[Mesa-dev] [PATCH v4 13/18] i965/surface_state: Emit the clear color address instead of value.

[Jason Ekstrand]

On Tue, Mar 27, 2018 at 4:37 AM, Pohjolainen, Topi <
topi.pohjolainen at gmail.com> wrote:

> On Thu, Mar 08, 2018 at 08:49:06AM -0800, Rafael Antognolli wrote:
> > On Gen10, when emitting the surface state, use the value stored in the
> > clear color entry buffer by using a clear color address in the surface
> > state.
> >
> > v4: Use the clear color offset from the clear_color_bo, when available.
> >
> > Signed-off-by: Rafael Antognolli <rafael.antognolli at intel.com>
> > ---
> >  src/mesa/drivers/dri/i965/brw_wm_surface_state.c | 29
> +++++++++++++++++++++++-
> >  1 file changed, 28 insertions(+), 1 deletion(-)
> >
> > diff --git a/src/mesa/drivers/dri/i965/brw_wm_surface_state.c
> b/src/mesa/drivers/dri/i965/brw_wm_surface_state.c
> > index caa92d7d878..268effef397 100644
> > --- a/src/mesa/drivers/dri/i965/brw_wm_surface_state.c
> > +++ b/src/mesa/drivers/dri/i965/brw_wm_surface_state.c
> > @@ -152,7 +152,7 @@ brw_emit_surface_state(struct brw_context *brw,
> >
> >     union isl_color_value clear_color = { .u32 = { 0, 0, 0, 0 } };
> >
> > -   struct brw_bo *aux_bo;
> > +   struct brw_bo *aux_bo = NULL;
> >     struct isl_surf *aux_surf = NULL;
> >     uint64_t aux_offset = 0;
> >     switch (aux_usage) {
> > @@ -186,6 +186,20 @@ brw_emit_surface_state(struct brw_context *brw,
> >                                   brw->isl_dev.ss.align,
> >                                   surf_offset);
> >
> > +   bool use_clear_address = devinfo->gen >= 10 && aux_surf;
> > +
> > +   struct brw_bo *clear_bo = NULL;
> > +   uint32_t clear_offset = 0;
> > +   if (use_clear_address) {
> > +      if (mt->mcs_buf && mt->mcs_buf->clear_color_bo) {
> > +         clear_bo = mt->mcs_buf->clear_color_bo;
> > +         clear_offset = mt->mcs_buf->clear_color_offset;
>
> In case of "mt->mcs_buf" holding the clear value one needs to use
> "mt->mcs_buf->clear_color_offset" instead of "aux_surf->size" as that
> includes the space reserved for clear value and hence points past it
> (aux_offset doesn't matter is it will be zero in that case):
>

I'm pretty sure the second case is now dead and should be deleted.


>
>         } else if (mt->mcs_buf) {
>            clear_bo = aux_bo;
>            clear_offset = mt->mcs_buf->clear_color_offset;
>
> > +      } else {
> > +         clear_bo = aux_bo;
> > +         clear_offset = aux_offset + aux_surf->size;
> > +      }
> > +   }
> > +
> >     isl_surf_fill_state(&brw->isl_dev, state, .surf = &surf, .view =
> &view,
> >                         .address = brw_state_reloc(&brw->batch,
> >                                                    *surf_offset +
> brw->isl_dev.ss.addr_offset,
> > @@ -194,6 +208,8 @@ brw_emit_surface_state(struct brw_context *brw,
> >                         .aux_address = aux_offset,
> >                         .mocs = brw_get_bo_mocs(devinfo, mt->bo),
> >                         .clear_color = clear_color,
> > +                       .use_clear_address = use_clear_address,
> > +                       .clear_address = clear_offset,
> >                         .x_offset_sa = tile_x, .y_offset_sa = tile_y);
> >     if (aux_surf) {
> >        /* On gen7 and prior, the upper 20 bits of surface state DWORD 6
> are the
> > @@ -223,6 +239,17 @@ brw_emit_surface_state(struct brw_context *brw,
> >
> >        }
> >     }
> > +
> > +   if (use_clear_address) {
> > +      /* Make sure the offset is aligned with a cacheline. */
> > +      assert((clear_offset & 0x3f) == 0);
> > +      uint32_t *clear_address =
> > +            state + brw->isl_dev.ss.clear_color_state_offset;
> > +      *clear_address = brw_state_reloc(&brw->batch,
> > +                                       *surf_offset +
> > +                                       brw->isl_dev.ss.clear_color_
> state_offset,
> > +                                       clear_bo, *clear_address,
> reloc_flags);
> > +   }
> >  }
> >
> >  static uint32_t
> > --
> > 2.14.3
> >
> > _______________________________________________
> > mesa-dev mailing list
> > mesa-dev at lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/mesa-dev
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20180327/51cab00f/attachment.html>