On 05/26/2015 02:03 PM, Jose Fonseca wrote:
On 25/05/15 15:41, Brian Paul wrote:
If the glPushAttrib() mask value was zero we didn't actually push
anything onto the attribute stack. A subsequent glPopAttrib() call
would generate a GL_STACK_UNDERFLOW error. Now push a dummy attribute
in that case to prevent the error.
Mesa now matches nvidia's behavior.
---
src/mesa/main/attrib.c | 17 +++++++++++++++++
1 file changed, 17 insertions(+)
diff --git a/src/mesa/main/attrib.c b/src/mesa/main/attrib.c
index b163c0a..365a79d 100644
--- a/src/mesa/main/attrib.c
+++ b/src/mesa/main/attrib.c
@@ -177,6 +177,10 @@ struct texture_state
};
+/** An unused GL_*_BIT value */
+#define DUMMY_BIT 0x10000000
+
+
/**
* Allocate new attribute node of given type/kind. Attach payload
data.
* Insert it into the linked list named by 'head'.
@@ -253,6 +257,15 @@ _mesa_PushAttrib(GLbitfield mask)
/* groups specified by the mask. */
head = NULL;
+ if (mask == 0) {
+ /* if mask is zero we still need to push something so that we
+ * don't get a GL_STACK_UNDERFLOW error in glPopAttrib().
+ */
+ GLuint dummy = 0;
+ if (!push_attrib(ctx, &head, DUMMY_BIT, sizeof(dummy), &dummy))
+ goto end;
+ }
+
if (mask & GL_ACCUM_BUFFER_BIT) {
if (!push_attrib(ctx, &head, GL_ACCUM_BUFFER_BIT,
sizeof(struct gl_accum_attrib),
@@ -928,6 +941,10 @@ _mesa_PopAttrib(void)
}
switch (attr->kind) {
+ case DUMMY_BIT:
+ /* do nothing */
+ break;
+
case GL_ACCUM_BUFFER_BIT:
{
const struct gl_accum_attrib *accum;
If I'm not mistaken, you could use 0 for DUMMY_BIT (instead of
0x10000000), and therefore completely avoid the possibility of a clash.
Yes, I just thought it was more obvious this way what's going on.
Either way
Reviewed-by: Jose Fonseca <jfons...@vmware.com>
Thanks.
-Brian
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
