drm/i915: Fix EIO/wedged handling in gem fault handler
authorDaniel Vetter <daniel.vetter@ffwll.ch>
Thu, 4 Sep 2014 07:36:18 +0000 (09:36 +0200)
committerJani Nikula <jani.nikula@intel.com>
Mon, 8 Sep 2014 05:38:50 +0000 (08:38 +0300)
commit2232f0315c6688f5ff6b2067ea88d97542034873
treea123e6a523803d2f40eac9f6858480a9db495eb0
parentad46cb533d586fdb256855437af876617c6cf609
drm/i915: Fix EIO/wedged handling in gem fault handler

In

commit 1f83fee08d625f8d0130f9fe5ef7b17c2e022f3c
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Thu Nov 15 17:17:22 2012 +0100

    drm/i915: clear up wedged transitions

I've accidentally inverted the EIO/wedged handling in the fault
handler: We want to return the EIO as a SIGBUS only if it's not
because of the gpu having died, to prevent userspace from unduly
dying.

In my defence the comment right above is completely misleading, so fix
both.

v2: Drop the WARN_ON, it's not actually a bug to e.g. receive an -EIO
when swap-in fails.

v3: Don't remove too much ... oops.

Reported-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@vger.kernel.org
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
drivers/gpu/drm/i915/i915_gem.c