[Piglit] [PATCH 3/3] fbo-blending-formats: Fix expected color values for DST_ALPHA blending

[Carl Worth]

In the case where there are 0 bits of destination alpha, we need to
compute different expected results for the color when rendering with
DST_ALPHA. The results are initialized with calculations that blend
src and destination color such as:

	res4[X] = dst4[X]*(1-dst4[ALPHA]) + src4[X]*dst4[ALPHA]

But when there is no destination alpha channel, instead of the blend
factor dst4[ALPHA] in the above, the result should be computed with
this value replaced with 1.0. In other words, the correct calculation
for the expected result color becomes:

	res4[X] = src4[X]
---
 tests/fbo/fbo-blending-formats.c |   12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/tests/fbo/fbo-blending-formats.c b/tests/fbo/fbo-blending-formats.c
index e8528e9..d238966 100644
--- a/tests/fbo/fbo-blending-formats.c
+++ b/tests/fbo/fbo-blending-formats.c
@@ -217,12 +217,24 @@ static enum piglit_result test_format(const struct format_desc *format)
 			res5[2] = 0;
 		}
 		if (!a) {
+			/* When there are no bits for the alpha channel, we
+			 * always expect to read an alpha value of 1.0.
+			 */
 			res0[3] = 1;
 			res1[3] = 1;
 			res2[3] = 1;
 			res3[3] = 1;
 			res4[3] = 1;
 			res5[3] = 1;
+
+			/* Also blending with DST_ALPHA/ONE_MINUS_DST_ALPHA
+			 * (as in case 4) with an implicit destination alpha
+			 * value of 1.0 means that the result color should be
+			 * identical to the source color.
+			 */
+			res4[0] = src4[0];
+			res4[1] = src4[1];
+			res4[2] = src4[2];
 		}
         }
 
-- 
1.7.10.4