On 01/14/2013 10:47 PM, Vinson Lee wrote:
On Mon, Jan 14, 2013 at 7:24 AM, Brian Paul<[email protected]> wrote:
On 01/11/2013 12:28 AM, Vinson Lee wrote:
Fixes sizeof not portable defect reported by Coverity.
Fixes allocator sizeof operand mismatch bug reported by Clang Static
Analyzer.
Signed-off-by: Vinson Lee<[email protected]>
---
tests/texturing/shaders/texelFetch.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/texturing/shaders/texelFetch.c
b/tests/texturing/shaders/texelFetch.c
index 78a4162..50e91a6 100644
--- a/tests/texturing/shaders/texelFetch.c
+++ b/tests/texturing/shaders/texelFetch.c
@@ -297,7 +297,7 @@ generate_texture()
expected_colors = calloc(miplevels, sizeof(float **));
for (l = 0; l< miplevels; l++) {
- expected_colors[l] = calloc(level_size[l][2], sizeof(float
**));
+ expected_colors[l] = calloc(level_size[l][2], sizeof(float
*));
i_ptr = i_level =
malloc(level_size[l][0] * level_size[l][1] *
Since expected_colors is declared as
float ***expected_colors;
sizeof(float **) would seem to be the right expression. But sizeof(float *)
is the same size so I guess this is OK. Maybe put a comment there saying
"use sizeof(float *) to silence Clang Static Analyzer warning".
Reviewed-by: Brian Paul<[email protected]>
I think both Coverity and Clang are correct here and no comment is needed.
The calloc in question is for expected_colors[l] and not for
expected_colors. expected_colors[l] is of type float** and the
allocation should be a multiple of sizeof(float *)
Yes, you're right. Thanks.
-Brian
_______________________________________________
Piglit mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/piglit
