Jason Ekstrand [Tue, 14 Aug 2018 19:37:39 +0000 (14:37 -0500)]
nir/algebraic: Add more extract_[iu](8|16) optimizations
This adds the "(a << N) >> M" family of mask or sign-extensions. Not a
huge win right now but this pattern will soon be generated by NIR format
lowering code.
Shader-db results on Kaby Lake:
total instructions in shared programs:
15166918 ->
15166916 (<.01%)
instructions in affected programs: 36 -> 34 (-5.56%)
helped: 2
HURT: 0
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jason Ekstrand [Tue, 14 Aug 2018 20:10:22 +0000 (15:10 -0500)]
nir/algebraic: Be more careful converting ushr to extract_u8/16
If it's not the right bit-size, it may not actually be the correct
extraction. For now, we'll only worry about 32-bit versions.
Fixes:
905ff8619824 "nir: Recognize open-coded extract_u16"
Fixes:
76289fbfa84a "nir: Recognize open-coded extract_u8"
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Sagar Ghuge [Wed, 29 Aug 2018 17:52:41 +0000 (10:52 -0700)]
intel/tools: new i965_disasm tool
Adds a new i965 instruction disassemble tool
v2: 1) fix a few nits (Matt Turner)
2) Remove i965_disasm header (Matt Turner)
v3: 1) Redirect output to correct file descriptors (Matt Turner)
2) Refactor code (Matt Turner)
3) Use better formatting style (Matt Turner)
Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Kenneth Graunke [Sat, 25 Aug 2018 06:45:27 +0000 (23:45 -0700)]
st/mesa: Disable blending for integer formats.
Blending isn't valid for integer formats. Rather than having drivers
worry about this, just disable blending in this case. This hopefully
will increase hits in the CSO cache as well, by eliminating most of the
meaningless fields in this case.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Brian Paul [Fri, 24 Aug 2018 02:48:04 +0000 (20:48 -0600)]
svga: add missing switch cases for shadow textures
This doesn't seem to make any difference in testing, but it fixes a
failed assertion when dumping sm3 shaders.
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Brian Paul [Thu, 23 Aug 2018 15:25:15 +0000 (09:25 -0600)]
svga: fix vgpu9 sprite coordinate bug
Setting GL_POINT_SPRITE_COORD_ORIGIN to GL_LOWER_LEFT did not work for
vgpu9. We can use the rasterizer sprite_coord_enable bitfield as-is.
We need to index into it using the TGSI semantic index, not the
register index.
This fixes the Piglit fbo-gl_pointcoord and glsl-fs-pointcoord tests.
Testing done: Piglit, Mesa sprite demos
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Brian Paul [Wed, 22 Aug 2018 17:11:22 +0000 (11:11 -0600)]
svga: fix PIPE_TEXTURE_RECT/BUFFER const buffer issue
The flag_rect and flag_buffer fields didn't sufficiently capture
the state changes needed for those resource types. For example,
if a texture binding was changed from a 500x500 rect texture to a
400x400 rect texture we didn't set SVGA_NEW_TEXTURE_CONSTS. But
we need to do that to emit the new texcoord scale factors to the
constant buffers. Rather than track the sizes of all bound
resources, just set the flag if the resource is a rect. Same
story with texture buffers.
Also, since rect/buffer textures are usable with VS/GS shaders,
add SVGA_NEW_TEXTURE_CONSTS to the flags we check for emitting
VS/GS constants.
This seems to help with XFCE / xfwm4 desktop scaling.
VMware issue 2156696.
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Brian Paul [Fri, 24 Aug 2018 21:17:20 +0000 (15:17 -0600)]
svga: minor improvements in svga_state_constants.c
Add const qualifiers. Add 'f' suffix on floats to avoid double
promotion.
Remove unneeded shader type assertion since the switch statement
handled it already.
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Jason Ekstrand [Wed, 29 Aug 2018 15:06:56 +0000 (10:06 -0500)]
anv: Free the app and engine name
Fixes:
8c048af5890d4 "anv: Copy the appliation info into the instance"
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Rhys Kidd [Fri, 10 Aug 2018 16:44:37 +0000 (12:44 -0400)]
nv50/ir: silence partitionLoadStore() unused function warning
Move this now-unused function into the existing comment block, which was its only prior use.
../../../../../src/gallium/drivers/nouveau/codegen/nv50_ir_from_tgsi.cpp:2645:1: warning:
unused function 'partitionLoadStore' [-Wunused-function]
partitionLoadStore(uint8_t comp[2], uint8_t size[2], uint8_t mask)
Fixes: ("
86e4440361 nouveau: codegen: Disable more old resource handling code")
Signed-off-by: Rhys Kidd <rhyskidd@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
vadym.shovkoplias [Tue, 28 Aug 2018 07:32:18 +0000 (10:32 +0300)]
glsl/linker: Link all out vars from a shader objects on a single stage
During intra stage linking some out variables can be dropped because
it is not used in a shader with the main function. But these out vars
can be referenced on later stages which can lead to further linking
errors.
Signed-off-by: Vadym Shovkoplias <vadym.shovkoplias@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105731
Lionel Landwerlin [Tue, 28 Aug 2018 10:16:33 +0000 (11:16 +0100)]
anv: blorp: support multiple aspect blits
Newer blit tests are enabling depth&stencils blits. We currently don't
support it but can do by iterating over the aspects masks (copy some
logic from the CopyImage function).
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes:
9f44745eca0e41 ("anv: Use blorp to implement VkBlitImage")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tapani Pälli [Mon, 27 Aug 2018 11:40:41 +0000 (14:40 +0300)]
mesa: allow GL_UNSIGNED_BYTE type for SNORM reads
OpenGL ES spec states:
"For normalized fixed-point rendering surfaces, the combination format
RGBA and type UNSIGNED_BYTE is accepted."
This fixes following failing VK-GL-CTS tests:
KHR-GLES3.packed_pixels.pbo_rectangle.rgba8_snorm
KHR-GLES3.packed_pixels.rectangle.rgba8_snorm
KHR-GLES3.packed_pixels.varied_rectangle.rgba8_snorm
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
https://bugs.freedesktop.org/show_bug.cgi?id=107658
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Tested-by: Andres Gomez <agomez@igalia.com>
Timothy Arceri [Sat, 7 Jul 2018 07:56:26 +0000 (17:56 +1000)]
nir: add loop unroll support for wrapper loops
This adds support for unrolling the classic
do {
// ...
} while (false)
that is used to wrap multi-line macros. GLSL IR also wraps switch
statements in a loop like this.
shader-db results IVB:
total loops in shared programs: 2515 -> 2512 (-0.12%)
loops in affected programs: 33 -> 30 (-9.09%)
helped: 3
HURT: 0
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Timothy Arceri [Wed, 11 Jul 2018 00:50:16 +0000 (10:50 +1000)]
nir/opt_loop_unroll: Remove unneeded phis if we make progress
Now that SSA values can be derefs and they have special rules, we have
to be a bit more careful about our LCSSA phis. In particular, we need
to clean up in case LCSSA ended up creating a phi node for a deref.
This avoids validation issues with some CTS tests with the following
patch, but its possible this we could also see the same problem with
the existing unrolling passes.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Timothy Arceri [Sat, 7 Jul 2018 02:09:26 +0000 (12:09 +1000)]
nir: add complex_loop bool to loop info
In order to be sure loop_terminator_list is an accurate
representation of all the jumps in the loop we need to be sure we
didn't encounter any other complex behaviour such as continues,
nested breaks, etc during analysis.
This will be used in the following patch.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Timothy Arceri [Sat, 7 Jul 2018 02:02:08 +0000 (12:02 +1000)]
nir: always attempt to find loop terminators
This will help later patches with unrolling loops that end with a
break i.e. loops the always exit on their first interation.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Marek Olšák [Tue, 28 Aug 2018 18:39:09 +0000 (14:39 -0400)]
ac/surface: fix CMASK fast clear for NPOT textures with mipmapping on SI/CI/VI
This fixes VM faults and corruption.
Cc: 18.1 18.2 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Ian Romanick [Sat, 25 Aug 2018 00:24:36 +0000 (17:24 -0700)]
i965/vec4: Emit BRW_AOP_INC or BRW_AOP_DEC for atomicAdd of +1 or -1
No shader-db changes on any Intel platform.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Ian Romanick [Sat, 25 Aug 2018 00:41:01 +0000 (17:41 -0700)]
i965/fs: Emit BRW_AOP_INC or BRW_AOP_DEC for imageAtomicAdd of +1 or -1
v2: Refactor selection of atomic opcode to a separate function.
Suggested by Jason.
No changes on any other Intel platforms.
Skylake
total instructions in shared programs:
14304261 ->
14304241 (<.01%)
instructions in affected programs: 1625 -> 1605 (-1.23%)
helped: 4
HURT: 0
helped stats (abs) min: 1 max: 8 x̄: 5.00 x̃: 5
helped stats (rel) min: 1.01% max: 14.29% x̄: 5.86% x̃: 4.07%
95% mean confidence interval for instructions value: -10.66 0.66
95% mean confidence interval for instructions %-change: -15.91% 4.19%
Inconclusive result (value mean confidence interval includes 0).
total cycles in shared programs:
527531226 ->
527531194 (<.01%)
cycles in affected programs: 92204 -> 92172 (-0.03%)
helped: 2
HURT: 0
Haswell and Broadwell had similar results. (Broadwell shown)
total instructions in shared programs:
14615730 ->
14615710 (<.01%)
instructions in affected programs: 1838 -> 1818 (-1.09%)
helped: 4
HURT: 0
helped stats (abs) min: 1 max: 8 x̄: 5.00 x̃: 5
helped stats (rel) min: 0.89% max: 13.04% x̄: 5.37% x̃: 3.78%
95% mean confidence interval for instructions value: -10.66 0.66
95% mean confidence interval for instructions %-change: -14.59% 3.85%
Inconclusive result (value mean confidence interval includes 0).
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Ian Romanick [Sat, 25 Aug 2018 00:23:26 +0000 (17:23 -0700)]
i965/fs: Refactor image atomics to be a bit more like other atomics
This greatly simplifies the next patch.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Ian Romanick [Thu, 23 Aug 2018 03:31:11 +0000 (20:31 -0700)]
i965/fs: Emit BRW_AOP_INC or BRW_AOP_DEC for atomicAdd of +1 or -1
Funny story... a single shader was hurt for instructions, spills, fills.
That same shader was also the most helped for cycles. #GPUsAreWeird
No changes on any other Intel platform.
v2: Refactor selection of atomic opcode to a separate function.
Suggested by Jason.
Haswell, Broadwell, and Skylake had similar results. (Skylake shown)
total instructions in shared programs:
14304116 ->
14304261 (<.01%)
instructions in affected programs: 12776 -> 12921 (1.13%)
helped: 19
HURT: 1
helped stats (abs) min: 1 max: 16 x̄: 2.32 x̃: 1
helped stats (rel) min: 0.05% max: 7.27% x̄: 0.92% x̃: 0.55%
HURT stats (abs) min: 189 max: 189 x̄: 189.00 x̃: 189
HURT stats (rel) min: 4.87% max: 4.87% x̄: 4.87% x̃: 4.87%
95% mean confidence interval for instructions value: -12.83 27.33
95% mean confidence interval for instructions %-change: -1.57% 0.31%
Inconclusive result (value mean confidence interval includes 0).
total cycles in shared programs:
527552861 ->
527531226 (<.01%)
cycles in affected programs: 1459195 -> 1437560 (-1.48%)
helped: 16
HURT: 2
helped stats (abs) min: 2 max: 21328 x̄: 1353.69 x̃: 6
helped stats (rel) min: 0.01% max: 5.29% x̄: 0.36% x̃: 0.03%
HURT stats (abs) min: 12 max: 12 x̄: 12.00 x̃: 12
HURT stats (rel) min: 0.03% max: 0.03% x̄: 0.03% x̃: 0.03%
95% mean confidence interval for cycles value: -3699.81 1295.92
95% mean confidence interval for cycles %-change: -0.94% 0.30%
Inconclusive result (value mean confidence interval includes 0).
total spills in shared programs: 8025 -> 8033 (0.10%)
spills in affected programs: 208 -> 216 (3.85%)
helped: 1
HURT: 1
total fills in shared programs: 10989 -> 11040 (0.46%)
fills in affected programs: 444 -> 495 (11.49%)
helped: 1
HURT: 1
Ivy Bridge
total instructions in shared programs:
11709181 ->
11709153 (<.01%)
instructions in affected programs: 3505 -> 3477 (-0.80%)
helped: 3
HURT: 0
helped stats (abs) min: 1 max: 23 x̄: 9.33 x̃: 4
helped stats (rel) min: 0.11% max: 1.16% x̄: 0.63% x̃: 0.61%
total cycles in shared programs:
254741126 ->
254738801 (<.01%)
cycles in affected programs: 919067 -> 916742 (-0.25%)
helped: 3
HURT: 0
helped stats (abs) min: 21 max: 2144 x̄: 775.00 x̃: 160
helped stats (rel) min: 0.03% max: 0.90% x̄: 0.32% x̃: 0.03%
total spills in shared programs: 4536 -> 4533 (-0.07%)
spills in affected programs: 40 -> 37 (-7.50%)
helped: 1
HURT: 0
total fills in shared programs: 4819 -> 4813 (-0.12%)
fills in affected programs: 94 -> 88 (-6.38%)
helped: 1
HURT: 0
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com> [v1]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Ian Romanick [Sat, 25 Aug 2018 00:24:59 +0000 (17:24 -0700)]
intel/compiler: Silence unused parameter warnings in brw_eu.h
All of the other brw_*_desc functions take a devinfo parameter, and all
of the others at least have an assert that uses it. Keep the parameter,
but mark it as unused.
Silences 37 warnings like:
In file included from src/intel/common/gen_disasm.c:27:0:
src/intel/compiler/brw_eu.h: In function ‘brw_pixel_interp_desc’:
src/intel/compiler/brw_eu.h:377:53: warning: unused parameter ‘devinfo’ [-Wunused-parameter]
brw_pixel_interp_desc(const struct gen_device_info *devinfo,
^~~~~~~
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Sagar Ghuge [Tue, 21 Aug 2018 21:28:17 +0000 (14:28 -0700)]
i965: enable AMD_depth_clamp_separate
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Sagar Ghuge [Tue, 21 Aug 2018 21:25:17 +0000 (14:25 -0700)]
i965: add functional changes for AMD_depth_clamp_separate
Gen >= 9 have ability to control clamping of depth values separately at
near and far plane.
z_w is clamped to the range [min(n,f), 0] if clamping at near plane is
enabled, [0, max(n,f)] if clamping at far plane is enabled and [min(n,f)
max(n,f)] if clamping at both plane is enabled.
v2: 1) Use better coding style (Ian Romanick)
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Sagar Ghuge [Fri, 27 Jul 2018 22:03:54 +0000 (15:03 -0700)]
mesa: add EXTRA_EXT for AMD_depth_clamp_separate
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Sagar Ghuge [Tue, 21 Aug 2018 20:40:55 +0000 (13:40 -0700)]
mesa: add support for GL_AMD_depth_clamp_separate tokens
_mesa_set_enable() and _mesa_IsEnabled() extended to accept new two
tokens GL_DEPTH_CLAMP_NEAR_AMD and GL_DEPTH_CLAMP_FAR_AMD.
v2: Remove unnecessary parentheses (Marek Olsak)
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Sagar Ghuge [Fri, 27 Jul 2018 21:55:57 +0000 (14:55 -0700)]
mesa: Add support for AMD_depth_clamp_separate
Enable _mesa_PushAttrib() and _mesa_PopAttrib() to handle
GL_DEPTH_CLAMP_NEAR_AMD and GL_DEPTH_CLAMP_FAR_AMD tokens.
Remove DepthClamp, because DepthClampNear + DepthClampFar replaces it,
as suggested by Marek Olsak.
Driver that enables AMD_depth_clamp_separate will only ever look at
DepthClampNear and DepthClampFar, as suggested by Ian Romanick.
v2: 1) Remove unnecessary parentheses (Marek Olsak)
2) if AMD_depth_clamp_separate is unsupported, TEST_AND_UPDATE
GL_DEPTH_CLAMP only (Marek Olsak)
3) Clamp against near and far plane separately (Marek Olsak)
4) Clip point separately for near and far Z clipping plane (Marek
Olsak)
v3: Clamp raster position zw to the range [min(n,f), 0] for near plane
and [0, max(n,f)] for far plane (Marek Olsak)
v4: Use MIN2 and MAX2 instead of CLAMP (Marek Olsak)
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Sagar Ghuge [Fri, 27 Jul 2018 03:08:44 +0000 (20:08 -0700)]
mesa: Add types for AMD_depth_clamp_separate.
Add some basic types and storage for the AMD_depth_clamp_separate
extension.
v2: 1) Drop unnecessary definition (Marek Olsak)
2) Expose extension in compatibility profile (Marek Olsak)
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Sagar Ghuge [Fri, 27 Jul 2018 21:42:36 +0000 (14:42 -0700)]
glapi: define AMD_depth_clamp_separate
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Jason Ekstrand [Tue, 30 Jan 2018 02:41:15 +0000 (18:41 -0800)]
anv: Claim to support depthBounds for ID games
Cc: "18.2" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Tue, 30 Jan 2018 02:12:04 +0000 (18:12 -0800)]
anv: Copy the appliation info into the instance
Cc: "18.2" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Tue, 30 Jan 2018 02:11:38 +0000 (18:11 -0800)]
vulkan/alloc: Add a vk_strdup helper
Cc: "18.2" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Dylan Baker [Fri, 24 Aug 2018 14:05:36 +0000 (07:05 -0700)]
meson: Actually load translation files
Currently we run the script but don't actually load any files, even in a
tarball where they exist.
Fixes:
3218056e0eb375eeda470058d06add1532acd6d4
("meson: Build i965 and dri stack")
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Caio Marcelo de Oliveira Filho [Mon, 27 Aug 2018 22:18:10 +0000 (15:18 -0700)]
nir: Remove outdated comment
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Kevin Rogovin [Mon, 27 Aug 2018 06:54:24 +0000 (09:54 +0300)]
i965: Add INTEL_fragment_shader_ordering support.
Adds suppport for INTEL_fragment_shader_ordering. We achieve
the fragment ordering by using the same instruction as for
beginInvocationInterlockARB() which is by issuing a memory
fence via sendc.
Signed-off-by: Kevin Rogovin <kevin.rogovin@intel.com>
Reviewed-by: Plamena Manolova <plamena.manolova@intel.com>
Kevin Rogovin [Mon, 27 Aug 2018 06:54:23 +0000 (09:54 +0300)]
mesa: Add GL/GLSL plumbing for INTEL_fragment_shader_ordering
This extension provides new GLSL built-in function
beginFragmentShaderOrderingIntel() that guarantees
(taking wording of GL_INTEL_fragment_shader_ordering
extension) that any memory transactions issued by
shader invocations from previous primitives mapped to
same xy window coordinates (and same sample when
per-sample shading is active), complete and are visible
to the shader invocation that called
beginFragmentShaderOrderingINTEL().
One advantage of INTEL_fragment_shader_ordering over
ARB_fragment_shader_interlock is that it provides a
function that operates as a memory barrie (instead
of a defining a critcial section) that can be called
under arbitary control flow from any function (in
contrast the begin/end of ARB_fragment_shader_interlock
may only be called once, from main(), under no control
flow.
Signed-off-by: Kevin Rogovin <kevin.rogovin@intel.com>
Reviewed-by: Plamena Manolova <plamena.manolova@intel.com>
Andrii Simiklit [Wed, 15 Aug 2018 15:20:32 +0000 (18:20 +0300)]
i965/gen6/xfb: handle case where transform feedback is not active
When the SVBI Payload Enable is false I guess the register R1.4
which contains the Maximum Streamed Vertex Buffer Index is filled by zero
and GS stops to write transform feedback when the transform feedback
is not active.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107579
Signed-off-by: Andrii Simiklit <andrii.simiklit@globallogic.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Rhys Perry [Tue, 21 Aug 2018 10:08:17 +0000 (11:08 +0100)]
docs: add forgotten features to 18.2.0 release notes
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewied-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: 18.2: <mesa-stable@lists.freedesktop.org>
Erik Faye-Lund [Mon, 20 Aug 2018 11:46:32 +0000 (12:46 +0100)]
virgl: add debug-switch to output TGSI
This is quite useful for debugging shader-transpiling issues in
virglrenderer.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-By: Gert Wollny <gert.wollny@collabora.com>
Erik Faye-Lund [Mon, 20 Aug 2018 11:08:55 +0000 (13:08 +0200)]
virgl: introduce $VIRGL_DEBUG=verbose
This adds an environment-varaible that can be used for driver-specific
flags, as well as a flag for it to enable verbose output.
While we're at it, quiet some overly chatty debug-output by default.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-By: Gert Wollny <gert.wollny@collabora.com>
Erik Faye-Lund [Mon, 20 Aug 2018 10:48:51 +0000 (12:48 +0200)]
virgl: replace fprintf-call with debug_printf
This is the only direct call-site for fprintf in virgl; all other
call-sites call debug_printf instead. So let's follow in style here.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-By: Gert Wollny <gert.wollny@collabora.com>
Erik Faye-Lund [Mon, 20 Aug 2018 10:45:14 +0000 (12:45 +0200)]
virgl: delete commented out fprintf-call
This is just debug-cruft left over. Let's just get rid of it.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-By: Gert Wollny <gert.wollny@collabora.com>
Guido Günther [Sun, 26 Aug 2018 20:24:00 +0000 (22:24 +0200)]
meson: Don't enable any vulkan drivers on arm, aarch64
There's no Vulkan support for arm atm.
Signed-off-by: Guido Günther <guido.gunther@puri.sm>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Guido Günther [Sun, 26 Aug 2018 20:23:59 +0000 (22:23 +0200)]
meson: Be a bit more helpful when arch or OS is unknown
V2: Add one missing @0@
Signed-off-by: Guido Günther <guido.gunther@puri.sm>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Sagar Ghuge [Mon, 27 Aug 2018 17:23:19 +0000 (10:23 -0700)]
intel/eu: print bytes instead of 32 bit hex value
INTEL_DEBUG=hex prints 32 bit hex value and due to endianness of CPU
byte order is reversed. In order to disassemble binary files, print
each byte instead of 32 bit hex value.
v2: Print blank spaces in order to vertically align output of compacted
instructions hex value with uncompacted instructions hex value.
(Matt Turner)
v3: Fix line wrap at correct length
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Lionel Landwerlin [Sat, 25 Aug 2018 17:22:00 +0000 (18:22 +0100)]
intel: decoder: handle 0 sized structs
Gen7.5 has a BLEND_STATE of size 0 which includes a variable length
group. We did not deal with that very well, leading to an endless
loop.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107544
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Rhys Perry [Fri, 3 Aug 2018 21:11:28 +0000 (22:11 +0100)]
nv50/ir,nvc0: use constant buffers for compute when possible on Kepler+
Gives a +7.79% increase in FPS with Hitman on lowest quality settings on
my GTX 1060.
total instructions in shared programs : 5787979 -> 5748677 (-0.68%)
total gprs used in shared programs : 669901 -> 669373 (-0.08%)
total shared used in shared programs : 548832 -> 548832 (0.00%)
total local used in shared programs : 21068 -> 21064 (-0.02%)
local shared gpr inst bytes
helped 1 0 152 274 274
hurt 0 0 0 0 0
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Rhys Perry [Sat, 18 Aug 2018 14:06:50 +0000 (15:06 +0100)]
nv50/ir: optimize multiplication by 16-bit immediates into two xmads
Rather than the usual three that would be created.
total instructions in shared programs : 5796385 -> 5786560 (-0.17%)
total gprs used in shared programs : 670103 -> 669968 (-0.02%)
total shared used in shared programs : 548832 -> 548832 (0.00%)
total local used in shared programs : 21164 -> 21068 (-0.45%)
local shared gpr inst bytes
helped 1 0 64 1040 1040
hurt 0 0 27 0 0
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Rhys Perry [Sat, 18 Aug 2018 14:06:01 +0000 (15:06 +0100)]
nv50/ir: optimize near power-of-twos into shladd
total instructions in shared programs : 5819319 -> 5796385 (-0.39%)
total gprs used in shared programs : 670571 -> 670103 (-0.07%)
total shared used in shared programs : 548832 -> 548832 (0.00%)
total local used in shared programs : 21164 -> 21164 (0.00%)
local shared gpr inst bytes
helped 0 0 318 1758 1758
hurt 0 0 63 0 0
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Rhys Perry [Wed, 13 Jun 2018 15:30:01 +0000 (16:30 +0100)]
nv50/ir: move a * b -> a << log2(b) code into createMul()
With this commit, OP_MAD is handled on nv50 too. This commit is also
useful for later commits.
Also, instead of creating a shladd, it relies on LateAlgebraicOpt to
create one. This simplifies the code and helps shader-db slightly overall.
total instructions in shared programs : 5820882 -> 5819319 (-0.03%)
total gprs used in shared programs : 670595 -> 670571 (-0.00%)
total shared used in shared programs : 548832 -> 548832 (0.00%)
total local used in shared programs : 21164 -> 21164 (0.00%)
local shared gpr inst bytes
helped 0 0 18 230 230
hurt 0 0 8 263 263
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Rhys Perry [Wed, 13 Jun 2018 15:25:23 +0000 (16:25 +0100)]
nv50/ir: optimize imul/imad to xmads
This hits the shader-db numbers a good bit, though a few xmads is way
faster than an imul or imad and the cost is mitigated by the next commit,
which optimizes many multiplications by immediates into shorter and less
register heavy instructions than the xmads.
total instructions in shared programs : 5768871 -> 5820882 (0.90%)
total gprs used in shared programs : 669919 -> 670595 (0.10%)
total shared used in shared programs : 548832 -> 548832 (0.00%)
total local used in shared programs : 21068 -> 21164 (0.46%)
local shared gpr inst bytes
helped 0 0 38 0 0
hurt 1 0 365 3076 3076
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Rhys Perry [Wed, 13 Jun 2018 15:21:56 +0000 (16:21 +0100)]
gm107/ir: add support for OP_XMAD on GM107+
v4: make the immediate field 16 bits
v5: don't ever emit h1 flags for immediates
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Rhys Perry [Wed, 13 Jun 2018 15:21:20 +0000 (16:21 +0100)]
nv50/ir: add preliminary support for OP_XMAD
v4: remove uint16_t(...)
v4: don't allow immediates outside [0,65535] in insnCanLoad()
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
vadym.shovkoplias [Thu, 23 Aug 2018 10:12:16 +0000 (13:12 +0300)]
glsl/linker: Allow unused in blocks which are not declated on previous stage
>From Section 4.3.4 (Inputs) of the GLSL 1.50 spec:
"Only the input variables that are actually read need to be written
by the previous stage; it is allowed to have superfluous
declarations of input variables."
Fixes:
* interstage-multiple-shader-objects.shader_test
v2:
Update comment in ir.h since the usage of "used" field
has been extended.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101247
Signed-off-by: Vadym Shovkoplias <vadym.shovkoplias@globallogic.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Jason Ekstrand [Fri, 24 Aug 2018 14:34:05 +0000 (09:34 -0500)]
nir: Pull block_ends_in_jump into nir.h
We had two different implementations in different files. May as well
have one and put it in nir.h.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Samuel Iglesias Gonsálvez [Fri, 24 Aug 2018 10:11:49 +0000 (12:11 +0200)]
anv: Add support for protected memory properties on anv_GetPhysicalDeviceProperties2()
VkPhysicalDeviceProtectedMemoryProperties structure is new on Vulkan 1.1.
Fixes Vulkan CTS CL#2849.
Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Sat, 25 Aug 2018 22:34:17 +0000 (17:34 -0500)]
intel/tools: Add 0x in front of a couple of hex values
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Sat, 25 Aug 2018 22:08:04 +0000 (17:08 -0500)]
anv: Fill holes in the VF VUE to zero
This fixes a GPU hang in DOOM 2016 running under wine.
Cc: mesa-stable@lists.freedesktop.org
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104809
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Kai Wasserbäch [Sat, 25 Aug 2018 10:00:30 +0000 (12:00 +0200)]
intel: tools: Fix aubinator_error's fprintf call (format-security)
The recent commit
4616639b49b4bbc91e503c1c27632dccc1c2b5be introduced
the new function aubinator_error() which is a trivial wrapper around
fprintf() to STDERR. The call to fprintf() however is passed the message
msg directly:
fprintf(stderr, msg);
This is a format-security violation and leads to an FTBFS with
-Werror=format-security (GCC 8):
../../../src/intel/tools/aubinator.c: In function 'aubinator_error':
../../../src/intel/tools/aubinator.c:74:4: error: format not a string literal and no format arguments [-Werror=format-security]
fprintf(stderr, msg);
^~~~~~~
This patch fixes this trivially by introducing a catch-all "%s" format
argument.
Fixes:
4616639b49b ("intel: tools: split aub parsing from aubinator")
Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Signed-off-by: Kai Wasserbäch <kai@dev.carbon-project.org>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Fri, 24 Aug 2018 21:05:08 +0000 (16:05 -0500)]
intel/batch_decoder: Print blend states properly
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Fri, 24 Aug 2018 21:04:03 +0000 (16:04 -0500)]
intel/batch_decoder: Fix dynamic state printing
Instead of printing addresses like everyone else, we were accidentally
printing the offset from state base address. Also, state_map is a void
pointer so we were incrementing in bytes instead of dwords and every
state other than the first was wrong.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Fri, 24 Aug 2018 20:27:38 +0000 (15:27 -0500)]
intel/decoder: Print ISL formats for vertex elements
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Fri, 24 Aug 2018 20:23:04 +0000 (15:23 -0500)]
intel/decoder: Clean up field iteration and fix sub-dword fields
First of all, setting iter->name in advance_field is unnecessary because
it gets set by gen_decode_field which gets called immediately after
gen_decode_field in the one call-site. Second, we weren't properly
initializing start_bit and end_bit in the initial condition of
gen_field_iterator_next so the first field of a struct would get printed
wrong if it doesn't start on the first bit. This is fixed by adding a
iter_start_field helper which sets the field and also sets up the other
bits we need. This fixes decoding of 3DSTATE_SBE_SWIZ.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Kenneth Graunke [Sun, 24 Jun 2018 00:26:47 +0000 (17:26 -0700)]
gallium: Split out PIPE_CAP_TEXTURE_MIRROR_CLAMP_TO_EDGE.
Some hardware can do PIPE_TEX_WRAP_MIRROR_REPEAT but not
PIPE_TEX_WRAP_MIRROR_CLAMP and PIPE_TEX_WRAP_MIRROR_CLAMP_TO_BORDER.
Drivers for such hardware would like to advertise support for
ARB_texture_mirror_clamp_to_edge but not EXT_texture_mirror_clamp.
This commit adds a new PIPE_CAP_TEXTURE_MIRROR_CLAMP_TO_EDGE bit,
changes the extension enable to be based on that, and enables it
in all upstream drivers which supported PIPE_CAP_TEXTURE_MIRROR_CLAMP
(so they continue supporting this mode).
Lionel Landwerlin [Tue, 14 Aug 2018 10:22:12 +0000 (11:22 +0100)]
intel: decoder: unify MI_BB_START field naming
The batch decoder looks for a field with a particular name to decide
whether an MI_BB_START leads into a second batch buffer level. Because
the names are different between Gen7.5/8 and the newer generation we
fail that test and keep on reading (invalid) instructions.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107544
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Dylan Baker [Fri, 24 Aug 2018 16:35:24 +0000 (09:35 -0700)]
docs: Update calendar, news, relnotes for 18.1.7
Dylan Baker [Fri, 24 Aug 2018 16:29:07 +0000 (09:29 -0700)]
docs: Add mesa 18.1.7 notes
Dylan Baker [Thu, 23 Aug 2018 16:39:20 +0000 (09:39 -0700)]
docs: Add mesa 18.1.7 docs
Andres Gomez [Fri, 24 Aug 2018 15:58:00 +0000 (18:58 +0300)]
docs: update calendar 18.2.0-rc4 is out, extend to 18.2.0-rc5
Signed-off-by: Andres Gomez <agomez@igalia.com>
Kevin Rogovin [Fri, 24 Aug 2018 06:00:46 +0000 (09:00 +0300)]
docs/relnotes: Mark NV_fragment_shader_interlock support in i965
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Emil Velikov [Thu, 9 Aug 2018 14:13:07 +0000 (15:13 +0100)]
egl/drm: use gbm_dri_bo() wrapper
Remove the explicit cast, using the appropriate wrapper instead.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Emil Velikov [Thu, 9 Aug 2018 14:11:38 +0000 (15:11 +0100)]
egl/drm: use gbm_dri_surface() wrapper
Remove the explicit cast, using the appropriate wrapper instead.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Emil Velikov [Thu, 9 Aug 2018 14:05:58 +0000 (15:05 +0100)]
egl/drm: use gbm_dri_device() wrapper
Remove the explicit cast, using the appropriate wrapper instead.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Emil Velikov [Wed, 8 Aug 2018 14:40:56 +0000 (15:40 +0100)]
egl/android: simplify device open/probe
Currently droid_probe_device, does not do any 'probing' but filtering
out a device if it doesn't match the vendor string given.
Rename the function, straighten the return type and call it only as
needed - an actual vendor string is provided.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Tomasz Figa <tfiga@chromium.org>
Emil Velikov [Wed, 8 Aug 2018 14:13:20 +0000 (15:13 +0100)]
egl/android: remove drmVersion::name NULL check
The name string is guaranteed to be non-NULL.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Tomasz Figa <tfiga@chromium.org>
Emil Velikov [Wed, 8 Aug 2018 14:05:56 +0000 (15:05 +0100)]
egl/android: remove droid_probe_driver()
The function name is misleading - it effectively checks if
loader_get_driver_for_fd fails. Which can happen only only on strdup
error - a close to impossible scenario.
Drop the function - we call the loader API at at later stage.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Tomasz Figa <tfiga@chromium.org>
Emil Velikov [Wed, 8 Aug 2018 13:56:00 +0000 (14:56 +0100)]
egl/android: use strcmp with drmVersion::name
The name string is guaranteed to be NULL terminated. Drop the explicit
length check that comes with strncmp().
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Tomasz Figa <tfiga@chromium.org>
Emil Velikov [Wed, 8 Aug 2018 13:19:05 +0000 (14:19 +0100)]
egl/android: use drmDevice instead of the manual /dev/dri iteration
Replace the manual handling of /dev/dri in favor of the drmDevice API.
The latter provides a consistent way of enumerating the devices,
providing device details as needed.
v2:
- Use ARRAY_SIZE (Frank)
- s/famour/favor/ typo (Frank)
- Make MAX_DRM_DEVICES a macro - fix vla errors (RobF)
- Remove left-over dev_path instance (RobF)
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Robert Foss <robert.foss@collabora.com> (v1)
Reviewed-by: Tomasz Figa <tfiga@chromium.org>
Emil Velikov [Fri, 24 Aug 2018 10:14:15 +0000 (11:14 +0100)]
Revert "configure: allow building with python3"
This reverts commit
ae7898dfdbe5c8dab7d11c71862353f1ae43feb0.
Turns out the python scripts are _not_ fully python 3 compatible.
As Ilia reported using get_xmlpool.py with LANG=C produces some weird
output - see the link for details.
Even though the issue was spotted with the autoconf build, it exposes a
genuine problem with the script (and lack of lang handling of the meson
build.)
https://lists.freedesktop.org/archives/mesa-dev/2018-August/203508.html
Emil Velikov [Fri, 24 Aug 2018 10:10:24 +0000 (11:10 +0100)]
Revert "travis: use python3 for the autoconf builds"
This reverts commit
855af9a5a209f061355513b92f3ba4576f48d091.
Turns out the python scripts are _not_ fully python 3 compatible.
As Ilia reported using get_xmlpool.py with LANG=C produces some weird
output - see the link for details.
Even though the issue was spotted with the autoconf build, it exposes a
genuine problem with the script (and lack of lang handling of the meson
build.)
https://lists.freedesktop.org/archives/mesa-dev/2018-August/203508.html
Kenneth Graunke [Fri, 24 Aug 2018 07:32:09 +0000 (00:32 -0700)]
Revert "mesa: bump GL_MAX_ELEMENTS_INDICES and GL_MAX_ELEMENTS_VERTICES"
This reverts commit
095515e16ca3cb2c9f1813b6602ee57ae28325a8.
This breaks KHR-GL46.map_buffer_alignment.functional on i965.
This code was apparently not reviewed and I don't know why we would
move from a driver configurable constant to a hardcoded value for all
drivers. This really looks like an accidental hack push.
Kenneth Graunke [Fri, 24 Aug 2018 03:58:32 +0000 (20:58 -0700)]
Revert recent changes about not including compute in combined limits.
As far as I can tell, no one reviewed these changes, they made i965
assert fail on driver load, and I am not certain they are correct.
(Hopefully reverting these does not break radeonsi too badly...)
The uniform related changes seem fine and reasonable, but the texture
image units change is possibly incorrect. According to the
OES_tessellation_shader spec issue 5:
(5) How are aggregate shader limits computed?
RESOLVED: Following the GL 4.4 model, but we restrict uniform
buffer bindings to 12/stage instead of 14, this results in
MAX_UNIFORM_BUFFER_BINDINGS = 72
This is 12 bindings/stage * 6 shader stages, allowing a static
partitioning of the bindings even though at most 5 stages can
appear in a program object).
MAX_COMBINED_UNIFORM_BLOCKS = 60
This is 12 blocks/stage * 5 stages, since compute shaders can't
be mixed with other stages.
MAX_COMBINED_TEXTURE_IMAGE_UNITS = 96
This is 16 textures/stage * 6 stages.
which definitely is including compute shaders in that last limit.
Not including compute shaders breaks the following test:
dEQP-GLES31.functional.state_query.integer.max_combined_texture_image_units_getinteger
There was enough breakage that I figured we should just send this back
to the drawing board.
Revert "i965: don't include compute resources in "Combined" limits"
Revert "st/mesa: don't include compute resources in "Combined" limits"
Revert "mesa: don't include compute resources in MAX_COMBINED_* limits"
This reverts commit
b03dcb1e5f507c5950d0de053a6f76e6306ee71f.
This reverts commit
cff290df4c09547cd2cb3b129ec59bdebdadba90.
This reverts commit
45f87a48f94148b484961f18a4f1ccf86f066b1c.
Roland Scheidegger [Thu, 23 Aug 2018 17:07:05 +0000 (19:07 +0200)]
gallivm: don't use saturated unsigned add/sub intrinsics for llvm 8.0
These have been removed. Unfortunately auto-upgrade doesn't work for
jit. (Worse, it seems we don't get a compilation error anymore when
compiling the shader, rather llvm will just do a call to a null
function in the jitted shaders making it difficult to detect when
intrinsics vanish.)
Luckily the signed ones are still there, I helped convincing llvm
removing them is a bad idea for now, since while the unsigned ones have
sort of agreed-upon simplest patterns to replace them with, this is not
the case for the signed ones, and they require _significantly_ more
complex patterns - to the point that the recognition is IMHO probably
unlikely to ever work reliably in practice (due to other optimizations
interfering). (Even for the relatively trivial unsigned patterns, llvm
already added test cases where recognition doesn't work, unsaturated
add followed by saturated add may produce atrocious code.)
Nevertheless, it seems there's a serious quest to squash all
cpu-specific intrinsics going on, so I'd expect patches to nuke them as
well to resurface.
Adapt the existing fallback code to match the simple patterns llvm uses
and hope for the best. I've verified with lp_test_blend that it does
produce the expected saturated assembly instructions. Though our
cmp/select build helpers don't use boolean masks, but it doesn't seem
to interfere with llvm's ability to recognize the pattern.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106231
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Marek Olšák [Mon, 6 Aug 2018 02:16:48 +0000 (22:16 -0400)]
st/mesa: expose KHR_texture_compression_astc_sliced_3d
This is ASTC 2D LDR allowing texture arrays and 3D, compressing each
slice as a separate 2D image. Tested by piglit. Trivial.
Marek Olšák [Mon, 6 Aug 2018 01:41:11 +0000 (21:41 -0400)]
st/mesa: expose EXT_disjoint_timer_query
same cap as ARB_timer_query, no changes needed, tested by piglit
Marek Olšák [Mon, 6 Aug 2018 06:48:12 +0000 (02:48 -0400)]
mesa: expose EXT_vertex_attrib_64bit
because the closed driver exposes it.
It's the same as the ARB extension.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Marek Olšák [Mon, 6 Aug 2018 06:26:09 +0000 (02:26 -0400)]
mesa: expose AMD_query_buffer_object
it's a subset of the ARB extension.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Marek Olšák [Mon, 6 Aug 2018 05:55:59 +0000 (01:55 -0400)]
mesa: expose AMD_multi_draw_indirect
because the closed driver exposes it.
This is equivalent to the ARB extension.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Marek Olšák [Mon, 6 Aug 2018 04:56:35 +0000 (00:56 -0400)]
mesa: expose AMD_gpu_shader_int64
because the closed driver exposes it.
It's equivalent to ARB_gpu_shader_int64.
In this patch, I did everything the same as we do for ARB_gpu_shader_int64.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Marek Olšák [Mon, 6 Aug 2018 05:25:51 +0000 (01:25 -0400)]
mesa: expose ARB_post_depth_coverage in the Compatibility profile
It only contains GLSL changes.
v2: allow the layout qualifier on GLSL <= 1.30
Jason Ekstrand [Tue, 24 Jul 2018 05:20:41 +0000 (22:20 -0700)]
intel/nir: Enable nir_opt_find_array_copies
We have to be a bit careful with this one because we want it to run in
the optimization loop but only in the first brw_nir_optimize call.
Later calls assume that we've lowered away copy_deref instructions and
we don't want to introduce any more.
Shader-db results on Kaby Lake:
total instructions in shared programs:
15176942 ->
15176942 (0.00%)
instructions in affected programs: 0 -> 0
helped: 0
HURT: 0
In spite of the lack of any shader-db improvement, this patch completely
eliminates spilling in the Batman: Arkham City tessellation shaders.
This is because we are now able to detect that the temporary array
created by DXVK for storing TCS inputs is a copy of the input arrays and
use indirect URB reads instead of making a copy of 4.5 KiB of input data
and then indirecting on it with if-ladders.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Tue, 24 Jul 2018 02:16:56 +0000 (19:16 -0700)]
nir: Add an array copy optimization
This peephole optimization looks for a series of load/store_deref or
copy_deref instructions that copy an array from one variable to another
and turns it into a copy_deref that copies the entire array. The
pattern it looks for is extremely specific but it's good enough to pick
up on the input array copies in DXVK and should also be able to pick up
the sequence generated by spirv_to_nir for a OpLoad of a large composite
followed by OpStore. It can always be improved later if needed.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Wed, 25 Jul 2018 15:54:09 +0000 (08:54 -0700)]
intel/nir: Use nir_shrink_vec_array_vars
Shader-db results on Kaby Lake:
total instructions in shared programs:
15177605 ->
15176765 (<.01%)
instructions in affected programs: 4259 -> 3419 (-19.72%)
helped: 1
HURT: 0
total spills in shared programs: 10954 -> 10855 (-0.90%)
spills in affected programs: 295 -> 196 (-33.56%)
helped: 1
HURT: 0
total fills in shared programs: 22222 -> 22117 (-0.47%)
fills in affected programs: 417 -> 312 (-25.18%)
helped: 1
HURT: 0
The helped shader is from the OglCSDof synmark test. On my Kaby Lake
laptop, the actual framerate of the benchmark didn't appear to improve
beyond the noise.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Wed, 25 Jul 2018 02:32:27 +0000 (19:32 -0700)]
nir: Add a array-of-vector variable shrinking pass
This pass looks for variables with vector or array-of-vector types and
narrows the type to only the components used.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Tue, 24 Jul 2018 17:08:20 +0000 (10:08 -0700)]
intel/nir: Use the new structure and array splitting passes
We call structure splitting once because it is guaranteed to split all
the structures in the entire shader in one go. We call array splitting
in the loop in case future optimizations turn indirects into direct
dereferences and we can split more arrays.
Shader-db results on Kaby Lake:
total instructions in shared programs:
15177605 ->
15177605 (0.00%)
instructions in affected programs: 0 -> 0
helped: 0
HURT: 0
This is unsurprising because nir_lower_vars_to_ssa already effectively
does structure and array splitting internally. It doesn't actually
split the variables but it's ability to reason about aliasing in the
presence of arrays and structures and pick out scalars or vectors to be
lowered to SSA values is fairly advanced.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Tue, 24 Jul 2018 19:33:46 +0000 (12:33 -0700)]
nir: Add an array splitting pass
This pass looks for array variables where at least one level of the
array is never indirected and splits it into multiple smaller variables.
This pass doesn't really do much now because nir_lower_vars_to_ssa can
already see through arrays of arrays and can detect indirects on just
one level or even see that arr[i][0][5] does not alias arr[i][1][j].
This pass exists to help other passes more easily see through arrays of
arrays. If a back-end does implement arrays using scratch or indirects
on registers, having more smaller arrays is likely to have better memory
efficiency.
v2 (Jason Ekstrand):
- Better comments and naming (some from Caio)
- Rework to use one hash map instead of two
v2.1 (Jason Ekstrand):
- Fix a couple of bugs that were added in the rework including one
which basically prevented it from running
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Tue, 24 Jul 2018 17:08:06 +0000 (10:08 -0700)]
nir: Add a structure splitting pass
This pass doesn't really do much now because nir_lower_vars_to_ssa can
already see through structures and considers them to be "split". This
pass exists to help other passes more easily see through structure
variables. If a back-end does implement arrays using scratch or
indirects on registers, having more smaller arrays is likely to have
better memory efficiency.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Wed, 25 Jul 2018 15:53:58 +0000 (08:53 -0700)]
nir/types: Add array_or_matrix helpers
Reviewed-by: Thomas Helland<thomashelland90@gmail.com>
Kenneth Graunke [Fri, 24 Aug 2018 00:24:19 +0000 (17:24 -0700)]
i965: don't include compute resources in "Combined" limits
The combined limits should only include shader stages that can be active
at the same time. We don't need to include compute.
See also
cff290df4c09547cd2cb3b129ec59bdebdadba90 for st/mesa.
Unbreaks i965 from assert failing on driver load since Marek's
45f87a48f94148b484961f18a4f1ccf86f066b1c, which dropped the core
Mesa capabilities before adjusting driver limits down to match.