vrend: reduce number of re-allocations in constant copying

Also use memcpy to copy the constats. These two things might shave of some
cycles but the measureable performance benefit is marginal.

v2: Avoid the memcpy that ralloc does when increasing the size, the data
   is overwritten anyway.

Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Signed-off-by: Jakob Bornecrantz <jakob@collabora.com>

src/vrend_renderer.c

@@ -420,6 +420,7 @@ struct vrend_vertex_element_array {
 struct vrend_constants {
    unsigned int *consts;
    uint32_t num_consts;
+   uint32_t num_allocated_consts;
 };
 
 struct vrend_shader_view {
@@ -2379,18 +2380,21 @@ void vrend_set_constants(struct vrend_context *ctx,
                          float *data)
 {
    struct vrend_constants *consts;
-   uint i;
 
    consts = &ctx->sub->consts[shader];
    ctx->sub->const_dirty[shader] = true;
 
-   consts->consts = realloc(consts->consts, num_constant * sizeof(float));
+   /* avoid reallocations by only growing the buffer */
+   if (consts->num_allocated_consts < num_constant) {
+      free(consts->consts);
+      consts->consts = malloc(num_constant * sizeof(float));
       if (!consts->consts)
          return;
+      consts->num_allocated_consts = num_constant;
+   }
 
+   memcpy(consts->consts, data, num_constant * sizeof(unsigned int));
    consts->num_consts = num_constant;
-   for (i = 0; i < num_constant; i++)
-      consts->consts[i] = ((unsigned int *)data)[i];
 }
 
 void vrend_set_uniform_buffer(struct vrend_context *ctx,