On 16/03/16 16:00, Chris Wilson wrote:
On Wed, Mar 16, 2016 at 03:40:54PM +0000, Tvrtko Ursulin wrote:
From: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
Majority of the code calls these ringbuf so fix up a few
places which use something else to be consistent.
I absolutely detest ringbuf as a name. The point of calling the engine
the engine was that we could call the ring state the ring.
:) I knew it. It is easily doable but it would be a somewhat bigger and
again disruptive patch. Maybe someone else should do it so I am not the
only one causing rebasing pain:
@@
expression E;
identifier r;
@@
- struct intel_ringbuffer *r = E;
+ struct intel_ringbuffer *ring = E;
<+...
- r
+ ring
...+>
@@
identifier r;
@@
- struct intel_ringbuffer *r;
+ struct intel_ringbuffer *ring;
<+...
- r
+ ring
...+>
@@
identifier func;
identifier r;
@@
func(..., struct intel_ringbuffer *
- r
+ ring
, ...)
{
<...
- r
+ ring
...>
}
@@
identifier func, r;
type T;
@@
T func(..., struct intel_ringbuffer *
- r
+ ring
, ...);
# cd $drm-intel/drivers/gpu/drm/i915
# spatch -c --no-includes --include-headers --sp-file <above> -j
<numcpus> --in-place .
Could maybe be all that is required.
Regards,
Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
