[Mesa-dev] [PATCH 2/2] mesa: free gl_uniform_storage::name

parcel_out_uniform_storage::visit_field() assigns a strdup()'d string into gl_uniform_storage::name, but it is never freed. Free gl_uniform_storage::name, fixes some Valgrind reported memory leaks. Signed-off-by: Pekka Paalanen --- src/mesa/main/shaderobj.c |4 1 files changed, 4 inser

[Mesa-dev] [PATCH 1/2] linker: fix strdup memory leak

string_to_uint_map::put() already does a strdup() for the key argument, so we leak the memory allocated by strdup() in link_uniforms.cpp. Remove the extra strdup(), fixes a few Valgrind detected leaks. Signed-off-by: Pekka Paalanen --- src/glsl/link_uniforms.cpp |3 +-- 1 files changed, 1 i

Re: [Mesa-dev] [PATCH 1/3] egl: add EGL_NV_post_sub_buffer

On Thu, Dec 15, 2011 at 5:21 AM, Ian Romanick wrote: > On 12/14/2011 12:24 PM, Fredrik Höglund wrote: >> >> v2: Handle EGL_POST_SUB_BUFFER_SUPPORTED_NV in >>     _eglParseSurfaceAttribList() >> >> Signed-off-by: Fredrik Höglund >> --- >>  include/EGL/eglext.h      |    9 + >>  src/egl/main

Re: [Mesa-dev] [PATCH 1/2] st/egl: Add support for EGL_NOK_swap_region

On Thu, Dec 15, 2011 at 4:06 AM, Fredrik Höglund wrote: > Backends indicate that they support this extension by returning > EGL_TRUE when native_display::get_param() is called with > NATIVE_PARAM_PRESENT_REGION and NATIVE_PARAM_PRESERVE_BUFFER. > > native_present_control is extended to include the

Re: [Mesa-dev] [PATCH 1/2] mesa: Set the correct ctx->NewState bitfield for rasterizer discard.

On 12/14/2011 04:10 PM, Marek Olšák wrote: > On Thu, Dec 15, 2011 at 12:18 AM, Paul Berry wrote: >> On 14 December 2011 15:00, Marek Olšák wrote: >>> >>> On Wed, Dec 14, 2011 at 11:25 PM, Paul Berry >>> wrote: On 14 December 2011 13:42, Marek Olšák wrote: > > I think RASTERIZER_DIS

Re: [Mesa-dev] [PATCH 1/2] mesa: Set the correct ctx->NewState bitfield for rasterizer discard.

On Thu, Dec 15, 2011 at 12:18 AM, Paul Berry wrote: > On 14 December 2011 15:00, Marek Olšák wrote: >> >> On Wed, Dec 14, 2011 at 11:25 PM, Paul Berry >> wrote: >> > On 14 December 2011 13:42, Marek Olšák wrote: >> >> >> >> I think RASTERIZER_DISCARD has nothing to do with transform feedback. >

Re: [Mesa-dev] [PATCH 1/2] mesa, gallium: add a cap for GPUs without unified color+generic varying slots

On Thu, Dec 15, 2011 at 12:01 AM, Ian Romanick wrote: >> Simple. I cannot pass the test glsl-max-varyings if I report 40 >> varying components, because I have only 32 texcoord components + 8 >> color components. I could re-use the color varyings, but it's not full >> single-precision. r300 uses a

Re: [Mesa-dev] [PATCH 1/2] mesa: Set the correct ctx->NewState bitfield for rasterizer discard.

On 14 December 2011 15:00, Marek Olšák wrote: > On Wed, Dec 14, 2011 at 11:25 PM, Paul Berry > wrote: > > On 14 December 2011 13:42, Marek Olšák wrote: > >> > >> I think RASTERIZER_DISCARD has nothing to do with transform feedback. > >> I think it's part of the same spec because it's not useful

Re: [Mesa-dev] [PATCH 13/14] i965/vs: Implement EXT_texture_swizzle support for VS texturing.

On 12/13/2011 01:19 PM, Eric Anholt wrote: > On Thu, 8 Dec 2011 17:08:04 -0800, Kenneth Graunke > wrote: >> Signed-off-by: Kenneth Graunke >> + if (one_mask) { >> + swizzled_result.writemask = one_mask; >> + emit(MOV(swizzled_result, src_reg(1.0f))); >> + } >> } > > I think this

Re: [Mesa-dev] [PATCH v2 1/8] mesa: Track changes to transform feedback state.

On 14 December 2011 13:54, Marek Olšák wrote: > On Wed, Dec 7, 2011 at 8:09 PM, Paul Berry > wrote: > > This patch adds a new bit to the ctx->NewState bitfield, > > _NEW_TRANSFORM_FEEDBACK, to track state changes that affect > > ctx->TransformFeedback. This bit can be used by driver back-ends t

Re: [Mesa-dev] [PATCH 1/2] mesa, gallium: add a cap for GPUs without unified color+generic varying slots

On 12/14/2011 01:47 PM, Marek Olšák wrote: On Wed, Dec 14, 2011 at 6:30 PM, Ian Romanick wrote: On 12/13/2011 04:32 PM, Marek Olšák wrote: --- src/gallium/drivers/r300/r300_screen.c |3 ++- src/gallium/include/pipe/p_defines.h |3 ++- src/glsl/linker.cpp|

Re: [Mesa-dev] [PATCH 1/2] mesa: Set the correct ctx->NewState bitfield for rasterizer discard.

On Wed, Dec 14, 2011 at 11:25 PM, Paul Berry wrote: > On 14 December 2011 13:42, Marek Olšák wrote: >> >> I think RASTERIZER_DISCARD has nothing to do with transform feedback. >> I think it's part of the same spec because it's not useful without it. >> As I understand it, _NEW_TRANSFORM_FEEDBACK

Re: [Mesa-dev] [PATCH] i965: Flush pipeline on EndTransformFeedback.

On 12/14/2011 02:54 PM, Paul Berry wrote: > A common use case for transform feedback is to perform one draw > operation that writes transform feedback output to a buffer, followed > by a second draw operation that consumes that buffer as vertex input. > Since vertex input is consumed at an earlier

Re: [Mesa-dev] [PATCH] i965: Flush pipeline on EndTransformFeedback.

A common use case for transform feedback is to perform one draw operation that writes transform feedback output to a buffer, followed by a second draw operation that consumes that buffer as vertex input. Since vertex input is consumed at an earlier pipeline stage than writing transform feedback out

Re: [Mesa-dev] [PATCH 1/2] mesa: Set the correct ctx->NewState bitfield for rasterizer discard.

On 14 December 2011 13:42, Marek Olšák wrote: > I think RASTERIZER_DISCARD has nothing to do with transform feedback. > I think it's part of the same spec because it's not useful without it. > As I understand it, _NEW_TRANSFORM_FEEDBACK is dirty when transform > feedback buffer bindings are chang

Re: [Mesa-dev] [PATCH v2 1/8] mesa: Track changes to transform feedback state.

On Wed, Dec 7, 2011 at 8:09 PM, Paul Berry wrote: > This patch adds a new bit to the ctx->NewState bitfield, > _NEW_TRANSFORM_FEEDBACK, to track state changes that affect > ctx->TransformFeedback.  This bit can be used by driver back-ends to > avoid expensive recomputations when transform feedback

Re: [Mesa-dev] [PATCH 1/2] mesa, gallium: add a cap for GPUs without unified color+generic varying slots

On Wed, Dec 14, 2011 at 6:30 PM, Ian Romanick wrote: > On 12/13/2011 04:32 PM, Marek Olšák wrote: >> >> --- >>  src/gallium/drivers/r300/r300_screen.c |    3 ++- >>  src/gallium/include/pipe/p_defines.h   |    3 ++- >>  src/glsl/linker.cpp                    |    6 ++ >>  src/mesa/main/mtypes.

Re: [Mesa-dev] [PATCH 1/2] mesa: Set the correct ctx->NewState bitfield for rasterizer discard.

I think RASTERIZER_DISCARD has nothing to do with transform feedback. I think it's part of the same spec because it's not useful without it. As I understand it, _NEW_TRANSFORM_FEEDBACK is dirty when transform feedback buffer bindings are changed or just enabled/disabled. On the other hand, RASTERIZ

Re: [Mesa-dev] [PATCH] meta: rework dest image allocation in mipmap generation code

On 12/14/2011 02:13 PM, Eric Anholt wrote: On Tue, 13 Dec 2011 18:07:05 -0700, Brian Paul wrote: From: Brian Paul This fixes two things: 1. If the texture object was created with glTexStorage2D, the call to _mesa_TexImage2D() would generate INVALID_OPERATION since the texture is marked

Re: [Mesa-dev] [PATCH 1/3] egl: add EGL_NV_post_sub_buffer

On 12/14/2011 12:24 PM, Fredrik Höglund wrote: v2: Handle EGL_POST_SUB_BUFFER_SUPPORTED_NV in _eglParseSurfaceAttribList() Signed-off-by: Fredrik Höglund --- include/EGL/eglext.h |9 + src/egl/main/eglapi.c | 24 src/egl/main/eglapi.h

Re: [Mesa-dev] [PATCH] meta: rework dest image allocation in mipmap generation code

On Tue, 13 Dec 2011 18:07:05 -0700, Brian Paul wrote: > From: Brian Paul > > This fixes two things: > 1. If the texture object was created with glTexStorage2D, the call >to _mesa_TexImage2D() would generate INVALID_OPERATION since the >texture is marked as immutable. > 2. _mesa_TexImage2

Re: [Mesa-dev] [PATCH 1/8] mesa: Record transform feedback strides/offsets in linker output.

Reviewed-by: Marek Olšák Marek On Wed, Dec 14, 2011 at 6:00 PM, Paul Berry wrote: > This patch adds two new fields to the gl_transform_feedback_info > struct: > > - BufferStride records the total number of components (per vertex) >  that transform feedback is being instructed to store in each b

[Mesa-dev] [PATCH 1/3] egl: add EGL_NV_post_sub_buffer

v2: Handle EGL_POST_SUB_BUFFER_SUPPORTED_NV in _eglParseSurfaceAttribList() Signed-off-by: Fredrik Höglund --- include/EGL/eglext.h |9 + src/egl/main/eglapi.c | 24 src/egl/main/eglapi.h |8 src/egl/main/egldisplay.h |2 ++

[Mesa-dev] [PATCH 2/3] egl_dri2/x11: Add support for EGL_NV_post_sub_buffer

Signed-off-by: Fredrik Höglund --- src/egl/drivers/dri2/platform_x11.c | 17 + 1 files changed, 17 insertions(+), 0 deletions(-) diff --git a/src/egl/drivers/dri2/platform_x11.c b/src/egl/drivers/dri2/platform_x11.c index 8dd231a..08a2c8d 100644 --- a/src/egl/drivers/dri2/plat

[Mesa-dev] [PATCH 3/3] st/egl: Add support for EGL_NV_post_sub_buffer

Signed-off-by: Fredrik Höglund --- src/gallium/state_trackers/egl/common/egl_g3d.c|8 - .../state_trackers/egl/common/egl_g3d_api.c| 26 2 files changed, 32 insertions(+), 2 deletions(-) diff --git a/src/gallium/state_trackers/egl/common/egl_g3d.c b/sr

Re: [Mesa-dev] [PATCH 1/8] intel: Drop check for wrapped_depth in RB mapping.

On 12/08/2011 01:39 PM, Eric Anholt wrote: > This used to be needed because irb->mt would be unset for fake packed > depth/stencil, but no longer. > --- > src/mesa/drivers/dri/intel/intel_fbo.c |2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/src/mesa/drivers/dri/int

[Mesa-dev] [PATCH 2/2] st/egl: Implement EGL_NOK_swap_region for x11

v2: inline x11_drawable_copy_buffers(). Signed-off-by: Fredrik Höglund --- src/gallium/state_trackers/egl/x11/native_dri2.c | 36 -- src/gallium/state_trackers/egl/x11/x11_screen.c | 21 +++- src/gallium/state_trackers/egl/x11/x11_screen.h | 14 - 3 fi

[Mesa-dev] [PATCH 1/2] st/egl: Add support for EGL_NOK_swap_region

Backends indicate that they support this extension by returning EGL_TRUE when native_display::get_param() is called with NATIVE_PARAM_PRESENT_REGION and NATIVE_PARAM_PRESERVE_BUFFER. native_present_control is extended to include the region that should be presented. When native_present_control::num

[Mesa-dev] [PATCH 2/2] i965 gen6: Implement rasterizer discard.

This patch enables rasterizer discard functionality (a part of transform feedback) in Gen6, by generating an alternate GS program when rasterizer discard is active. Instead of forwarding vertices down the pipeline, the alternate GS program uses a URB Write message to deallocate the URB entry that

[Mesa-dev] [PATCH 1/2] mesa: Set the correct ctx->NewState bitfield for rasterizer discard.

Previously, we were setting the _NEW_TRANSFORM bit when enabling or disabling RASTERIZER_DISCARD. This is incorrect, since _NEW_TRANSFORM flags changes to ctx->Transform, but the rasterizer discard flag is in ctx->TransformFeedback. This patch sets the correct bit, _NEW_TRANSFORM_FEEDBACK. --- s

Re: [Mesa-dev] [PATCH 1/8] mesa: Record transform feedback stride in linker output.

On 12/14/2011 05:48 AM, Marek Olšák wrote: > The strides are useful, although they can be computed by summing up > all the NumComponents fields for each buffer. Note that I also take > into account ARB_transform_feedback3, which allows having interleaved > attribs in more than one buffer. For the E

[Mesa-dev] [PATCH] i965: Flush pipeline on EndTransformFeedback.

From: Paul Berry A common use case for transform feedback is to perform one draw operation that writes transform feedback output to a buffer, followed by a second draw operation that consumes that buffer as vertex input. Since vertex input is consumed at an earlier pipeline stage than writing tra

Re: [Mesa-dev] [PATCH 6/8] i965 gen6: Turn on transform feedback extension.

On 12/14/2011 08:20 AM, Paul Berry wrote: > On 14 December 2011 02:51, Kenneth Graunke > wrote: > > On 12/13/2011 03:35 PM, Paul Berry wrote: > > This patch advertises support for EXT_transform_feedback on Intel Gen6 > > and higher. > > > > Since

Re: [Mesa-dev] [PATCH 7/8] i965 gen6+: Make intel_batchbuffer_emit_mi_flush() actually flush.

On 12/14/2011 08:28 AM, Paul Berry wrote: > On 14 December 2011 02:59, Kenneth Graunke > wrote: > > On 12/13/2011 03:35 PM, Paul Berry wrote: > > Previous to this patch, the function intel_batchbuffer_emit_mi_flush() > > was a bit of a misnomer. On Gen4+

Re: [Mesa-dev] [PATCH 3/8] i965 gen6+: Use 1-wide null operands for IF instructions

On Wed, 14 Dec 2011 08:06:20 -0800, Paul Berry wrote: > On 14 December 2011 02:33, Kenneth Graunke wrote: > > I think the point is that the ARF null register doesn't have a value, so > > null<0,1,0> is just as good as null<4,4,1> or null<8,8,1>. Either way > > you get nothing and write nothing.

Re: [Mesa-dev] [PATCH 1/2] mesa, gallium: add a cap for GPUs without unified color+generic varying slots

On 12/13/2011 04:32 PM, Marek Olšák wrote: --- src/gallium/drivers/r300/r300_screen.c |3 ++- src/gallium/include/pipe/p_defines.h |3 ++- src/glsl/linker.cpp|6 ++ src/mesa/main/mtypes.h |9 + src/mesa/state_tracker/st_exten

Re: [Mesa-dev] [PATCH 1/8] mesa: Record transform feedback strides/offsets in linker output.

This patch adds two new fields to the gl_transform_feedback_info struct: - BufferStride records the total number of components (per vertex) that transform feedback is being instructed to store in each buffer. - Outputs[i].DstOffset records the offset within the interleaved structure of each t

Re: [Mesa-dev] [PATCH 1/8] mesa: Record transform feedback stride in linker output.

On 14 December 2011 05:48, Marek Olšák wrote: > The strides are useful, although they can be computed by summing up > all the NumComponents fields for each buffer. Note that I also take > into account ARB_transform_feedback3, which allows having interleaved > attribs in more than one buffer. For

Re: [Mesa-dev] [PATCH 1/8] mesa: Record transform feedback stride in linker output.

On 13 December 2011 15:53, Brian Paul wrote: > On Tue, Dec 13, 2011 at 4:35 PM, Paul Berry > wrote: > > This patch adds the field gl_transform_feedback_info::BufferStride, > > which records the total number of components (per vertex) that > > transform feedback is being instructed to store in ea

Re: [Mesa-dev] [PATCH 7/8] i965 gen6+: Make intel_batchbuffer_emit_mi_flush() actually flush.

On 14 December 2011 02:59, Kenneth Graunke wrote: > On 12/13/2011 03:35 PM, Paul Berry wrote: > > Previous to this patch, the function intel_batchbuffer_emit_mi_flush() > > was a bit of a misnomer. On Gen4+, when not using the blit engine, it > > didn't actually flush the pipeline--it simply gen

Re: [Mesa-dev] [PATCH 6/8] i965 gen6: Turn on transform feedback extension.

On 14 December 2011 02:51, Kenneth Graunke wrote: > On 12/13/2011 03:35 PM, Paul Berry wrote: > > This patch advertises support for EXT_transform_feedback on Intel Gen6 > > and higher. > > > > Since transform feedback support is not completely finished yet, for > > now we only advertise support f

Re: [Mesa-dev] [PATCH 3/8] i965 gen6+: Use 1-wide null operands for IF instructions

On 14 December 2011 02:33, Kenneth Graunke wrote: > On 12/13/2011 03:35 PM, Paul Berry wrote: > > The Sandy Bridge PRM, volume 4, part 2, section 5.3.10 ("5.3.10 > > Register Region Restrictions") contains the following restriction on > > the execution size and operand width of instructions: > >

Re: [Mesa-dev] [PATCH 8/8] i965: Flush pipeline on EndTransformFeedback.

A common use case for transform feedback is to perform one draw operation that writes transform feedback output to a buffer, followed by a second draw operation that consumes that buffer as vertex input. Since vertex input is consumed at an earlier pipeline stage than writing transform feedback out

Re: [Mesa-dev] [PATCH 8/8] i965: Flush pipeline on EndTransformFeedback.

On 14 December 2011 01:44, Kenneth Graunke wrote: > On 12/13/2011 03:35 PM, Paul Berry wrote: > > A common use case for transform feedback is to perform one draw > > operation that writes transform feedback output to a buffer, followed > > by a second draw operation that consumes that buffer as v

Re: [Mesa-dev] [PATCH 1/2] st/egl: Add support for EGL_NOK_swap_region

On Tuesday 13 December 2011, Chia-I Wu wrote: > On Fri, Dec 9, 2011 at 11:36 PM, Fredrik Höglund wrote: > > Backends indicate that they support this extension by returning > > EGL_TRUE when native_display::get_param() is called with > > NATIVE_PARAM_PRESENT_REGION. > > > > native_present_control i

Re: [Mesa-dev] [PATCH 1/8] mesa: Record transform feedback stride in linker output.

The strides are useful, although they can be computed by summing up all the NumComponents fields for each buffer. Note that I also take into account ARB_transform_feedback3, which allows having interleaved attribs in more than one buffer. For the EXT_transform_feedback case and separate attribs, Bu

Re: [Mesa-dev] shaders for g3dvl compositor (was vdpau: Handle destination rectangles correctly)

Hi Andy, On 12.12.2011 19:15, Andy Furniss wrote: Christian König wrote: @Andy: Could you please confirm that they are also working? (They should apply ontop of current master). Yes - they are working OK for me. Ok, so I committed them just a couple of minutes ago. One thing I noticed wh

Re: [Mesa-dev] [PATCH 7/8] i965 gen6+: Make intel_batchbuffer_emit_mi_flush() actually flush.

On 12/13/2011 03:35 PM, Paul Berry wrote: > Previous to this patch, the function intel_batchbuffer_emit_mi_flush() > was a bit of a misnomer. On Gen4+, when not using the blit engine, it > didn't actually flush the pipeline--it simply generated a > _3DSTATE_PIPE_CONTROL command with the necessary

Re: [Mesa-dev] [PATCH 6/8] i965 gen6: Turn on transform feedback extension.

On 12/13/2011 03:35 PM, Paul Berry wrote: > This patch advertises support for EXT_transform_feedback on Intel Gen6 > and higher. > > Since transform feedback support is not completely finished yet, for > now we only advertise support for it when MESA_GL_VERSION_OVERRIDE is > 3.0 or greater (since

Re: [Mesa-dev] [PATCH 4/8] i965 gs: Move vue_map to brw_gs_compile.

On 12/13/2011 03:35 PM, Paul Berry wrote: > This patch stores the geometry shader VUE map from a local variable in > compile_gs_prog() to a field in the brw_gs_compile struct, so that it > will be available while compiling the geometry shader. This is > necessary in order to support transform feed

Re: [Mesa-dev] [PATCH 3/8] i965 gen6+: Use 1-wide null operands for IF instructions

On 12/13/2011 03:35 PM, Paul Berry wrote: > The Sandy Bridge PRM, volume 4, part 2, section 5.3.10 ("5.3.10 > Register Region Restrictions") contains the following restriction on > the execution size and operand width of instructions: > >"3. ExecSize must be equal to or greater than Width." >

Re: [Mesa-dev] [PATCH 2/8] mesa: Fix off-by-one error in transform feedback size check.

On 12/13/2011 03:35 PM, Paul Berry wrote: > In _mesa_BindBufferRange(), we need to verify that the offset and size > specified by the client do not exceed the size of the underlying > buffer. We were accidentally doing this check using ">=" rather than > ">", so we were generating a bogus error if

Re: [Mesa-dev] [PATCH 1/8] mesa: Record transform feedback stride in linker output.

On 12/13/2011 03:35 PM, Paul Berry wrote: > This patch adds the field gl_transform_feedback_info::BufferStride, > which records the total number of components (per vertex) that > transform feedback is being instructed to store in each buffer. The > i965 gen6 back-end needs this information in orde

Re: [Mesa-dev] [PATCH 8/8] i965: Flush pipeline on EndTransformFeedback.

On 12/13/2011 03:35 PM, Paul Berry wrote: > A common use case for transform feedback is to perform one draw > operation that writes transform feedback output to a buffer, followed > by a second draw operation that consumes that buffer as vertex input. > Since vertex input is consumed at an earlier