
Build Operating System: Linux 3.19.0-rc7+ x86_64
Current Operating System: Linux x-ivb9 4.0.0_drm-intel-nightly_92bb36_20150424+ #38 SMP Fri Apr 24 08:59:42 EDT 2015 x86_64
Kernel command line: BOOT_IMAGE=kernels//nightly_parents/2015_04_27/drm-intel-nightly/92bb36c80e561f82b1f4b63cc269a71833137841/bzImage_x86_64 root=/dev/sda2 drm.debug=0xe hostname=x-ivb9 modules_path=kernels//nightly_parents/2015_04_27/drm-intel-nightly/92bb36c80e561f82b1f4b63cc269a71833137841/modules_x86_64/lib/modules/4.0.0_drm-intel-nightly_92bb36_20150424+ kexec_jump_back_entry=0xcccccccc
Build Date: 26 April 2015  05:20:34PM

Current version of pixman: 0.33.1
        Before reporting problems, check http://wiki.x.org
        to make sure that you have the latest version.
Markers: (--) probed, (**) from config file, (==) default setting,
        (++) from command line, (!!) notice, (II) informational,
        (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(==) Log file: "/opt/X11R7/var/log/Xorg.0.log", Time: Mon Apr 27 02:47:22 2015
(==) Using config directory: "/etc/X11/xorg.conf.d"
(==) Using system config directory "/opt/X11R7/share/X11/xorg.conf.d"
xterm: cannot load font '-misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso10646-1'

[root@x-ivb9 ~]# glxinfo
name of display: :0.0
libGL: OpenDriver: trying /opt/X11R7/lib/dri/tls/i965_dri.so
libGL: OpenDriver: trying /opt/X11R7/lib/dri/i965_dri.so

dmesg:
[  159.708718] [drm] stuck on render ring
[  159.709116] [drm] GPU HANG: ecode 7:0:0xabcff7fb, in glxinfo [3813], reason: Ring hung, action: reset
[  159.709118] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[  159.709119] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[  159.709119] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[  159.709120] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[  159.709121] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[  159.709140] [drm:i915_reset_and_wakeup] resetting chip
[  159.710708] drm/i915: Resetting chip after gpu hang


==Reproduce steps==
---------------------------- 
1. clean boot system
2. xinit
3. glxinfo