i965/vec4: Make with_writemask() non-static.

[mesa.git] / src / mesa / drivers / dri / i965 / brw_fs.cpp

diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp b/src/mesa/drivers/dri/i965/brw_fs.cpp
index 69e544aa4e1b20202e5acd7d2941c39099e5dc1b..96cb2ee5a60cbd76e20f7316e3a9843e638c8cf2 100644 (file)
--- a/src/mesa/drivers/dri/i965/brw_fs.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs.cpp
@@ -179,6 +179,7 @@ ALU3(BFI2)
 ALU1(FBH)
 ALU1(FBL)
 ALU1(CBIT)
+ALU3(MAD)
 
 /** Gen4 predicated IF. */
 fs_inst *
@@ -1358,7 +1359,11 @@ fs_visitor::split_virtual_grfs()
        * the send is reading the whole thing.
        */
       if (inst->is_send_from_grf()) {
-         split_grf[inst->src[0].reg] = false;
+         for (int i = 0; i < 3; i++) {
+            if (inst->src[i].file == GRF) {
+               split_grf[inst->src[i].reg] = false;
+            }
+         }
       }
    }
 
@@ -2118,6 +2123,20 @@ fs_visitor::register_coalesce()
            }
         }
 
+         if (has_source_modifiers) {
+            for (int i = 0; i < 3; i++) {
+               if (scan_inst->src[i].file == GRF &&
+                   scan_inst->src[i].reg == inst->dst.reg &&
+                   scan_inst->src[i].reg_offset == inst->dst.reg_offset &&
+                   inst->dst.type != scan_inst->src[i].type)
+               {
+                 interfered = true;
+                 break;
+               }
+            }
+         }
+
+
         /* The gen6 MATH instruction can't handle source modifiers or
          * unusual register regions, so avoid coalescing those for
          * now.  We should do something more specific.
@@ -3143,7 +3162,8 @@ brw_fs_precompile(struct gl_context *ctx, struct gl_shader_program *prog)
 
    key.clamp_fragment_color = ctx->API == API_OPENGL_COMPAT;
 
-   for (int i = 0; i < MAX_SAMPLERS; i++) {
+   unsigned sampler_count = _mesa_fls(fp->Base.SamplersUsed);
+   for (unsigned i = 0; i < sampler_count; i++) {
       if (fp->Base.ShadowSamplers & (1 << i)) {
          /* Assume DEPTH_TEXTURE_MODE is the default: X, X, X, 1 */
          key.tex.swizzles[i] =