2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
Nice find! Series is:
Reviewed-by: Paul Berry
>
> diff --git a/src/mesa/drivers/dri/i965/brw_blorp_clear.cpp
> b/src/mesa/drivers/dri/i965/brw_blorp_clear.cpp
> index 45b9fa0..7e19c2c 100644
> --- a/src/mesa/drivers/dri/i9
On 17 March 2014 14:54, Eric Anholt wrote:
> ---
> src/mesa/drivers/dri/i965/brw_draw_upload.c | 9 -
> 1 file changed, 9 deletions(-)
>
For future cross-referencing it would be nice to mention the SHA of Marek's
change in the commit message.
>
> diff --git a/src/mesa/drivers/dri/i965
On 10 March 2014 17:23, Ian Romanick wrote:
> I had a pretty similar patch on the top of my pow-optimization branch.
> I also expand x**3 and x**4. I had hoped that would enable some cases
> to expand then merge to MADs. It should also be faster on older GENs
> where POW perf sucks. I didn't s
On 24 February 2014 00:36, Ilia Mirkin wrote:
> Before you read any further, this is nowhere close to working. However
> it's in
> a state where I think most of the structure is there, albeit with a lot of
> XXX
> comments. And I haven't actually implemented the new opcodes I've added.
>
> I was
_samples > 0 || dst_mt->num_samples > 0)
> + return false;
> +
>
You might consider moving this check up to the top of the function, since
it doesn't depend on src_format or dst_format. But it's not a big deal.
Either way, the patch is:
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
| 16 +---
> 5 files changed, 48 insertions(+), 3 deletions(-)
>
Hopefully I will get a chance to look at this in more detail soon, but for
now, consider this:
Acked-by: Paul Berry
>
> diff --git a/src/mesa/drivers/dri/i965/brw_defines.h
> b/src/mesa/drivers/dri/i965/br
On 4 February 2014 10:03, Jordan Justen wrote:
> >>
> >> What about trying to make use of
> >> MESA_EXTENSION_OVERRIDE=GL_ARB_compute_shader?
> >>
> >> We could add
> >> extensions.c:bool _mesa_is_extension_override_enabled(char *)
> >>
> >> And then
> >> if (_mesa_is_extension_override_enabl
0; i < num_state_slots; i++) {
> + mem.write_int32_t(state_slots[i].swizzle);
> + for (unsigned j = 0; j < 5; j++) {
> + mem.write_int32_t(state_slots[i].tokens[j]);
> + }
> + }
> +
> + if (constant_value)
> + constant_value->serializ
On 4 February 2014 19:52, Paul Berry wrote:
> Whoops, I discovered another issue:
>
>
> On 29 January 2014 01:24, Tapani Pälli wrote:
>
>> Class will be used by the shader binary cache implementation.
>>
>> Signed-off-by: Tapani Pälli
>>
Whoops, I discovered another issue:
On 29 January 2014 01:24, Tapani Pälli wrote:
> Class will be used by the shader binary cache implementation.
>
> Signed-off-by: Tapani Pälli
> ---
> src/glsl/memory_writer.h | 188
> +++
> 1 file changed, 188 ins
it. If you would prefer to just add a comment explaining why you're
serializing the null terminator, I would be ok with that too.
> + }
> +
> + unsigned position()
> + {
> + return pos;
> + }
> +
> + /**
> +* check if some data was written
> +*/
> + bool data_was_written(void *data, uint32_t id)
>
For functions that ask questions, I recommend using a word order that's
unambiguously a question (e.g. "was_data_written"). "data_was_written"
sounds more like a statement, so someone might misinterpret this function
as a synonym for mark_data_written().
But I'm nit picking again here. With the other issues addressed, this
patch is:
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On 1 February 2014 23:21, Jordan Justen wrote:
> On Thu, Jan 9, 2014 at 6:19 PM, Paul Berry
> wrote:
> > This will allow testing of compute shader functionality before it is
> > completed.
> >
> > To enable ARB_compute_shader functionality in the i965 driver,
On 1 February 2014 22:37, Jordan Justen wrote:
> On Thu, Jan 9, 2014 at 6:19 PM, Paul Berry
> wrote:
> > ---
> > src/mesa/drivers/dri/i965/brw_context.h | 8
> > src/mesa/drivers/dri/i965/brw_program.c | 11 +++
> > 2 files changed, 19 insertio
On 1 February 2014 22:28, Jordan Justen wrote:
> On Thu, Jan 9, 2014 at 6:19 PM, Paul Berry
> wrote:
> > ---
> > src/mesa/main/shaderapi.c | 18 ++
> > 1 file changed, 18 insertions(+)
> >
> > diff --git a/src/mesa/main/shaderapi.c b/src/mes
On 1 February 2014 21:25, Jordan Justen wrote:
> On Thu, Jan 9, 2014 at 6:19 PM, Paul Berry
> wrote:
> > ---
> > src/glsl/main.cpp | 1 +
> > src/glsl/standalone_scaffolding.cpp | 1 +
> > src/mesa/main/context.c | 1 +
> > src
+* so without spilling. If the GS invocations count > 1, then we can't
> use
> +* dual object mode.
> */
> - if (likely(!(INTEL_DEBUG & DEBUG_NO_DUAL_OBJECT_GS))) {
> + if (c->prog_data.invocations <= 1 ||
> + likely(!(INTEL_DEBUG & DEBUG_NO_DUAL_OBJ
On 28 January 2014 11:22, Jordan Justen wrote:
> Signed-off-by: Jordan Justen
> ---
> src/mesa/drivers/dri/i965/brw_defines.h | 5 +
> src/mesa/drivers/dri/i965/brw_vec4_gs_visitor.cpp | 24
> ---
> 2 files changed, 26 insertions(+), 3 deletions(-)
>
> diff --
+ if (check_gs_query(ctx, shProg))
> + *params = shProg->Geom.Invocations;
> + return;
> case GL_GEOMETRY_INPUT_TYPE:
>if (!has_core_gs)
> break;
> --
> 1.8.5.3
>
With that fixed, the patch is:
Reviewed-by: Paul Berry
__
GLint VerticesIn;
>GLint VerticesOut;
> + GLint Invocations;
>
Same comment applies here.
>GLenum InputType; /**< GL_POINTS, GL_LINES, GL_LINES_ADJACENCY_ARB,
>GL_TRIANGLES, or GL_TRIANGLES_ADJACENCY_ARB
> */
>GLenum OutputType; /**< GL_POINTS, GL_LINE_STRIP or
> GL_TRIANGLE_STRIP */
> --
> 1.8.5.3
>
With those minor issues addressed, this patch is:
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On 28 January 2014 11:22, Jordan Justen wrote:
> _mesa_glsl_parse_state::gs_invocations will store the
> invocation count.
>
> Signed-off-by: Jordan Justen
>
This looks like it will work if the shader contains a line like this:
layout(triangles, invocations=6) in;
But it won't handle this:
In addition to making it public, we also need to change its first
argument from an ir_loop * to an exec_list *, so that it can be used
to insert the condition anywhere in the IR (rather than just in the
body of the loop).
This will be necessary in order to make continue statements work
properly in
>From the GLSL 4.40 spec, section 6.4 (Jumps):
The continue jump is used only in loops. It skips the remainder of
the body of the inner most loop of which it is inside. For while
and do-while loops, this jump is to the next evaluation of the
loop condition-expression from which the
gsi can't handle.
>
I had a detailed look at your patches and I really like where you're going
with this. I sent a lot of feedback in response to patches 4, 5, 7, 9, and
10; I don't think any of my feedback would require a major change to your
overall plan. Nice work! Consider pat
t; +foreach_list(sig_node, &f->signatures) {
> + ir_function_signature *sig = (ir_function_signature *)
> sig_node;
> +
> + convert_from_ssa_function(&sig->body);
> +}
> + }
> + }
> +}
>
It looks like the only reason you need to manually walk through the
functions and signatures is so that you can make sure that
ir_from_ssa_visitor::base_instrs always points to the correct function's
instruction list. Normally the way we achieve that sort of thing is to do:
ir_visitor_status
ir_from_ssa_visitor::visit_enter(ir_function_signature *ir)
{
this->base_instrs = &ir->body;
return visit_continue;
}
ir_visitor_status
ir_from_ssa_visitor::visit_leave(ir_function_signature *ir)
{
this->base_instrs = NULL;
return visit_continue;
}
That way you could get rid of convert_from_ssa_function() and
convert_from_ssa() would just become:
void
convert_from_ssa(exec_list *instructions)
{
ir_from_ssa_visitor v;
v.run(instructions);
}
I don't have really strong feelings about it, though. Provided that the
confusing comment in eliminate_phi_if() is fixed, this patch is:
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On 28 January 2014 13:15, Carl Worth wrote:
> Paul Berry writes:
> > Previous to this patch, the _mesa_{Begin,Resume}TransformFeedback
> > functions were using ctx->Shader.CurrentProgram[MESA_SHADER_VERTEX]
>
> I see that this commit was nominated for the stable branch
On 22 January 2014 09:16, Connor Abbott wrote:
> diff --git a/src/glsl/ir_clone.cpp b/src/glsl/ir_clone.cpp
> index cb732a5..7075579 100644
> --- a/src/glsl/ir_clone.cpp
> +++ b/src/glsl/ir_clone.cpp
> @@ -40,7 +40,15 @@ ir_rvalue::clone(void *mem_ctx, struct hash_table *ht)
> const
> ir_variabl
On 22 January 2014 09:16, Connor Abbott wrote:
> This patch introduces all the changes to the IR that are necessary for
> representing programs in the SSA form. This consists of a new variable
> mode, the SSA temporary, which is guarenteed to be written to exactly
> once, and classes to represent
ead of storing the ir_loop *. But I don't feel terribly strongly about
it.
Either way, the patch is:
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
r_visitor_status visit(ir_loop_jump *);
> + virtual ir_visitor_status visit_enter(ir_return *);
> + virtual ir_visitor_status visit_enter(ir_discard *);
> +
> + ir_dead_branches *get_dead_branches(ir_if *ir);
> +
> +private:
> + void visit_return();
> +
> +
The "shader" field in fs_generator, vec4_generator, and gen8_generator
was only used for one purpose; to figure out if we were compiling an
assembly shader or a GLSL shader. And it wasn't being used properly:
in vec4 shaders we were always initializing it based on
prog->_LinkedShaders[MESA_SHADER_
erts didn't get compiled. If that's the case, I'd like to encourage you
to do a debug build and double-check that piglit tests still pass. You can
do that by adding the "--enable-debug" option when you run ./autogen.sh.
With that fixed, the series is:
Re
DEX' to the
> generator in fs_inst::target instead of hardcoding it
>
> CC: Paul Berry
> Signed-off-by: Topi Pohjolainen
>
Thanks, Topi. The whole series is now:
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@
On 22 January 2014 08:20, Kenneth Graunke wrote:
> On 01/22/2014 06:07 AM, Paul Berry wrote:
> > @@ -376,25 +376,48 @@ _mesa_compute_max_transform_feedback_vertices(
> > **/
> >
> >
> > +/**
> > + * Figure out which stage of the pipeline is the source of
Transform feedback may come from either the geometry shader or the
vertex shader, so we can't use
ctx->Shader.CurrentProgram[MESA_SHADER_VERTEX] to find the current
post-link transform feedback information. Fortunately we can use
ctx->TransformFeedback.CurrentObject->shader_program.
---
src/mesa/
Previous to this patch, the _mesa_{Begin,Resume}TransformFeedback
functions were using ctx->Shader.CurrentProgram[MESA_SHADER_VERTEX] to
find the program that would be the source of transform feedback data.
This isn't correct--if there's a geometry shader present it should be
ctx->Shader.CurrentPro
On 22 January 2014 04:50, Francisco Jerez wrote:
> Paul Berry writes:
>
> > On 15 January 2014 11:42, Francisco Jerez wrote:
> >
> >> v2: Reuse the glsl_sampler_dim enum for images. Reuse the
> >> glsl_type::sampler_* fields instead of creating new o
On 15 January 2014 14:01, Francisco Jerez wrote:
> Paul Berry writes:
>
> > On 2 December 2013 11:31, Francisco Jerez wrote:
> >
> >> Until now it was only being taken into account in the VEC4 back-end
> >> but not in the FS back-end. Do it in both cases.
On 21 January 2014 19:21, Paul Berry wrote:
>
> not realize that is is safe to call it on a register with 32 bits per
> component, because they might erroneously think "oh, this function only
> works on 16-component registers; that is, registers that
>
Oops, sorry about the
On 15 January 2014 13:53, Francisco Jerez wrote:
> Paul Berry writes:
>
> > On 2 December 2013 11:31, Francisco Jerez wrote:
> >
> >> +/**
> >> + * Get either of the 8-component halves of a 16-component register.
> >> + */
> >> +static
On 15 January 2014 11:42, Francisco Jerez wrote:
> v2: Reuse the glsl_sampler_dim enum for images. Reuse the
> glsl_type::sampler_* fields instead of creating new ones specific
> to image types. Reuse the same constructor as for samplers adding
> a new 'base_type' argument.
> ---
>
On 21 January 2014 14:54, Timothy Arceri wrote:
> Hi Paul
>
> Thanks for the thorough review and taking the time to explain how to make
> improvements. I can already see a number of improvements that can be made
> to patch 5 based on the suggestions you have already made so you could hold
> off r
On 21 January 2014 04:19, Timothy Arceri wrote:
> Signed-off-by: Timothy Arceri
> ---
> docs/GL3.txt | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@lists.freed
them can be combined to:
snprintf(n + idx, name_length - idx, "[%u]%s", length, array->name + idx);
Good catch on noticing that it's necessary to reverse the order of the
dimensions. With that fixed, this patch is:
Reviewed-by: Paul Berry
> + } else {
> +
On 21 January 2014 18:23, Paul Berry wrote:
> On 21 January 2014 04:19, Timothy Arceri wrote:
>
>> Signed-off-by: Timothy Arceri
>> ---
>> src/glsl/ast_to_hir.cpp | 4
>> 1 file changed, 4 insertions(+)
>>
>> diff --git a/src/glsl/ast_to_hir.cpp
>is_array())
check_type = check_type->element_type();
Then we can leave the switch statement alone.
With that fixed, this patch is:
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On 21 January 2014 04:19, Timothy Arceri wrote:
> Signed-off-by: Timothy Arceri
> ---
> src/glsl/ast.h | 19 +++-
> src/glsl/ast_function.cpp | 14 +++--
> src/glsl/ast_to_hir.cpp | 29 +-
> src/glsl/glsl_parser.yy | 36
Most of the time it is not necessary to perform type inference to
compile GLSL; the type of every expression can be inferred from the
contents of the expression itself (and previous type declarations).
The exception is aggregate initializers: their type is determined by
the LHS of the variable bein
On 21 January 2014 04:19, Timothy Arceri wrote:
> Signed-off-by: Timothy Arceri
> ---
> src/glsl/ast_to_hir.cpp | 6 --
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
Can you say more in the commit message about why you're making this
change? Is this fixing an existing bug? Is it pr
On 21 January 2014 04:19, Timothy Arceri wrote:
> Signed-off-by: Timothy Arceri
> ---
> src/glsl/ast.h | 44 +++
> src/glsl/ast_array_index.cpp| 13
> src/glsl/ast_to_hir.cpp | 160
> +++-
> src/glsl/ast_type.cpp
20" from the error message here.
> + }
> +
>
I think we need to add:
$$ = $1;
here.
> + $$->array_dimensions.push_tail(& $3->link);
> + $$->dimension_count++;
> + $$->set_location(yylloc);
>
Assuming I'm right about adding &qu
(in glcpp-parse.y) to call add_builtin_define(parser,
"GL_ARB_arrays_of_arrays") if extensions->ARB_arrays_of_arrays is set.
With that fixed, this patch is:
Reviewed-by: Paul Berry
Personally I would also add the extension to the
_mesa_glsl_supported_extensions[] table in this
5/test_blorp_blit_eu_gen.cpp| 30 +++---
> 2 files changed, 47 insertions(+), 31 deletions(-)
>
Series is:
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
he
> visitor. This patch does that - and it is indeed much nicer.)
>
> --Ken
>
Looks good, thanks!
Reviewed-by: Paul Berry
>
> diff --git a/src/mesa/drivers/dri/i965/brw_vec4_gs.c
> b/src/mesa/drivers/dri/i965/brw_vec4_gs.c
> index 018b0b6..c40112b 100644
> --- a
he changes are
correct, and in the unlikely event that we later discover that they aren't,
it'll make it easier to bisect to the problem.
With this patch split, it is:
Reviewed-by: Paul Berry
I've sent comments on patches 24, 27, 37, 38, 39, and 41. All the others
are:
Rev
On 20 January 2014 19:36, Paul Berry wrote:
> On 20 December 2013 06:38, Topi Pohjolainen wrote:
>
>> diff --git a/src/mesa/drivers/dri/i965/brw_blorp_blit_eu.cpp
>> b/src/mesa/drivers/dri/i965/brw_blorp_blit_eu.cpp
>> index b189aa2..dcfd82b 100644
>> -
On 20 December 2013 06:39, Topi Pohjolainen wrote:
> This is rather ugly but as I couldn't think of anything better
> for now and wanted to get the rest of the series under review,
> I left it as it is.
> Even though immediately surrounding code has tabs this piece is
> written space-indented.
>
>
ely, we
could store it in inst->target.
With that issue addressed, this patch is:
Reviewed-by: Paul Berry
> +inst->mlen,
> +0,
> +true,
> +inst->header_present);
> +}
> +
>
___
a default value of NULL, and only dump instructions if
it is non-NULL?
Either way, the patch is:
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
+ brw_CMP(&func, vec16(brw_null_reg()), op, x, y);
> + brw_IF(&func, BRW_EXECUTE_16);
> + }
>
Can we maybe call tihs "emit_cmp_if()", so that it's clear when looking at
the caller that it performs both a CMP and an IF?
With that change, this patch is:
Reviewed-by:
ALIGN_1);
> + for (int k = 0; k < 8; k += 2)
> + emit_lrp(vec8(SAMPLE(0, k)),
> + offset(y_frac, k & 1),
>
Same comment applies here.
> + vec8(SAMPLE(2, k)),
> + vec8(SAMPLE(0, k)));
> #undef SAMPLE
> }
>
With those two things fixed, this patch is:
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
eading
emt_combine() won't have to look at manual_blend_average() to understand
why we use ADD for floats and AVG for ints.
With that change, this patch is:
Reviewed-by: Paul Berry
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Previously, time spent in geometry shaders would be counted as part of
the vertex shader time.
---
src/mesa/drivers/dri/i965/brw_context.h | 3 +++
src/mesa/drivers/dri/i965/brw_program.c | 10 +-
src/mesa/drivers/dri/i965/brw_vec4.cpp
These messages are in code that is shared between the VS and GS
back-ends, so use the terminology "vec4" to avoid confusion.
---
src/mesa/drivers/dri/i965/brw_vec4_generator.cpp | 8
src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp | 2 +-
2 files changed, 5 insertions(+), 5 deletions(-)
On 14 January 2014 03:35, Tapani Pälli wrote:
> On 01/14/2014 01:24 AM, Paul Berry wrote:
>
> On 2 January 2014 03:58, Tapani Pälli wrote:
>
> + var->state_slots = NULL;
>> +
>> + if (var->num_state_slots > 0) {
>> + var->
On 14 January 2014 02:35, Tapani Pälli wrote:
> On 01/13/2014 08:27 PM, Paul Berry wrote:
>
> On 2 January 2014 03:58, Tapani Pälli wrote:
>
>> Class will be used by the shader binary cache implementation.
>>
>> Signed-off-by: Tapani Pälli
>> --
On 15 January 2014 02:46, Tapani wrote:
> On 01/14/2014 07:53 PM, Paul Berry wrote:
>
> On 2 January 2014 03:58, Tapani Pälli wrote:
>
>> +static void
>>> +serialize_uniform_storage(gl_uniform_storage *uni, memory_writer &blob)
>>>
On 16 January 2014 05:37, Tapani Pälli wrote:
> On 01/15/2014 06:13 PM, Paul Berry wrote:
>
> On 2 January 2014 03:58, Tapani Pälli wrote:
>
>> Signed-off-by: Tapani Pälli
>> ---
>> src/mesa/main/shaderapi.c | 44
>> +++
for string_to_uint_map
> glsl: functions to serialize gl_shader and gl_shader_program
> mesa: OES_get_program_binary extension functionality
>
I sent comments on all patches but 4, 5, and 7. Those patches are:
Reviewed-by: Paul Berry
I se
On 2 January 2014 03:58, Tapani Pälli wrote:
> Signed-off-by: Tapani Pälli
> ---
> src/mesa/main/shaderapi.c | 44
> ++--
> 1 file changed, 38 insertions(+), 6 deletions(-)
>
> diff --git a/src/mesa/main/shaderapi.c b/src/mesa/main/shaderapi.c
> index 575
On 2 January 2014 03:58, Tapani Pälli wrote:
> diff --git a/src/glsl/shader_cache.cpp b/src/glsl/shader_cache.cpp
> new file mode 100644
> index 000..4b5de45
> --- /dev/null
> +++ b/src/glsl/shader_cache.cpp
> @@ -0,0 +1,489 @@
> +/* -*- c++ -*- */
> +/*
> + * Copyright © 2013 Intel Corporati
On 2 January 2014 03:58, Tapani Pälli wrote:
> +
> +
> +/**
> + * Reads header part of the binary blob. Main purpose of this header is to
> + * validate that cached shader was produced with same Mesa driver version.
> + */
> +int
> +ir_deserializer::read_header(struct gl_shader *shader)
> +{
> +
On 2 January 2014 03:58, Tapani Pälli wrote:
> Class will be used by the shader binary cache implementation.
>
> Signed-off-by: Tapani Pälli
> ---
> src/glsl/memory_map.h | 174
> ++
> 1 file changed, 174 insertions(+)
> create mode 100644 src/gl
mall integer, so that we could
serialize each type just once (see
http://lists.freedesktop.org/archives/mesa-dev/2013-November/047740.html).
At the time, it sounded like you liked that idea. Have you made that
change? It looks to me like you've stopped serializing the built-in types,
bu
On 2 January 2014 03:58, Tapani Pälli wrote:
> Class will be used by the shader binary cache implementation.
>
> Signed-off-by: Tapani Pälli
> ---
> src/glsl/memory_writer.h | 147
> +++
> 1 file changed, 147 insertions(+)
> create mode 100644 src/gl
Prior to this patch, if we ran out of aperture space during
brw_try_draw_prims(), we would rewind the batch buffer pointer
(potentially throwing some state that may have been emitted by
brw_upload_state()), flush the batch, and then try again. However, we
wouldn't reset the dirty bits to the state
On 10 January 2014 11:41, Kenneth Graunke wrote:
> On 01/09/2014 06:19 PM, Paul Berry wrote:
> > This is possible now that ctx->Shader.CurrentProgram is an array.
> > ---
> > src/mesa/main/texstate.c | 75
> +++-
> &g
On 9 January 2014 22:32, Chris Forbes wrote:
> Minor nit, but could CS be done after the ordered pipeline stages, for
> consistency?
>
Sure, no problem.
In fact, I just realized that if I change the "type" parameter of
use_shader_program to gl_shader_stage, I can call it in a loop. I'll do
tha
On 9 January 2014 20:03, Chris Forbes wrote:
> This is a slightly odd construction (although copied from the existing
> code):
>
> > + if ((shProg == NULL) || (shProg->_LinkedShaders[stage] == NULL))
> > +shProg = NULL;
>
You're right. This would be much better as:
if ((shProg != NUL
On 9 January 2014 22:17, Chris Forbes wrote:
> This is a nice cleanup; I like that this brings both writes to
> prog->LastClipDistanceArraySize together -- but it looks like the
> behavior changes slightly.
>
> Previously, if there was no VS and no GS, then we would never write
> prog->LastClipDi
On 9 January 2014 18:19, Chris Forbes wrote:
> I'm not convinced this is correct/safe, at least for the meta clear.
>
> The GL 3.2 spec says, on page 221 (page 235 of the PDF):
>
> * If the number of layers of each attachment are not all identical,
> rendering
> will be limited to the small
This will allow testing of compute shader functionality before it is
completed.
To enable ARB_compute_shader functionality in the i965 driver, set
INTEL_COMPUTE_SHADER=1.
---
src/mesa/drivers/dri/i965/brw_context.c | 11 ++-
src/mesa/drivers/dri/i965/intel_extensions.c | 2 ++
2 fil
---
src/glsl/glcpp/glcpp-parse.y| 3 +++
src/glsl/glsl_parser_extras.cpp | 1 +
src/glsl/glsl_parser_extras.h | 2 ++
src/glsl/standalone_scaffolding.cpp | 1 +
src/mesa/main/extensions.c | 1 +
src/mesa/main/mtypes.h | 1 +
6 files changed, 9 insertions(+)
---
src/glsl/ast_to_hir.cpp | 7 +++
1 file changed, 7 insertions(+)
diff --git a/src/glsl/ast_to_hir.cpp b/src/glsl/ast_to_hir.cpp
index c1b1d6a..ed3a4f9 100644
--- a/src/glsl/ast_to_hir.cpp
+++ b/src/glsl/ast_to_hir.cpp
@@ -2220,6 +2220,13 @@ apply_type_qualifier_to_variable(const struct
a
This patch replaces code in _mesa_new_shader() and delete_shader_cb()
that checks the type of a shader with calls to
_mesa_validate_shader_target(). This has two advantages: it allows
for a more thorough check (since _mesa_validate_shader_target()
doesn't permit shader targets that aren't supporte
---
src/mesa/main/shaderapi.c | 18 ++
1 file changed, 18 insertions(+)
diff --git a/src/mesa/main/shaderapi.c b/src/mesa/main/shaderapi.c
index 053f27b..680d449 100644
--- a/src/mesa/main/shaderapi.c
+++ b/src/mesa/main/shaderapi.c
@@ -663,6 +663,24 @@ get_programiv(struct gl_con
---
src/glsl/builtin_variables.cpp | 4
src/glsl/glsl_parser_extras.cpp | 2 ++
src/glsl/glsl_parser_extras.h | 1 +
src/glsl/main.cpp | 3 +++
src/glsl/standalone_scaffolding.cpp | 3 +++
src/mesa/main/context.c | 3 +++
src/mesa/main/get.c
This is possible now that ctx->Shader.CurrentProgram is an array.
---
src/mesa/main/shaderapi.c | 39 +--
1 file changed, 9 insertions(+), 30 deletions(-)
diff --git a/src/mesa/main/shaderapi.c b/src/mesa/main/shaderapi.c
index d57748c..61ac0e3 100644
--- a/src
---
src/glsl/main.cpp | 1 +
src/glsl/standalone_scaffolding.cpp | 1 +
src/mesa/main/context.c | 1 +
src/mesa/main/get.c | 1 +
src/mesa/main/get_hash_params.py| 3 +++
src/mesa/main/mtypes.h | 1 +
6 files changed, 8 insertions(+)
---
src/glsl/ast.h | 34 +++
src/glsl/ast_to_hir.cpp | 79 ++
src/glsl/ast_type.cpp | 14 ++
src/glsl/builtin_variables.cpp | 20 +
src/glsl/glsl_lexer.ll | 3 +-
src/glsl/glsl_parser.yy |
---
src/glsl/builtin_variables.cpp | 27 +++
src/glsl/glsl_parser_extras.cpp | 4
src/glsl/glsl_parser_extras.h | 3 +++
src/glsl/main.cpp | 3 +++
src/glsl/standalone_scaffolding.cpp | 3 +++
src/mesa/main/context.c |
---
src/glsl/linker.cpp | 64 +++
src/mesa/main/mtypes.h| 17 +
src/mesa/main/shaderapi.c | 7 ++
3 files changed, 88 insertions(+)
diff --git a/src/glsl/linker.cpp b/src/glsl/linker.cpp
index 7461b17..11e0651 100644
--- a/src
This is the first of several planned patch series to implement the
extension ARB_compute_shader in Mesa. This series allows the Mesa
front-end to parse and compile a "do-nothing" compute shader--that is,
one which contains nothing but a compute shader input layout
declaration and an empty main() f
---
src/mesa/main/shaderapi.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/mesa/main/shaderapi.c b/src/mesa/main/shaderapi.c
index 519b200..5188e9c 100644
--- a/src/mesa/main/shaderapi.c
+++ b/src/mesa/main/shaderapi.c
@@ -993,6 +993,7 @@ _mesa_use_program(struct gl_context *ctx, struct
---
src/glsl/main.cpp | 4
src/glsl/standalone_scaffolding.cpp | 4
2 files changed, 8 insertions(+)
diff --git a/src/glsl/main.cpp b/src/glsl/main.cpp
index 03b7c78..afc15cb 100644
--- a/src/glsl/main.cpp
+++ b/src/glsl/main.cpp
@@ -50,6 +50,10 @@ initialize_context(s
---
src/mesa/drivers/dri/i965/brw_context.h | 8
src/mesa/drivers/dri/i965/brw_program.c | 11 +++
2 files changed, 19 insertions(+)
diff --git a/src/mesa/drivers/dri/i965/brw_context.h
b/src/mesa/drivers/dri/i965/brw_context.h
index df32ccb..abc1783 100644
--- a/src/mesa/drive
Since ctx->Shader.Current{Vertex,Geometry,Fragment}Program is an
array, this allows some meta code to be rolled up into loops.
---
src/mesa/drivers/common/meta.c | 30 ++
1 file changed, 14 insertions(+), 16 deletions(-)
diff --git a/src/mesa/drivers/common/meta.c b/sr
---
src/mapi/glapi/gen/ARB_compute_shader.xml | 40 +++
src/mapi/glapi/gen/Makefile.am| 1 +
src/mapi/glapi/gen/gl_API.xml | 4 ++-
src/mapi/glapi/gen/gl_genexec.py | 1 +
src/mesa/Makefile.sources | 1 +
src/mesa/SConscript
Fixes piglit test:
spec/ARB_compute_shader/linker/mix_compute_and_non_compute
---
src/glsl/linker.cpp | 7 +++
1 file changed, 7 insertions(+)
diff --git a/src/glsl/linker.cpp b/src/glsl/linker.cpp
index 11e0651..f1344ea 100644
--- a/src/glsl/linker.cpp
+++ b/src/glsl/linker.cpp
@@ -2107,6 +2
This is possible now that ctx->Shader.CurrentProgram is an array.
---
src/mesa/main/texstate.c | 75 +++-
1 file changed, 29 insertions(+), 46 deletions(-)
diff --git a/src/mesa/main/texstate.c b/src/mesa/main/texstate.c
index b9c76da..905a9d5 100644
--
This patch adds MESA_SHADER_COMPUTE to the gl_shader_stage enum.
Also, where it is trivial to do so, it adds a compute shader case to
switch statements that switch based on the type of shader. This
avoids "unhandled switch case" compiler warnings.
---
src/glsl/ast_to_hir.cpp | 9 ++
1 - 100 of 2245 matches
Mail list logo
