[Mesa-dev] [PATCH 14/23] i965: Make brw_update_sampler_state() use brw_emit_sampler_state().

Pohjolainen, Topi topi.pohjolainen at intel.com
Wed Jul 30 01:38:28 PDT 2014


On Tue, Jul 29, 2014 at 04:29:19PM -0700, Kenneth Graunke wrote:
> Instead of stuffing bits directly into the brw_sampler_state structure,
> we now store them in local variables, then use brw_emit_sampler_state()
> to assemble the packet.  This separates the decision about what values
> to use from the actual packet emission, which makes the code more
> reusable across generations.
> 
> Signed-off-by: Kenneth Graunke <kenneth at whitecape.org>

Apart from the few small nits looks good to me.

> ---
>  src/mesa/drivers/dri/i965/brw_sampler_state.c | 206 +++++++++++---------------
>  1 file changed, 90 insertions(+), 116 deletions(-)
> 
> diff --git a/src/mesa/drivers/dri/i965/brw_sampler_state.c b/src/mesa/drivers/dri/i965/brw_sampler_state.c
> index c631b76..ae4694b 100644
> --- a/src/mesa/drivers/dri/i965/brw_sampler_state.c
> +++ b/src/mesa/drivers/dri/i965/brw_sampler_state.c
> @@ -286,100 +286,107 @@ upload_default_color(struct brw_context *brw,
>  static void
>  brw_update_sampler_state(struct brw_context *brw,
>                           int unit,
> -                         struct brw_sampler_state *sampler,
> +                         uint32_t *sampler_state,
>                           uint32_t batch_offset_for_sampler_state)
>  {
>     struct gl_context *ctx = &brw->ctx;
>     struct gl_texture_unit *texUnit = &ctx->Texture.Unit[unit];
>     struct gl_texture_object *texObj = texUnit->_Current;
> -   struct gl_sampler_object *gl_sampler = _mesa_get_samplerobj(ctx, unit);
> -   bool using_nearest = false;
> +   struct gl_sampler_object *sampler = _mesa_get_samplerobj(ctx, unit);

This is only used for reading, right? I guess we could declare it constant
to make that clearer.

> +
> +   unsigned min_filter, mag_filter, mip_filter;
> +   unsigned max_anisotropy = BRW_ANISORATIO_2;
> +   unsigned address_rounding = 0;
> +   unsigned wrap_s, wrap_t, wrap_r;
> +   unsigned min_lod, max_lod, base_level;
> +   unsigned shadow_function = 0;
> +   int lod_bias;
> +   bool non_normalized_coordinates = texObj->Target == GL_TEXTURE_RECTANGLE;

I would make this constant and also move it just before the final call.

>  
>     /* These don't use samplers at all. */
>     if (texObj->Target == GL_TEXTURE_BUFFER)
>        return;
>  
> -   switch (gl_sampler->MinFilter) {
> +   /* Select min and mip filters. */
> +   switch (sampler->MinFilter) {
>     case GL_NEAREST:
> -      sampler->ss0.min_filter = BRW_MAPFILTER_NEAREST;
> -      sampler->ss0.mip_filter = BRW_MIPFILTER_NONE;
> -      using_nearest = true;
> +      min_filter = BRW_MAPFILTER_NEAREST;
> +      mip_filter = BRW_MIPFILTER_NONE;
>        break;
>     case GL_LINEAR:
> -      sampler->ss0.min_filter = BRW_MAPFILTER_LINEAR;
> -      sampler->ss0.mip_filter = BRW_MIPFILTER_NONE;
> +      min_filter = BRW_MAPFILTER_LINEAR;
> +      mip_filter = BRW_MIPFILTER_NONE;
>        break;
>     case GL_NEAREST_MIPMAP_NEAREST:
> -      sampler->ss0.min_filter = BRW_MAPFILTER_NEAREST;
> -      sampler->ss0.mip_filter = BRW_MIPFILTER_NEAREST;
> +      min_filter = BRW_MAPFILTER_NEAREST;
> +      mip_filter = BRW_MIPFILTER_NEAREST;
>        break;
>     case GL_LINEAR_MIPMAP_NEAREST:
> -      sampler->ss0.min_filter = BRW_MAPFILTER_LINEAR;
> -      sampler->ss0.mip_filter = BRW_MIPFILTER_NEAREST;
> +      min_filter = BRW_MAPFILTER_LINEAR;
> +      mip_filter = BRW_MIPFILTER_NEAREST;
>        break;
>     case GL_NEAREST_MIPMAP_LINEAR:
> -      sampler->ss0.min_filter = BRW_MAPFILTER_NEAREST;
> -      sampler->ss0.mip_filter = BRW_MIPFILTER_LINEAR;
> +      min_filter = BRW_MAPFILTER_NEAREST;
> +      mip_filter = BRW_MIPFILTER_LINEAR;
>        break;
>     case GL_LINEAR_MIPMAP_LINEAR:
> -      sampler->ss0.min_filter = BRW_MAPFILTER_LINEAR;
> -      sampler->ss0.mip_filter = BRW_MIPFILTER_LINEAR;
> +      min_filter = BRW_MAPFILTER_LINEAR;
> +      mip_filter = BRW_MIPFILTER_LINEAR;
>        break;
>     default:
>        break;
>     }
>  
> -   /* Set Anisotropy:
> -    */
> -   if (gl_sampler->MaxAnisotropy > 1.0) {
> -      sampler->ss0.min_filter = BRW_MAPFILTER_ANISOTROPIC;
> -      sampler->ss0.mag_filter = BRW_MAPFILTER_ANISOTROPIC;
> +   /* Select mag filter. */
> +   if (sampler->MagFilter == GL_LINEAR)
> +      mag_filter = BRW_MAPFILTER_LINEAR;
> +   else
> +      mag_filter = BRW_MAPFILTER_NEAREST;
>  
> -      if (gl_sampler->MaxAnisotropy > 2.0) {
> -	 sampler->ss3.max_aniso = MIN2((gl_sampler->MaxAnisotropy - 2) / 2,
> -				       BRW_ANISORATIO_16);
> -      }
> -   }
> -   else {
> -      switch (gl_sampler->MagFilter) {
> -      case GL_NEAREST:
> -	 sampler->ss0.mag_filter = BRW_MAPFILTER_NEAREST;
> -	 using_nearest = true;
> -	 break;
> -      case GL_LINEAR:
> -	 sampler->ss0.mag_filter = BRW_MAPFILTER_LINEAR;
> -	 break;
> -      default:
> -	 break;
> +   /* Enable anisotropic filtering if desired. */
> +   if (sampler->MaxAnisotropy > 1.0) {
> +      min_filter = BRW_MAPFILTER_ANISOTROPIC;
> +      mag_filter = BRW_MAPFILTER_ANISOTROPIC;
> +
> +      if (sampler->MaxAnisotropy > 2.0) {
> +	 max_anisotropy =
> +            MIN2((sampler->MaxAnisotropy - 2) / 2, BRW_ANISORATIO_16);
>        }
>     }
>  
> -   sampler->ss1.r_wrap_mode = translate_wrap_mode(brw, gl_sampler->WrapR,
> -						  using_nearest);
> -   sampler->ss1.s_wrap_mode = translate_wrap_mode(brw, gl_sampler->WrapS,
> -						  using_nearest);
> -   sampler->ss1.t_wrap_mode = translate_wrap_mode(brw, gl_sampler->WrapT,
> -						  using_nearest);
> +   /* Set address rounding bits if not using nearest filtering. */
> +   if (min_filter != BRW_MAPFILTER_NEAREST) {
> +      address_rounding |= BRW_ADDRESS_ROUNDING_ENABLE_U_MIN |
> +                          BRW_ADDRESS_ROUNDING_ENABLE_V_MIN |
> +                          BRW_ADDRESS_ROUNDING_ENABLE_R_MIN;
> +   }
> +   if (mag_filter != BRW_MAPFILTER_NEAREST) {
> +      address_rounding |= BRW_ADDRESS_ROUNDING_ENABLE_U_MAG |
> +                          BRW_ADDRESS_ROUNDING_ENABLE_V_MAG |
> +                          BRW_ADDRESS_ROUNDING_ENABLE_R_MAG;
> +   }
>  
> -   if (brw->gen >= 6 &&
> -       sampler->ss0.min_filter != sampler->ss0.mag_filter)
> -	sampler->ss0.min_mag_neq = 1;
> +   bool either_nearest =
> +      sampler->MinFilter == GL_NEAREST || sampler->MagFilter == GL_NEAREST;

Could be constant as well.

> +   wrap_s = translate_wrap_mode(brw, sampler->WrapS, either_nearest);
> +   wrap_t = translate_wrap_mode(brw, sampler->WrapT, either_nearest);
> +   wrap_r = translate_wrap_mode(brw, sampler->WrapR, either_nearest);
>  
> -   /* Cube-maps on 965 and later must use the same wrap mode for all 3
> -    * coordinate dimensions.  Futher, only CUBE and CLAMP are valid.
> -    */
>     if (texObj->Target == GL_TEXTURE_CUBE_MAP ||
>         texObj->Target == GL_TEXTURE_CUBE_MAP_ARRAY) {
> -      if ((ctx->Texture.CubeMapSeamless || gl_sampler->CubeMapSeamless) &&
> -	  (gl_sampler->MinFilter != GL_NEAREST ||
> -	   gl_sampler->MagFilter != GL_NEAREST)) {
> -	 sampler->ss1.r_wrap_mode = BRW_TEXCOORDMODE_CUBE;
> -	 sampler->ss1.s_wrap_mode = BRW_TEXCOORDMODE_CUBE;
> -	 sampler->ss1.t_wrap_mode = BRW_TEXCOORDMODE_CUBE;
> +      /* Cube maps must use the same wrap mode for all three coordinate
> +       * dimensions.  Prior to Haswell, only CUBE and CLAMP are valid.
> +       */
> +      if ((ctx->Texture.CubeMapSeamless || sampler->CubeMapSeamless) &&
> +         (sampler->MinFilter != GL_NEAREST ||
> +          sampler->MagFilter != GL_NEAREST)) {
> +	 wrap_s = BRW_TEXCOORDMODE_CUBE;
> +	 wrap_t = BRW_TEXCOORDMODE_CUBE;
> +	 wrap_r = BRW_TEXCOORDMODE_CUBE;
>        } else {
> -	 sampler->ss1.r_wrap_mode = BRW_TEXCOORDMODE_CLAMP;
> -	 sampler->ss1.s_wrap_mode = BRW_TEXCOORDMODE_CLAMP;
> -	 sampler->ss1.t_wrap_mode = BRW_TEXCOORDMODE_CLAMP;
> +	 wrap_s = BRW_TEXCOORDMODE_CLAMP;
> +	 wrap_t = BRW_TEXCOORDMODE_CLAMP;
> +	 wrap_r = BRW_TEXCOORDMODE_CLAMP;
>        }
>     } else if (texObj->Target == GL_TEXTURE_1D) {
>        /* There's a bug in 1D texture sampling - it actually pays
> @@ -387,66 +394,34 @@ brw_update_sampler_state(struct brw_context *brw,
>         * Override the wrap_t value here to GL_REPEAT to keep
>         * any nonexistent border pixels from floating in.
>         */
> -      sampler->ss1.t_wrap_mode = BRW_TEXCOORDMODE_WRAP;
> -   }
> -
> -
> -   /* Set shadow function:
> -    */
> -   if (gl_sampler->CompareMode == GL_COMPARE_R_TO_TEXTURE_ARB) {
> -      /* Shadowing is "enabled" by emitting a particular sampler
> -       * message (sample_c).  So need to recompile WM program when
> -       * shadow comparison is enabled on each/any texture unit.
> -       */
> -      sampler->ss0.shadow_function =
> -	 intel_translate_shadow_compare_func(gl_sampler->CompareFunc);
> +      wrap_t = BRW_TEXCOORDMODE_WRAP;
>     }
>  
> -   /* Set LOD bias:
> -    */
> -   sampler->ss0.lod_bias = S_FIXED(CLAMP(texUnit->LodBias +
> -					 gl_sampler->LodBias, -16, 15), 6);
> -
> -   sampler->ss0.lod_preclamp = 1; /* OpenGL mode */
> -   sampler->ss0.default_color_mode = 0; /* OpenGL/DX10 mode */
> -
> -   sampler->ss0.base_level = U_FIXED(0, 1);
> -
> -   sampler->ss1.max_lod = U_FIXED(CLAMP(gl_sampler->MaxLod, 0, 13), 6);
> -   sampler->ss1.min_lod = U_FIXED(CLAMP(gl_sampler->MinLod, 0, 13), 6);
> -
> -   /* On Gen6+, the sampler can handle non-normalized texture
> -    * rectangle coordinates natively
> -    */
> -   if (brw->gen >= 6 && texObj->Target == GL_TEXTURE_RECTANGLE) {
> -      sampler->ss3.non_normalized_coord = 1;
> -   }
> -
> -   uint32_t sdc_offset;
> -   upload_default_color(brw, gl_sampler, unit, &sdc_offset);
> -
> -   if (brw->gen >= 6) {
> -      sampler->ss2.default_color_pointer = sdc_offset >> 5;
> -   } else {
> -      /* reloc */
> -      sampler->ss2.default_color_pointer =
> -         (brw->batch.bo->offset64 + sdc_offset) >> 5;
> -
> -      drm_intel_bo_emit_reloc(brw->batch.bo,
> -			      batch_offset_for_sampler_state +
> -			      offsetof(struct brw_sampler_state, ss2),
> -			      brw->batch.bo, sdc_offset,
> -			      I915_GEM_DOMAIN_SAMPLER, 0);
> +   /* Set shadow function. */
> +   if (sampler->CompareMode == GL_COMPARE_R_TO_TEXTURE_ARB) {
> +      shadow_function =
> +	 intel_translate_shadow_compare_func(sampler->CompareFunc);
>     }
>  
> -   if (sampler->ss0.min_filter != BRW_MAPFILTER_NEAREST)
> -      sampler->ss3.address_round |= BRW_ADDRESS_ROUNDING_ENABLE_U_MIN |
> -                                    BRW_ADDRESS_ROUNDING_ENABLE_V_MIN |
> -                                    BRW_ADDRESS_ROUNDING_ENABLE_R_MIN;
> -   if (sampler->ss0.mag_filter != BRW_MAPFILTER_NEAREST)
> -      sampler->ss3.address_round |= BRW_ADDRESS_ROUNDING_ENABLE_U_MAG |
> -                                    BRW_ADDRESS_ROUNDING_ENABLE_V_MAG |
> -                                    BRW_ADDRESS_ROUNDING_ENABLE_R_MAG;
> +   min_lod = U_FIXED(CLAMP(sampler->MinLod, 0, 13), 6);
> +   max_lod = U_FIXED(CLAMP(sampler->MaxLod, 0, 13), 6);
> +   lod_bias = S_FIXED(CLAMP(texUnit->LodBias + sampler->LodBias, -16, 15), 6);
> +   base_level = U_FIXED(0, 1);

Declarations for these four could be moved here and made constant.

> +
> +   uint32_t border_color_offset;
> +   upload_default_color(brw, sampler, unit, &border_color_offset);
> +
> +   brw_emit_sampler_state(brw,
> +                          sampler_state,
> +                          batch_offset_for_sampler_state,
> +                          min_filter, mag_filter, mip_filter,
> +                          max_anisotropy,
> +                          address_rounding,
> +                          wrap_s, wrap_t, wrap_r,
> +                          min_lod, max_lod, lod_bias, base_level,
> +                          shadow_function,
> +                          non_normalized_coordinates,
> +                          border_color_offset);
>  }
>  
>  
> @@ -484,7 +459,6 @@ brw_upload_sampler_state_table(struct brw_context *brw,
>                                           sampler_state);
>              } else {
>                 brw_update_sampler_state(brw, unit,
> -                                        (struct brw_sampler_state *)
>                                          sampler_state,
>                                          batch_offset_for_sampler_state);
>              }
> -- 
> 2.0.2
> 
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/mesa-dev


More information about the mesa-dev mailing list