Various things can go wrong during initialisation and teardown, but they
usually don't, so the error-handling paths go largely untested. This
collection of patches fixes some things I recently noticed. Some might
lead to a kernel OOPS, but mostly they're leaks and other inconsistencies.
Includes Nick Hoath's patch [6/6] to swap context/engine teardown, because
that helps make setup & teardown more consistent too; but more importantly,
that patch [6/6] is dependent on patch [5/6] of this set to work correctly.
Applying [6/6] without at least [5/6] as well will cause a fault on driver
unload!
v2:
Patches reordered and the previous [1/4] split into [4..6/6],
so that the bugs that it's fixing are more clearly visible
(Mika Kuoppala, although he said it wasn't actually *necessary*).
v3:
Patches reordered again; the bug fixes are now in patches 3 and 5.
The former could stand alone; the latter depends on patch 4 and is
a prerequisite for Nick's patch [6/6] to function.
Dave Gordon (5):
drm/i915: tidy up initialisation failure paths (legacy)
drm/i915: tidy up initialisation failure paths (GEM & LRC)
drm/i915: unmap the correct page in intel_logical_ring_cleanup()
drm/i915: consolidate LRC mode HWSP setup & teardown
drm/i915: HWSP should be unmapped earlier in LRC teardown sequence
Nick Hoath (1):
drm/i915: fix context/engine cleanup order
drivers/gpu/drm/i915/i915_dma.c | 4 +-
drivers/gpu/drm/i915/i915_drv.h | 2 +-
drivers/gpu/drm/i915/i915_gem.c | 28 ++++++++------
drivers/gpu/drm/i915/intel_lrc.c | 65 ++++++++++++++++++++-------------
drivers/gpu/drm/i915/intel_ringbuffer.c | 49 ++++++++++++-------------
5 files changed, 82 insertions(+), 66 deletions(-)
--
1.9.1
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
