Mauro Rossi [Fri, 27 Mar 2015 22:25:45 +0000 (22:25 +0000)]
android: add inital NIR build
Required by the i965 driver.
v2:
- Split out the nir_builder_opcodes.h rules.
- Do not unconditionally hide the python command - use $(hide)
- Use LOCAL_EXPORT_C_INCLUDE_DIRS to manage includes for the generated
sources.
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
[Emil Velikov: Split from a larger commit, v2]
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
Emil Velikov [Fri, 27 Mar 2015 21:39:15 +0000 (21:39 +0000)]
android: dri: link against libmesa_util
The dri modules depend on symbols provided by it.
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
Emil Velikov [Sat, 28 Mar 2015 01:11:54 +0000 (01:11 +0000)]
android: add $(mesa_top)/src/mesa/main to the includes list
Required by the format_{un,}pack rework. Otherwise the build will fail
to locate the respective headers - format_{un,}pack.h
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
Emil Velikov [Fri, 27 Mar 2015 20:10:35 +0000 (20:10 +0000)]
android: add HAVE__BUILTIN_* and HAVE_FUNC_ATTRIBUTE_* defines
All of those are available on gcc 4.5 and later with the current android
build using gcc 4.7.
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
Emil Velikov [Sat, 28 Mar 2015 01:10:52 +0000 (01:10 +0000)]
android: add gallium dirs to more places in the tree
Similar to
e8c5cbfd921(mesa: Add gallium include dirs to more parts of
the tree.)
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
Emil Velikov [Fri, 27 Mar 2015 18:57:46 +0000 (18:57 +0000)]
android: dri/common: conditionally include drm_cflags/set __NOT_HAVE_DRM_H
Otherwise we'll fail to find the drm.h header.
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Emil Velikov [Fri, 27 Mar 2015 18:36:10 +0000 (18:36 +0000)]
android: egl: add libsync_cflags to the build
... via local_shared_libraries. Otherwise the sync/sync.h header won't
be found.
Note: 10.5 and earlier will need similar change in st/egl.
v2: Append the library to the local_shared_libraries list. (Chih-Wei)
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
Mauro Rossi [Fri, 27 Mar 2015 18:20:53 +0000 (18:20 +0000)]
android: mesa: generate the format_{un,}pack.[ch] sources
Missed out with commit
e1fdcddafe9(mesa: Autogenerate format_unpack.c)
v2: Conditionaly print the python commands - s/@/$(hide) / (Chih-Wei)
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
[Emil Velikov: Split our from a larger commit.]
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Emil Velikov [Fri, 27 Mar 2015 16:13:50 +0000 (16:13 +0000)]
android: add $(mesa_top)/src include to the whole of mesa
Many parts of mesa already have the include with others depending on it
but it's missing. Add it once at the top makefile and be done with it.
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
Emil Velikov [Fri, 27 Mar 2015 14:18:24 +0000 (14:18 +0000)]
android: use := operator for assigning MESA_VERSION
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Chih-Wei Huang [Wed, 21 Jan 2015 17:21:00 +0000 (01:21 +0800)]
util: android: optimize the rules to generate format_srgb.c
Signed-off-by: Chih-Wei Huang <cwhuang@linux.org.tw>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Chih-Wei Huang [Wed, 21 Jan 2015 17:17:42 +0000 (01:17 +0800)]
android: simplify the subdirs including rules
Use the macro defined in the Android build system.
Signed-off-by: Chih-Wei Huang <cwhuang@linux.org.tw>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Emil Velikov [Fri, 27 Mar 2015 11:40:42 +0000 (11:40 +0000)]
android: use LOCAL_SHARED_LIBRARIES over TARGET_OUT_HEADERS
... to manage the LIBDRM*_CFLAGS. The former is the recommended approach
by the Android build system developers while the latter has been
depreciated for quite some time.
Cc: "10.4 10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Emil Velikov [Fri, 27 Mar 2015 11:13:21 +0000 (11:13 +0000)]
ilo: remove unused include from Android.mk
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Chih-Wei Huang <cwhuang@linux.org.tw>
Kenneth Graunke [Fri, 10 Apr 2015 17:24:33 +0000 (10:24 -0700)]
drirc: Add "Second Life" quirk (allow_glsl_extension_directive_midshader).
Appears to fix shader compilation. Tested by starting the client,
dragging the "quality and speed" slider back and forth, and watching the
console output - instead of piles of "shader failed to compile", the CPU
seems to be busy compiling shaders. I haven't actually tried to play.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=69226
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=71591
Cc: mesa-stable@lists.freedesktop.org
Kenneth Graunke [Fri, 17 Apr 2015 21:08:14 +0000 (14:08 -0700)]
nir: Fix per-component negation in prog_to_nir's SWZ handling.
I missed the fact that the ARB_fragment_program SWZ instruction allows
per-component negation. To fix this, move Abs/Negate handling into both
the simple case and the SWZ case's per-component loop.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90000
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Tapani Pälli [Tue, 21 Apr 2015 17:11:43 +0000 (20:11 +0300)]
glsl: correct indentation of comment, Trivial.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Matt Turner [Tue, 14 Apr 2015 20:17:38 +0000 (13:17 -0700)]
i965/fs: Combine pixel center calculation into one inst.
The X and Y values come interleaved in g1 (.4-.11 inclusive), so we can
calculate them together with a single add(32) instruction on some
platforms like Broadwell and newer or in SIMD8 elsewhere.
Note that I also moved the PIXEL_X/PIXEL_Y virtual opcodes from before
LINTERP to after it. That's because the writes_accumulator_implicitly()
function in backend_instruction tests for <= LINTERP for determining
whether the instruction indeed writes the accumulator implicitly. The
old FS_OPCODE_PIXEL_X/Y emitted ADD instructions, which did, but the new
opcodes just emit MOVs, which don't. It doesn't matter, since we don't
use these opcodes on Gen4/5 anymore, but in the case that we do...
On Broadwell:
total instructions in shared programs:
7192355 ->
7186224 (-0.09%)
instructions in affected programs:
1190700 ->
1184569 (-0.51%)
helped: 6131
On Haswell:
total instructions in shared programs:
6155979 ->
6152800 (-0.05%)
instructions in affected programs: 652362 -> 649183 (-0.49%)
helped: 3179
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Matt Turner [Tue, 7 Apr 2015 00:44:40 +0000 (17:44 -0700)]
i965/fs: Calculate delta_x and delta_y together.
This lets SIMD16 programs on G45 and Gen5 use the PLN instruction.
On Ironlake:
total instructions in shared programs:
5634757 ->
5518055 (-2.07%)
instructions in affected programs:
1745837 ->
1629135 (-6.68%)
helped: 11439
HURT: 4
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Matt Turner [Thu, 2 Apr 2015 23:15:53 +0000 (16:15 -0700)]
i965/fs: Emit ADDs for gl_FragCoord, not virtual opcodes.
These were used only on Gen4 and 5. emit_interpolation_setup_gen6() emits
ADDs directly. The virtual opcodes weren't providing anything useful.
I'm going to repurpose these opcodes, so deleting and readding them makes
it simpler to see what's going on.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Matt Turner [Tue, 7 Apr 2015 04:46:54 +0000 (21:46 -0700)]
i965/fs: Manually set source regioning on PLN instructions.
Like LINE (commit
92346db0), src0 must have a scalar region. Setting
src1's region to <8,8,1> lets us pass a properly sized combined delta_xy
argument in a few commits without getting a bogus <16,16,1> region.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Matt Turner [Sat, 11 Apr 2015 22:00:19 +0000 (15:00 -0700)]
i965/fs: Add LINTERP's src0 to fs_inst::regs_read().
LINTERP's src0 is PLN's src1, and PLN's src1 reads exec_size / 4
registers.
Having that information lets us drop the delta_x/y special case code in
split_virtual_grfs().
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Matt Turner [Sat, 11 Apr 2015 21:51:13 +0000 (14:51 -0700)]
i965/fs: Set compression only if writing two registers.
We don't want to set compression control on a SIMD16 instruction
operating on words or smaller.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Matt Turner [Sat, 4 Apr 2015 00:44:06 +0000 (17:44 -0700)]
i965/fs: Allow an execution size of 32.
In a few commits, we'll start emitting an add(32) instruction on some
platforms.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Matt Turner [Sat, 11 Apr 2015 21:49:50 +0000 (14:49 -0700)]
i965: Make type_sz() return unsigned.
Avoids annoying warnings when comparing with sizeof(...).
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Matt Turner [Tue, 14 Apr 2015 19:40:34 +0000 (12:40 -0700)]
i965: Replace guess_execution_size with something simpler.
guess_execution_size() does two things:
1. Cope with small destination registers.
2. Cope with SIMD8 vs SIMD16 mode.
This patch replaces the first with a simple if block in brw_set_dest: if
the destination register width is less than 8, you probably want the
execution size to match. (I didn't put this in the 3src block because
it doesn't seem to matter.)
Since only the FS compiler cares about SIMD16 mode, it's easy to just
set the default execution size there.
This pattern was already been proven in the Gen8+ generator, but we
didn't port it back to the existing generator when we combined the two.
This is based on a patch from Ken from about a year ago. I've rebased it
and and fixed a few bugs.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Matt Turner [Thu, 2 Apr 2015 23:57:10 +0000 (16:57 -0700)]
i965/fs: Ensure delta_x/y are even-aligned registers on Gen6.
The BSpec says this applies to Gen6 as well.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Marius Predut [Thu, 16 Apr 2015 10:36:07 +0000 (13:36 +0300)]
main: remove __FUNCTION__ defined because it is obsolete
Consistently just use C99's __func__ everywhere.
No functional changes.
Signed-off-by: Marius Predut <marius.predut@intel.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Marius Predut [Thu, 16 Apr 2015 10:35:50 +0000 (13:35 +0300)]
radeon: replace __FUNCTION__ with __func__
Consistently just use C99's __func__ everywhere.
No functional changes.
Signed-off-by: Marius Predut <marius.predut@intel.com>
Acked-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Tapani Pälli [Tue, 21 Apr 2015 05:26:03 +0000 (08:26 +0300)]
mesa: add missing break in switch statement
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-By: Martin Peres <martin.peres@linux.intel.com>
Tapani Pälli [Tue, 21 Apr 2015 05:25:16 +0000 (08:25 +0300)]
glsl: add fallthrough comment on switch
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-By: Martin Peres <martin.peres@linux.intel.com>
Tapani Pälli [Mon, 20 Apr 2015 12:41:06 +0000 (15:41 +0300)]
mesa: fix UBO queries for active uniforms
Commit
34df5eb introduced regression to GetActiveUniformBlockiv
when querying one of the following properties:
GL_UNIFORM_BLOCK_ACTIVE_UNIFORMS
GL_UNIFORM_BLOCK_ACTIVE_UNIFORM_INDICES
Implementation counted all uniforms in ubo directly while query should
check first if the uniform in question is _active_.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90109
Reviewed-By: Martin Peres <martin.peres@linux.intel.com>
Neil Roberts [Wed, 18 Feb 2015 18:41:58 +0000 (18:41 +0000)]
i965/skl: Fix the qpitch value
On Skylake the qpitch value is uploaded as part of the surface state
so we don't need to add the extra rows that are done for other
generations. However for 3D textures it needs to be aligned to the
tile height and for depth/stencil textures it needs to be a multiple
of 8. Unlike previous generations the qpitch is measured as a multiple
of the block size for compressed surfaces. When the horizontal mipmap
layout is used for 1D textures then the qpitch is measured in pixels
instead of rows.
v2: Align the depth/stencil textures to a multiple of 8
v3: Add an assert that ALL_SLICES_AT_EACH_LOD is not used. Ignore the
vertical alignment when picking the qpitch for 1D_ARRAY textures.
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Neil Roberts [Fri, 20 Feb 2015 19:11:46 +0000 (19:11 +0000)]
i965/skl: Don't use ALL_SLICES_AT_EACH_LOD
The render surface state command for Skylake doesn't have the surface
array spacing bit so it's not possible to select this layout. I think
it was only used in order to make it pick a tightly-packed qpitch
value that doesn't include space for the mipmaps. However this won't
be necessary after the next patch because it will automatically pick a
packed qpitch value whenever first_level==last_level. It is better to
remove this layout entirely on Gen8+ because although it can
effectively be implemented with a small qpitch value when there are no
mipmaps it isn't possible to support the case where there are mipmaps
because in that case the layout is very different.
It could be good to make a similar change for Gen8 if we also change
the layouting code to pick the qpitch value in a similar way.
v2: Make the commit message and comments more convincing
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Tested-by: Ben Widawsky <ben@bwidawsk.net>
EdB [Sun, 19 Apr 2015 08:50:19 +0000 (10:50 +0200)]
clover: remove pre llvm 3.5.0 compatibility code
Acked-by: Francisco Jerez <currojerez@riseup.net>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
EdB [Mon, 20 Apr 2015 18:28:09 +0000 (20:28 +0200)]
clover: make llvm >= 3.5.0 and c++11 mandatory
Clover not longer compile with llvm <= 3.5.0 since
e1d363b3.
e1d363b3 implies c++11 and llvm 3.5.0 CXXFLAGS provided it.
No one seems to have noticed it, it's now official.
Acked-by: Francisco Jerez <currojerez@riseup.net>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Dave Airlie [Mon, 20 Apr 2015 08:41:52 +0000 (18:41 +1000)]
docs/GL3.txt: update ARB_shader_subroutine status
Admit to having started working on this, I don't admit to ever finishing it
Signed-off-by: Dave Airlie <airlied@redhat.com>
Nick Sarnie [Mon, 20 Apr 2015 03:51:26 +0000 (23:51 -0400)]
gallivm: Fix build against LLVM 3.7 SVN r235265
LLVM removed JITEmitDebugInfo from TargetOptions since they weren't used
v2: Be consistent with the LLVM version check (Aaron Watry)
Signed-off-by: Nick Sarnie <commendsarnex@gmail.com>
Reviewed-and-Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Ian Romanick [Thu, 16 Apr 2015 19:17:19 +0000 (12:17 -0700)]
doc: Add GL_ARB_shader_image_size dependency for OpenGL ES 3.1
imageSize() is in the GLSL ES 3.1 spec. Trivial.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Ilia Mirkin [Tue, 7 Apr 2015 05:42:05 +0000 (01:42 -0400)]
indices: fix provoking vertex for quads/quadstrips
This allows drivers to provide consistent flat shading for quads.
Otherwise a driver that only supported tris would have to force last
provoking vertex when drawing quads (and would have to say that quads
don't follow the provoking vertex convention).
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Ilia Mirkin [Tue, 7 Apr 2015 15:43:52 +0000 (11:43 -0400)]
primconvert: select pv convention only from flatshade_first
This should match to how drivers program hardware. flatshade relates to
whether color inputs are interpolated, not the provoking vertex
convention.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Ilia Mirkin [Tue, 7 Apr 2015 04:57:34 +0000 (00:57 -0400)]
freedreno/a3xx: enable polymode setting with non-fill modes
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Ilia Mirkin [Mon, 6 Apr 2015 05:15:09 +0000 (01:15 -0400)]
freedreno/a3xx: fix integer and 32-bit float border colors
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Ilia Mirkin [Mon, 6 Apr 2015 01:38:57 +0000 (21:38 -0400)]
freedreno/a3xx: add support for float R/RG render targets
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Connor Abbott [Fri, 17 Apr 2015 16:53:53 +0000 (12:53 -0400)]
mesa: add .mesa-install-links files to gitignore
Signed-off-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Connor Abbott [Fri, 17 Apr 2015 16:50:30 +0000 (12:50 -0400)]
mesa/main: add autogenerated format-info.c to gitignore
v2: move to right after format-info.h
Signed-off-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Kenneth Graunke [Wed, 25 Feb 2015 05:34:30 +0000 (21:34 -0800)]
i965: Issue perf_debug messages for unsynchronized maps on !LLC systems.
We haven't implemented proper unsynchronized map support on !LLC systems
(pre-SNB, Atom). MapBufferRange with GL_MAP_UNSYNCHRONIZE_BIT will
actually do a synchronized map, probably killing performance.
Also warn on BufferSubData, when we should be doing an unsynchronized
upload, but instead have to do a synchronous map.
v2: Only complain if the buffer is actually busy - we use unsynchronized
maps internally for vertex upload and such, but expect those to not
be busy.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Tested-by: Ben Widawsky <ben@bwidawsk.net>
Kenneth Graunke [Wed, 15 Apr 2015 09:16:47 +0000 (02:16 -0700)]
i965: Make shader_time store names/ids instead of referencing shaders.
Jason noticed that shader_time was bumping the reference count on the
gl_shader_program and gl_program structures, in code called during
compilation.
Not only were these never unreferenced, but it meant fragment shaders
might be referenced twice (SIMD8 and SIMD16)...or only once.
We don't actually need the programs. We just need their numeric ID and
their language (GLSL/ARB/FF) or KHR_debug label. If there's a label, we
have to strdup it since the underlying program could be deleted.
To be fair, we're not exactly cleaning that up either, but we at least
ralloc it out of the shader_time arrays, so if we ever bother cleaning
those up, they'll go away properly.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Acked-by: Anuj Phogat <anuj.phogat@gmail.com>
Kenneth Graunke [Wed, 15 Apr 2015 09:04:17 +0000 (02:04 -0700)]
i965: Delete some unnecessary code in brw_report_shader_time().
It is true that a gl_shader_program with ID 0 will be a fixed-function
fragment program; a gl_program with ID 0 but NULL gl_shader_program
means that it's a fixed-function vertex shader.
But that's not terribly interesting or relevant to what we're doing.
We just need to know that ID 0 means "fixed function".
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Kenneth Graunke [Wed, 15 Apr 2015 08:57:52 +0000 (01:57 -0700)]
i965: Make shader_time use 0 instead of -1 for "no meaningful ID".
0 is not a valid GLSL shader or ARB program ID. For some reason,
shader_time used -1 instead...so we had code to detect 0, then override
it to -1.
We can just delete that.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Tobias Nygren [Fri, 17 Apr 2015 18:27:55 +0000 (20:27 +0200)]
adjust a couple of ifdefs to handle NetBSD correctly
Acked-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Tobias Nygren <tnn@NetBSD.org>
Tobias Nygren [Fri, 17 Apr 2015 18:18:48 +0000 (20:18 +0200)]
configure.ac: fix bashism
Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Tobias Nygren <tnn@NetBSD.org>
Anuj Phogat [Fri, 10 Apr 2015 11:35:24 +0000 (04:35 -0700)]
i965: Render R16G16B16X16 as R16G16B16A16
This enables using _mesa_meta_pbo_TexSubImage() to upload data
to R16G16B16X16 texture. Earlier it fell back to slower paths.
Jenkins run shows no piglit regressions.
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Anuj Phogat [Mon, 13 Apr 2015 17:20:29 +0000 (10:20 -0700)]
i965: Update the comment about platforms supporting blorp
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Chad Versace <chad.versace@intel.com>
Matt Turner [Thu, 2 Apr 2015 17:21:16 +0000 (10:21 -0700)]
nir: Allow abs/neg in select peephole pass.
total instructions in shared programs:
4314531 ->
4308949 (-0.13%)
instructions in affected programs: 429085 -> 423503 (-1.30%)
helped: 1680
HURT: 0
GAINED: 0
LOST: 111
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Jason Ekstrand [Fri, 3 Apr 2015 18:07:47 +0000 (11:07 -0700)]
i965/fs: Change SEL and MOV types as needed to propagate source modifiers
SEL and MOV instructions, as long as they don't have source modifiers, are
just copying bits around. This commit adds support to copy propagation to
switch the type of a SEL or MOV instruction as needed so that it can
propagate source modifiers. This is needed because NIR generates integer
SEL and MOV instructions whenver it doesn't know what else to generate.
shader-db results with NIR:
total FS instructions in shared programs:
4360910 ->
4360186 (-0.02%)
FS instructions in affected programs: 59094 -> 58370 (-1.23%)
helped: 341
HURT: 0
GAINED: 2
LOST: 0
Signed-off-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Jason Ekstrand [Fri, 3 Apr 2015 19:15:48 +0000 (12:15 -0700)]
i965/fs: Use the source type when looking for UD negations in copy prop
There can be problems with floats and conditional modifiers when
copy-propagating a negated UD source. The problem arises when a source
modifier is applied to a UD value. In this case, a 33-bit representation
is internally used. If you do the following:
1: mov foo:UD 7U
2: mov bar:UD -foo:UD
3: mov out:F bar:UD
the out register will have the value (float)(unt32_t)-7 which is some very
large floating-point number. However, if we allow copy-propagation of the
second mov, we get
1: mov foo:UD 7U
3: mov out:f -bar:UD
and, since the negation is computed in 33-bits, we get a value of -7.0f
which is clearly not the same. This is a similar problem if the
instruction has a conditional modifier where the 33-bit value is used in
the comparison and not the 32-bit version.
Previously, we checked the source to be copied for the negate and then
checked the source being propagated to for the type. This isn't quite what
we want because we are really just looking for negated UD sources. A check
later in the file ensures that both ends of the propagate have the right
type so it works. However, if we relax the restriction that both ends of
the propagation have the same type, it ends up causing us to bail early in
cases we don't want.
Signed-off-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Rob Clark [Thu, 16 Apr 2015 19:35:50 +0000 (15:35 -0400)]
freedreno/ir3/nir: few little fixes
isaml needs to scale up coords based on LoD. Also fix bogus bary.f
varying # when there are non-bary frag shader inputs. And use sub.s of
a positive immediate rather than add.s of negative (since CP is better
about figuring out that those can be collapsed into the cat2 instr).
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Wed, 15 Apr 2015 14:16:31 +0000 (10:16 -0400)]
freedreno/ir3/nir: lower if/else
For now, completely flatten if/else blocks. That will almost certainly
change once we have flow control.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Sun, 12 Apr 2015 16:58:52 +0000 (12:58 -0400)]
freedreno/a4xx: support for large shaders
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Sun, 12 Apr 2015 18:25:29 +0000 (14:25 -0400)]
freedreno: update generated headers
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Sun, 12 Apr 2015 13:47:40 +0000 (09:47 -0400)]
freedreno/ir3/nir: UBO support
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Sun, 12 Apr 2015 13:46:34 +0000 (09:46 -0400)]
freedreno/ir3: move out helper
We'll also want it in NIR f/e for implementing UBO support.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Sat, 11 Apr 2015 16:15:17 +0000 (12:15 -0400)]
freedreno/a4xx: sysvals and UBOs
Basically just sync up the cmdstream emit parts to match the changes
already done on a3xx.
Also, fix scheduling for mem instructions. This is needed on a4xx, and
I am a bit surprised it isn't needed for a3xx.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Wed, 15 Apr 2015 19:49:15 +0000 (15:49 -0400)]
nir/builder: add nir_builder_insert_after_instr()
For lowering if/else, I need a way to insert at the end of the previous
block.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Rob Clark [Thu, 16 Apr 2015 19:16:12 +0000 (15:16 -0400)]
gallium/ttn: fix TXF
There is a level param stashed away in the .w component of the first
src.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Rob Clark [Sun, 12 Apr 2015 12:15:33 +0000 (08:15 -0400)]
gallium/ttn: add UBO support
v2: move ishl into ttn (instead of driver backend) to keep the units
consistent between immediate and indirect offsets
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Rob Clark [Sat, 11 Apr 2015 19:25:58 +0000 (15:25 -0400)]
gallium/ttn: minor cleanup
v2: also use ttn_src_for_indirect() everywhere for addr access, rather
than open-coding it for INPUT/CONST srcs
v3: move ralloc out of ttn_src_for_indirect() into the one call site
that needs a ptr
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Rob Clark [Sat, 11 Apr 2015 15:37:12 +0000 (11:37 -0400)]
gallium/ttn: add support for TXL2
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Rob Clark [Sat, 11 Apr 2015 15:35:29 +0000 (11:35 -0400)]
gallium/ttn: add support for texture offsets
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Eric Anholt <eric@anholt.net>
Brian Paul [Fri, 17 Apr 2015 14:14:23 +0000 (15:14 +0100)]
mesa/st: Free st_translate with FREE macro.
To match CALLOC_STRUCT macro.
Fixes memory corruption on Windows when u_memory's memory debugging is
enabled.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Jose Fonseca [Thu, 16 Apr 2015 12:08:56 +0000 (13:08 +0100)]
libgl-gdi: Prevent "pure virtual method called" error when.
When running piglit w/ llvmpipe on Windows several tests terminate
abnormally just when the test exits.
The problem was that LLVMContextDispose was being called
after LLVM global destructors.
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Ville Syrjälä [Thu, 16 Apr 2015 16:21:07 +0000 (19:21 +0300)]
i965: Add marketing names for CHV
All CHV devices will be branded as "Intel(r) HD Graphics".
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Ian Romanick [Wed, 15 Apr 2015 22:20:57 +0000 (15:20 -0700)]
nir: Convert the if-test for num_inputs == 2 to an assertion
Suggested by Jason on a different patch after some comments /
questions by Ilia.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Connor Abbott <cwabott0@gmail.com>
Marek Olšák [Mon, 13 Apr 2015 19:44:02 +0000 (21:44 +0200)]
configure.ac: print LLVM_LDFLAGS
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Emil Velikov <emil.l.velilkov@gmail.com>
Marek Olšák [Sat, 11 Apr 2015 18:05:41 +0000 (20:05 +0200)]
glsl_to_tgsi: only associate the uniform storage once at link time
This hack is no longer needed. (see the previous commit)
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sat, 11 Apr 2015 18:01:22 +0000 (20:01 +0200)]
glsl_to_tgsi: add STATE_FB_WPOS_Y_TRANSFORM at link time
This will allow removing the uniform storage re-association during
TGSI generation at draw time.
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sat, 11 Apr 2015 12:40:09 +0000 (14:40 +0200)]
glsl_to_tgsi: add assertions for detecting out-of-bounds immediates access
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sat, 11 Apr 2015 12:55:26 +0000 (14:55 +0200)]
glsl_to_tgsi: don't use a potentially-undefined immediate for ir_query_levels
Cc: 10.4 10.5 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sat, 11 Apr 2015 11:49:38 +0000 (13:49 +0200)]
glsl_to_tgsi: fix out-of-bounds constant access and crash for uniforms
This fixes piglit shaders@glsl-fs-uniform-array-loop-unroll with immediate
shader compilation - it's a compiler test, so it has never been translated
to TGSI before.
Cc: 10.4 10.5 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Wed, 18 Mar 2015 11:38:19 +0000 (12:38 +0100)]
glsl_to_tgsi: cleanup includes
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sat, 11 Apr 2015 17:35:40 +0000 (19:35 +0200)]
mesa/program: remove dead code
Reviewed-by: Matt Turner <mattst88@gmail.com>
Marek Olšák [Fri, 10 Apr 2015 21:58:34 +0000 (23:58 +0200)]
radeonsi: add a debug option to compile shaders when they're created
Tested-by: Tom Stellard <thomas.stellard@amd.com>
Marek Olšák [Fri, 10 Apr 2015 21:58:34 +0000 (23:58 +0200)]
st/mesa: add a debug option to compile shaders at link time
v2: fix crashes
Tested-by: Tom Stellard <thomas.stellard@amd.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Kristian Høgsberg [Tue, 14 Apr 2015 15:02:18 +0000 (15:02 +0000)]
i965: Rewrite ir_tex to ir_txl with lod 0 for vertex shaders
The ir_tex opcode turns into a sample or sample_c message, which will try to
compute derivatives to determine the lod. This produces garbage for
non-fragment shaders where the sample coordinates don't correspond to
subspans.
We fix this by rewriting the opcode from ir_tex to ir_txl and setting the
lod to 0.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89457
Cc: "10.5" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Kristian Høgsberg <kristian.h.kristensen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Emil Velikov [Wed, 8 Apr 2015 18:27:02 +0000 (19:27 +0100)]
radeonsi: remove bogus r600-- triple
As mentioned by Michel Dänzer for LLVM >= 3.6 we create the
LLVMTargetMachine (with triple amdgcn--), as we setup the radeonsi
context. For older LLVM or hardware (r600) the triple is always r600--
and is created at a later stage - radeon_llvm_compile()
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Neil Roberts [Tue, 24 Mar 2015 15:52:20 +0000 (15:52 +0000)]
i965/skl: Add the header for constant loads outside of the generator
Commit
5a06ee738 added a step to the generator to set up the message
header when generating the VS_OPCODE_PULL_CONSTANT_LOAD_GEN7
instruction. That pseudo opcode is implemented in terms of multiple
actual opcodes, one of which writes to one of the source registers in
order to set up the message header. This causes problems because the
scheduler isn't aware that the source register is written to and it
can end up reorganising the instructions incorrectly such that the
write to the source register overwrites a needed value from a previous
instruction. This problem was presenting itself as a rendering error
in the weapon in Enemy Territory: Quake Wars.
Since commit
588859e1 there is an additional problem that the double
register allocated to include the message header would end up being
split into two. This wasn't happening previously because the code to
split registers was explicitly avoided for instructions that are
sending from the GRF.
This patch fixes both problems by splitting the code to set up the
message header into a new pseudo opcode so that it will be done
outside of the generator. This new opcode has the header register as a
destination so the scheduler can recognise that the register is
written to. This has the additional benefit that the scheduler can
optimise the message header slightly better by moving the mov
instructions further away from the send instructions.
On Skylake it appears to fix the following three Piglit tests without
causing any regressions:
gs-float-array-variable-index
gs-mat3x4-row-major
gs-mat4x3-row-major
I think we actually may need to do something similar for the fs
backend and possibly for message headers from regular texture sampling
but I'm not entirely sure.
v2: Make sure the exec-size is retained as 8 for the mov instruction
to initialise the header from g0. This was accidentally lost
during a rebase on top of
07c571a39fa1.
Split the patch into two so that the helper function is a separate
change.
Fix emitting the MOV instruction on Gen7.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89058
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Neil Roberts [Wed, 15 Apr 2015 13:28:26 +0000 (14:28 +0100)]
i965/vec4: Add a helper function to emit VS_OPCODE_PULL_CONSTANT_LOAD
There were three places in the visitor that had a similar chunk of
code to emit the VS_OPCODE_PULL_CONSTANT_LOAD opcode using a register
for the offset. This patch combines the chunks into a helper function
to reduce the code duplication. It will also be useful in the next
patch to expand what happens on Gen9+. This shouldn't introduce any
functional changes.
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Jose Fonseca [Thu, 16 Apr 2015 09:19:57 +0000 (10:19 +0100)]
mesa,glsl: rename `interface` to `programInterface`.
`interface` is a define on Windows -- an alias for `struct` keyword,
used when declaring COM interfaces in C or C++.
So use instead `programInterface`, therefore matching the name used
in GL_ARB_program_interface_query spec/headers, which was renamed exactly
for the same reason:
"Revision 10, May 10, 2012 (pbrown)
- Rename the formal parameter <interface> used by the functions in this
extension to <programInterface>. Certain versions of the Microsoft
C/C++ compiler and/or its headers cause "interface" to be treated as a
reserved keyword."
Trivial.
Flora Cui [Tue, 24 Mar 2015 10:34:29 +0000 (18:34 +0800)]
gbm: Add GBM_BO_USE_LINEAR flag
Signed-off-by: Flora Cui <flora.cui@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Tapani Pälli [Fri, 13 Mar 2015 06:53:23 +0000 (08:53 +0200)]
mesa: refactor GetUniformBlockIndex
Use _mesa_program_resource_index to get index.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Martin Peres <martin.peres@linux.intel.com>
Tapani Pälli [Thu, 12 Mar 2015 12:37:20 +0000 (14:37 +0200)]
mesa: refactor GetUniformIndices
Use _mesa_program_resource_index to get indices.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Martin Peres <martin.peres@linux.intel.com>
Tapani Pälli [Thu, 12 Mar 2015 12:31:15 +0000 (14:31 +0200)]
mesa: refactor GetUniformLocation
Use _mesa_program_resource_location to get location.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Martin Peres <martin.peres@linux.intel.com>
Tapani Pälli [Thu, 12 Mar 2015 11:33:25 +0000 (13:33 +0200)]
mesa: refactor GetActiveUniformBlockName
Use _mesa_get_program_resource_name to get name.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Martin Peres <martin.peres@linux.intel.com>
Tapani Pälli [Thu, 12 Mar 2015 11:53:07 +0000 (13:53 +0200)]
mesa: remove unused _mesa_get_uniform_name
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Martin Peres <martin.peres@linux.intel.com>
Tapani Pälli [Thu, 12 Mar 2015 11:22:16 +0000 (13:22 +0200)]
mesa: refactor GetActiveUniformName
Use _mesa_get_program_resource_name to get name.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Martin Peres <martin.peres@linux.intel.com>
Tapani Pälli [Thu, 12 Mar 2015 10:55:18 +0000 (12:55 +0200)]
mesa: refactor GetActiveUniform
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Martin Peres <martin.peres@linux.intel.com>
Tapani Pälli [Thu, 12 Mar 2015 10:08:56 +0000 (12:08 +0200)]
mesa: refactor GetTransformFeedbackVarying
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Martin Peres <martin.peres@linux.intel.com>
Tapani Pälli [Thu, 12 Mar 2015 08:17:09 +0000 (10:17 +0200)]
mesa: refactor GetActiveUniformsiv, use _mesa_program_resource_prop
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Tapani Pälli [Thu, 12 Mar 2015 05:58:48 +0000 (07:58 +0200)]
mesa: mesa_bufferiv utility function for buffer objects
Patch adds new function 'mesa_bufferiv' and refactors existing
GetActiveUniformBlockiv and GetActiveAtomicCounterBufferiv to
use it.
corresponding Piglit tests:
arb_uniform_buffer_object*
arb_shader_atomic_counters*
(Many tests hit the corresponding queries.)
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Martin Peres <martin.peres@linux.intel.com>