Use one intrinsic for all sorts of interpolation.
Use two separate unexpanded instructions to represent INTERP_XY and _ZW -
this will allow to eliminate one part if it's not used.
Track liveness of special interpolation regs instead of reserving them -
this will allow to reuse those regs, lowering
Get rid of special handling for reserved regs.
Use one intrinsic for all kinds of interpolation.
Signed-off-by: Vadim Girlin
---
src/gallium/drivers/r600/r600_llvm.c | 200 +++
src/gallium/drivers/r600/r600_shader.c | 159
src/gallium/dr
On 12/23/2012 11:43 PM, Eric Anholt wrote:
Improves GLBenchmark 2.1 offscreen performance by 3.2% +/- 1.5% (n=52).
---
src/mesa/drivers/dri/intel/intel_mipmap_tree.c |3 +++
src/mesa/drivers/dri/intel/intel_tex.c |4
src/mesa/drivers/dri/intel/intel_tex_image.c|
On 12/24/2012 09:53 AM, Eric Anholt wrote:
x before
+ after
+--+
| xx + |
| x* x + + + +
We don't need them now that our set of parameter pointers points at the
GL core storage for them. This should save memory/bandwidth/overhead in
uniform updates.
---
src/mesa/drivers/dri/i965/brw_shader.cpp |3 ---
1 file changed, 3 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_shad
There's no reason to use an external copy if the relayout in the
external copy isn't serving us.
---
src/mesa/drivers/dri/i965/brw_fs.cpp | 70 +++---
src/mesa/drivers/dri/i965/brw_fs.h |2 +-
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp |2 +-
3 fil
NumParameters used to be an upper bound on the number of vec4s to be
uploaded, which was basically safe (unless your buffer was bound near
the top of address space *and* you array indexed outside the buffer, in
which case I think you might GPU hang). As I migrate the driver away
from ParameterValu
Now that ParameterValues doesn't change across the visitor, we don't
need to go through this.
---
src/mesa/drivers/dri/i965/brw_fs.cpp | 34 +++---
src/mesa/drivers/dri/i965/brw_fs.h |6 -
src/mesa/drivers/dri/i965/brw_fs_fp.cpp |5 ++--
src
---
src/glsl/ir_uniform.h |4
1 file changed, 4 insertions(+)
diff --git a/src/glsl/ir_uniform.h b/src/glsl/ir_uniform.h
index 913c537..30e6f26 100644
--- a/src/glsl/ir_uniform.h
+++ b/src/glsl/ir_uniform.h
@@ -80,6 +80,10 @@ struct gl_uniform_driver_storage {
struct gl_uniform_storag
Like in the FS, there's no reason to use an external copy if the
ParameterValues[] relayout of it isn't the layout we need.
---
src/mesa/drivers/dri/i965/brw_vec4.h |2 +-
src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp | 87 +---
2 files changed, 35 insertions(+
---
src/mesa/drivers/dri/i965/brw_shader.cpp |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mesa/drivers/dri/i965/brw_shader.cpp
b/src/mesa/drivers/dri/i965/brw_shader.cpp
index add3d96..bbde12c 100644
--- a/src/mesa/drivers/dri/i965/brw_shader.cpp
+++ b/src/mesa/drive
If adding scale parameters during program compile caused a realloc of
ParameterValues, then the driver uniform storage set up by
_mesa_associate_uniform_storage() would point to potentially freed
memory.
Note that this uses TexturesUsed, which may change at runtime for GLSL
when sampler uniforms c
---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp |4
1 file changed, 4 insertions(+)
diff --git a/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
b/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
index 8967a6c..4bde6c7 100644
--- a/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
+++ b/src/mesa/d
x before
+ after
+--+
| xx + |
| x* x + + + + |
| +
Things are even more restrictive than they used to be, so I've made
mistakes in this area.
---
src/mesa/drivers/dri/i965/brw_fs.cpp |8
src/mesa/drivers/dri/i965/brw_fs.h |1 +
src/mesa/drivers/dri/i965/brw_vec4.cpp |9 +
src/mesa/drivers/dri/i965/brw_vec4.h |
https://bugs.freedesktop.org/show_bug.cgi?id=58716
Sergey Anikin changed:
What|Removed |Added
See Also||https://bugs.freedesktop.or
https://bugs.freedesktop.org/show_bug.cgi?id=58718
Sergey Anikin changed:
What|Removed |Added
See Also||https://bugs.freedesktop.or
https://bugs.freedesktop.org/show_bug.cgi?id=58718
Priority: medium
Bug ID: 58718
Assignee: mesa-dev@lists.freedesktop.org
Summary: Crash in src_register() during glClear() call
Severity: critical
Classification: Unclassified
https://bugs.freedesktop.org/show_bug.cgi?id=58716
--- Comment #4 from Sergey Anikin ---
Created attachment 72074
--> https://bugs.freedesktop.org/attachment.cgi?id=72074&action=edit
Sample application to reproduce the crash, p.3
--
You are receiving this mail because:
You are the assignee fo
https://bugs.freedesktop.org/show_bug.cgi?id=58716
--- Comment #3 from Sergey Anikin ---
Created attachment 72073
--> https://bugs.freedesktop.org/attachment.cgi?id=72073&action=edit
Sample application to reproduce the crash, p.2
--
You are receiving this mail because:
You are the assignee fo
https://bugs.freedesktop.org/show_bug.cgi?id=58716
Sergey Anikin changed:
What|Removed |Added
Attachment #72072|Sample application to |Incorrect attachment -
descriptio
https://bugs.freedesktop.org/show_bug.cgi?id=58716
--- Comment #2 from Sergey Anikin ---
Created attachment 72072
--> https://bugs.freedesktop.org/attachment.cgi?id=72072&action=edit
Sample application to reproduce the crash, p.2
--
You are receiving this mail because:
You are the assignee fo
https://bugs.freedesktop.org/show_bug.cgi?id=58716
--- Comment #1 from Sergey Anikin ---
Created attachment 72071
--> https://bugs.freedesktop.org/attachment.cgi?id=72071&action=edit
Sample application to reproduce the crash, p.1
--
You are receiving this mail because:
You are the assignee fo
https://bugs.freedesktop.org/show_bug.cgi?id=58716
Priority: medium
Bug ID: 58716
Assignee: mesa-dev@lists.freedesktop.org
Summary: Crash in _mesa_unpack_rgba_row() during glDrawArrays()
call
Severity: critical
Classifica
24 matches
Mail list logo
