drm/i915: Disable use of hwsp_cacheline for kernel_context
authorChris Wilson <chris@chris-wilson.co.uk>
Mon, 10 Feb 2020 20:57:20 +0000 (20:57 +0000)
committerChris Wilson <chris@chris-wilson.co.uk>
Tue, 11 Feb 2020 17:42:17 +0000 (17:42 +0000)
commitf16ccb6445d3c2038998375a8491686c43e6026e
tree8e45527a62aeffef39696561f6baea39c503d096
parent6313e78e720f82ab135f8cce809e0259ab54f7f3
drm/i915: Disable use of hwsp_cacheline for kernel_context

Currently on execlists, we use a local hwsp for the kernel_context,
rather than the engine's HWSP, as this is the default for execlists.
However, seqno wrap requires allocating a new HWSP cacheline, and may
require pinning a new HWSP page in the GGTT. This operation requiring
pinning in the GGTT is not allowed within the kernel_context timeline,
as doing so may require re-entering the kernel_context in order to evict
from the GGTT. As we want to avoid requiring a new HWSP for the
kernel_context, we can use the permanently pinned engine's HWSP instead.
However to do so we must prevent the use of semaphores reading the
kernel_context's HWSP, as the use of semaphores do not support rollover
onto the same cacheline. Fortunately, the kernel_context is mostly
isolated, so unlikely to give benefit to semaphores.

Reported-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200210205722.794180-5-chris@chris-wilson.co.uk
drivers/gpu/drm/i915/gt/intel_lrc.c
drivers/gpu/drm/i915/gt/selftest_lrc.c
drivers/gpu/drm/i915/i915_request.c