[Mesa-dev] [PATCH 39/42] i965/fs: introduce blorp specific rt-write for fs_generator

Paul Berry stereotype441 at gmail.com
Mon Jan 20 20:05:14 PST 2014


On 20 December 2013 06:39, Topi Pohjolainen <topi.pohjolainen at intel.com>wrote:

> diff --git a/src/mesa/drivers/dri/i965/brw_fs_generator.cpp
> b/src/mesa/drivers/dri/i965/brw_fs_generator.cpp
> index df91235..4c159e6 100644
> --- a/src/mesa/drivers/dri/i965/brw_fs_generator.cpp
> +++ b/src/mesa/drivers/dri/i965/brw_fs_generator.cpp
> @@ -190,6 +190,21 @@ fs_generator::generate_fb_write(fs_inst *inst)
>     mark_surface_used(surf_index);
>  }
>
> +void
> +fs_generator::generate_blorp_fb_write(fs_inst *inst)
> +{
> +   brw_fb_WRITE(p,
> +                16 /* dispatch_width */,
> +                inst->base_mrf,
> +                brw_reg_from_fs_reg(&inst->src[0]),
> +                BRW_DATAPORT_RENDER_TARGET_WRITE_SIMD16_SINGLE_SOURCE,
> +                1 /* BRW_BLORP_RENDERBUFFER_BINDING_TABLE_INDEX */,
>

This seems like it would lead to a maintenance burden if anyone ever
decides to change BRW_BLORP_RENDERBUFFER_BINDING_TABLE_INDEX to a number
other than 1.  Can we move the declaration of
BRW_BLORP_RENDERBUFFER_BINDING_TABLE_INDEX to somewhere that's accessible
to this file so that we can just pass it in directly?  Alternatively, we
could store it in inst->target.

With that issue addressed, this patch is:

Reviewed-by: Paul Berry <stereotype441 at gmail.com>


> +                inst->mlen,
> +                0,
> +                true,
> +                inst->header_present);
> +}
> +
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/mesa-dev/attachments/20140120/42c11b86/attachment.html>


More information about the mesa-dev mailing list