platform/upstream/mesa.git
2 years agonir: Remove gather_info after removing dead vars
Konstantin Seurer [Sat, 16 Jul 2022 11:33:41 +0000 (13:33 +0200)]
nir: Remove gather_info after removing dead vars

Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16593>

2 years agonir: fix typo in lower_double options handling
Timothy Arceri [Thu, 10 Nov 2022 10:24:48 +0000 (21:24 +1100)]
nir: fix typo in lower_double options handling

Seems the intention was to check that both flags were not enabled
instead we were checking that the floor flag was both set and not
set so the result would always be false.

Fixes: 3749a6ecd282 ("nir: honor lower_double options for ffloor and ffract")
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19642>

2 years agofreedreno/cffdec: Add NOP debug messages
Mark Collins [Thu, 20 Oct 2022 18:35:57 +0000 (18:35 +0000)]
freedreno/cffdec: Add NOP debug messages

We want to encode special messages into the CS that can be printed
by cffdec, these messages have identifiers that describe their
usage (message, scope begin and scope end) which allow for an
improved trace navigation experience due to the additional
information.

Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18271>

2 years agotu: Implement utrace CS marker support
Mark Collins [Thu, 20 Oct 2022 18:28:33 +0000 (18:28 +0000)]
tu: Implement utrace CS marker support

Adds support for emitting utrace markers into the CS, this allows
for useful debug information that can be decoded from a recorded
command stream.

Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18271>

2 years agocommon/utrace: Add CS logging support
Mark Collins [Thu, 20 Oct 2022 18:26:47 +0000 (18:26 +0000)]
common/utrace: Add CS logging support

Viewing CS traces retrieved from the driver is common practice to
determine driver bugs but there is no way to determine what
function a certain part of the CS was emitted by. This is crucial
information to determine what function is responsible for emitting
broken CS packets and to help with navigation of the CS trace.

Signed-off-by: Mark Collins <mark@igalia.com>
Reviewed-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Ack-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18271>

2 years agocommon/utrace: Prefix all environment variables with `MESA_`
Mark Collins [Thu, 20 Oct 2022 12:16:39 +0000 (12:16 +0000)]
common/utrace: Prefix all environment variables with `MESA_`

To be more consistent with other environment variables and ensure
better scoping, all environment variables in utrace have now been
prefixed with `MESA_`.

Signed-off-by: Mark Collins <mark@igalia.com>
Reviewed-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Ack-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18271>

2 years agocommon/utrace: Rename `u_trace_context_actively_tracing` to `u_trace_should_process`
Mark Collins [Thu, 20 Oct 2022 12:13:51 +0000 (12:13 +0000)]
common/utrace: Rename `u_trace_context_actively_tracing` to `u_trace_should_process`

Signed-off-by: Mark Collins <mark@igalia.com>
Reviewed-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Ack-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18271>

2 years agocommon/utrace: Refactor and combine all envvars into `GPU_TRACES`
Mark Collins [Wed, 19 Oct 2022 11:29:36 +0000 (19:29 +0800)]
common/utrace: Refactor and combine all envvars into `GPU_TRACES`

All environment variables involved in utrace usage were very
fragmented and convoluted to decode the meaning of, this commit has
simplified them down into easier to understand flags which directly
indicate the resulting behavior (such as `perfetto` enabling queued
logs rather than needing to set a `queued` flag) while combining
them into a single envvar `GPU_TRACES` and updating existing
terminology in utrace to match up with the new options.

Signed-off-by: Mark Collins <mark@igalia.com>
Reviewed-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Ack-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18271>

2 years agotu: Move utrace arguments to `start` from `end`
Mark Collins [Fri, 16 Sep 2022 14:18:42 +0000 (14:18 +0000)]
tu: Move utrace arguments to `start` from `end`

All arguments in Turnip code are fit to be moved to the start
event where they fit better as any sequential logging should print
the arguments with the scope start as it makes more sense than
printing arguments with the end of a scope.

Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18271>

2 years agotu/perfetto: Handle tracepoint args from start and end
Mark Collins [Fri, 16 Sep 2022 14:06:52 +0000 (14:06 +0000)]
tu/perfetto: Handle tracepoint args from start and end

Perfetto's tracing backend was designed to only handle arguments
at the end event, we want to cover arguments both at the start
and end of the trace.

Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18271>

2 years agov3dv: ignore imported BOs when tracking BO memory usage
Iago Toral Quiroga [Fri, 11 Nov 2022 11:58:10 +0000 (12:58 +0100)]
v3dv: ignore imported BOs when tracking BO memory usage

Imported BOs are not allocated by the device so we don't
update BO stats when they are imported. Therefore, we should
not be updating them when they are freed either.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19675>

2 years agodraw: Prevent index buffer overflow.
Jose Fonseca [Wed, 9 Nov 2022 09:22:06 +0000 (09:22 +0000)]
draw: Prevent index buffer overflow.

This prevents segfaults running wgf11draw DrawMethod=DrawIndexed tests
cases.

This change is not, however, sufficient to pass all tests.  That is,
there must remain code paths where indices read beyond the end of the
index buffer do not return zero index.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19648>

2 years agov3dv: fix debug dump on BO free
Iago Toral Quiroga [Fri, 11 Nov 2022 10:36:29 +0000 (11:36 +0100)]
v3dv: fix debug dump on BO free

We were resetting the BO struct right before dumping its data. Fix
this by moving the reset later.

Fixes: 44fa8304d45 ('v3dv: add a refcount mechanism to BOs')
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19672>

2 years agopvr: Implement new firmware stream interface
Matt Coster [Wed, 3 Aug 2022 08:52:17 +0000 (09:52 +0100)]
pvr: Implement new firmware stream interface

Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19530>

2 years agopvr: Set all compute sub-command registers in the same place
Matt Coster [Tue, 5 Jul 2022 09:09:12 +0000 (10:09 +0100)]
pvr: Set all compute sub-command registers in the same place

This moves the first group of registers from
pvr_sub_cmd_compute_job_init() to just before the second group in
pvr_compute_job_ws_submit_info_init().

Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19530>

2 years agodisable zinks shader cache when the needed functions do not exist
noasakurajin [Tue, 8 Nov 2022 09:16:22 +0000 (09:16 +0000)]
disable zinks shader cache when the needed functions do not exist

Fixes: 4e14da056d6 ("zink: Enable mesa/st frontend shader caching.")
Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19592>

2 years agoanv: remove dg2 condition for Wa_22011440098
Tapani Pälli [Thu, 10 Nov 2022 16:27:14 +0000 (18:27 +0200)]
anv: remove dg2 condition for Wa_22011440098

We need same workaround for MTL.

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19636>

2 years agoiris: remove dg2 condition for Wa_22011440098
Tapani Pälli [Thu, 10 Nov 2022 16:25:15 +0000 (18:25 +0200)]
iris: remove dg2 condition for Wa_22011440098

We need same workaround for MTL.

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19636>

2 years agoanv: setup stage bitmask for Wa_22011440098
Tapani Pälli [Thu, 10 Nov 2022 16:23:21 +0000 (18:23 +0200)]
anv: setup stage bitmask for Wa_22011440098

Fixes: 40b66a44998 ("anv, iris: Add Wa_22011440098 for DG2")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19636>

2 years agoiris: setup stage bitmask for Wa_22011440098
Tapani Pälli [Thu, 10 Nov 2022 16:22:12 +0000 (18:22 +0200)]
iris: setup stage bitmask for Wa_22011440098

Fixes: 40b66a44998 ("anv, iris: Add Wa_22011440098 for DG2")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19636>

2 years agoanv: split internal surface states from descriptors
Lionel Landwerlin [Mon, 24 Oct 2022 11:12:28 +0000 (14:12 +0300)]
anv: split internal surface states from descriptors

On Intel HW we use the same mechanism for internal operations surfaces
as well as application surfaces (VkDescriptor).

This change splits the surface pool in 2, one part dedicated to
internal allocations, the other to application VkDescriptors.

To do so, the STATE_BASE_ADDRESS::SurfaceStateBaseAddress points to a
4Gb area, with the following layout :
   - 1Gb of binding table pool
   - 2Gb of internal surface states
   - 1Gb of bindless surface states

That way any entry from the binding table can refer to both internal &
bindless surface states but none of the driver allocations interfere
with the allocation of the application.

Based off a change from Sviatoslav Peleshko.

v2: Allocate image view null surface state from bindless heap (Sviatoslav)
    Removed debug stuff (Sviatoslav)

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7110
Cc: mesa-stable
Tested-by: Sviatoslav Peleshko <sviatoslav.peleshko@globallogic.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19275>

2 years agonir/algeraic_opt: use double options too for lowering ftrunc@64
Gert Wollny [Thu, 10 Nov 2022 12:06:08 +0000 (13:06 +0100)]
nir/algeraic_opt: use double options too for lowering ftrunc@64

ftrunc@64 also might need lowering on fp64 only, especially now
that it might be introduced by nir_lower_int64.

Fixes: 29da9856826fa6a4b5117c43c78b4301a49bc6dd
   nir/lower_int64: Enable lowering of 64-bit float to 64-bit integer conversions.

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19657>

2 years agoradv: advertise extendedDynamicState3ProvokingVertexMode with NGG
Samuel Pitoiset [Wed, 19 Oct 2022 13:55:00 +0000 (15:55 +0200)]
radv: advertise extendedDynamicState3ProvokingVertexMode with NGG

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19168>

2 years agoradv: add support for dynamic provoking vertex mode with NGG
Samuel Pitoiset [Wed, 19 Oct 2022 13:57:52 +0000 (15:57 +0200)]
radv: add support for dynamic provoking vertex mode with NGG

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19168>

2 years agoradv: add lowering support for load_provoking_vtx_in_prim_amd when dynamic
Samuel Pitoiset [Wed, 19 Oct 2022 13:45:21 +0000 (15:45 +0200)]
radv: add lowering support for load_provoking_vtx_in_prim_amd when dynamic

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19168>

2 years agoradv: declare shader arguments/locations for provoking vertex
Samuel Pitoiset [Wed, 19 Oct 2022 13:44:54 +0000 (15:44 +0200)]
radv: declare shader arguments/locations for provoking vertex

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19168>

2 years agoradv: add radv_pipeline_key::dynamic_provoking_vtx_mode
Samuel Pitoiset [Wed, 19 Oct 2022 12:25:55 +0000 (14:25 +0200)]
radv: add radv_pipeline_key::dynamic_provoking_vtx_mode

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19168>

2 years agoradv: re-emit NGG culling settings when conservative rast mode is dynamic
Samuel Pitoiset [Thu, 10 Nov 2022 14:24:24 +0000 (15:24 +0100)]
radv: re-emit NGG culling settings when conservative rast mode is dynamic

Found by inspection.

Fixes: fbed3aed4aa ("radv: add support for dynamic conservative rasterization mode")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19651>

2 years agomeson: only enable intel-clc for x86_64 builds
Luis Felipe Strano Moraes [Thu, 10 Nov 2022 23:57:05 +0000 (15:57 -0800)]
meson: only enable intel-clc for x86_64 builds

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19662>

2 years agoutil: Replace the usage of boolean with c11 bool in u_cpu_detect.c
Yonggang Luo [Sun, 6 Nov 2022 14:35:30 +0000 (22:35 +0800)]
util: Replace the usage of boolean with c11 bool in u_cpu_detect.c

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19649>

2 years agomicrosoft: Fixes error: unknown type name ‘boolean’ in nir_to_dxil.c
Yonggang Luo [Mon, 7 Nov 2022 05:46:48 +0000 (13:46 +0800)]
microsoft: Fixes error: unknown type name ‘boolean’ in nir_to_dxil.c

error message:
../../src/microsoft/compiler/nir_to_dxil.c:1711:8: error: unknown type name ‘boolean’
 1711 | static boolean

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19649>

2 years agoutil: Replace usage of boolean with c11 bool in src/util/format/* and src/util/tests...
Yonggang Luo [Mon, 7 Nov 2022 05:34:30 +0000 (13:34 +0800)]
util: Replace usage of boolean with c11 bool in src/util/format/* and src/util/tests/format/*

This is done by find and replace:
boolean -> bool
TRUE -> true
FALSE -> false

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19649>

2 years agoutil: Replace the usage of boolean with c11 bool in u_debug_symbol.c
Yonggang Luo [Wed, 9 Nov 2022 09:43:16 +0000 (17:43 +0800)]
util: Replace the usage of boolean with c11 bool in u_debug_symbol.c

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19649>

2 years agoutil: Replace the usage of boolean with c11 bool in u_debug_memory.c
Yonggang Luo [Sun, 6 Nov 2022 13:23:22 +0000 (21:23 +0800)]
util: Replace the usage of boolean with c11 bool in u_debug_memory.c

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19649>

2 years agoutil: Replace all usage of __FUNCTION__ with __func__ in src/util/*
Yonggang Luo [Thu, 10 Nov 2022 08:31:11 +0000 (16:31 +0800)]
util: Replace all usage of __FUNCTION__ with __func__ in src/util/*

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19646>

2 years agoutil: Remove va_copy fixup because of c11 is required
Yonggang Luo [Thu, 10 Nov 2022 08:26:09 +0000 (16:26 +0800)]
util: Remove va_copy fixup because of c11 is required

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19646>

2 years agoutil: Replace TAB with space in compiler.h
Yonggang Luo [Thu, 10 Nov 2022 08:24:16 +0000 (16:24 +0800)]
util: Replace TAB with space in compiler.h

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19646>

2 years agoutil: Cleanup util/compiler.h
Yonggang Luo [Thu, 10 Nov 2022 08:01:35 +0000 (16:01 +0800)]
util: Cleanup util/compiler.h

Remove MESA_*_ENDIAN
Use UTIL_ARCH_*_ENDIAN to define CPU_TO_LE32

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19646>

2 years agowsi: Use UTIL_ARCH_LITTLE_ENDIAN instead MESA_LITTLE_ENDIAN
Yonggang Luo [Thu, 10 Nov 2022 08:00:41 +0000 (16:00 +0800)]
wsi: Use UTIL_ARCH_LITTLE_ENDIAN instead MESA_LITTLE_ENDIAN

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19646>

2 years agomeson: Fixes name_prefix for clover on mingw
Yonggang Luo [Wed, 9 Nov 2022 03:24:32 +0000 (11:24 +0800)]
meson: Fixes name_prefix for clover on mingw

Cc: mesa-stable
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19610>

2 years agoclover: Fixes building with mingw-x86
Yonggang Luo [Tue, 8 Nov 2022 11:07:35 +0000 (19:07 +0800)]
clover: Fixes building with mingw-x86

Cc: mesa-stable
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19610>

2 years agoclover: Rename *OpenCL.def to *OpenCL.def.in
Yonggang Luo [Wed, 9 Nov 2022 02:30:57 +0000 (10:30 +0800)]
clover: Rename *OpenCL.def to *OpenCL.def.in

Cc: mesa-stable
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19610>

2 years agomeson: fixes mingw-clang32 building
Yonggang Luo [Tue, 8 Nov 2022 12:48:30 +0000 (20:48 +0800)]
meson: fixes mingw-clang32 building

Cc: mesa-stable
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19610>

2 years agomeson: Refactoring shared gen_vs_module_defs_normal_command out
Yonggang Luo [Tue, 8 Nov 2022 12:32:07 +0000 (20:32 +0800)]
meson: Refactoring shared gen_vs_module_defs_normal_command out

Cc: mesa-stable
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19610>

2 years agoradv: Fixes prototypes
Yonggang Luo [Tue, 8 Nov 2022 12:15:10 +0000 (20:15 +0800)]
radv: Fixes prototypes

Cc: mesa-stable
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19610>

2 years agopvr: use common command buffer status
Dave Airlie [Wed, 8 Jun 2022 00:52:36 +0000 (10:52 +1000)]
pvr: use common command buffer status

Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16922>

2 years agopvr: Hook up vk_command_buffer_ops::reset
Jason Ekstrand [Fri, 9 Sep 2022 23:43:38 +0000 (18:43 -0500)]
pvr: Hook up vk_command_buffer_ops::reset

Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16922>

2 years agopvr: Call command buffer reset conditionally
Jason Ekstrand [Fri, 9 Sep 2022 23:36:39 +0000 (18:36 -0500)]
pvr: Call command buffer reset conditionally

Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16922>

2 years agoturnip: use common command buffer status code.
Dave Airlie [Wed, 8 Jun 2022 00:46:26 +0000 (10:46 +1000)]
turnip: use common command buffer status code.

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16922>

2 years agoturnip: Initialize push descriptors in tu_create_cmd_buffer()
Jason Ekstrand [Tue, 8 Nov 2022 20:32:56 +0000 (14:32 -0600)]
turnip: Initialize push descriptors in tu_create_cmd_buffer()

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16922>

2 years agopanvk: move to using common command buffer status
Dave Airlie [Wed, 8 Jun 2022 00:49:09 +0000 (10:49 +1000)]
panvk: move to using common command buffer status

Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16922>

2 years agopanvk: Use common code for command buffer lifecycle management
Jason Ekstrand [Sat, 10 Sep 2022 00:14:13 +0000 (19:14 -0500)]
panvk: Use common code for command buffer lifecycle management

Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16922>

2 years agopanvk: Drop panvk_cmd_buffer::queue_family_index
Jason Ekstrand [Mon, 12 Sep 2022 10:17:42 +0000 (05:17 -0500)]
panvk: Drop panvk_cmd_buffer::queue_family_index

Suggested-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16922>

2 years agolavapipe: use common command buffer status code.
Dave Airlie [Wed, 8 Jun 2022 00:46:43 +0000 (10:46 +1000)]
lavapipe: use common command buffer status code.

Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16922>

2 years agoradv: use common command buffer status code.
Dave Airlie [Wed, 8 Jun 2022 00:46:11 +0000 (10:46 +1000)]
radv: use common command buffer status code.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16922>

2 years agovulkan: Add common cmd buffer state tracking
Dave Airlie [Wed, 8 Jun 2022 00:45:44 +0000 (10:45 +1000)]
vulkan: Add common cmd buffer state tracking

For drivers already using vk_common_ResetCommandBuffer(), it now only
calls the driver's reset hook if the command buffer is not in the
INITIAL state.  Pulled this trick from the PowerVR driver.

v2 (Jason Ekstrand):
 - Rename from "status" to "state" since that's what's in the spec
 - Add vk_command_buffer_begin/end instead of drivers setting it all
   manually

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16922>

2 years agoradeonsi: move param assign offset to si_get_nir_shader
Qiang Yu [Thu, 27 Oct 2022 13:57:58 +0000 (21:57 +0800)]
radeonsi: move param assign offset to si_get_nir_shader

We need this info for gfx11 param export soon and nir vertex
export lowering in the future.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19429>

2 years agoradeonsi: pack code into si_assign_param_offsets
Qiang Yu [Thu, 3 Nov 2022 02:28:13 +0000 (10:28 +0800)]
radeonsi: pack code into si_assign_param_offsets

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19429>

2 years agoradeonsi: small refine for si_nir_assign_param_offsets
Qiang Yu [Thu, 3 Nov 2022 02:26:30 +0000 (10:26 +0800)]
radeonsi: small refine for si_nir_assign_param_offsets

Pass si_shader as arg directly, move primitive id pending
inside it.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19429>

2 years agoradeonsi: move si_nir_assign_param_offsets above
Qiang Yu [Thu, 3 Nov 2022 02:18:41 +0000 (10:18 +0800)]
radeonsi: move si_nir_assign_param_offsets above

It's going to be called by si_get_nir_shader.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19429>

2 years agoradeonsi: implement nir_load_ring_attr_amd
Qiang Yu [Thu, 27 Oct 2022 07:34:44 +0000 (15:34 +0800)]
radeonsi: implement nir_load_ring_attr_amd

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19429>

2 years agoac/llvm: implement intrinsic for gfx11 param export
Qiang Yu [Thu, 27 Oct 2022 07:08:56 +0000 (15:08 +0800)]
ac/llvm: implement intrinsic for gfx11 param export

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19429>

2 years agoradeonsi: use si_nir_clamp_vertex_color
Qiang Yu [Fri, 28 Oct 2022 03:42:05 +0000 (11:42 +0800)]
radeonsi: use si_nir_clamp_vertex_color

Clamp vertex color in nir. Now only GS copy shader use
si_vertex_color_clamping, so move it there. It will be
completely removed after we switch to nir GS copy shader.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19429>

2 years agoradeonsi: add si_nir_clamp_vertex_color pass
Qiang Yu [Fri, 28 Oct 2022 03:15:03 +0000 (11:15 +0800)]
radeonsi: add si_nir_clamp_vertex_color pass

To replace the LLVM si_vertex_color_clamping.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19429>

2 years agonir,ac/llvm,radeonsi: add nir_load_clamp_vertex_color_amd
Qiang Yu [Fri, 28 Oct 2022 02:15:21 +0000 (10:15 +0800)]
nir,ac/llvm,radeonsi: add nir_load_clamp_vertex_color_amd

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19429>

2 years agoradeonsi: remove lower_fsat nir option
Qiang Yu [Tue, 8 Nov 2022 02:05:20 +0000 (10:05 +0800)]
radeonsi: remove lower_fsat nir option

We implemented nir_op_fsat nir to llvm translation.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19429>

2 years agomeson: enable Wmisleading-indentation
Lucas Stach [Tue, 1 Nov 2022 13:42:12 +0000 (14:42 +0100)]
meson: enable Wmisleading-indentation

Clang enables it by default, but GCC does not. Enable it explicitly to
have some consistency between the supported compilers.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Acked-by: Eric Engestrom <eric@igalia.com>
Acked-by: Dylan Baker <dylan.c.baker@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19439>

2 years agofreedreno/drm/virtio: Pre-allocate cmdstream buffers
Rob Clark [Thu, 10 Nov 2022 15:36:22 +0000 (07:36 -0800)]
freedreno/drm/virtio: Pre-allocate cmdstream buffers

We know cmdstream buffers are immediately mmap'd, which is both
expensive on the host, and breaks the pipelining as guest is forced
to stall waiting for the host.  So pre-allocate some cmdestream
buffers, so that we have something that is (hopefully) already
allocated and mapped to guest's physical memory before we need it.
The older buffer from the head of the prealloc list replaces the
newly allocated buffer which is pushed to the tail of the prealloc
list.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19656>

2 years agofreedreno/drm: Use same-sized "chunks" for growable rings
Rob Clark [Wed, 9 Nov 2022 23:47:25 +0000 (15:47 -0800)]
freedreno/drm: Use same-sized "chunks" for growable rings

Using the same size that we suballoc from for suballoc'd streaming and
long-lived stateobjs should help improve bo cache usage, by making more
of the backing BOs the same size and interchangable.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19656>

2 years agoci/broadcom: v3dv-rpi4-vk:arm64 flakes too often
David Heidelberg [Thu, 10 Nov 2022 19:45:36 +0000 (20:45 +0100)]
ci/broadcom: v3dv-rpi4-vk:arm64 flakes too often

See https://gitlab.freedesktop.org/mesa/mesa/-/issues/7403

Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19659>

2 years agoutil/glsl2spirv: pass path to glslangValidator into the script
Dylan Baker [Tue, 1 Nov 2022 20:37:06 +0000 (13:37 -0700)]
util/glsl2spirv: pass path to glslangValidator into the script

This allows users to override the location of glslang using normal meson
mechanisms.

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agomeson: prog_glslang should always be for the build machine
Dylan Baker [Tue, 1 Nov 2022 20:35:24 +0000 (13:35 -0700)]
meson: prog_glslang should always be for the build machine

It doesn't produce different code when compiled for the host machine,
and this avoids needing two copies and possibly a host machine emulator
in cross compilation settings

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spirv: cleanup list extension
Dylan Baker [Tue, 1 Nov 2022 20:29:48 +0000 (13:29 -0700)]
util/glsl2spirv: cleanup list extension

- consistently use list.extend instead of list +=, which has gotchas
- condense list extension calls when possible

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spirv: use f-strings
Dylan Baker [Tue, 1 Nov 2022 20:29:21 +0000 (13:29 -0700)]
util/glsl2spirv: use f-strings

Which are both faster and easier to read

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spirv: use if `x in str` instead of `str.find`
Dylan Baker [Tue, 1 Nov 2022 20:28:18 +0000 (13:28 -0700)]
util/glsl2spirv: use if `x in str` instead of `str.find`

The latter is only idiomatically used when a start and/or stop position
is required.

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spirv: simplify subprocess handling
Dylan Baker [Tue, 1 Nov 2022 20:02:41 +0000 (13:02 -0700)]
util/glsl2spirv: simplify subprocess handling

Since we're not doing anything fancy, we can just use `subprocess.run`.
I've also removed the custom error class, we're not going to catch it,
so just printing and exiting is fine.

v2:
  - Print stdout as well as stderr in case of a glslang failure

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spirv: close resources as soon as possible
Dylan Baker [Tue, 1 Nov 2022 20:00:26 +0000 (13:00 -0700)]
util/glsl2spirv: close resources as soon as possible

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spirv: add type annotations
Dylan Baker [Tue, 1 Nov 2022 19:59:06 +0000 (12:59 -0700)]
util/glsl2spirv: add type annotations

Which are all clean

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spirv: add some error handling for unexpected code paths
Dylan Baker [Tue, 1 Nov 2022 19:58:19 +0000 (12:58 -0700)]
util/glsl2spirv: add some error handling for unexpected code paths

We expect that convert_to_static_variable and override_version will find
and replace something, so let's fail loudly if they don't.

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spriv: make --vn required
Dylan Baker [Tue, 1 Nov 2022 19:55:12 +0000 (12:55 -0700)]
util/glsl2spriv: make --vn required

I'm not 100% sure whether it's right to make --vn required, or to avoid
the static conversion, but this seems correct. Mypy (type checking
coming soon) points out that if --vn is None then the
convert_to_static_variable function will fail. Our one use of this sets
--vn, so there is no change there. Making --vn required
ensures that it will never be None, avoiding the problem.

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spirv: fix appending extra flags
Dylan Baker [Tue, 1 Nov 2022 19:49:53 +0000 (12:49 -0700)]
util/glsl2spirv: fix appending extra flags

The variable is called `extra`, but what's written is `extra - flags`,
and `flags` is undefined, so if the variable was ever passed there would
be an uncaught exception.

fixes: 9786d9ef2abb45a4e832cf1347581e3ca3aae9f0

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spirv: let argparse actually enforce the restrictions we've set
Dylan Baker [Tue, 1 Nov 2022 19:47:21 +0000 (12:47 -0700)]
util/glsl2spirv: let argparse actually enforce the restrictions we've set

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spirv: drop inconsistent use of `io.open`
Dylan Baker [Tue, 1 Nov 2022 19:41:44 +0000 (12:41 -0700)]
util/glsl2spirv: drop inconsistent use of `io.open`

In Python 3 (the only python we support) `io.open` is an alias of the
builtin `open` function, so it's not getting us anything, and we're not
using it consistently.

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoutil/glsl2spirv: fix type error in argument handling
Dylan Baker [Tue, 1 Nov 2022 19:40:21 +0000 (12:40 -0700)]
util/glsl2spirv: fix type error in argument handling

args.Olib is set to `store_true`, which means it will always be `True`
or `False`, this means that the we always, unconditionally, add
`--keep-uncalled` to the command line.

fixes: 9786d9ef2abb45a4e832cf1347581e3ca3aae9f0

Reviewed-by: Luis Felipe Strano Moraes <luis.strano@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19449>

2 years agoci/freedreno: disable antichambers trace
David Heidelberg [Thu, 10 Nov 2022 17:23:46 +0000 (18:23 +0100)]
ci/freedreno: disable antichambers trace

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7668
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19627
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19634>

2 years agoturnip: Don't disable LRZ for color write mask channels beyond the format's
Emma Anholt [Thu, 10 Nov 2022 00:41:29 +0000 (16:41 -0800)]
turnip: Don't disable LRZ for color write mask channels beyond the format's

aztec_ruins under ANGLE was getting LRZ writes disabled because 0xf out of
the 0x3 mask was enabled.  The goal was to see if there are partial writes
being done, though.  This caused a 2-3% performance regression.

Fixes: 85d0205db137 ("tu: Implement extendedDynamicState3ColorWriteMask")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19635>

2 years agoir3: Re-fuse ffmas after nir_lower_imul cleanup breaks them.
Emma Anholt [Mon, 3 Oct 2022 23:49:15 +0000 (16:49 -0700)]
ir3: Re-fuse ffmas after nir_lower_imul cleanup breaks them.

The nir_opt_algebraic() call to clean up nir_lower_imul's split up mul
operations (stuff like "the top 16 bits were 0, no need to mul and add
that part") would trigger the options->fuse_ffma_* early ffma splitting,
so you need to call nir_opt_algebraic_late() again after that (which, in
turn, requires a DCE).

Gets us a lot more ffmas in Aztec Ruins high under zink/angle, but doesn't
seem to change perf.

shader-db highlights:

total instructions in shared programs: 11574843 -> 10999629 (-4.97%)
instructions in affected programs: 3308870 -> 2733656 (-17.38%)

total dwords in shared programs: 24344722 -> 23230122 (-4.58%)
dwords in affected programs: 6569568 -> 5454968 (-16.97%)

total full in shared programs: 762616 -> 762224 (-0.05%)
full in affected programs: 15505 -> 15113 (-2.53%)

total stp in shared programs: 4046 -> 4050 (0.10%)
stp in affected programs: 3372 -> 3376 (0.12%)

total ldp in shared programs: 2166 -> 2170 (0.18%)
ldp in affected programs: 1716 -> 1720 (0.23%)

total (ss) in shared programs: 219541 -> 216261 (-1.49%)
(ss) in affected programs: 23227 -> 19947 (-14.12%)

total (sy) in shared programs: 101633 -> 101927 (0.29%)
(sy) in affected programs: 8611 -> 8905 (3.41%)

total waves in shared programs: 1501942 -> 1501772 (-0.01%)
waves in affected programs: 1880 -> 1710 (-9.04%)

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18946>

2 years agoir3/ra: Make sure we don't pick a preferred reg overflowing the file.
Emma Anholt [Wed, 9 Nov 2022 20:23:53 +0000 (12:23 -0800)]
ir3/ra: Make sure we don't pick a preferred reg overflowing the file.

If we're in handle_collect()'s dst allocation and are part of a merge set
near the end of the file, our check for reg_elem_size(reg) would let us
use the preferred reg when that would immediately lead to
allocate_dst_fixed() creating an interval extending thruogh reg_size(reg)
that overflows the file.

Avoids a regression on gfxbench5/gl_5_high_off/17.shader_test in the next
commit.  No change on shader-db.

Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18946>

2 years agointel/compiler: Use std::unique_ptr for tracking the fs_visitors
Caio Oliveira [Tue, 8 Nov 2022 22:14:37 +0000 (14:14 -0800)]
intel/compiler: Use std::unique_ptr for tracking the fs_visitors

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19605>

2 years agodocs: add linkcheck job
Erik Faye-Lund [Mon, 7 Nov 2022 14:24:27 +0000 (15:24 +0100)]
docs: add linkcheck job

This should make it easier to keep links in our docs up-to-date.

But, because links can die behind our backs, we can't really enable this
all over the place, or we'll risk blocking merge-requests due to
unrelated changes.

So let's just make this a periodic job on the main branch instead.

Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19574>

2 years agodocs: do not linkcheck relnotes
Erik Faye-Lund [Mon, 7 Nov 2022 13:46:27 +0000 (14:46 +0100)]
docs: do not linkcheck relnotes

There's a *lot* of relnotes, and we don't really actively maintain them.
Let's drop linkchecking them to speed things up a bit.

This does a whole lot of nothing unless you have Sphinx 4.4 or newer.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19574>

2 years agodocs: configure linkcheck
Erik Faye-Lund [Mon, 7 Nov 2022 11:03:57 +0000 (12:03 +0100)]
docs: configure linkcheck

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19574>

2 years agodocs/ci: allow overriding sphinx builder
Erik Faye-Lund [Fri, 4 Nov 2022 15:13:20 +0000 (16:13 +0100)]
docs/ci: allow overriding sphinx builder

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19574>

2 years agodocs/panfrost: drop anchor before channel-name
Erik Faye-Lund [Mon, 7 Nov 2022 14:08:46 +0000 (15:08 +0100)]
docs/panfrost: drop anchor before channel-name

The use of a hash symbol in a URI is an anchor, not to indicate an IRC
channel name. This confuses the Sphinx linkchecker.

Dropping the hash here still makes the link work fine, so let's just do
that.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19574>

2 years agodocs: apply some redirects
Erik Faye-Lund [Mon, 7 Nov 2022 12:30:03 +0000 (13:30 +0100)]
docs: apply some redirects

Here's a few redirects that we should apply, in case the redirects
gets removed in the future.

These are mostly of the 301 (moved permanently) kind, but also a
few where the site probably *should* have used the permanent
error-code.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19574>

2 years agodocs: update lavasoftware.org link
Erik Faye-Lund [Mon, 7 Nov 2022 13:32:22 +0000 (14:32 +0100)]
docs: update lavasoftware.org link

Seems lavasoftware.org now prefers using the www-version of the link.
Let's update to that, to quiet some link-checking.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19574>

2 years agodocs/perfetto: update documentation links
Erik Faye-Lund [Mon, 7 Nov 2022 14:05:17 +0000 (15:05 +0100)]
docs/perfetto: update documentation links

These links redirect to something seemingly random. Here's my best
effort to update them to the relevant pages in the current docs.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19574>

2 years agodocs: update link to GitLab CI docs
Erik Faye-Lund [Mon, 7 Nov 2022 13:53:57 +0000 (14:53 +0100)]
docs: update link to GitLab CI docs

This link is being redirected, but not really where we want to send
people. Let's update it, so they end up in the right place.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19574>

2 years agodocs: update link to Intel intrinsics-guide
Erik Faye-Lund [Mon, 7 Nov 2022 13:27:54 +0000 (14:27 +0100)]
docs: update link to Intel intrinsics-guide

This link is being redirected, but not really where we want to send
people. Let's update it, so they end up in the right place.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19574>

2 years agodocs: update links to SPECviewperf 11/12
Erik Faye-Lund [Mon, 7 Nov 2022 13:02:55 +0000 (14:02 +0100)]
docs: update links to SPECviewperf 11/12

These links redirect to a catch-all page, and not really where we want
people to end up. Let's correct these links and point to the benchmark
specific pages.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19574>