mesa.git
6 years agoradv: use EOP_DATA_SEL_* instead of magic numbers
Samuel Pitoiset [Wed, 20 Jun 2018 14:10:55 +0000 (16:10 +0200)]
radv: use EOP_DATA_SEL_* instead of magic numbers

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agor600: fix copy/paste bug for sampleMaskIn workaround
Roland Scheidegger [Fri, 15 Jun 2018 21:36:52 +0000 (23:36 +0200)]
r600: fix copy/paste bug for sampleMaskIn workaround

The sampleMaskIn workaround (b936f4d1ca0d2ab1e828ff6a6e617f12469687fa)
tries to figure out if the shader is running at per-sample frequency, but
there's a typo bug so it will only recognize per-sample linar inputs,
not per-sample perspective ones.

Spotted by Eric Engestrom <eric.engestrom@intel.com>

Fixes: b936f4d1ca0d2ab1e828a "r600: partly fix sampleMaskIn value"
6 years agov3d: Fix min vs mag determination when not doing mip filtering.
Eric Anholt [Wed, 20 Jun 2018 17:22:44 +0000 (10:22 -0700)]
v3d: Fix min vs mag determination when not doing mip filtering.

Fixes all 128 failing tests in
dEQP-GLES3.functional.texture.filtering.*.combinations

6 years agovulkan: EXT_acquire_xlib_display requires libXrandr headers to build
Keith Packard [Tue, 19 Jun 2018 22:58:30 +0000 (15:58 -0700)]
vulkan: EXT_acquire_xlib_display requires libXrandr headers to build

When VK_USE_PLATFORM_XLIB_XRANDR_EXT is defined, vulkan.h includes
X11/extensions/Xrandr.h for the RROutput typedef which is used in
the vkGetRandROutputDisplayEXT interface.

Make sure we have the required header by checking during the build,
and also set CFLAGS to point at the right directory.

We don't need to link against the library as we don't use any
functions from there, so don't add the _LIBS value in the autotools
build.

Signed-off-by: Keith Packard <keithp@keithp.com>
Fixes: dbac8e25f851ed44c51f "radv: Add EXT_acquire_xlib_display to radv driver [v2]"
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
6 years agov3d: Implement ALPHA_TO_COVERAGE.
Eric Anholt [Tue, 19 Jun 2018 23:27:17 +0000 (16:27 -0700)]
v3d: Implement ALPHA_TO_COVERAGE.

There's a convenient "FTOC" instruction for generating the coverage now,
unlike vc4.  This fixes
dEQP-GLES3.functional.multisample.fbo_4_samples.proportionality_alpha_to_coverage

6 years agov3d: Track write reference to the separate stencil buffer.
Eric Anholt [Tue, 19 Jun 2018 23:00:15 +0000 (16:00 -0700)]
v3d: Track write reference to the separate stencil buffer.

Otherwise, a blit from separate stencil may fail to flush the job that
initialized it, or new drawing could fail to flush a blit reading from
stencil.

Fixes:
dEQP-GLES3.functional.fbo.blit.depth_stencil.depth32f_stencil8_basic
dEQP-GLES3.functional.fbo.blit.depth_stencil.depth32f_stencil8_scale
dEQP-GLES3.functional.fbo.blit.depth_stencil.depth32f_stencil8_stencil_only
dEQP-GLES3.functional.fbo.msaa.2_samples.depth32f_stencil8
dEQP-GLES3.functional.fbo.msaa.4_samples.depth32f_stencil8

6 years agov3d: Add missing reference to the separate stencil buffer.
Eric Anholt [Tue, 19 Jun 2018 22:48:15 +0000 (15:48 -0700)]
v3d: Add missing reference to the separate stencil buffer.

Noticed while debugging a missing flush of rendering in the z32f_s8 case.

6 years agov3d: Fix return value from fence_finish.
Eric Anholt [Tue, 19 Jun 2018 18:24:56 +0000 (11:24 -0700)]
v3d: Fix return value from fence_finish.

We needed to convert from a -errno to a boolean success value.  Fixes:

GTF-GLES3.gtf.GL3Tests.sync.sync_functionality_clientwaitsync_flush
GTF-GLES3.gtf.GL3Tests.sync.sync_functionality_clientwaitsync_signaled

6 years agomesa/st: only do scalar lowerings if driver benefits
Christian Gmeiner [Fri, 15 Jun 2018 10:29:59 +0000 (12:29 +0200)]
mesa/st: only do scalar lowerings if driver benefits

As not every (upcoming) backend compiler is happy with
nir_lower_xxx_to_scalar lowerings do them only if the backend
is scalar (and not vec4) based.

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
6 years agogallium: add scalar isa shader cap
Christian Gmeiner [Fri, 15 Jun 2018 10:29:58 +0000 (12:29 +0200)]
gallium: add scalar isa shader cap

v1 -> v2:
 - nv30 is _NOT_ scalar as suggested by Ilia Mirkin.
 - Change from a screen cap to a shader cap as suggested
   by Eric Anholt.
 - radeonsi is scalar as suggested by Marek Olšák.
 - Change missing ones to be scalar.

v2 -> v3:
 - r600 prefers vec4 as suggested by Marek Olšák.

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoradv: Add VK_EXT_display_surface_counter to radv driver
Keith Packard [Wed, 11 Oct 2017 07:20:08 +0000 (00:20 -0700)]
radv: Add VK_EXT_display_surface_counter to radv driver

This extension is required to support EXT_display_control as it offers
a way to query whether the vblank counter is supported.

Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agoanv: Add VK_EXT_display_surface_counter to anv driver [v2]
Keith Packard [Wed, 11 Oct 2017 07:20:08 +0000 (00:20 -0700)]
anv: Add VK_EXT_display_surface_counter to anv driver [v2]

This extension is required to support EXT_display_control as it offers
a way to query whether the vblank counter is supported.

v2:
Add extension to list in alphabetical order

Suggested-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agoVulkan/wsi: Implement VK_EXT_display_surface_counter
Jason Ekstrand [Sat, 16 Jun 2018 17:44:11 +0000 (10:44 -0700)]
Vulkan/wsi: Implement VK_EXT_display_surface_counter

This extension is required to support EXT_display_control as it offers a
way to query whether the vblank counter is supported.  Internally, it is
implemented using a fake MESA extension which provides a chain-in to
GetSurfaceCapabilities2KHR which contains the one added field.  This has
the advantage of reducing number of callbacks needed in the back-ends.
It also means that anything chained into GetSurfaceCapabilities2EXT
through VkSurfaceCapabilities2KHR::pNext so we only need to handle
crawling the pNext chain once per back-end.

Reviewed-by: Keith Packard <keithp@keithp.com>
6 years agovulkan/wsi: Get rid of the get_capabilities hook
Jason Ekstrand [Sat, 16 Jun 2018 17:27:40 +0000 (10:27 -0700)]
vulkan/wsi: Get rid of the get_capabilities hook

Instead, we can just use get_capabilities2.  This way back-ends only
have to implement one hook.

Reviewed-by: Keith Packard <keithp@keithp.com>
6 years agointel/aubinator: drop unused functions
Eric Engestrom [Tue, 19 Jun 2018 14:55:26 +0000 (15:55 +0100)]
intel/aubinator: drop unused functions

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
6 years agoradv: always initialize the clear depth/stencil values to 0
Samuel Pitoiset [Tue, 19 Jun 2018 14:04:30 +0000 (16:04 +0200)]
radv: always initialize the clear depth/stencil values to 0

Similar to the clear color values.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoradv: always initialize the clear color values to 0
Samuel Pitoiset [Tue, 19 Jun 2018 13:56:19 +0000 (15:56 +0200)]
radv: always initialize the clear color values to 0

Having random data in there is probably not the best.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoradv: always initialize the DCC predicate to FALSE
Samuel Pitoiset [Tue, 19 Jun 2018 13:39:25 +0000 (15:39 +0200)]
radv: always initialize the DCC predicate to FALSE

This might eventually skip some useless DCC decompression
passes.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoradv: do not use an user SGPR for the sample position offset
Samuel Pitoiset [Tue, 19 Jun 2018 12:25:48 +0000 (14:25 +0200)]
radv: do not use an user SGPR for the sample position offset

We know the number of samples at compile time.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoradv: don't store the number of samples as log2
Samuel Pitoiset [Tue, 19 Jun 2018 12:25:47 +0000 (14:25 +0200)]
radv: don't store the number of samples as log2

Needed for the following patch.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agogallium/aux/util/u_cpu_detect.h: Fix -Wsign-compare warning in u_cpu_detect.c
Gert Wollny [Tue, 5 Jun 2018 11:59:06 +0000 (13:59 +0200)]
gallium/aux/util/u_cpu_detect.h: Fix -Wsign-compare warning in u_cpu_detect.c

Change the type of util_cpu_caps::nr_cpus to int because sysconfig
returns a signed value, fixes:

u_cpu_detect.c: In function 'util_cpu_detect':
u_cpu_detect.c:317:30: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
    if (util_cpu_caps.nr_cpus == -1)

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/util/u_debug.h: Fix "noreturn" warnings in debug mode
Gert Wollny [Tue, 5 Jun 2018 11:59:05 +0000 (13:59 +0200)]
gallium/aux/util/u_debug.h: Fix "noreturn" warnings in debug mode

Only decorate function as noreturn when DEBUG is not defined, because
when compiled in DEBUG mode the function actually executes an int3 and
may return, fixes:
u_debug.c: In function '_debug_assert_fail':
u_debug.c:309:1: warning: 'noreturn' function does return

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/util: Fix some warnings
Gert Wollny [Tue, 5 Jun 2018 11:59:04 +0000 (13:59 +0200)]
gallium/aux/util: Fix some warnings

util/u_cpu_detect.c: In function 'util_cpu_detect':
util/u_cpu_detect.c:377:30: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    if (util_cpu_caps.nr_cpus == ~0u)
                              ^~

util/u_hash_table.c:274:21: warning: unused parameter 'k' [-Wunused-
parameter]
 util_hash_inc(void *k, void *v, void *d)
                     ^
util/u_hash_table.c:274:30: warning: unused parameter 'v' [-Wunused-
parameter]
 util_hash_inc(void *k, void *v, void *d)
                              ^

util/u_tests.c: In function 'test_texture_barrier':
util/u_tests.c:652:25: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
       for (int i = 0; i < num_samples / 2; i++) {
                         ^

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_ureg.c: remove unused parameter from match_or_expand_immediate64
Gert Wollny [Tue, 5 Jun 2018 11:59:03 +0000 (13:59 +0200)]
gallium/aux/tgsi_ureg.c: remove unused parameter from match_or_expand_immediate64

remove "type" from "match_or_expand_immediate64", fixes:

tgsi/tgsi_ureg.c: In function 'match_or_expand_immediate64':
tgsi/tgsi_ureg.c:837:34: warning: unused parameter 'type' [-Wunused-
parameter]
                              int type,
                                  ^~~~

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_two_side.c: Fix -Wsign-compare warnings
Gert Wollny [Tue, 5 Jun 2018 11:59:02 +0000 (13:59 +0200)]
gallium/aux/tgsi_two_side.c: Fix -Wsign-compare warnings

Integer propagation rules can sometimes be irritating. With
"unsigned x" "x + 1" gets propagated to a signed integer, so explicitely
assign the sum to an unsigned and use that for comaprison.

In file included from tgsi/tgsi_two_side.c:41:0:
tgsi/tgsi_two_side.c: In function 'xform_decl':
./util/u_math.h:660:29: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
 #define MAX2( A, B )   ( (A)>(B) ? (A) : (B) )
                             ^
tgsi/tgsi_two_side.c:86:24: note: in expansion of macro 'MAX2'
       ts->num_inputs = MAX2(ts->num_inputs, decl->Range.Last + 1);
                        ^~~~
./util/u_math.h:660:40: warning: signed and unsigned type in conditional
expression [-Wsign-compare]
 #define MAX2( A, B )   ( (A)>(B) ? (A) : (B) )
                                        ^
tgsi/tgsi_two_side.c:86:24: note: in expansion of macro 'MAX2'
       ts->num_inputs = MAX2(ts->num_inputs, decl->Range.Last + 1);
                        ^~~~
./util/u_math.h:660:29: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
 #define MAX2( A, B )   ( (A)>(B) ? (A) : (B) )
                             ^
tgsi/tgsi_two_side.c:89:23: note: in expansion of macro 'MAX2'
       ts->num_temps = MAX2(ts->num_temps, decl->Range.Last + 1);
                       ^~~~
./util/u_math.h:660:40: warning: signed and unsigned type in conditional
expression [-Wsign-compare]
 #define MAX2( A, B )   ( (A)>(B) ? (A) : (B) )
                                        ^
tgsi/tgsi_two_side.c:89:23: note: in expansion of macro 'MAX2'
       ts->num_temps = MAX2(ts->num_temps, decl->Range.Last + 1);
                       ^~~~
tgsi/tgsi_two_side.c: In function 'xform_inst':
tgsi/tgsi_two_side.c:184:45: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
             if (inst->Src[i].Register.Index == ts-
>front_color_input[j]) {
                                             ^~

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_ureg.c: Fix various warnings
Gert Wollny [Tue, 5 Jun 2018 11:59:01 +0000 (13:59 +0200)]
gallium/aux/tgsi_ureg.c: Fix various warnings

tgsi/tgsi_ureg.c: In function 'ureg_DECL_sampler':
tgsi/tgsi_ureg.c:721:34: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
       if (ureg->sampler[i].Index == nr)
                                  ^~
tgsi/tgsi_ureg.c: In function 'match_or_expand_immediate64':
tgsi/tgsi_ureg.c:837:34: warning: unused parameter 'type' [-Wunused-
parameter]
                              int type,
                                  ^~~~
tgsi/tgsi_ureg.c: In function 'emit_decls':
tgsi/tgsi_ureg.c:1821:31: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
       if (ureg->properties[i] != ~0)
                               ^~
tgsi/tgsi_ureg.c: In function 'ureg_create_with_screen':
tgsi/tgsi_ureg.c:2193:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (i = 0; i < ARRAY_SIZE(ureg->properties); i++)
                  ^

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_text.c: Fix -Wsign-compare warnings
Gert Wollny [Tue, 5 Jun 2018 11:59:00 +0000 (13:59 +0200)]
gallium/aux/tgsi_text.c: Fix -Wsign-compare warnings

tgsi/tgsi_text.c: In function 'parse_identifier':
tgsi/tgsi_text.c:218:16: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
          if (i == len - 1)
                ^~
tgsi/tgsi_text.c: In function 'parse_optional_swizzle':
tgsi/tgsi_text.c:873:21: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
       for (i = 0; i < components; i++) {
                     ^
tgsi/tgsi_text.c: In function 'parse_instruction':
tgsi/tgsi_text.c:1103:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (i = 0; i < info->num_dst + info->num_src + info->is_tex; i++) {
                  ^
tgsi/tgsi_text.c:1118:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
       else if (i < info->num_dst + info->num_src) {
                  ^
tgsi/tgsi_text.c: In function 'parse_immediate':
tgsi/tgsi_text.c:1660:24: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (type = 0; type < ARRAY_SIZE(tgsi_immediate_type_names); ++type)
{
                        ^
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_point_sprite.c: Fix -Wsign-compare warnings
Gert Wollny [Tue, 5 Jun 2018 11:58:59 +0000 (13:58 +0200)]
gallium/aux/tgsi_point_sprite.c: Fix -Wsign-compare warnings

tgsi/tgsi_lowering.c: In function 'emit_twoside':
tgsi/tgsi_lowering.c:1179:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (i = 0; i < ctx->two_side_colors; i++) {
                  ^
tgsi/tgsi_lowering.c:1208:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (i = 0; i < ctx->two_side_colors; i++) {
                  ^
tgsi/tgsi_lowering.c:1216:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (i = 0; i < ctx->two_side_colors; i++) {
                  ^
tgsi/tgsi_lowering.c: In function 'emit_decls':
tgsi/tgsi_lowering.c:1280:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (i = 0; i < ctx->numtmp; i++) {
                  ^
tgsi/tgsi_lowering.c: In function 'rename_color_inputs':
tgsi/tgsi_lowering.c:1311:28: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
             if (src->Index == ctx->two_side_idx[j]) {
                            ^~

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_lowering.c: Fix -Wsign-compare warnings
Gert Wollny [Tue, 5 Jun 2018 11:58:58 +0000 (13:58 +0200)]
gallium/aux/tgsi_lowering.c: Fix -Wsign-compare warnings

tgsi/tgsi_lowering.c: In function 'emit_twoside':
tgsi/tgsi_lowering.c:1179:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (i = 0; i < ctx->two_side_colors; i++) {
                  ^
tgsi/tgsi_lowering.c:1208:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (i = 0; i < ctx->two_side_colors; i++) {
                  ^
tgsi/tgsi_lowering.c:1216:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (i = 0; i < ctx->two_side_colors; i++) {
                  ^
tgsi/tgsi_lowering.c: In function 'emit_decls':
tgsi/tgsi_lowering.c:1280:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (i = 0; i < ctx->numtmp; i++) {
                  ^
tgsi/tgsi_lowering.c: In function 'rename_color_inputs':
tgsi/tgsi_lowering.c:1311:28: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
             if (src->Index == ctx->two_side_idx[j]) {
                            ^~

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_build.c: Fix -Wsign-compare warnings
Gert Wollny [Tue, 5 Jun 2018 11:58:57 +0000 (13:58 +0200)]
gallium/aux/tgsi_build.c: Fix -Wsign-compare warnings

tgsi/tgsi_build.c: In function 'tgsi_build_full_immediate':
tgsi/tgsi_build.c:622:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for( i = 0; i < full_imm->Immediate.NrTokens - 1; i++ ) {
                  ^
tgsi/tgsi_build.c: In function 'tgsi_build_full_property':
tgsi/tgsi_build.c:1393:18: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for( i = 0; i < full_prop->Property.NrTokens - 1; i++ ) {
                  ^

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_build.c: Remove now unused variable
Gert Wollny [Tue, 5 Jun 2018 11:58:56 +0000 (13:58 +0200)]
gallium/aux/tgsi_build.c: Remove now unused variable

Removing the unused prev_tocken from the function calls made this local
variable also unused.

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_build.c: Remove unused parameters prev_token from various functions
Gert Wollny [Tue, 5 Jun 2018 11:58:55 +0000 (13:58 +0200)]
gallium/aux/tgsi_build.c: Remove unused parameters prev_token from various functions

remove parameter prev_token unused in
   tgsi_build_instruction_label
   tgsi_build_instruction_texture
   tgsi_build_instruction_memory
   tgsi_build_texture_offset

This fixes the following warnings:

tgsi/tgsi_build.c: In function 'tgsi_build_instruction_label':
tgsi/tgsi_build.c:716:24: warning: unused parameter 'prev_token' [-
Wunused-parameter]
    struct tgsi_token  *prev_token,
                        ^~~~~~~~~~
tgsi/tgsi_build.c: In function 'tgsi_build_instruction_texture':
tgsi/tgsi_build.c:749:23: warning: unused parameter 'prev_token' [-
Wunused-parameter]
    struct tgsi_token *prev_token,
                       ^~~~~~~~~~
tgsi/tgsi_build.c: In function 'tgsi_build_instruction_memory':
tgsi/tgsi_build.c:784:23: warning: unused parameter 'prev_token' [-
Wunused-parameter]
    struct tgsi_token *prev_token,
                       ^~~~~~~~~~
tgsi/tgsi_build.c: In function 'tgsi_build_texture_offset':
tgsi/tgsi_build.c:819:23: warning: unused parameter 'prev_token' [-
Wunused-parameter]
    struct tgsi_token *prev_token,
                       ^~~~~~~~~~

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_exec.c: Fix various -Wsign-compare
Gert Wollny [Tue, 5 Jun 2018 11:58:54 +0000 (13:58 +0200)]
gallium/aux/tgsi_exec.c: Fix various -Wsign-compare

tgsi/tgsi_exec.c: In function 'exec_tex':
tgsi/tgsi_exec.c:2254:46: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
       assert(shadow_ref >= dim && shadow_ref < ARRAY_SIZE(args));
                                              ^
./util/u_debug.h:189:30: note: in definition of macro 'debug_assert'
 #define debug_assert(expr) ((expr) ? (void)0 : _debug_assert_fail(#expr,
__FILE__, __LINE__, __FUNCTION__))
                              ^~~~
tgsi/tgsi_exec.c:2254:7: note: in expansion of macro 'assert'
       assert(shadow_ref >= dim && shadow_ref < ARRAY_SIZE(args));
       ^~~~~~
tgsi/tgsi_exec.c:2290:23: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
       for (i = dim; i < ARRAY_SIZE(args); i++)
                       ^
In file included from ./util/u_memory.h:39:0,
                 from tgsi/tgsi_exec.c:62:
tgsi/tgsi_exec.c: In function 'exec_lodq':
tgsi/tgsi_exec.c:2357:15: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    assert(dim <= ARRAY_SIZE(coords));
               ^
./util/u_debug.h:189:30: note: in definition of macro 'debug_assert'
 #define debug_assert(expr) ((expr) ? (void)0 : _debug_assert_fail(#expr,
__FILE__, __LINE__, __FUNCTION__))
                              ^~~~
tgsi/tgsi_exec.c:2357:4: note: in expansion of macro 'assert'
    assert(dim <= ARRAY_SIZE(coords));
    ^~~~~~
tgsi/tgsi_exec.c:2363:20: warning: comparison between signed and
unsigned integer expressions [-Wsign-compare]
    for (i = dim; i < ARRAY_SIZE(coords); i++) {
                    ^

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_exec.c: remove superfluous parameter from etch_source_d
Gert Wollny [Tue, 5 Jun 2018 11:58:53 +0000 (13:58 +0200)]
gallium/aux/tgsi_exec.c: remove superfluous parameter from etch_source_d

Remove unused parameter src_datatype from fetch_source_d, fixes warning;

tgsi/tgsi_exec.c: In function 'fetch_source_d':
tgsi/tgsi_exec.c:1594:40: warning: unused parameter 'src_datatype' [-Wunused-parameter]
                enum tgsi_exec_datatype src_datatype)
                                        ^~~~~~~~~~~~
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_exec.c: remove superfluous parameter from store_dest_dstret
Gert Wollny [Tue, 5 Jun 2018 11:58:52 +0000 (13:58 +0200)]
gallium/aux/tgsi_exec.c: remove superfluous parameter from store_dest_dstret

remove unused parameter inst from store_dest_dstret (and consequently also from
store_dest_double), fixes warning:

tgsi/tgsi_exec.c: In Funktion »store_dest_dstret«:
tgsi/tgsi_exec.c:1765:47: Warning: unused parameter »inst« [-Wunused-parameter]
           const struct tgsi_full_instruction *inst)
                                               ^~~~

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_exec.c: Remove unused parameter from fetch_src_file_channel
Gert Wollny [Tue, 5 Jun 2018 11:58:51 +0000 (13:58 +0200)]
gallium/aux/tgsi_exec.c: Remove unused parameter from fetch_src_file_channel

remove unused parameter chan_index from fetch_src_file_channel, fixes warning:

tgsi/tgsi_exec.c: In Funktion »fetch_src_file_channel«:
tgsi/tgsi_exec.c:1480:35: Warning: unused parameter »chan_index« [-Wunused-parameter]
                        const uint chan_index,
                                   ^~~~~~~~~~

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_exec.c: Remove paramater inst from exec_kill
Gert Wollny [Tue, 5 Jun 2018 11:58:50 +0000 (13:58 +0200)]
gallium/aux/tgsi_exec.c: Remove paramater inst from exec_kill

Fixes warning:
tgsi/tgsi_exec.c: In Funktion »exec_kill«:
tgsi/tgsi_exec.c:2049:47: Warning: unused parameter »inst« [-Wunused-parameter]
           const struct tgsi_full_instruction *inst)
                                               ^~~~

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_aa_point.c: Fix -Wsign-compare warnings
Gert Wollny [Tue, 5 Jun 2018 11:58:49 +0000 (13:58 +0200)]
gallium/aux/tgsi_aa_point.c: Fix -Wsign-compare warnings

tgsi/tgsi_aa_point.c:32:0:
tgsi/tgsi_aa_point.c: In Funktion »aa_decl«:
./util/u_math.h:660:29: Comparison between signed and unsigned in
conditional expressions [-Wsign-compare]
 #define MAX2( A, B )   ( (A)>(B) ? (A) : (B) )
                             ^
tgsi/tgsi_aa_point.c:76:21: Remark: when substituting of the macro
»MAX2«
       ts->num_tmp = MAX2(ts->num_tmp, decl->Range.Last + 1);
                     ^~~~
./util/u_math.h:660:40: Warning: signed and unsigned type in conditional
expression [-Wsign-compare]
 #define MAX2( A, B )   ( (A)>(B) ? (A) : (B) )
                                        ^
tgsi/tgsi_aa_point.c:76:21: Remark: when substituting of the macro
»MAX2«
       ts->num_tmp = MAX2(ts->num_tmp, decl->Range.Last + 1);
                     ^~~~
tgsi/tgsi_aa_point.c: In Funktion »aa_inst«:
tgsi/tgsi_aa_point.c:220:31: Comparison between signed and unsigned in
conditional expressions [-Wsign-compare]
           dst->Register.Index == ts->color_out) {

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi_sanity.c: Fix -Wsign-compare warnings
Gert Wollny [Tue, 5 Jun 2018 11:58:48 +0000 (13:58 +0200)]
gallium/aux/tgsi_sanity.c: Fix -Wsign-compare warnings

tgsi_sanity.c: In function 'iter_instruction':
tgsi_sanity.c:316:29: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
       if (ctx->index_of_END != ~0) {
                             ^~
tgsi_sanity.c: In function 'epilog':
tgsi_sanity.c:488:26: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
    if (ctx->index_of_END == ~0) {

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi/tgsi_parse.c: Fix two warnings
Gert Wollny [Tue, 5 Jun 2018 11:58:47 +0000 (13:58 +0200)]
gallium/aux/tgsi/tgsi_parse.c: Fix two warnings

tgsi_parse.c: In function 'tgsi_parse_free':
tgsi_parse.c:54:31: warning: unused parameter 'ctx' [-Wunused-parameter]
    struct tgsi_parse_context *ctx )
                               ^~~
tgsi_parse.c: In function 'tgsi_parse_end_of_tokens':
tgsi_parse.c:62:25: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
    return ctx->Position >=

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/tgsi/tgsi_dump.c: Fix -Wsign-compare warnings
Gert Wollny [Tue, 5 Jun 2018 11:58:46 +0000 (13:58 +0200)]
gallium/aux/tgsi/tgsi_dump.c: Fix -Wsign-compare warnings

tgsi_dump.c: In function 'iter_property':
tgsi_dump.c:443:18: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
    for (i = 0; i < prop->Property.NrTokens - 1; ++i) {
                  ^
tgsi_dump.c:459:13: warning: comparison between signed and unsigned
integer expressions [-Wsign-compare]
       if (i < prop->Property.NrTokens - 2)

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agogallium/aux/cso_cache: Fix various warnings
Gert Wollny [Tue, 5 Jun 2018 11:58:45 +0000 (13:58 +0200)]
gallium/aux/cso_cache: Fix various warnings

cso_cache.c: In Function »delete_blend_state«:
cso_cache/cso_cache.c:90:51: Warning: unused parameter »data« [-Wunused-
parameter]
 static void delete_blend_state(void *state, void *data)
                                                   ^~~~
cso_cache/cso_cache.c: In Funktion »delete_depth_stencil_state«:
cso_cache/cso_cache.c:98:59: Warning: unused parameter »data« [-Wunused-
parameter]
 static void delete_depth_stencil_state(void *state, void *data)
                                                           ^~~~
cso_cache/cso_cache.c: In Funktion »delete_sampler_state«:
cso_cache/cso_cache.c:106:53: Warning: unused parameter »data« [-
Wunused-parameter]
 static void delete_sampler_state(void *state, void *data)
                                                     ^~~~
cso_cache/cso_cache.c: In Funktion »delete_rasterizer_state«:
cso_cache/cso_cache.c:114:56: Warning: unused parameter »data« [-
Wunused-parameter]
 static void delete_rasterizer_state(void *state, void *data)
                                                        ^~~~
cso_cache/cso_cache.c: In Funktion »delete_velements«:
cso_cache/cso_cache.c:122:49: Warning: unused parameter »data« [-
Wunused-parameter]
 static void delete_velements(void *state, void *data)
                                                 ^~~~
cso_cache/cso_cache.c: In Funktion »sanitize_cb«:
cso_cache/cso_cache.c:166:52: Warning: unused parameter »user_data« [-
Wunused-parameter]
                                int max_size, void *user_data)
                                                    ^~~~~~~~~
gallium/aux/cso_context.c: a -Wunused-parameter warning

cso_cache/cso_context.c: In Funktion »delete_sampler_state«:
cso_cache/cso_context.c:163:57: Warning: unused parameter »ctx« [-
Wunused-parameter]
 static boolean delete_sampler_state(struct cso_context *ctx, void
*state)
                                                         ^~~

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
6 years agoconfigure.ac: Add CFLAG -Wno-missing-field-initializers (v5)
Gert Wollny [Mon, 11 Jun 2018 16:24:39 +0000 (18:24 +0200)]
configure.ac: Add CFLAG -Wno-missing-field-initializers (v5)

This warning is misleading: When a struct is partially initialized without
assigning to the structure members by name, then the remaining fields
will be zeroed out, and this warning will be issued (if enabled). If, on the
other hand, the partial initialization is done by assigning to named members,
the remaining structure elements may hold random data, but the warning is not
issued. Since in Mesa the first approach to initialize structure elements is
used very often, and it is usually assumed that the remaining elements are
zeroed out, heeding this warning would be counter-productive.

v2: - add -Wno-missing-field-initializers to meson-build
    - fix empty line error
    (both Eric Engestrom)

v3: * check for -Wmissing-field-initializers warning and then disable it
      because gcc and clang always accept -Wno-* (Dylan Baker)
    * Also disable this warning for C++

v4: * meson.build add -Wno-missing-field-initializers to
      c_args instead of no_override_init_args (Eric Engstrom)

v5: * configure.ac: Correct copy/paste error with CFLAGS/CXXFLAGS

Reviewed-by: Marek Olšák <marek.olsak@amd.com> (v1)
Reviewed-by: Emil Velikov <emil.velikov@collabora.com> (v2)
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
6 years agoradv: remove unnecessary code around CACHE_FLUSH_AND_INV_TS_EVENT
Samuel Pitoiset [Tue, 19 Jun 2018 13:24:39 +0000 (15:24 +0200)]
radv: remove unnecessary code around CACHE_FLUSH_AND_INV_TS_EVENT

AMDVLK also always uses CACHE_FLUSH_AND_INV_TS_EVENT. The other
workaround is to flush DB metadata after emitting the framebuffer,
but that seems slower.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoradv: Fix flush_bits being used uninitialized.
Bas Nieuwenhuizen [Tue, 19 Jun 2018 21:48:46 +0000 (23:48 +0200)]
radv: Fix flush_bits being used uninitialized.

A case of making things worse while trying to fix something minor ...

Fixes: ef79457004e "radv: Merge the flush bits of CMASK & DCC clear."
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
6 years agoradv: Add EXT_acquire_xlib_display to radv driver [v2]
Keith Packard [Fri, 9 Feb 2018 15:45:58 +0000 (07:45 -0800)]
radv: Add EXT_acquire_xlib_display to radv driver [v2]

This extension adds the ability to borrow an X RandR output for
temporary use directly by a Vulkan application to the radv driver.

v2:
Simplify addition of VK_USE_PLATFORM_XLIB_XRANDR_KHR to
vulkan_wsi_args

Suggested-by: Eric Engestrom <eric.engestrom@imgtec.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agoanv: Add EXT_acquire_xlib_display to anv driver [v3]
Keith Packard [Fri, 9 Feb 2018 15:45:58 +0000 (07:45 -0800)]
anv: Add EXT_acquire_xlib_display to anv driver [v3]

This extension adds the ability to borrow an X RandR output for
temporary use directly by a Vulkan application to the anv driver.

v2:
Simplify addition of VK_USE_PLATFORM_XLIB_XRANDR_KHR to
vulkan_wsi_args

Suggested-by: Eric Engestrom <eric.engestrom@imgtec.com>
v3:
Add extension to list in alphabetical order

Suggested-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agovulkan: Add EXT_acquire_xlib_display [v5]
Keith Packard [Fri, 9 Feb 2018 15:45:58 +0000 (07:45 -0800)]
vulkan: Add EXT_acquire_xlib_display [v5]

This extension adds the ability to borrow an X RandR output for
temporary use directly by a Vulkan application. For DRM, we use the
Linux resource leasing mechanism.

v2:
Clean up xlib_lease detection

* Use separate temporary '_xlib_lease' variable to hold the
  option value to avoid changin the type of a variable.

* Use boolean expressions instead of additional if statements
  to compute resulting with_xlib_lease value.

* Simplify addition of VK_USE_PLATFORM_XLIB_XRANDR_KHR to
          vulkan_wsi_args

Suggested-by: Eric Engestrom <eric.engestrom@imgtec.com>
Move mode list from wsi_display to wsi_display_connector

Fix scope for wsi_display_mode and wsi_display_connector allocs

Suggested-by: Jason Ekstrand <jason@jlekstrand.net>
v3:
Adopt Jason Ekstrand's coding conventions

Declare variables at first use, eliminate extra whitespace
between types and names. Wrap lines to 80 columns.

Explicitly forbid multiple DRM leases. Making the code support
this looks tricky and will require additional thought.

Use xcb_randr_output_t throughout the internals of the
implementation. Convert at the public API
(wsi_get_randr_output_display).

Clean up check for usable active_crtc (possible when only the
desired output is connected to the crtc).

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
v4:
Move output resource fetching closer to use in
wsi_display_get_output. This simplifies the error returns in
earlier parts of the code a bit.

Return VK_ERROR_INITIALIZATION_FAILED from
wsi_acquire_xlib_display. Jason says this is the right error
message.

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
v5:
randr doesn't pass vscan over the wire, so we set vscan to 0
for randr-acquired modes, and test wsi modes for vscan <= 1
when comparing against randr modes.

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agoradv: Add EXT_direct_mode_display to radv driver
Keith Packard [Fri, 9 Feb 2018 15:38:32 +0000 (07:38 -0800)]
radv: Add EXT_direct_mode_display to radv driver

Add support for the EXT_direct_mode_display extension. This just
provides the vkReleaseDisplayEXT function.

Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agoanv: Add EXT_direct_mode_display to anv driver [v2]
Keith Packard [Fri, 9 Feb 2018 15:38:32 +0000 (07:38 -0800)]
anv: Add EXT_direct_mode_display to anv driver [v2]

Add support for the EXT_direct_mode_display extension. This just
provides the vkReleaseDisplayEXT function.

v2: Add extension to list in alphabetical order

Suggested-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agovulkan: Add EXT_direct_mode_display [v2]
Keith Packard [Fri, 9 Feb 2018 15:38:32 +0000 (07:38 -0800)]
vulkan: Add EXT_direct_mode_display [v2]

Add support for the EXT_direct_mode_display extension. This just
provides the vkReleaseDisplayEXT function.

v2:
Adopt Jason Ekstrand's coding conventions

Declare variables at first use, eliminate extra whitespace
between types and names. Wrap lines to 80 columns.

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agoradv: Add KHR_display extension to radv [v5]
Keith Packard [Wed, 7 Feb 2018 18:31:44 +0000 (10:31 -0800)]
radv: Add KHR_display extension to radv [v5]

This adds support for the KHR_display extension to the radv Vulkan
driver. The driver now attempts to open the master DRM node when the
KHR_display extension is requested so that the common winsys code can
perform the necessary operations.

v2:
* Simplify addition of VK_USE_PLATFORM_DISPLAY_KHR to
          vulkan_wsi_args

Suggested-by: Eric Engestrom <eric.engestrom@imgtec.com>
v3:
Adapt to new wsi_device_init API (added display_fd)

v4:
Adopt Jason Ekstrand's coding conventions

Declare variables at first use, eliminate extra whitespace
between types and names. Wrap lines to 80 columns.

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
v5:
Add vkCreateDisplayModeKHR. This doesn't actually create
new modes, it only looks to see if the requested parameters
matches an existing mode and returns that.

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agoanv: Add KHR_display extension to anv [v7]
Keith Packard [Wed, 7 Feb 2018 18:31:44 +0000 (10:31 -0800)]
anv: Add KHR_display extension to anv [v7]

This adds support for the KHR_display extension to the anv Vulkan
driver. The driver now attempts to open the master DRM node when the
KHR_display extension is requested so that the common winsys code can
perform the necessary operations.

v2: Make sure primary fd is usable

When KHR_display is selected, we try to open the primary node
instead of the render node in case the user wants to use
KHR_display for presentation. However, if we're actually going
to end up using RandR leases, then we don't care if the
resulting fd can't be used for display, but the kernel also
prevents us from using it for drawing when someone else has
master.

v3:
Simplify addition of VK_USE_PLATFORM_DISPLAY_KHR to vulkan_wsi_args

Suggested-by: Eric Engestrom <eric.engestrom@imgtec.com>
v4:
Adapt primary node usage to new wsi_device_init API

v5:
Adopt Jason Ekstrand's coding conventions

        Declare variables at first use, eliminate extra whitespace between
        types and names. Wrap lines to 80 columns.

Remove spurious MM_PER_PIXEL define

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
v6:
Open DRM master before initializing WSI layer.

The DRM master FD is passed to the WSI layer during
initialization, so we need to open the device slightly earlier
in the function.

Close DRM master in device_finish.

Use anv_gem_get_param to detect working master_fd instead of
directly using the ioctl.

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
v7:
Add vkCreateDisplayModeKHR. This doesn't actually create
new modes, it only looks to see if the requested parameters
matches an existing mode and returns that.

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agovulkan: Add KHR_display extension using DRM [v10]
Keith Packard [Wed, 7 Feb 2018 18:31:44 +0000 (10:31 -0800)]
vulkan: Add KHR_display extension using DRM [v10]

This adds support for the KHR_display extension support to the vulkan
WSI layer. Driver support will be added separately.

v2:
* fix double ;; in wsi_common_display.c

* Move mode list from wsi_display to wsi_display_connector

* Fix scope for wsi_display_mode andwsi_display_connector
          allocs

* Switch all allocations to vk_zalloc instead of vk_alloc.

* Fix DRM failure in
          wsi_display_get_physical_device_display_properties

  When DRM fails, or when we don't have a master fd
  (presumably due to application errors), just return 0
  properties from this function, which is at least a valid
  response.

* Use vk_outarray for all property queries

  This is a bit less error-prone than open-coding the same
  stuff.

* Remove VK_COMPOSITE_ALPHA_INHERIT_BIT_KHR from surface caps

  Until we have multi-plane support, we shouldn't pretend to
  have any multi-plane semantics, even if undefined.

Suggested-by: Jason Ekstrand <jason@jlekstrand.net>
* Simplify addition of VK_USE_PLATFORM_DISPLAY_KHR to
          vulkan_wsi_args

Suggested-by: Eric Engestrom <eric.engestrom@imgtec.com>
v3:
Add separate 'display_fd' and 'render_fd' arguments to
wsi_device_init API. This allows drivers to use different FDs
for the different aspects of the device.

Use largest mode as display size when no preferred mode.

If the display doesn't provide a preferred mode, we'll assume
that the largest supported mode is the "physical size" of the
device and report that.

v4:
Make wsi_image_state enumeration values uppercase.
Follow more common mesa conventions.

Remove 'render_fd' from wsi_device_init API.  The
wsi_common_display code doesn't use this fd at all, so stop
passing it in. This avoids any potential confusion over which
fd to use when creating display-relative object handles.

Remove call to wsi_create_prime_image which would never have
been reached as the necessary condition (use_prime_blit) is
never set.

whitespace cleanups in wsi_common_display.c

Suggested-by: Jason Ekstrand <jason@jlekstrand.net>
Add depth/bpp info to available surface formats.  Instead of
hard-coding depth 24 bpp 32 in the drmModeAddFB call, use the
requested format to find suitable values.

Destroy kernel buffers and FBs when swapchain is destroyed. We
were leaking both of these kernel objects across swapchain
destruction.

Note that wsi_display_wait_for_event waits for anything to
happen.  wsi_display_wait_for_event is simply a yield so that
the caller can then check to see if the desired state change
has occurred.

Record swapchain failures in chain for later return. If some
asynchronous swapchain activity fails, we need to tell the
application eventually. Record the failure in the swapchain
and report it at the next acquire_next_image or queue_present
call.

Fix error returns from wsi_display_setup_connector.  If a
malloc failed, then the result should be
VK_ERROR_OUT_OF_HOST_MEMORY. Otherwise, the associated ioctl
failed and we're either VT switched away, or our lease has
been revoked, in which case we should return
VK_ERROR_OUT_OF_DATE_KHR.

Make sure both sides of if/else brace use matches

Note that we assume drmModeSetCrtc is synchronous. Add a
comment explaining why we can idle any previous displayed
image as soon as the mode set returns.

Note that EACCES from drmModePageFlip means VT inactive.  When
vt switched away drmModePageFlip returns EACCES. Poll once a
second waiting until we get some other return value back.

Clean up after alloc failure in
wsi_display_surface_create_swapchain. Destroy any created
images, free the swapchain.

Remove physical_device from wsi_display_init_wsi. We never
need this value, so remove it from the API and from the
internal wsi_display structure.

Use drmModeAddFB2 in wsi_display_image_init.  This takes a drm
format instead of depth/bpp, which provides more control over
the format of the data.

v5:
Set the 'currentStackIndex' member of the
VkDisplayPlanePropertiesKHR record to zero, instead of
indexing across all displays. This value is the stack depth of
the plane within an individual display, and as the current
code supports only a single plane per display, should be set
to zero for all elements

Discovered-by: David Mao <David.Mao@amd.com>
v6:
Remove 'platform_display' bits from the build and use the
existing 'platform_drm' instead.

v7:
Ensure VK_ICD_WSI_PLATFORM_MAX is large enough by
setting to VK_ICD_WSI_PLATFORM_DISPLAY + 1

v8:
Simplify wsi_device_init failure from wsi_display_init_wsi
by using the same pattern as the other wsi layers.

    Adopt Jason Ekstrand's white space and variable declaration
suggestions. Declare variables at first use, eliminate extra
whitespace between types and names, add list iterator helpers,
switch to lower-case list_ macros.

    Respond to Jason's April 8 review:

* Create a function to convert relative to absolute timeouts
          to catch overflow issues in one place

* use VK_NULL_HANDLE to clear prop->currentDisplay

* Get rid of available_present_modes array.

* return OUT_OF_DATE_KHR when display_queue_next called after
  display has been released.

* Make errors from mode setting fatal in display_queue_next

* Remove duplicate pthread_mutex_init call

* Add wsi_init_pthread_cond_monotonic helper function to
  isolate pthread error handling from wsi_display_init_wsi

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
v9:
Fix vscan handling by using MAX2(vscan, 1) everywhere. Vscan
can be zero anywhere, which is treated the same as 1.

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
v10:
Respond to Vulkan CTS failures.

1. Initialize planeReorderPossible in display_properties code

2. Only report connected displays in
   get_display_plane_supported_displays

3. Return VK_ERROR_OUT_OF_HOST_MEMORY when pthread cond
   initialization fails.

Signed-off-by: Jason Ekstrand <jason.ekstrand@intel.com>
4. Add vkCreateDisplayModeKHR. This doesn't actually create
   new modes, it only looks to see if the requested parameters
   matches an existing mode and returns that.

Suggested-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Keith Packard <keithp@keithp.com>
6 years agoradv: Merge the flush bits of CMASK & DCC clear.
Bas Nieuwenhuizen [Tue, 19 Jun 2018 08:05:20 +0000 (10:05 +0200)]
radv: Merge the flush bits of CMASK & DCC clear.

Probably won't be much different in practice, but still wrong.

Fixes Coverity issue 1435002.

Not CC'ing to stable since this is only hit if you enable MSAA
DCC via RADV_DEBUG.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
6 years agoradv: Don't check for pipeline being set in draw.
Bas Nieuwenhuizen [Tue, 19 Jun 2018 08:03:20 +0000 (10:03 +0200)]
radv: Don't check for pipeline being set in draw.

Draws without pipeline are definitely not allowed.

Fixes Coverity issue 1434216.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
6 years agoradeonsi: rename r600_texture -> si_texture, rxxx -> xxx or sxxx
Marek Olšák [Tue, 19 Jun 2018 01:34:57 +0000 (21:34 -0400)]
radeonsi: rename r600_texture -> si_texture, rxxx -> xxx or sxxx

Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agoamd,radeonsi: rename radeon_winsys_cs -> radeon_cmdbuf
Marek Olšák [Tue, 19 Jun 2018 01:07:10 +0000 (21:07 -0400)]
amd,radeonsi: rename radeon_winsys_cs -> radeon_cmdbuf

Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
6 years agofreedreno/a5xx: move emit_marker5() into a5xx backend
Rob Clark [Mon, 18 Jun 2018 22:22:29 +0000 (18:22 -0400)]
freedreno/a5xx: move emit_marker5() into a5xx backend

The scratch registers move again in a6xx.. so for post-a4xx let's just
move this into the backend, and move the one place it used to be needed
in core into fd5_emit_ib().  For a6xx we will do similar, calling
emit_marker6() from fd6_emit_ib().

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno/a5xx: fix crash in dEQP-GLES31.stress.vertex_attribute_binding.buffer_bound...
Rob Clark [Thu, 14 Jun 2018 13:34:11 +0000 (09:34 -0400)]
freedreno/a5xx: fix crash in dEQP-GLES31.stress.vertex_attribute_binding.buffer_bounds.bind_vertex_buffer_offset_near_wrap_10

This is kind of a hack, but really the only problem is the
debug_assert() in OUT_RELOC().  But the debug_assert() is
useful to catch real issues.  So just add some #ifdef DEBUG
code to filter things out before we hit the assert.

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno/a5xx: don't crash if compute shader compile fails
Rob Clark [Wed, 13 Jun 2018 16:46:17 +0000 (12:46 -0400)]
freedreno/a5xx: don't crash if compute shader compile fails

It is impolite, and a bit annoying with dEQP (all tests running in
single process).

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno/ir3: fix missing recursion into block condition
Rob Clark [Wed, 13 Jun 2018 14:50:37 +0000 (10:50 -0400)]
freedreno/ir3: fix missing recursion into block condition

Fixes a problem seen with dEQP-GLES31.functional.ssbo.layout.single_basic_array.shared.row_major_mat4

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno/a5xx: better FOUR_QUAD/TWO_QUAD decision for compute
Rob Clark [Wed, 13 Jun 2018 13:50:34 +0000 (09:50 -0400)]
freedreno/a5xx: better FOUR_QUAD/TWO_QUAD decision for compute

If we aren't going to get full occupancy, then use TWO_QUAD.

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno/a5xx: bordercolor fixes
Rob Clark [Tue, 12 Jun 2018 13:56:12 +0000 (09:56 -0400)]
freedreno/a5xx: bordercolor fixes

Need a bit of hand-holding for stencil bordercolor, and add border color
values for sRGB.

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno: remove per-stateobj dirty_mask's
Rob Clark [Mon, 11 Jun 2018 18:05:19 +0000 (14:05 -0400)]
freedreno: remove per-stateobj dirty_mask's

These never got updated in fd_context_all_dirty() so actually trying to
rely on them (in the case of fd5_emit_images()) ends up in some cases
where state is not emitted but should be.  Best to just rip this out.

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno/a5xx: remove one image stateblock
Rob Clark [Sun, 10 Jun 2018 15:35:56 +0000 (11:35 -0400)]
freedreno/a5xx: remove one image stateblock

I think this ends up just setting uniform/const memory.  But we upload
x/y/z stride differently.  At best this is unneeded, at worst it could
possibly clobber other uniform/const memory.

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno/a5xx: cubemap image fixes
Rob Clark [Sun, 10 Jun 2018 15:34:37 +0000 (11:34 -0400)]
freedreno/a5xx: cubemap image fixes

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno/ir3: handle image buffer
Rob Clark [Thu, 7 Jun 2018 19:00:32 +0000 (15:00 -0400)]
freedreno/ir3: handle image buffer

Similar to txf case, we need to insert a 2nd coordinate (zero).

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno/ir3: handle arrays of images
Rob Clark [Sun, 3 Jun 2018 16:56:09 +0000 (12:56 -0400)]
freedreno/ir3: handle arrays of images

Unlike textures, this doesn't get lowered for us.  (Would be nice
if they were.. at least until we are ready to deal w/ indirect
indexing..)

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno/ir3: images can be arrays too
Rob Clark [Sat, 2 Jun 2018 00:20:43 +0000 (20:20 -0400)]
freedreno/ir3: images can be arrays too

Seems I previously toally forgot about 2d-arrays, etc..

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agofreedreno/ir3: use move_load_const pass
Rob Clark [Tue, 5 Jun 2018 17:42:21 +0000 (13:42 -0400)]
freedreno/ir3: use move_load_const pass

Signed-off-by: Rob Clark <robdclark@gmail.com>
6 years agonir: add pass to move load_const
Rob Clark [Wed, 6 Jun 2018 14:11:45 +0000 (10:11 -0400)]
nir: add pass to move load_const

Run this pass late (after opt loop) to move load_const instructions back
into the basic blocks which use the result, in cases where a load_const
is only consumed in a single block.

This helps reduce register usage in cases where the backend driver
cannot lower the load_const to a uniform.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
6 years agomesa/st/nir: fix driver_location for arrays of image/sampler
Rob Clark [Mon, 11 Jun 2018 18:49:12 +0000 (14:49 -0400)]
mesa/st/nir: fix driver_location for arrays of image/sampler

We can have arrays of images or samplers.  But I forgot to handle that
case long ago.  Suprised no one complained yet.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
6 years agonir: add comment for loop_unroll pass
Rob Clark [Fri, 15 Jun 2018 20:12:23 +0000 (16:12 -0400)]
nir: add comment for loop_unroll pass

Save the next person from digging through the code to figure out what
the indirect_mask parameter actually does.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
6 years agoglsl: fix random typo
Rob Clark [Fri, 15 Jun 2018 20:11:48 +0000 (16:11 -0400)]
glsl: fix random typo

Just something I stumbled across.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
6 years agoradeonsi: ignore PIPE_RESOURCE_FLAG_MAP_COHERENT
Marek Olšák [Sat, 9 Jun 2018 02:29:55 +0000 (22:29 -0400)]
radeonsi: ignore PIPE_RESOURCE_FLAG_MAP_COHERENT

We treat coherent and non-coherent buffers the same.

And move external_usage for better packing.

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
6 years agoradeonsi: always put persistent buffers into GTT on radeon
Marek Olšák [Sat, 9 Jun 2018 01:48:23 +0000 (21:48 -0400)]
radeonsi: always put persistent buffers into GTT on radeon

This improves performance for certain games.

Cc: 18.1 <mesa-stable@lists.freedesktop.org>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
6 years agoradeonsi: fix si_get_num_queries for radeon
Marek Olšák [Sat, 9 Jun 2018 01:34:55 +0000 (21:34 -0400)]
radeonsi: fix si_get_num_queries for radeon

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
6 years agoradeonsi: don't expose performance counters for non-existent blocks
Marek Olšák [Fri, 8 Jun 2018 23:21:09 +0000 (19:21 -0400)]
radeonsi: don't expose performance counters for non-existent blocks

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
6 years agoac/gpu_info: add radeon_info::num_tcc_blocks
Marek Olšák [Fri, 8 Jun 2018 23:09:02 +0000 (19:09 -0400)]
ac/gpu_info: add radeon_info::num_tcc_blocks

The values for the radeon winsys were copied from the kernel driver.

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
6 years agoradeonsi: set a better NUM_PATCHES hard limit
Marek Olšák [Tue, 5 Jun 2018 06:38:06 +0000 (02:38 -0400)]
radeonsi: set a better NUM_PATCHES hard limit

AMDVLK uses 64 (distributed) and 16 (non-distributed).
radeonsi will use 63 and 16.
* This might improve tessellation performance on Hawaii, Bonaire, Tahiti,
  Pitcairn. (they will use 16)
* I'm not sure if this matters for 1 SE configs.

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
6 years agoradeonsi: make sure LS-HS vector lanes are reasonably occupied
Marek Olšák [Tue, 5 Jun 2018 06:09:52 +0000 (02:09 -0400)]
radeonsi: make sure LS-HS vector lanes are reasonably occupied

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
6 years agoradeonsi: properly compute an LS-HS thread group size limit
Marek Olšák [Tue, 5 Jun 2018 05:20:23 +0000 (01:20 -0400)]
radeonsi: properly compute an LS-HS thread group size limit

"64 / max * 4" is less than "64 * 4 / max".

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
6 years agov3d: Fix blitting from a linear winsys BO.
Eric Anholt [Mon, 18 Jun 2018 22:46:53 +0000 (15:46 -0700)]
v3d: Fix blitting from a linear winsys BO.

This is the case for the simulator environment, and broke many blitter
tests by trying to texture from linear while the HW can only actually do
UIF/UBLINEAR/LT.  Just make a temporary and copy into it with the CPU,
then blit from that.

This is the kind of path that should use the TFU, but I haven't exposed
that hardware yet.

Fixes dEQP-GLES3.functional.fbo.blit.default_framebuffer.*

6 years agov3d: Add missing always_flush debug flag.
Eric Anholt [Mon, 18 Jun 2018 22:00:04 +0000 (15:00 -0700)]
v3d: Add missing always_flush debug flag.

The #define existed and was checked in the driver.

6 years agovirgl: Remove debugging left-overs
Tomeu Vizoso [Mon, 18 Jun 2018 12:50:51 +0000 (14:50 +0200)]
virgl: Remove debugging left-overs

Some fprintfs were probably left unintentionally a few years ago and are
a bit of a nuisance.

Fixes: 2d3301e4d513 ("virgl: fix reference counting of prime handles")
Cc: Rob Herring <robh@kernel.org>
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
6 years agoglsl: fix desktop glsl linking regression
Timothy Arceri [Tue, 19 Jun 2018 07:52:00 +0000 (17:52 +1000)]
glsl: fix desktop glsl linking regression

The prog->Shaders[i]->IsES check was accidentally removed causing
ES linking rules to be applied to desktop GLSL.

Fixes: 725b1a406dbe ("mesa/util: add allow_glsl_relaxed_es driconfig override")
6 years agoutil: add allow_glsl_relaxed_es to drirc for Google Earth VR
Timothy Arceri [Thu, 14 Jun 2018 01:00:25 +0000 (11:00 +1000)]
util: add allow_glsl_relaxed_es to drirc for Google Earth VR

Reviewed-by: Dave Airlie <airlied@redhat.com>
6 years agomesa/util: add allow_glsl_relaxed_es driconfig override
Timothy Arceri [Thu, 14 Jun 2018 01:00:24 +0000 (11:00 +1000)]
mesa/util: add allow_glsl_relaxed_es driconfig override

This relaxes a number of ES shader restrictions allowing shaders
to follow more desktop GLSL like rules.

This initial implementation relaxes the following:

 - allows linking ES shaders with desktop shaders
 - allows mismatching precision qualifiers
 - always enables standard derivative builtins

These relaxations allow Google Earth VR shaders to compile.

Reviewed-by: Dave Airlie <airlied@redhat.com>
6 years agoutil: add allow_glsl_builtin_const_expression to drirc for Google Earth VR
Timothy Arceri [Thu, 14 Jun 2018 01:00:23 +0000 (11:00 +1000)]
util: add allow_glsl_builtin_const_expression to drirc for Google Earth VR

Reviewed-by: Dave Airlie <airlied@redhat.com>
6 years agomesa/util: add allow_glsl_builtin_const_expression driconf override
Timothy Arceri [Thu, 14 Jun 2018 01:00:22 +0000 (11:00 +1000)]
mesa/util: add allow_glsl_builtin_const_expression driconf override

Google Earth VR shaders uses builtins in constant expressions with
GLSL 1.10. That feature wasn't allowed until GLSL 1.20.

Reviewed-by: Dave Airlie <airlied@redhat.com>
6 years agoutil: manually extract the program name from program_invocation_name
Timothy Arceri [Thu, 14 Jun 2018 01:00:21 +0000 (11:00 +1000)]
util: manually extract the program name from program_invocation_name

Glibc has the same code to get program_invocation_short_name. However
for some reason the short name gets mangled for some wine apps.

For example with Google Earth VR I get:

program_invocation_name:
"/home/tarceri/.local/share/Steam/steamapps/common/EarthVR/Earth.exe"

program_invocation_short_name:
"e"

Acked-by: Eric Engestrom <eric.engestrom@intel.com>
6 years agoac/surface: Set compressZ for stencil-only surfaces.
Bas Nieuwenhuizen [Mon, 18 Jun 2018 14:29:16 +0000 (16:29 +0200)]
ac/surface: Set compressZ for stencil-only surfaces.

We HTILE compress stencil-only surfaces too.

CC: 18.1 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agoanv: Use a single global API patch version
Jason Ekstrand [Sun, 17 Jun 2018 23:28:02 +0000 (16:28 -0700)]
anv: Use a single global API patch version

The Vulkan API has only one patch version shared among all of the
major.minor versions.  We should also advertise the same patch version
regardless of major.minor.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106941
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
6 years agoradeonsi: enable OpenGL 3.3 compat profile
Timothy Arceri [Mon, 18 Jun 2018 05:23:20 +0000 (15:23 +1000)]
radeonsi: enable OpenGL 3.3 compat profile

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
6 years agomesa: add ff fragment shader support for geom and tess shaders
Timothy Arceri [Mon, 18 Jun 2018 02:36:54 +0000 (12:36 +1000)]
mesa: add ff fragment shader support for geom and tess shaders

This is required for compatibility profile support.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
6 years agov3d: Set the SO offsets correctly if we have to re-emit.
Eric Anholt [Sat, 16 Jun 2018 00:08:29 +0000 (17:08 -0700)]
v3d: Set the SO offsets correctly if we have to re-emit.

This should fix TF across a glFlush() or TF pause/restart.  Fixes
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_float
and many, many others.

6 years agogallium/hud: = should rename the last added data source
Marek Olšák [Fri, 8 Jun 2018 23:49:22 +0000 (19:49 -0400)]
gallium/hud: = should rename the last added data source

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
6 years agoanv: Disable constant buffer 0 being relative.
Rafael Antognolli [Fri, 15 Jun 2018 18:44:28 +0000 (11:44 -0700)]
anv: Disable constant buffer 0 being relative.

If we are on gen8+ and have context isolation support, just make that
constant buffer address be absolute, so we can use it for push UBOs too.

v2: Do not duplicate constant_buffer_0_is_relative flag (Jason)

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
6 years agoanv/device: Check for kernel support of context isolation.
Rafael Antognolli [Fri, 15 Jun 2018 16:31:25 +0000 (09:31 -0700)]
anv/device: Check for kernel support of context isolation.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>