Due to the implicit move-from-GRF, unary math looks a lot like the Gen6+
math instruction: it's a single instruction (SEND) with a GRF source.
The difference is that it also implicitly clobbers a message register.
The only visible effect is that CSE will remove the MRF-clobbering from
later math o
On 14.10.2014 11:42, Ilia Mirkin wrote:
Reported by Coverity
Signed-off-by: Ilia Mirkin
---
No idea how to test this, but seems pretty straightforward.
src/gallium/state_trackers/gbm/gbm_drm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/gb
From: Connor Abbott
v2 (Ken): Refactor the Gen7 code separately; rebase.
Signed-off-by: Connor Abbott
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs.h | 3 ++-
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 2 +-
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 8
brw_lower_unnormalized_offset sets ir->offset to NULL if it applies the
texelFetchOffset workarounds, so there's no need to special case it
here---there won't be an offset for ir_txf.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 5 ++---
src/mesa/drivers/dr
Before, we had 3 different emit functions for various different gen's,
as well as some ancilliary work that was the same across all gen's which
was either contained in functions or duplicated across the GLSL IR and
Mesa IR backends. Now, we have a single method, emit_texture(), that
takes all the i
From: Connor Abbott
Now that we've made all the texture emit code mostly independent of GLSL
IR, this isn't necessary any more.
Signed-off-by: Connor Abbott
Reviewed-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 19 +--
1 file changed, 5 insertions(+), 14 d
From: Connor Abbott
Our new IR won't have ir_texture objects.
Signed-off-by: Connor Abbott
Reviewed-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 4 +++-
src/mesa/drivers/dri/i965/brw_shader.cpp | 13 +
src/mesa/drivers/dri/i965/brw_shader.h
From: Connor Abbott
Our new IR won't have ir_texture objects.
Signed-off-by: Connor Abbott
Reviewed-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs.h | 3 ++-
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 2 +-
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 12 ++
This is slightly clearer. Based on a patch by Connor Abbott.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
b/src/mesa/drivers/dri/i965
From: Connor Abbott
Our new IR won't have ir_texture objects, but using glsl_type is fine.
v2 (Ken): Drop redundant ir->coordinate NULL check; rebase.
Signed-off-by: Connor Abbott
Reviewed-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs.h | 2 +-
src/mesa/drivers/dri/i965
From: Connor Abbott
This drops a dependency on ir_texture objects.
Signed-off-by: Connor Abbott
Reviewed-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs.h | 9 +++--
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 9 +++--
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 52
From: Connor Abbott
Our new IR won't have ir_texture objects.
Signed-off-by: Connor Abbott
Reviewed-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs.h | 2 +-
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 7 +++
2 files changed, 4 insertions(+), 5 deletions(-)
diff --g
From: Connor Abbott
At this point, the only thing it's used for is the opcode.
Signed-off-by: Connor Abbott
Reviewed-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs.h | 6 ++---
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 6 ++---
src/mesa/drivers/dri/i965/brw_fs_visit
From: Connor Abbott
Our new IR won't have ir_texture objects.
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 2 +-
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 6 +++---
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/
From: Connor Abbott
This drops a dependency on ir_texture objects.
v2 (Ken): Rename lod_components to grad_components, as it only has a
meaningful value for ir_txd. We could set it to 1 for TXL,
but there's no real need.
Signed-off-by: Connor Abbott
Reviewed-by: Kenneth Gr
From: Connor Abbott
This happened to work before, but it would convert the output to a float
and then back to an integer which seems bad.
Signed-off-by: Connor Abbott
Reviewed-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 1 +
1 file changed, 1 insertion(+)
diff --gi
This moves the handling of non-constant texel offset subexpression trees
to the place where we visit other such subtrees. It also removes some
uses of ir->offset in emit_texture_gen7, which will be useful when we
write the backend for our new upcoming IR.
Based on a patch by Connor Abbott.
Signe
Eric's original code to work around TXF offset bugs contained a comment
explaining the problem, which was lost when Chris generalized it to an
IR transformation (in commit 598ca510b8a118c3c7e18b5d031a2b116120e0a6).
This commit adds the original comment to the newer code.
Signed-off-by: Kenneth Gr
Reported by Coverity
Signed-off-by: Ilia Mirkin
---
No idea how to test this, but seems pretty straightforward.
src/gallium/state_trackers/gbm/gbm_drm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/gbm/gbm_drm.c
b/src/gallium/state_trackers/gb
On Mon, Oct 13, 2014 at 5:12 PM, Kenneth Graunke
wrote:
> fs_visitor::grf_used is used to calculate prog_data->reg_blocks_16,
> which was getting populated with an uninitialized value. Normally,
> grf_used is set by the register allocator, but emit_repclear_shader
> bypasses that; it must set gr
This code is already in if (!variable->C->is_r500) so no need check
twice.
Signed-off-by: David Heidelberger
---
src/gallium/drivers/r300/compiler/radeon_pair_regalloc.c | 7 ++-
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/src/gallium/drivers/r300/compiler/radeon_pair_reg
fs_visitor::grf_used is used to calculate prog_data->reg_blocks_16,
which was getting populated with an uninitialized value. Normally,
grf_used is set by the register allocator, but emit_repclear_shader
bypasses that; it must set grf_used directly.
This bug was well hidden: reg_blocks and reg_blo
On Sat, Oct 11, 2014 at 3:28 PM, Chris Forbes wrote:
> Suppresses a bunch of warning noise about sample_map possibly being used
> uninitialized.
>
> Signed-off-by: Chris Forbes
> ---
> src/mesa/drivers/common/meta_blit.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/src/mesa/drivers/c
Ah yes that was it. We can't expose that yet though since it would also
require baseinstance and not just basevertex shader var support (plus
DrawID too actually). So I still don't know what caps should be added...
Roland
On 10/13/2014 11:52 AM, Ilia Mirkin wrote:
Roland, is this the extens
Roland, is this the extension you were looking for?
https://www.opengl.org/registry/specs/ARB/shader_draw_parameters.txt
On Mon, Oct 13, 2014 at 2:44 PM, Roland Scheidegger wrote:
> Oh I actually thought that BASEVERTEX was available as an vs input in some
> gl extension. Looks like that's not th
Oh I actually thought that BASEVERTEX was available as an vs input in
some gl extension. Looks like that's not the case.
Still, I guess we'd need some cap bit so we can tell the glsl compiler
if it actually should lower vertexid to vertexid_zerobase + basevertex
or not. And even then that still
On 10/10/2014 11:46 PM, Vinson Lee wrote:
This patch fixes this build error on DragonFly BSD.
CC os/os_misc.lo
os/os_misc.c: In function 'os_get_total_physical_memory':
os/os_misc.c:132:2: error: #error Unsupported *BSD
Signed-off-by: Vinson Lee
---
src/gallium/auxiliary/os/os_misc.
On 13/10/14 15:11, Julien Cristau wrote:
> On Mon, Oct 13, 2014 at 00:38:25 +0100, Emil Velikov wrote:
>
>> Mesa 10.3.1 has been released. Mesa 10.3.1 is a bug fix release
>> fixing bugs since the 10.3 release, (see below for a list of
>> changes).
>>
>> The tag in the git repository for Mesa 10.3
https://bugs.freedesktop.org/show_bug.cgi?id=84894
--- Comment #2 from shillshoc...@gmail.com ---
Discussion on the solution for LightDM are here:
https://bugs.launchpad.net/lightdm/+bug/1380069
--
You are receiving this mail because:
You are the assignee for the bug.
___
On Mon, Oct 13, 2014 at 00:38:25 +0100, Emil Velikov wrote:
> Mesa 10.3.1 has been released. Mesa 10.3.1 is a bug fix release
> fixing bugs since the 10.3 release, (see below for a list of
> changes).
>
> The tag in the git repository for Mesa 10.3.1 is 'mesa-10.3.1'.
>
Hi Emil,
I'm not seeing
Actually, util_blitter_cache_all_shaders was supposed to prevent all
crashes. Drivers that use the Draw AA stages should call it before the
stages are installed, so that no shaders need to be compiled later. I
guess some shaders are missing there. Just updating
util_blitter_cache_all_shaders would
https://bugs.freedesktop.org/show_bug.cgi?id=84711
--- Comment #2 from LoneVVolf ---
You are right, and libva pkgconfig on arch indeed has /usr/lib/dri as default
folder.
Thanks for the information.
--
You are receiving this mail because:
You are the assignee for the bug.
_
radeonsi supports VERTEXID_ZEROBASE and BASEVERTEX. VERTEXID can be
implemented trivially using the two. I don't think we need any cap
bits, because this is required by OpenGL. In the worst case, drivers
will have to store BASEVERTEX in a constant buffer.
Marek
On Mon, Oct 13, 2014 at 9:45 AM,
From: Roland Scheidegger
Just like in core mesa, expose a separate vertexid which doesn't include
the base vertex (aka d3d10 style). And expose basevertex too (vertexid is
basevertex + vertexid_zerobase). This will be needed to fix the piglit
vertexid failures in draw due to it following d3d10 se
From: Roland Scheidegger
So, gallium doesn't yet expose basevertex shader semantic. Neither does
it really handle the different vertexID options (GL-style including
basevertex and d3d10-style not including it). draw currently implements
the latter (on purpose) whereas most drivers probably actual
35 matches
Mail list logo
