mesa.git
6 years agor600g: Support emitting scratch ops
Glenn Kennard [Sun, 5 Mar 2017 17:26:49 +0000 (18:26 +0100)]
r600g: Support emitting scratch ops

Signed-off-by: Glenn Kennard <glenn.kennard@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
6 years agor600: fix texture gather swizzling.
Dave Airlie [Thu, 8 Feb 2018 06:19:28 +0000 (16:19 +1000)]
r600: fix texture gather swizzling.

This fixes:
KHR-GL45.texture_gather.swizzle
on cayman and redwood.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
6 years agoac: add 64bit support to ac_find_lsb()
Timothy Arceri [Tue, 6 Feb 2018 03:38:57 +0000 (14:38 +1100)]
ac: add 64bit support to ac_find_lsb()

v2: use LLVMBuildTrunc()

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoac: move get_elem_bits() to ac_llvm_build.c
Timothy Arceri [Tue, 6 Feb 2018 03:38:19 +0000 (14:38 +1100)]
ac: move get_elem_bits() to ac_llvm_build.c

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoac: add 64bit bitCount support
Timothy Arceri [Tue, 6 Feb 2018 03:34:55 +0000 (14:34 +1100)]
ac: add 64bit bitCount support

v2: use LLVMBuildTrunc()

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoac/nir: clean up handle_fs_outputs_post()
Samuel Pitoiset [Thu, 8 Feb 2018 13:56:48 +0000 (14:56 +0100)]
ac/nir: clean up handle_fs_outputs_post()

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoac/nir: add radv_load_output() helper
Samuel Pitoiset [Thu, 8 Feb 2018 13:56:47 +0000 (14:56 +0100)]
ac/nir: add radv_load_output() helper

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoac/shader: scan info about output PS declarations
Samuel Pitoiset [Thu, 8 Feb 2018 13:56:46 +0000 (14:56 +0100)]
ac/shader: scan info about output PS declarations

NIR->LLVM should only be a translation pass, and all scan stuff
should be done before.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoac/nir: add radv_export_param() helper
Samuel Pitoiset [Thu, 8 Feb 2018 13:56:45 +0000 (14:56 +0100)]
ac/nir: add radv_export_param() helper

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoac/nir: remove set but unused export_mask
Samuel Pitoiset [Thu, 8 Feb 2018 13:56:44 +0000 (14:56 +0100)]
ac/nir: remove set but unused export_mask

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoac/nir: remove dead code in handle_vs_outputs_post()
Samuel Pitoiset [Thu, 8 Feb 2018 13:56:43 +0000 (14:56 +0100)]
ac/nir: remove dead code in handle_vs_outputs_post()

The memcpy can't be reached because the condition is always false.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoac/nir: remove useless check in si_llvm_init_export_args()
Samuel Pitoiset [Thu, 8 Feb 2018 13:56:42 +0000 (14:56 +0100)]
ac/nir: remove useless check in si_llvm_init_export_args()

values can't be NULL because we use ac_build_export_null() now.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoac/nir: use ac_build_export_null()
Samuel Pitoiset [Wed, 7 Feb 2018 18:09:13 +0000 (19:09 +0100)]
ac/nir: use ac_build_export_null()

The number of enabled channels should be 0 when exporting null.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoac: add ac_build_export_null() helper
Samuel Pitoiset [Wed, 7 Feb 2018 18:09:12 +0000 (19:09 +0100)]
ac: add ac_build_export_null() helper

Imported from RadeonSI.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agomeson: Add build option for tools
Scott D Phillips [Thu, 8 Feb 2018 00:55:24 +0000 (16:55 -0800)]
meson: Add build option for tools

Add a build option to control building some of the misc tools we
have. Also set the executables to install, presumably you want
that if you're asking for the build.

v2: set 'install:' to the with_tools value, not true (Jordan)
    handle 'all' in a the comma list (Dylan)
    Add freedreno's tools (Dylan)

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
6 years agointel: Add Coffee Lake brand strings
Anuj Phogat [Mon, 29 Jan 2018 18:42:17 +0000 (10:42 -0800)]
intel: Add Coffee Lake brand strings

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
6 years agogallium/util: silence clang warning in blitter code
Brian Paul [Thu, 8 Feb 2018 17:00:21 +0000 (10:00 -0700)]
gallium/util: silence clang warning in blitter code

Silence "warning: comparison of constant 4294967295 with expression
of type 'ubyte'".

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
6 years agotgsi: s/unsigned/enum tgsi_semantic/ in ureg_DECL_output()
Brian Paul [Thu, 8 Feb 2018 16:54:52 +0000 (09:54 -0700)]
tgsi: s/unsigned/enum tgsi_semantic/ in ureg_DECL_output()

So the function matches the prototype.  Found with clang.
v2: fix copy&paste error

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
6 years agotgsi: use TGSI_INTERPOLATE_x arguments instead of zeros in ureg code
Brian Paul [Thu, 8 Feb 2018 01:29:12 +0000 (18:29 -0700)]
tgsi: use TGSI_INTERPOLATE_x arguments instead of zeros in ureg code

TGSI_INTERPOLATE_CONSTANT and TGSI_INTERPOLATE_LOC_CENTER have the
value zero so there's no change in behavior.  It seems funny to
declare these fs input registers with constant interpolation.  But
it looks like ureg_DECL_input_layout() is not called anywhere and
ureg_DECL_input() is only called from
util_make_geometry_passthrough_shader().

Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agogallium/util: s/uint/enum tgsi_semantic/ in simple shader code
Brian Paul [Thu, 8 Feb 2018 01:28:34 +0000 (18:28 -0700)]
gallium/util: s/uint/enum tgsi_semantic/ in simple shader code

Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agotgsi: s/unsigned/enum pipe_shader_type/ in ureg code
Brian Paul [Thu, 8 Feb 2018 01:18:39 +0000 (18:18 -0700)]
tgsi: s/unsigned/enum pipe_shader_type/ in ureg code

And add a default switch case to silence a compiler warning.

Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agogallium/util: s/uint/enum tgsi_semantic/ in u_blitter.c
Brian Paul [Wed, 7 Feb 2018 23:14:11 +0000 (16:14 -0700)]
gallium/util: s/uint/enum tgsi_semantic/ in u_blitter.c

And put static qualifier on const arrays.

Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agost/mesa: s/unsigned/enum tgsi_semantic/ st_cb_drawpixels.c
Brian Paul [Wed, 7 Feb 2018 23:12:59 +0000 (16:12 -0700)]
st/mesa: s/unsigned/enum tgsi_semantic/ st_cb_drawpixels.c

Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agovbo: add a comment on vbo_draw_transform_feedback()
Brian Paul [Wed, 7 Feb 2018 23:12:35 +0000 (16:12 -0700)]
vbo: add a comment on vbo_draw_transform_feedback()

Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agogallium/util: trivial whitespace/formatting fixes in u_blit.c
Brian Paul [Wed, 7 Feb 2018 16:28:16 +0000 (09:28 -0700)]
gallium/util: trivial whitespace/formatting fixes in u_blit.c

Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agovbo: improve comments on vbo_draw_func()
Brian Paul [Tue, 6 Feb 2018 22:35:30 +0000 (15:35 -0700)]
vbo: improve comments on vbo_draw_func()

And rename a parameter name.

Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agocso: add a couple sanity check assertions in cso_draw_vbo()
Brian Paul [Tue, 6 Feb 2018 22:33:37 +0000 (15:33 -0700)]
cso: add a couple sanity check assertions in cso_draw_vbo()

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agost/mesa: rename some vars related to indirect draw count
Brian Paul [Wed, 31 Jan 2018 22:01:09 +0000 (15:01 -0700)]
st/mesa: rename some vars related to indirect draw count

'indirect_params' was a bit vague.  Use the names that we use in
gallium's pipe_draw_indirect_info.

Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agost/mesa: remove out_num_textures from update_textures
Marek Olšák [Thu, 16 Nov 2017 15:01:11 +0000 (16:01 +0100)]
st/mesa: remove out_num_textures from update_textures

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
6 years agost/mesa: don't store non-fragment sampler states and views in st_context
Marek Olšák [Wed, 15 Nov 2017 23:32:22 +0000 (00:32 +0100)]
st/mesa: don't store non-fragment sampler states and views in st_context

those are unused.

st_context: 10120 -> 3704 bytes

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
6 years agoi965: perf: cleanup detection of kernel support for loadable configs
Lionel Landwerlin [Tue, 6 Feb 2018 23:28:24 +0000 (23:28 +0000)]
i965: perf: cleanup detection of kernel support for loadable configs

The initial revision of the patch adding loadable configs was testing
the feature's availability by adding a new config successfully and
then removing it.

A second version tested the availability just by exercising the
removal. But some unused code remained.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
6 years agoi965: perf: use drmIoctl() instead of ioctl()
Lionel Landwerlin [Tue, 6 Feb 2018 23:23:22 +0000 (23:23 +0000)]
i965: perf: use drmIoctl() instead of ioctl()

ioctl() might be interrupted, use drmIoctl() instead as it'll retry
automatically.

Fixes: 27ee83eaf7e "i965: perf: add support for userspace configurations"
Cc: "18.0" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Tested-by: Mark Janes <mark.a.janes@intel.com>
6 years agoi965: perf: add debug messages for loaded configs
Lionel Landwerlin [Tue, 6 Feb 2018 17:00:58 +0000 (17:00 +0000)]
i965: perf: add debug messages for loaded configs

This helps figuring out potential problems when metrics don't show up
on frameretrace for example.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
6 years agor600: implement tg4 integer workaround. (v2)
Dave Airlie [Thu, 8 Feb 2018 02:35:46 +0000 (12:35 +1000)]
r600: implement tg4 integer workaround. (v2)

This ports the texture gather integer workaround from radeonsi.

This fixes:
KHR-GL45.texture_gather.plain-gather-uint/int*

v2: add rect support, fix 2d array shadow
Reviewed-by: Roland Scheidegger <sroland@vmware.com> (on irc)
Signed-off-by: Dave Airlie <airlied@redhat.com>
6 years agor600: clean up initial shader register setup
Glenn Kennard [Tue, 6 Feb 2018 20:24:07 +0000 (06:24 +1000)]
r600: clean up initial shader register setup

This is taken from Glenn Kennards scratch series, but separated
out as a cleanup by me.

Reviewed-By: Gert Wollny <gw.fossdev@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
6 years agor600: partly fix sampleMaskIn value
Roland Scheidegger [Sun, 4 Feb 2018 22:54:26 +0000 (23:54 +0100)]
r600: partly fix sampleMaskIn value

The hw gives us coverage for pixel, not for individual fragment shader
invocations, in case execution isn't per pixel (eg, unlike cm, actually
cannot do "real" minSampleShading, it's either per-pixel or per-fragment,
but it doesn't really make a difference here).
Also, with msaa disabled, the hw still gives us a mask corresponding to
the number of samples, where GL requires this to be 1.
Fix this up by masking the sampleMaskIn bits with the bit corresponding to
the sampleID, if we know this shader is always executed at per-sample
granularity. (In case of a per-sample frequency shader and msaa disabled,
the sampleID will always be 0, so this works just fine there.)
Fixing this for the minSampleShading case will need a shader key (radeonsi
uses the prolog part for) (for eg, could get away with a single bit, cm
would need more bits depending on sample/invocation ratio, or read the
bits from a uniform), unless we'd want to always use a sample mask uniform
(which is probably not a good idea, as it would make the ordinary common
msaa case slower for no good reason).
This fixes some parts of piglit arb_sample_shading-samplemask (with fixed
test), in particular those which use a sampleID, still failing others
as expected.

Reviewed-by: Dave Airlie <airlied@redhat.com>
6 years agor600: clean up fragment shader input scan code
Roland Scheidegger [Sun, 4 Feb 2018 22:38:28 +0000 (23:38 +0100)]
r600: clean up fragment shader input scan code

For some reason, we were iterating through the code twice (first just for
instructions needing barycentrics, then for instructions and input dcls).
Move things around slightly so this is no longer necessary.
There also was a unnedeed enabling of the fixed_pt_position_gpr - this is only
needed if the per-sample interpolation comes from an input, not from an
instruction (just move the assert where it belongs) (since the sample id to
sample from comes from a tgsi src in this case, and isn't sampleID).
Otherwise there should be no functional change.

Reviewed-by: Dave Airlie <airlied@redhat.com>
6 years agomesa: (trivial) remove unused ignore_sample_qualifier_parameter
Roland Scheidegger [Sat, 3 Feb 2018 23:32:05 +0000 (00:32 +0100)]
mesa: (trivial) remove unused ignore_sample_qualifier_parameter

This parameter for _mesa_get_min_incations_per_fragment() was once used
by the intel driver, but it's long gone.

Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Dave Airlie <airlied@vmware.com>
6 years agor600/cm: (trivial) code cleanup for emitting msaa state
Roland Scheidegger [Sat, 3 Feb 2018 19:11:35 +0000 (20:11 +0100)]
r600/cm: (trivial) code cleanup for emitting msaa state

No functional change (compile tested only).

Reviewed-by: Dave Airlie <airlied@redhate.com>
6 years agotgsi: use tgsi_semantic enum type in ureg code
Brian Paul [Wed, 7 Feb 2018 05:17:10 +0000 (22:17 -0700)]
tgsi: use tgsi_semantic enum type in ureg code

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agost/mesa: use tgsi_semantic enum type
Brian Paul [Wed, 7 Feb 2018 05:16:41 +0000 (22:16 -0700)]
st/mesa: use tgsi_semantic enum type

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agotgsi: use TGSI enum types in ureg code
Brian Paul [Wed, 7 Feb 2018 05:11:41 +0000 (22:11 -0700)]
tgsi: use TGSI enum types in ureg code

v2: fix enum tgsi_interpolate_mode/loc typo.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agost/mesa: use TGSI enum types in st_glsl_to_tgsi.cpp
Brian Paul [Wed, 7 Feb 2018 05:10:59 +0000 (22:10 -0700)]
st/mesa: use TGSI enum types in st_glsl_to_tgsi.cpp

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agogallium/util: replace uint with tgsi enum types
Brian Paul [Wed, 7 Feb 2018 04:55:10 +0000 (21:55 -0700)]
gallium/util: replace uint with tgsi enum types

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agogallium/util: replace unsigned with tgsi enum types
Brian Paul [Wed, 7 Feb 2018 04:54:38 +0000 (21:54 -0700)]
gallium/util: replace unsigned with tgsi enum types

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
6 years agoradv: implement VK_EXT_external_memory_host
Fredrik Höglund [Thu, 25 Jan 2018 17:12:14 +0000 (18:12 +0100)]
radv: implement VK_EXT_external_memory_host

Ported from the radeonsi GL_AMD_pinned_memory implementation.

Signed-off-by: Fredrik Höglund <fredrik@kde.org>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agor600: fix rendering regression on r6/7 gpus
Dave Airlie [Wed, 7 Feb 2018 22:12:36 +0000 (08:12 +1000)]
r600: fix rendering regression on r6/7 gpus

Fixes: 2d5b5d267e (r600: work out target mask at framebuffer bind.)
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104989

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
6 years agoradeonsi: avoid int-to-pointer-cast warnings on 32bit
Grazvydas Ignotas [Sat, 3 Feb 2018 21:54:28 +0000 (23:54 +0200)]
radeonsi: avoid int-to-pointer-cast warnings on 32bit

I hope the actual dropping of MSB is ok, but that's what's already
happened before this change.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agogallium/hud: update some query functions
Grazvydas Ignotas [Sat, 3 Feb 2018 21:42:28 +0000 (23:42 +0200)]
gallium/hud: update some query functions

It seems these were missed when struct pipe_context * argument was
added to hud_graph::query_new_value.

Fixes: 3132afdf4c "gallium/hud: pass pipe_context explicitly to most functions"
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoRevert "gallium: build ddebug, noop, rbug, trace as part of auxiliary"
Roland Scheidegger [Wed, 7 Feb 2018 22:47:39 +0000 (23:47 +0100)]
Revert "gallium: build ddebug, noop, rbug, trace as part of auxiliary"

This reverts commit 6f82b8d8d0a986aac28e7bec47fc313fb950475c.

This broke scons build, and reportedly clover with autotools/meson too.

6 years agogallium: build ddebug, noop, rbug, trace as part of auxiliary
Marek Olšák [Mon, 4 Sep 2017 20:36:34 +0000 (22:36 +0200)]
gallium: build ddebug, noop, rbug, trace as part of auxiliary

Building gallium is faster by 7.5 seconds on a 4core/8thread 3GHz CPU.
(gallium build time is reduced by 15% when building only radeonsi)

Non-recursive makefiles are great!

6 years agou_blit: (trivial) fix bogus argument order for set_fragment_shader
Roland Scheidegger [Wed, 7 Feb 2018 21:02:54 +0000 (22:02 +0100)]
u_blit: (trivial) fix bogus argument order for set_fragment_shader

Amazingly this still worked sometimes, albeit I'm not even sure why...
This fixes d7bec6f7a6a2a35c80be939db8532011af1e9b67.

6 years agomesa: fix incorrect type when allocating arrays
Andres Rodriguez [Wed, 7 Feb 2018 19:38:52 +0000 (14:38 -0500)]
mesa: fix incorrect type when allocating arrays

The array members are have type 'struct gl_buffer_object *'

Found by coverity.

Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
6 years agou_blit,u_simple_shaders: add shader to convert from xrbias format
Roland Scheidegger [Wed, 7 Feb 2018 04:18:17 +0000 (05:18 +0100)]
u_blit,u_simple_shaders: add shader to convert from xrbias format

We need this to handle some oddball dx10 format
(DXGI_FORMAT_R10G10B10_XR_BIAS_A2_UNORM). What you can do with this
format is very limited, hence we don't want to add it as a gallium
format (we could not express the properties of this format as
ordinary format properties neither, so like all special formats
it would need specific code for handling it in any case).
While here, also nuke the array for different shaders for different
writemasks, as it was not actually used (always full masks are
passed in for generating shaders).

Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
6 years agou_simple_shaders: fix mask handling in util_make_fragment_tex_shader_writemask
Roland Scheidegger [Wed, 7 Feb 2018 04:03:42 +0000 (05:03 +0100)]
u_simple_shaders: fix mask handling in util_make_fragment_tex_shader_writemask

The writemask handling was busted, since writing defaults to output
meant they got overwritten by the tex sampling anyway. Albeit the
affected components were undefined, so maybe with some luck it
still would have worked with some drivers - if not could as well
kill it... (This would have affected u_blitter but not u_blit since
the latter always used xyzw mask.)

Reviewed-by: Brian Paul <brianp@vmware.com>
6 years agoautotools: Only build libmesa-st-tests-common.a for tests.
Bas Nieuwenhuizen [Fri, 2 Feb 2018 15:59:23 +0000 (16:59 +0100)]
autotools: Only build libmesa-st-tests-common.a for tests.

We don't need the library if we don't build tests, and building
it adds a dependency on gtest which adds a dependency on cxxabi.h.

Fixes: 6569b33b6e "mesa/st/tests: unify MockCodeLine* classes"
Reviewed-By: Gert Wollny <gw.fossdev@gmail.com>
6 years agoi965: add __DRI2_BLOB support and set cache functions
Tapani Pälli [Wed, 17 Jan 2018 09:43:59 +0000 (11:43 +0200)]
i965: add __DRI2_BLOB support and set cache functions

v2: adjust to change that moved cache from ctx to screen

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agodisk cache: add callback functionality
Tapani Pälli [Wed, 7 Feb 2018 06:13:00 +0000 (08:13 +0200)]
disk cache: add callback functionality

v2: add disk_cache_has_key, disk_cache_put_key support
    using blob cache (Nicolai, Jordan)

v3: rename set_cb as put_cb to match existing naming (Timothy)

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agodisk cache: initialize cache path and index only when used
Tapani Pälli [Mon, 22 Jan 2018 09:55:06 +0000 (11:55 +0200)]
disk cache: initialize cache path and index only when used

This patch makes disk_cache initialize path and index lazily so
that we can utilize disk_cache without a path using callback
functionality introduced by next patch.

v2: unmap mmap and destroy queue only if index_mmap exists

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agoglsl/tests: changes to test_disk_cache_create test
Tapani Pälli [Thu, 1 Feb 2018 07:08:52 +0000 (09:08 +0200)]
glsl/tests: changes to test_disk_cache_create test

Next patch will allow disk_cache instance to be created without
path set for it, modify some test cases that assume disk_cache
creation to fail with invalid path. Creation should succeed but
simple put/get test fail.

v2: leave tests as is but check that both cache struct exists
    and try simple put/get that should fail with invalid path set
    (Emil)

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com> (v1)
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agoglsl/tests: move utility functions in cache_test
Tapani Pälli [Tue, 30 Jan 2018 09:42:55 +0000 (11:42 +0200)]
glsl/tests: move utility functions in cache_test

Patch moves functions higher so that we can utilize them from
test_disk_cache_create which is modified by next patch.

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agoegl: add support for EGL_ANDROID_blob_cache
Tapani Pälli [Thu, 28 Dec 2017 08:51:11 +0000 (10:51 +0200)]
egl: add support for EGL_ANDROID_blob_cache

v2: cleanup, move callbacks to _egl_display struct (Emil Velikov)
    adapt to earlier ctx->screen changes

v3: remove useless checking, add _eglSetFuncName (Emil Velikov)

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com> (v2)
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agodri: add interface for EGL_ANDROID_blob_cache extension
Tapani Pälli [Thu, 28 Dec 2017 07:27:24 +0000 (09:27 +0200)]
dri: add interface for EGL_ANDROID_blob_cache extension

v2: move from __DRIcontext to __DRIscreen (Emil Velikov)

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agoac/nir: use new pknorm_i16/u16 and pk_i16/u16 LLVM intrinsics
Samuel Pitoiset [Tue, 6 Feb 2018 21:06:11 +0000 (22:06 +0100)]
ac/nir: use new pknorm_i16/u16 and pk_i16/u16 LLVM intrinsics

Ported from RadeonSI.

Only one F1 2017 shader is affected, code size decreased
from 532 to 488 on both Polaris10 and Vega10.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoac/nir: avoid loading unused VS input components
Samuel Pitoiset [Thu, 1 Feb 2018 10:35:06 +0000 (11:35 +0100)]
ac/nir: avoid loading unused VS input components

Polaris10:
Totals from affected shaders:
SGPRS: 122840 -> 120984 (-1.51 %)
VGPRS: 78812 -> 78440 (-0.47 %)
Spilled SGPRs: 177 -> 129 (-27.12 %)
Code Size: 2950028 -> 2941276 (-0.30 %) bytes
Max Waves: 17899 -> 17976 (0.43 %)

Vega10:
Totals from affected shaders:
SGPRS: 117144 -> 115776 (-1.17 %)
VGPRS: 77580 -> 77532 (-0.06 %)
Spilled SGPRs: 0 -> 152 (0.00 %)
Code Size: 3352656 -> 3347860 (-0.14 %) bytes
Max Waves: 19756 -> 19866 (0.56 %)

This increases SGPRs spilling a bit with Talos, but I have
some other ideas that might reduce it.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoac/shader: scan vertex inputs usage mask
Samuel Pitoiset [Thu, 1 Feb 2018 10:32:32 +0000 (11:32 +0100)]
ac/shader: scan vertex inputs usage mask

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoi965: allocate a SGVS element when VertexID or InstanceID are read
Iago Toral Quiroga [Thu, 4 Jan 2018 02:55:13 +0000 (03:55 +0100)]
i965: allocate a SGVS element when VertexID or InstanceID are read

Although on gen8+ platforms we can in theory use 3DSTATE_VF_SGVS
to put these beyond the last vertex element it seems that we still
need to allocate the SVGS element, otherwise we have observed cases
where we end up reading garbage. Specifically, the CTS test mentioned
below was flaky with a fail rate of ~1% on some gen9+ platforms caused
by reading garbage for the gl_InstanceID value. The flakyness goes
away as soon as we start allocating the SVGS element.

v2:
  - Do this for gen8+, not just gen9+, and pull the boolean
    outside the #if block (Jason)

Fixes flaky test:
KHR-GL45.vertex_attrib_64bit.limits_test

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104335
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agoglapi: fix check_table test for non-shared glapi with meson
Dylan Baker [Mon, 20 Nov 2017 22:55:13 +0000 (14:55 -0800)]
glapi: fix check_table test for non-shared glapi with meson

v2: - Add glapitable_h generated source to requirements

Fixes: 3218056e0eb3 ("meson: Build i965 and dri stack")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com> (v1)
Reviewed-by: Emil Velikov <emil.velikov@collabora.com> (v1)
6 years agoglapi: Don't search through subdirs from glapitable.h
Dylan Baker [Mon, 20 Nov 2017 22:54:30 +0000 (14:54 -0800)]
glapi: Don't search through subdirs from glapitable.h

Because meson won't put it in that folder.

Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agostate_tracker: Don't build st-renumerate-test without shared glapi
Dylan Baker [Mon, 20 Nov 2017 22:53:09 +0000 (14:53 -0800)]
state_tracker: Don't build st-renumerate-test without shared glapi

Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agoglapi: remove APPLE extensions from test
Dylan Baker [Mon, 20 Nov 2017 22:39:27 +0000 (14:39 -0800)]
glapi: remove APPLE extensions from test

Fixes: 7009955281260fbb ("mesa: Remove GL_APPLE_vertex_array_object stubs")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Acked-by: Ian Romanick <ian.d.romanick@intel.com>
6 years agoglapi/check_table: Remove 'extern "C"' block
Dylan Baker [Mon, 20 Nov 2017 18:05:25 +0000 (10:05 -0800)]
glapi/check_table: Remove 'extern "C"' block

Using 'extern "C"' around includes is always incorrect, as the header may
contain C++ symbols (as it does in this case), which means it cannot use
C linkage. In this case the header has a template in it, which obviously
cannot be linked with C linkage rules.

Fixes: a29ad2b421b75a1727b ("mesa/tests: Add tests for the generated dispatch table")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agomeson: fix test source name for static glapi
Dylan Baker [Wed, 8 Nov 2017 00:00:34 +0000 (16:00 -0800)]
meson: fix test source name for static glapi

fixes: 43a6e84927e3 ("meson: build mesa test.")
Signed-off-by: Dylan Baker <dylanx.c.baker@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agoglapi: don't walk backwards for includes
Dylan Baker [Tue, 7 Nov 2017 17:40:06 +0000 (09:40 -0800)]
glapi: don't walk backwards for includes

Instead just set the proper -I flags and include it from a more standard
path. In this case we'll add -Isrc/mesa (which is common), and #include
main/foo.h.

Signed-off-by: Dylan Baker <dylanx.c.baker@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agomesa: rename gl_vertex_array_object::_VertexAttrib -> _VertexArray
Brian Paul [Mon, 5 Feb 2018 16:33:58 +0000 (09:33 -0700)]
mesa: rename gl_vertex_array_object::_VertexAttrib -> _VertexArray

Since the type is gl_vertex_array.  Update comment to explain that
these arrays are only used by the VBO module.

Also rename some local variables in _mesa_update_vao_derived_arrays().

Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
6 years agomesa: minor whitespace fixes, line wrapping in texcompress.c
Brian Paul [Tue, 6 Feb 2018 22:12:58 +0000 (15:12 -0700)]
mesa: minor whitespace fixes, line wrapping in texcompress.c

Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
6 years agomesa: simplify _mesa_get_compressed_formats()
Brian Paul [Tue, 6 Feb 2018 16:55:13 +0000 (09:55 -0700)]
mesa: simplify _mesa_get_compressed_formats()

Instead of testing for formats==NULL everywhere, just point formats at
a dummy array which will be discarded.

Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
6 years agoutil: remove redundant check for the __clang__ macro
Vlad Golovkin [Tue, 6 Feb 2018 13:48:00 +0000 (06:48 -0700)]
util: remove redundant check for the __clang__ macro

Clang defines __GNUC__ macro, so one doesn't need to check __clang__
macro in this particular case.

v2: added comment as per Brian Paul's suggestion

Reviewed-by: Brian Paul <brianp@vmware.com>
6 years agost/mesa: use st_access_flags_to_transfer_flags() helper in more places
Brian Paul [Fri, 2 Feb 2018 16:21:44 +0000 (09:21 -0700)]
st/mesa: use st_access_flags_to_transfer_flags() helper in more places

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agost/mesa: refactor st_bufferobj_map_range()
Brian Paul [Fri, 2 Feb 2018 15:54:19 +0000 (08:54 -0700)]
st/mesa: refactor st_bufferobj_map_range()

Use a new helper function, st_access_flags_to_transfer_flags(), to
convert the GL_MAP_x flags to PIPE_TRANSFER_x flags.

We'll be able to use this function in a couple other places.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agost/mesa: refactor bufferobj_data()
Brian Paul [Fri, 2 Feb 2018 15:38:50 +0000 (08:38 -0700)]
st/mesa: refactor bufferobj_data()

Split out some of the code into three new helper functions:
buffer_target_to_bind_flags(), storage_flags_to_buffer_flags(),
buffer_usage() to make the code more managable.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoradv: run nir_opt_shrink_load
Samuel Pitoiset [Mon, 29 Jan 2018 16:19:18 +0000 (17:19 +0100)]
radv: run nir_opt_shrink_load

LLVM can't shrink loads.

Polaris10:
Totals from affected shaders:
SGPRS: 62528 -> 59955 (-4.11 %)
VGPRS: 44708 -> 44616 (-0.21 %)
Spilled SGPRs: 16 -> 8 (-50.00 %)
Code Size: 1355504 -> 1355172 (-0.02 %) bytes
Max Waves: 11710 -> 11670 (-0.34 %)

Vega10:
Totals from affected shaders:
SGPRS: 51448 -> 50371 (-2.09 %)
VGPRS: 39140 -> 39048 (-0.24 %)
Spilled SGPRs: 16 -> 16 (0.00 %)
Code Size: 1307188 -> 1304296 (-0.22 %) bytes
Max Waves: 11312 -> 11292 (-0.18 %)

This reduces SGPRs spilling in MadMax, and it also reduces
number of SGPRs in DOW3 and F12017. The number of waves slightly
decreases in F1 but I don't see any performance changes after
benchmarking it. Talos and Serious Sam are not affected because
they don't use any push constants.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agonir: add nir_opt_shrink_load pass
Samuel Pitoiset [Mon, 29 Jan 2018 16:19:00 +0000 (17:19 +0100)]
nir: add nir_opt_shrink_load pass

This is a very simple pass that just shrinks load_push_constant
intrinsics when some components are unused. For now, it can just
shrink vec4 to vec3, vec3 to vec2 and so on.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoradeonsi/nir: add nir support for compiling compute shaders
Timothy Arceri [Thu, 1 Feb 2018 23:09:47 +0000 (10:09 +1100)]
radeonsi/nir: add nir support for compiling compute shaders

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoac/radeonsi: add num_work_groups to the abi
Timothy Arceri [Fri, 2 Feb 2018 03:33:06 +0000 (14:33 +1100)]
ac/radeonsi: add num_work_groups to the abi

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoac: implement nir_intrinsic_shader_clock
Timothy Arceri [Fri, 2 Feb 2018 02:55:25 +0000 (13:55 +1100)]
ac: implement nir_intrinsic_shader_clock

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoac/radeonsi: create ac_build_shader_clock() helper
Timothy Arceri [Fri, 2 Feb 2018 02:54:48 +0000 (13:54 +1100)]
ac/radeonsi: create ac_build_shader_clock() helper

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoac/radeonsi: add load_local_group_size() to the abi
Timothy Arceri [Fri, 2 Feb 2018 02:14:41 +0000 (13:14 +1100)]
ac/radeonsi: add load_local_group_size() to the abi

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoradeonsi: add get_block_size() helper
Timothy Arceri [Fri, 2 Feb 2018 02:06:02 +0000 (13:06 +1100)]
radeonsi: add get_block_size() helper

This will be reused by the nir backend in a later patch.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoac: don't call emit_outputs() for compute
Timothy Arceri [Thu, 1 Feb 2018 23:24:16 +0000 (10:24 +1100)]
ac: don't call emit_outputs() for compute

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoac/radeonsi: add local_invocation_ids to the abi
Timothy Arceri [Thu, 1 Feb 2018 23:23:46 +0000 (10:23 +1100)]
ac/radeonsi: add local_invocation_ids to the abi

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoac/radeonsi: add workgroup_ids to the abi
Timothy Arceri [Thu, 1 Feb 2018 23:08:52 +0000 (10:08 +1100)]
ac/radeonsi: add workgroup_ids to the abi

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoradeonsi/nir: gather some compute info in si_nir_scan_shader()
Timothy Arceri [Thu, 1 Feb 2018 23:07:47 +0000 (10:07 +1100)]
radeonsi/nir: gather some compute info in si_nir_scan_shader()

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoradeonsi/nir: always set input_usage_mask as using all components
Timothy Arceri [Mon, 5 Feb 2018 22:52:47 +0000 (09:52 +1100)]
radeonsi/nir: always set input_usage_mask as using all components

This fixes a regression for now, in the future we should gather
the used components properly.

V2: just set for VS and correctly handle doubles

Fixes: be973ed21f6e "radeonsi: load the right number of components for VS inputs and TBOs"
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoi965: remove unused brw_nir_lower_cs_shared()
Timothy Arceri [Sun, 4 Feb 2018 22:15:52 +0000 (09:15 +1100)]
i965: remove unused brw_nir_lower_cs_shared()

This has been unused since 8761a04d0d93.

Reviewed-by: Elie Tournier <elie.tournier@collabora.com>
6 years agovulkan/wsi: Fix OOM behavior with prime images.
Bas Nieuwenhuizen [Tue, 6 Feb 2018 14:43:36 +0000 (15:43 +0100)]
vulkan/wsi: Fix OOM behavior with prime images.

Fixes: d50937f137 "vulkan/wsi: Implement prime in a completely generic way"
Reviewed-by: Dave Airlie <airlied@redhat.com>
6 years agoac/nir: fix GS load input type.
Bas Nieuwenhuizen [Tue, 6 Feb 2018 11:52:43 +0000 (12:52 +0100)]
ac/nir: fix GS load input type.

Fixes: df1d5174fc "ac/nir: replace SI.buffer.load.dword with amdgcn.buffer.load"
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
6 years agomesa: Factor out _mesa_disable_vertex_array_attrib.
Mathias Fröhlich [Sat, 3 Feb 2018 19:25:39 +0000 (20:25 +0100)]
mesa: Factor out _mesa_disable_vertex_array_attrib.

And use it in the enable code path.
Move _mesa_update_attribute_map_mode into its only remaining file.

Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Reviewed-by: Brian Paul <brianp@vmware.com>
6 years agovbo: Move vbo_rebase into its only caller module tnl.
Mathias Fröhlich [Sat, 16 Dec 2017 09:57:47 +0000 (10:57 +0100)]
vbo: Move vbo_rebase into its only caller module tnl.

Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Reviewed-by: Brian Paul <brianp@vmware.com>
6 years agomesa: Use atomics for buffer objects reference counts.
Mathias Fröhlich [Sat, 3 Feb 2018 14:06:16 +0000 (15:06 +0100)]
mesa: Use atomics for buffer objects reference counts.

The mutex is currently used for reference counting and updating
the minmax index cache.
The change uses atomics directly for reference counting and
the mutex for the minmax cache.
This is safe since the reference count is not modified beside
in _mesa_reference_buffer_object where atomics aim to be used.
While using the minmax cache, the calling code holds a reference
to the buffer object. Thus unreferencing or even referencing the
buffer object does not need to be serialized with accessing
the minmax cache.
The change reduces the time _mesa_reference_buffer_object_ takes
by about a factor of two when looking at perf results for some
of my favorite use cases.

Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Reviewed-by: Brian Paul <brianp@vmware.com>