Re: [Mesa-dev] [PATCH 13/32] i965/fs: Fix lower_load_payload() not to use an incorrect half for immediates and uniforms.

[Jason Ekstrand]

Yeah... More proof that the lower_load_payload code is plenty bogus...

Reviewed-by: Jason Ekstrand jason.ekstr...@intel.com

On Fri, Feb 6, 2015 at 9:42 AM, Francisco Jerez curroje...@riseup.net
wrote:

 ---
  src/mesa/drivers/dri/i965/brw_fs.cpp | 8 
  1 file changed, 8 insertions(+)

 diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp
 b/src/mesa/drivers/dri/i965/brw_fs.cpp
 index 8da1f47..e2ebf7e 100644
 --- a/src/mesa/drivers/dri/i965/brw_fs.cpp
 +++ b/src/mesa/drivers/dri/i965/brw_fs.cpp
 @@ -3116,6 +3116,14 @@ fs_visitor::lower_load_payload()
  inst-src[i].reg_offset;
mov-force_sechalf = metadata[src_reg].force_sechalf;
mov-force_writemask_all =
 metadata[src_reg].force_writemask_all;
 +   } else {
 +  /* We don't have any useful metadata for immediates or
 +   * uniforms.  Assume that any of the channels of the
 +   * destination may be used.
 +   */
 +  assert(inst-src[i].file == IMM ||
 + inst-src[i].file == UNIFORM);
 +  mov-force_writemask_all = true;
 }

 if (dst.file == GRF) {
 --
 2.1.3

 ___
 mesa-dev mailing list
 mesa-dev@lists.freedesktop.org
 http://lists.freedesktop.org/mailman/listinfo/mesa-dev

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH 13/32] i965/fs: Fix lower_load_payload() not to use an incorrect half for immediates and uniforms.

[Matt Turner]

On Fri, Feb 6, 2015 at 12:53 PM, Matt Turner matts...@gmail.com wrote:
 Patches 12 and 13 seem plausible, but I'd like Jason to review them.
 (And #11 pending my question about it)

Sorry, I meant 11-13 and #10 pending my question.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH 13/32] i965/fs: Fix lower_load_payload() not to use an incorrect half for immediates and uniforms.

[Matt Turner]

Patches 12 and 13 seem plausible, but I'd like Jason to review them.
(And #11 pending my question about it)
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [PATCH 13/32] i965/fs: Fix lower_load_payload() not to use an incorrect half for immediates and uniforms.

[Francisco Jerez]

---
 src/mesa/drivers/dri/i965/brw_fs.cpp | 8 
 1 file changed, 8 insertions(+)

diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp 
b/src/mesa/drivers/dri/i965/brw_fs.cpp
index 8da1f47..e2ebf7e 100644
--- a/src/mesa/drivers/dri/i965/brw_fs.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs.cpp
@@ -3116,6 +3116,14 @@ fs_visitor::lower_load_payload()
 inst-src[i].reg_offset;
   mov-force_sechalf = metadata[src_reg].force_sechalf;
   mov-force_writemask_all = 
metadata[src_reg].force_writemask_all;
+   } else {
+  /* We don't have any useful metadata for immediates or
+   * uniforms.  Assume that any of the channels of the
+   * destination may be used.
+   */
+  assert(inst-src[i].file == IMM ||
+ inst-src[i].file == UNIFORM);
+  mov-force_writemask_all = true;
}
 
if (dst.file == GRF) {
-- 
2.1.3

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev