Fwiw I remember having trouble finding justification for the comments in
that issue, and iirc the opposite seemed to be indicated by the spec text.
(E.g.that array indices should be maintained.) I suspect there are also AoA
interactions.
I guess my point here is to double check this, as well as what other
drivers do in this and similar situations.
On Aug 1, 2016 12:31 PM, "Kenneth Graunke" <kenn...@whitecape.org> wrote:
Issue 16 of the ARB_program_interface_query spec gives an example:
For example, in the following code:
uniform Block1 {
int member1;
};
uniform Block2 {
int member2;
} instance2;
uniform Block3 {
int member3;
} instance3[2]; // uses two separate buffer bindings
the three uniforms (if active) are enumerated as "member1",
"Block2.member2", and "Block3.member3".
>From this it's pretty clear that the array index should not be included.
Cc: mesa-sta...@lists.freedesktop.org
Signed-off-by: Kenneth Graunke <kenn...@whitecape.org>
---
src/compiler/glsl/linker.cpp | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/compiler/glsl/linker.cpp b/src/compiler/glsl/linker.cpp
index 6d45a02..bf11cb4 100644
--- a/src/compiler/glsl/linker.cpp
+++ b/src/compiler/glsl/linker.cpp
@@ -3800,7 +3800,8 @@ add_shader_variable(struct gl_shader_program *shProg,
unsigned stage_mask,
*/
const char *prefixed_name = (var->data.from_named_ifc_block &&
!is_gl_identifier(var->name))
- ? ralloc_asprintf(shProg, "%s.%s",
var->get_interface_type()->name,
+ ? ralloc_asprintf(shProg, "%s.%s",
+
var->get_interface_type()->without_array()->name,
name)
: name;
--
2.9.2
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
