]> git.hungrycats.org Git - linux/commit
drm/i915: Insert a command barrier on BLT/BSD cache flushes
authorChris Wilson <chris@chris-wilson.co.uk>
Thu, 22 Jan 2015 13:42:00 +0000 (13:42 +0000)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 18 Mar 2015 13:10:57 +0000 (14:10 +0100)
commitb16a8497a0342524a343d07a234508cc12b68443
treeb109d80ba44d1e40a8da66d7ff779fb168bcdd32
parent12bc2f3df1d5c9445202a1eb9119460ca6928042
drm/i915: Insert a command barrier on BLT/BSD cache flushes

commit f0a1fb10e5f79f5aaf8d7e94b9fa6bf2fa9aeebf upstream.

This looked like an odd regression from

commit ec5cc0f9b019af95e4571a9fa162d94294c8d90b
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Jun 12 10:28:55 2014 +0100

    drm/i915: Restrict GPU boost to the RCS engine

but in reality it undercovered a much older coherency bug. The issue that
boosting the GPU frequency on the BCS ring was masking was that we could
wake the CPU up after completion of a BCS batch and inspect memory prior
to the write cache being fully evicted. In order to serialise the
breadcrumb interrupt (and so ensure that the CPU's view of memory is
coherent) we need to perform a post-sync operation in the MI_FLUSH_DW.

v2: Fix all the MI_FLUSH_DW (bsd plus the duplication in execlists).

Also fix the invalidate_domains mask in gen8_emit_flush() for ring !=
VCS.

Testcase: gpuX-rcs-gpu-read-after-write
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Daniel Vetter <daniel@ffwll.ch>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/gpu/drm/i915/intel_lrc.c
drivers/gpu/drm/i915/intel_ringbuffer.c