On 07.02.2017 04:42, Timothy Arceri wrote:
From: Timothy Arceri <timothy.arc...@collabora.com>
Pointers will have different lengths so we simply create a different
sha1 for each platform.
In theory we should be able to share cached shaders as we cache all
pointer as uint64_t however if a pointer is ever added to one of the
structs we write to file with blob_write_bytes() we run the risk of
introducing a bug that would be difficult to reproduce or report from
a user point of veiw. It's also very unlikely that Mesa developers will
regularly regression test the interaction of cache sharing between
platforms.
I think we should avoid storing pointers in the first place...
Nicolai
---
src/compiler/glsl/shader_cache.cpp | 2 +-
src/compiler/glsl/shader_cache.h | 6 ++++++
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/src/compiler/glsl/shader_cache.cpp
b/src/compiler/glsl/shader_cache.cpp
index 44ca5a4..ddcaeee 100644
--- a/src/compiler/glsl/shader_cache.cpp
+++ b/src/compiler/glsl/shader_cache.cpp
@@ -1348,7 +1348,7 @@ shader_cache_read_program_metadata(struct gl_context *ctx,
/* Include bindings when creating sha1. These bindings change the resulting
* binary so they are just as important as the shader source.
*/
- char *buf = ralloc_strdup(NULL, "vb: ");
+ char *buf = ralloc_strdup(NULL, CACHED_PROGRAM"\n vb: ");
prog->AttributeBindings->iterate(create_binding_str, &buf);
ralloc_strcat(&buf, "fb: ");
prog->FragDataBindings->iterate(create_binding_str, &buf);
diff --git a/src/compiler/glsl/shader_cache.h b/src/compiler/glsl/shader_cache.h
index 1596c33..2994b66 100644
--- a/src/compiler/glsl/shader_cache.h
+++ b/src/compiler/glsl/shader_cache.h
@@ -27,6 +27,12 @@
#include "util/disk_cache.h"
+#if __x86_64__
+#define CACHED_PROGRAM "program64:"
+#else
+#define CACHED_PROGRAM "program32:"
+#endif
+
static uint64_t inline
ptr_to_uint64_t(void *ptr)
{
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
