On Sun, 16 Oct 2011 10:23:31 +0200
Daniel Vetter wrote:
> From: Jesse Barnes
>
> v2 by danvet: Use a new flag to flush the render target cache on gen6+
> (hw reuses the old write flush bit), as suggested by Ben Widawsdy.
>
> Signed-off-by: Jesse Barnes
> Signed-off-by: Kenneth Graunke
> [dan
From: Jesse Barnes
v2 by danvet: Use a new flag to flush the render target cache on gen6+
(hw reuses the old write flush bit), as suggested by Ben Widawsdy.
Signed-off-by: Jesse Barnes
Signed-off-by: Kenneth Graunke
[danvet: this seems to fix cairo-perf-trace hangs on my snb]
Signed-Off-by: Da
On Mon, 03 Oct 2011 15:59:29 -0700
Eric Anholt wrote:
> On Mon, 3 Oct 2011 13:00:16 -0700, Jesse Barnes
> wrote:
> > On Mon, 26 Sep 2011 11:59:23 -0700
> > Kenneth Graunke wrote:
> >
> > > + /* Just flush everything for now */
> > > + flags |= PIPE_CONTROL_WC_FLUSH;
> > > + flags |= PIPE_CONT
On Mon, 3 Oct 2011 13:00:16 -0700, Jesse Barnes
wrote:
> On Mon, 26 Sep 2011 11:59:23 -0700
> Kenneth Graunke wrote:
>
> > + /* Just flush everything for now */
> > + flags |= PIPE_CONTROL_WC_FLUSH;
> > + flags |= PIPE_CONTROL_IS_FLUSH;
> > + flags |= PIPE_CONTROL_TC_FLUSH;
> > + flag
On Mon, 03 Oct 2011 14:14:53 -0700
Keith Packard wrote:
> On Mon, 3 Oct 2011 13:00:16 -0700, Jesse Barnes
> wrote:
>
> > This is the only bit I'd like to see changed. While we still have the
> > domain tracking code we may as well try to honor it and limit our
> > flushing here like we do wit
On Mon, 3 Oct 2011 13:00:16 -0700, Jesse Barnes
wrote:
> This is the only bit I'd like to see changed. While we still have the
> domain tracking code we may as well try to honor it and limit our
> flushing here like we do with MI_FLUSH.
I'd like to see this patch put in place, and then any 'op
On Mon, 26 Sep 2011 11:59:23 -0700
Kenneth Graunke wrote:
> + /* Just flush everything for now */
> + flags |= PIPE_CONTROL_WC_FLUSH;
> + flags |= PIPE_CONTROL_IS_FLUSH;
> + flags |= PIPE_CONTROL_TC_FLUSH;
> + flags |= PIPE_CONTROL_DEPTH_FLUSH;
> + flags |= PIPE_CONTROL_VF
On Mon, 26 Sep 2011 13:38:26 -0700, Kenneth Graunke
wrote:
> On 09/26/2011 01:31 PM, Daniel Vetter wrote:
> > On Mon, Sep 26, 2011 at 11:59:23AM -0700, Kenneth Graunke wrote:
> >> +static int
> >> +gen6_render_ring_flush(struct intel_ring_buffer *ring,
> >> + u32 invalidat
On 09/26/2011 01:31 PM, Daniel Vetter wrote:
> On Mon, Sep 26, 2011 at 11:59:23AM -0700, Kenneth Graunke wrote:
>> +static int
>> +gen6_render_ring_flush(struct intel_ring_buffer *ring,
>> + u32 invalidate_domains, u32 flush_domains)
>> +{
>> +u32 flags = 0;
>> +stru
On Mon, Sep 26, 2011 at 11:59:23AM -0700, Kenneth Graunke wrote:
> +static int
> +gen6_render_ring_flush(struct intel_ring_buffer *ring,
> + u32 invalidate_domains, u32 flush_domains)
> +{
> + u32 flags = 0;
> + struct pipe_control *pc = ring->private;
> + u32 sc
On 09/26/2011 12:16 PM, Chris Wilson wrote:
> On Mon, 26 Sep 2011 11:59:23 -0700, Kenneth Graunke
> wrote:
>> From: Jesse Barnes
>
> From the school of "If ain't broke, don't fix it" there needs to be a real
> explanation of why this change is required here.
>
> PIPE_CONTROL and its workaround
On Mon, 26 Sep 2011 20:16:04 +0100, Chris Wilson
wrote:
> PIPE_CONTROL and its workarounds is a very bitter pill to swallow if
> MI_FLUSH continues to function.
If you look at the simulator source code, you'll see that it implements
MI_FLUSH as a specific kind of PIPE_CONTROL. And that MI_FLUSH
On Mon, 26 Sep 2011 21:23:02 +0200, Daniel Vetter wrote:
> On Mon, Sep 26, 2011 at 08:16:04PM +0100, Chris Wilson wrote:
> > On Mon, 26 Sep 2011 11:59:23 -0700, Kenneth Graunke
> > wrote:
> > > From: Jesse Barnes
> >
> > From the school of "If ain't broke, don't fix it" there needs to be a rea
On Mon, Sep 26, 2011 at 08:16:04PM +0100, Chris Wilson wrote:
> On Mon, 26 Sep 2011 11:59:23 -0700, Kenneth Graunke
> wrote:
> > From: Jesse Barnes
>
> From the school of "If ain't broke, don't fix it" there needs to be a real
> explanation of why this change is required here.
>
> PIPE_CONTROL
On Mon, 26 Sep 2011 11:59:23 -0700, Kenneth Graunke
wrote:
> From: Jesse Barnes
>From the school of "If ain't broke, don't fix it" there needs to be a real
explanation of why this change is required here.
PIPE_CONTROL and its workarounds is a very bitter pill to swallow if
MI_FLUSH continues t
From: Jesse Barnes
Signed-off-by: Jesse Barnes
Signed-off-by: Kenneth Graunke
Tested-by: Kenneth Graunke
---
drivers/gpu/drm/i915/i915_reg.h |7 +-
drivers/gpu/drm/i915/intel_ringbuffer.c | 131 ---
2 files changed, 124 insertions(+), 14 deletions(-)
16 matches
Mail list logo
