Eric Engestrom [Mon, 27 Feb 2023 16:50:50 +0000 (16:50 +0000)]
glapi/meson: drop duplicate line in deps
Fixes:
0842bc879b34414de54b ("meson: wire the new generator for es1 and es2")
Signed-off-by: Eric Engestrom <eric@igalia.com>
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21562>
Samuel Pitoiset [Mon, 27 Feb 2023 09:49:16 +0000 (10:49 +0100)]
radv: fix flushing non-coherent images in EndCommandBuffer()
The condition was inverted.
This doesn't fix anything known.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21549>
Samuel Pitoiset [Mon, 27 Feb 2023 09:45:08 +0000 (10:45 +0100)]
radv: fix flushing non-coherent images inside secondaries on GFX9+
Fixes
dEQP-VK.draw.dynamic_rendering.complete_secondary_cmd_buff.multi_draw.mosaic.*
on VEGA10 (related to the use of HTILE).
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21549>
Eric Engestrom [Mon, 27 Feb 2023 15:24:36 +0000 (15:24 +0000)]
glsl: align definition of _mesa_problem with the one in main/error.h
The ctx pointer not used by that function anyway, so const'ing it makes
no difference.
Signed-off-by: Eric Engestrom <eric@igalia.com>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21557>
Dylan Baker [Thu, 23 Feb 2023 17:50:33 +0000 (09:50 -0800)]
util: rzalloc and free hash_table_u64
Otherwise we're prone to leaking the table itself, since it's not freed
in the destroy function
CID: 1516552
fixes:
6649b840c34016b4753e69d4513a8d09da9febb2
("mesa/util: add a hash table wrapper which support 64-bit keys")
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21499>
Tapani Pälli [Mon, 27 Feb 2023 11:19:26 +0000 (13:19 +0200)]
anv: emit 3DSTATE_HS for each primitive on gfx12
This is Wa_16011107343, same workaround as commit
880a3efe6c2
but for gfx12.
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/21551>
Tapani Pälli [Mon, 27 Feb 2023 11:13:20 +0000 (13:13 +0200)]
iris: emit 3DSTATE_HS for each primitive on gfx12
This is Wa_16011107343, same workaround as commit
a043ae8e249
but for gfx12.
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/21551>
Qiang Yu [Mon, 26 Dec 2022 05:01:03 +0000 (13:01 +0800)]
aco: only ls and ps use store output now
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21435>
Qiang Yu [Mon, 26 Dec 2022 03:38:56 +0000 (11:38 +0800)]
ac/llvm: only init outputs when fragment shader for radv
LS pass output to TCS by reg is not enabled when LLVM.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21435>
Qiang Yu [Mon, 26 Dec 2022 02:03:59 +0000 (10:03 +0800)]
radeonsi: only init llvm output when needed.
These are the cases which still uses nir_store_output.
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/21435>
Emma Anholt [Wed, 22 Feb 2023 21:47:37 +0000 (13:47 -0800)]
glsl: Move lower_vector_insert to GLSL-to-NIR.
We already have a nir_builder equivalent for generating this code, just
use that instead of doing it in GLSL.
No change on r300 shader-db.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21476>
Emma Anholt [Mon, 27 Feb 2023 23:07:34 +0000 (15:07 -0800)]
panfrost/midgard: Drop redundant arg to emit_explicit_constant.
Every caller passed the same value twice. Just reuse it?
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21476>
Emma Anholt [Mon, 27 Feb 2023 22:24:07 +0000 (14:24 -0800)]
panfrost/midgard: Fix handling of csel with a vector constant condition.
If it's not all true or all false, then you'll have a csel with a vector
constant, and the backend failed to translate appropriately. Expand the
constant to fix it.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21476>
Yiwei Zhang [Fri, 17 Feb 2023 09:35:26 +0000 (01:35 -0800)]
venus: fix device memory export alloc info
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Fri, 17 Feb 2023 04:52:13 +0000 (20:52 -0800)]
venus: fix external image creation
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Fri, 17 Feb 2023 05:29:54 +0000 (21:29 -0800)]
venus: simplify support for non-AHB external images
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Wed, 15 Feb 2023 23:07:01 +0000 (15:07 -0800)]
venus: simplify ahb image creation
Those excessive asserts are only useful when bringing up Android.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Wed, 15 Feb 2023 22:11:02 +0000 (14:11 -0800)]
venus: refactor image create info pnext tracking
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Wed, 15 Feb 2023 21:44:30 +0000 (13:44 -0800)]
venus: remove redundant abstractions for wsi struct search
Doing that won't avoid linking wsi headers, and in fact we have already
included both android and common wsi headers. For swapchain info, it's
currently disabled by the swapchain spec version advertised on Android.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Mon, 13 Feb 2023 00:08:45 +0000 (16:08 -0800)]
venus: fix external buffer creation
Move buffer create info scrub into core venus.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Wed, 8 Feb 2023 21:50:23 +0000 (13:50 -0800)]
venus: distinguish external memory from mappable memory support
Mappable memory support is a must for Venus core, but the support of
such can be transparent to the driver. Thus the renderer external memory
type won't expose opaque fd type.
External memory over vtest can be exposed and the wsi support on top can
be made explicit as long as masking out the importable bit.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Wed, 15 Feb 2023 17:48:49 +0000 (09:48 -0800)]
venus: tighten up the sync fd requirements for Android wsi
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Wed, 15 Feb 2023 17:22:29 +0000 (09:22 -0800)]
venus: refactor sync fd fence and sempahore features
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Wed, 15 Feb 2023 16:56:00 +0000 (08:56 -0800)]
venus: clean up globalFencing
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Tue, 14 Feb 2023 07:23:28 +0000 (23:23 -0800)]
venus: clean up memoryResourceAllocationSize
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Tue, 14 Feb 2023 07:18:39 +0000 (23:18 -0800)]
venus: start requiring all experimental features
This is to prepare for promoting exp features to core.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Mon, 13 Feb 2023 04:30:53 +0000 (20:30 -0800)]
venus: replace binary search with hardcode for max buffer size
Avoid the couple test buffer creations without maintenance4.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Yiwei Zhang [Mon, 13 Feb 2023 03:45:28 +0000 (19:45 -0800)]
venus: further disallow sparse resource
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21379>
Georg Lehmann [Sun, 26 Feb 2023 14:54:18 +0000 (15:54 +0100)]
aco/optimizer_postRA: assume all registers are untrackable in loop headers
Register writes from the pre-header might not be correct for any but
the first loop iteration because they can be clobbered inside the loop.
Foz-DB Navi21:
Totals from 18 (0.01% of 134913) affected shaders:
CodeSize: 251384 -> 251508 (+0.05%)
Instrs: 47644 -> 47664 (+0.04%)
Latency: 801801 -> 801852 (+0.01%)
InvThroughput: 177579 -> 177593 (+0.01%)
Copies: 4752 -> 4771 (+0.40%)
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8376
Fixes:
d3b0f781103 ("aco/optimizer_postRA: Initialize loop header with preheader information")
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21540>
Mike Blumenkrantz [Mon, 20 Feb 2023 16:40:00 +0000 (11:40 -0500)]
zink: skip buffer barriers for ACCESS_NONE -> ACCESS_READ / ACCESS_WRITE
these do nothing and may cause unnecessary driver flushing
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21427>
Emma Anholt [Wed, 22 Feb 2023 23:13:53 +0000 (15:13 -0800)]
ci: Update traces expectations for gutting glsl opt_algebraic.
All look like harmless changes.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 21:12:24 +0000 (13:12 -0800)]
glsl: Remove unused as_rvalue_to_saturate().
This is not where saturate recognition happens. Dead code since
5598458e6923 ("i965/vec4: Remove try_emit_saturate") in 2014!
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 21:16:40 +0000 (13:16 -0800)]
glsl/opt_algebraic: Drop the flrp recognizer.
No change to r300.
freedreno looks mixed but slightly positive in instructions:
total instructions in shared programs:
11012472 ->
11012453 (<.01%)
instructions in affected programs: 8250 -> 8231 (-0.23%)
helped: 16
HURT: 50
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 21:01:13 +0000 (13:01 -0800)]
glsl/opt_algebraic: Drop the ftrunc pattern recognizer.
Now that it's in NIR, there's no change to r300 or freedreno shader-db
when we do.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:59:16 +0000 (12:59 -0800)]
nir: Port a floor->truncate algebraic opt pattern from GLSL.
Prevents regression when dropping code from the GLSL optimizer.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:44:40 +0000 (12:44 -0800)]
glsl/opt_algebraic: Drop some fmul simplifications.
Looks like mostly noise, trending slightly positively.
freedreno:
total instructions in shared programs:
11012781 ->
11012472 (<.01%)
instructions in affected programs: 114072 -> 113763 (-0.27%)
helped: 123
HURT: 153
r300:
total instructions in shared programs: 1338236 -> 1337897 (-0.03%)
instructions in affected programs: 3460 -> 3121 (-9.80%)
helped: 61
HURT: 11
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:39:41 +0000 (12:39 -0800)]
glsl/opt_algebraic: Drop the flrp/ffma simplifiers.
NIR seems to do a better job. Freedreno:
total instructions in shared programs:
11013096 ->
11012781 (<.01%)
instructions in affected programs: 258358 -> 258043 (-0.12%)
helped: 470
HURT: 269
r300:
total instructions in shared programs: 1338237 -> 1338236 (<.01%)
instructions in affected programs: 161 -> 160 (-0.62%)
helped: 1
HURT: 0
total presub in shared programs: 45127 -> 44881 (-0.55%)
presub in affected programs: 1719 -> 1473 (-14.31%)
helped: 246
HURT: 0
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:29:37 +0000 (12:29 -0800)]
glsl/opt_algebraic: Drop no-op pack/unpack optimization.
No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:27:07 +0000 (12:27 -0800)]
glsl/opt_algebraic: Drop the eq/neq add-removal optimization.
No change on freedreno or r300 shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:24:49 +0000 (12:24 -0800)]
glsl/opt_algebraic: Drop scalar all_eq/any_neq -> eq/neq opt.
No change in r300 or freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:10:32 +0000 (12:10 -0800)]
glsl/opt_algebraic: Drop fdot 0-channel optimizations.
No change on i915g shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:22:03 +0000 (12:22 -0800)]
nir: Add optimization for fdot(x, 0) -> 0.
We had all these nice fdot opts to drop individual channels that were 0,
but nothing handling it being entirely 0! Avoids r300g regression when
dropping them from GLSL.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:08:55 +0000 (12:08 -0800)]
glsl/opt_algebraic: Drop csel(true/false, x, y) optimization.
No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:04:34 +0000 (12:04 -0800)]
glsl/opt_algebraic: Drop x + -x -> 0 optimization.
No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:03:34 +0000 (12:03 -0800)]
glsl/opt_algebraic: Drop add/sub with 0 optimizations.
Looks like minor instruction selection noise in freedreno shader-db:
total instructions in shared programs:
11013100 ->
11013096 (<.01%)
instructions in affected programs: 2714 -> 2710 (-0.15%)
helped: 8
HURT: 6
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 20:00:24 +0000 (12:00 -0800)]
glsl/opt_algebraic: Drop fdiv(1,x) -> frcp(x) and fdiv(x,1) -> x optimizations.
No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 19:59:18 +0000 (11:59 -0800)]
glsl/opt_algebraic: Drop and/or/xor optimizations.
NIR has them, and if anything freedreno shader-db prefers that NIR sees them:
total instructions in shared programs:
11013112 ->
11013100 (<.01%)
instructions in affected programs: 26266 -> 26254 (-0.05%)
helped: 4
HURT: 0
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 19:56:08 +0000 (11:56 -0800)]
glsl/opt_algebraic: Drop rcp optimizations.
No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 19:54:15 +0000 (11:54 -0800)]
glsl/opt_algebraic: Drop pow optimizations.
These should all be covered by NIR. Minor shader-db changes on freedreno,
which appear to be scheduling noise.
total instructions in shared programs:
11013132 ->
11013112 (<.01%)
instructions in affected programs: 3408 -> 3388 (-0.59%)
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 19:48:05 +0000 (11:48 -0800)]
glsl/opt_algebraic: Drop shifts of 0 optimizations.
No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 19:45:20 +0000 (11:45 -0800)]
glsl/opt_algebraic: drop fsat(fadd(b2f(x), b2f(y))) -> b2f(ior(x, y)) opt.
No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 19:43:23 +0000 (11:43 -0800)]
glsl/opt_algebraic: Drop f2i(trunc(x)) -> f2i(x) optimization.
No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 19:39:18 +0000 (11:39 -0800)]
glsl/opt_algebraic: Drop -(-x) -> x optimization.
No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 19:38:14 +0000 (11:38 -0800)]
glsl/opt_algebraic: Drop abs(-x) -> abs(x) and abs(abs(x)) -> abs(x).
NIR does this. No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 19:36:24 +0000 (11:36 -0800)]
glsl/opt_algebraic: Drop pow-recognizer.
NIR handles pow recognizing, too. No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 19:34:57 +0000 (11:34 -0800)]
glsl/opt_algebraic: Drop log(exp(x)) -> x and exp(log(x)) -> x optimisations.
No change on freedreno shader-db.
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Emma Anholt [Wed, 22 Feb 2023 19:33:06 +0000 (11:33 -0800)]
glsl/opt_algebraic: Drop ~~x == x transformation.
No change on freedreno shader-db.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21475>
Mike Blumenkrantz [Wed, 22 Feb 2023 20:20:15 +0000 (15:20 -0500)]
zink: use tc renderpass optimizing to more optimally start queries
if tc has determined that no queries are ended inside a renderpass,
it becomes possible to start queries outside of a renderpass, which
is more performant on some hw
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 20:19:33 +0000 (15:19 -0500)]
zink: always start/stop/resume queries inside renderpasses
this avoids potentially splitting renderpasses by ensuring that
all (non-cs) query operations always occur inside renderpasses
zink_query_update_gs_states() now has to be called inside renderpass
to catch the active queries
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 22:44:15 +0000 (17:44 -0500)]
zink: pull 'was_line_loop' into ctx for query updating
make this a bit more flexible
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 21:05:15 +0000 (16:05 -0500)]
zink: use more consistent check for deleting zink_query::stats_list links
the linkage of this may vary depending on the query's active state,
so avoid issues by just checking the linkage directly
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 20:18:01 +0000 (15:18 -0500)]
zink: break out query suspend functionality for reuse
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 20:17:16 +0000 (15:17 -0500)]
zink: track whether queries were started in a renderpass
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 18:03:47 +0000 (13:03 -0500)]
zink: rework query pool overflow
now when a query pool is full, a new query pool can be created and the
previous one can be dropped from reuse to be freed at a later time
this has the added benefit of avoiding yet another place where a renderpass
might get split
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 17:59:46 +0000 (12:59 -0500)]
zink: create/use query pools dynamically
this fits the concept of "pools are shared between queries" a bit
better and leaves the pool ownership/destruction to their users
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 17:47:29 +0000 (12:47 -0500)]
zink: rework find_or_allocate_qp()
this and its caller are now a bit more streamlined with regards to
functionality of the surrounding code
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 17:46:52 +0000 (12:46 -0500)]
zink: reorder some query code
no functional changes
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 17:28:42 +0000 (12:28 -0500)]
zink: try updating qbos on query resume if !in_rp
this is another logical place to do the update since it won't split
a renderpass
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Thu, 23 Feb 2023 01:24:21 +0000 (20:24 -0500)]
zink: only update qbo for TIME_ELAPSED on start if !in_rp
this otherwise splits the current renderpass for no reason
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Thu, 23 Feb 2023 01:05:35 +0000 (20:05 -0500)]
zink: handle null query results for conditional render
this just needs to have a zero blasted into the buffer
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 21:27:40 +0000 (16:27 -0500)]
zink: handle null query results
it's possible that a query may have no results at all, so shortcut
everything and return zero as fast as possible for those cases
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 17:28:09 +0000 (12:28 -0500)]
zink: only try doing qbo updates on query suspend if !in_rp
this otherwise creates infinite recursion trying to end a currently
ending renderpass
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 15:49:37 +0000 (10:49 -0500)]
zink: refuse to start cs invocation queries in renderpass
these can't produce data in renderpasses, so ensure they aren't accidentally
started inside a renderpass where they'll have to later be restarted outside
of one
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 17:13:05 +0000 (12:13 -0500)]
zink: don't double suspend queries
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 17:01:22 +0000 (12:01 -0500)]
zink: un-suspend queries if they end while suspended
ensure they don't accidentally resume themselves after suspend
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 15:44:16 +0000 (10:44 -0500)]
zink: add zink_query::suspended to indicate suspended state of queries
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 15:42:36 +0000 (10:42 -0500)]
zink: don't auto-sync qbos on query end
these can now be synced later all at once
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 14:52:07 +0000 (09:52 -0500)]
zink: handle multiple query starts in qbo update
this still isn't used due to how queries are started/stopped, but it should
be a bit more robust
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Tue, 21 Feb 2023 21:59:26 +0000 (16:59 -0500)]
zink: delete zink_query::last_start_idx
this was always zero
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Tue, 21 Feb 2023 21:24:41 +0000 (16:24 -0500)]
zink: make zink_vk_query unref consistent
the no-check free seems suspicious
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Fri, 24 Feb 2023 12:49:29 +0000 (07:49 -0500)]
zink: fix possible query destroy leak
if the starts array has been reset, then the counter will be inaccurate,
and some of the members will leak, so this needs to iterate over the capacity
of the array instead of the contents
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 18:13:27 +0000 (13:13 -0500)]
zink: fix zink_query_start initialization
not sure if this was a bug, but it would have fully zeroed the struct
and prevented previous zink_vk_query members from being unrefed
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Wed, 22 Feb 2023 18:12:34 +0000 (13:12 -0500)]
zink: rewrite zink_query_start struct
making the non-pool members easier to zero
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Tue, 21 Feb 2023 21:11:45 +0000 (16:11 -0500)]
zink: reset queries on the promoted cmdbuf when possible
this avoids splitting renderpasses just to reset queries
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Thu, 23 Feb 2023 13:20:08 +0000 (08:20 -0500)]
zink: remove suspended queries from list before resuming
this avoids double-starting primgen queries from recursion when
needs_rast_discard_workaround is true
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Mike Blumenkrantz [Tue, 21 Feb 2023 20:14:41 +0000 (15:14 -0500)]
zink: move zink_batch_no_rp call for query reset
this could otherwise pointlessly terminate a renderpass
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21534>
Patrick Lerda [Sat, 25 Feb 2023 13:08:53 +0000 (14:08 +0100)]
mesa/shaderapi: fix path memory leak
For instance, with "piglit/arb_shading_language_include-api -auto -fbo" or
"piglit/shader_runner tests/spec/arb_shading_language_include/execution/replacement.shader_test -auto -fbo":
Direct leak of 66 byte(s) in 6 object(s) allocated from:
#0 0x7fa4b59050 in __interceptor_strdup (/usr/lib64/libasan.so.6+0x59050)
#1 0x7f9a098fe0 in validate_and_tokenise_sh_incl ../src/mesa/main/shaderapi.c:3383
#2 0x7f9a0a43e8 in _mesa_NamedStringARB ../src/mesa/main/shaderapi.c:3547
Signed-off-by: Patrick Lerda <patrick9876@free.fr>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21553>
Giancarlo Devich [Mon, 27 Feb 2023 19:03:42 +0000 (11:03 -0800)]
d3d12: Don't loop in update_draw_indirect_with_sysvals
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21567>
Giancarlo Devich [Fri, 24 Feb 2023 21:31:59 +0000 (13:31 -0800)]
d3d12: Create varying structures as necessary, reference them
This changes instances of d3d12_varying_info to d3d12_varying_info*,
significantly reducing the size of the d3d12_shader_key,
d3d12_gs_variant_key, and d3d12_tcs_variant_key.
Associated changes to key fill, compare, hashing, and gs and tcs variant
maps significantly reduce the amount of time spent clearing and
comparing memory.
The biggest win here is not having to re-zero _or_ re-fill varyings in
d3d12_fill_shader_key, validate_geometry_shader_variant, and
validate_tess_ctrl_shader_variant.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21527>
Giancarlo Devich [Fri, 24 Feb 2023 21:26:46 +0000 (13:26 -0800)]
d3d12: Don't memcmp gs/tcs keys
This is unnecessary and incomplete. Not only is it covered by the hash
map lookup, but also does not consider the other key members.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21527>
Giancarlo Devich [Fri, 24 Feb 2023 20:00:16 +0000 (12:00 -0800)]
d3d12: Don't unnecessarily zero out gs/tcs keys
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21527>
Giancarlo Devich [Fri, 24 Feb 2023 19:50:57 +0000 (11:50 -0800)]
d3d12: Track max varying slot, set and compare less bytes
Often, the full range of available slots don't need to participate in
the comparison or be zeroed out.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21527>
Giancarlo Devich [Fri, 24 Feb 2023 19:30:07 +0000 (11:30 -0800)]
d3d12: Don't recompute has_flat_varyings or missing_dual_src_outputs
Per shader selection, these two flags can be expensive to compute.
Instead, recompute their values as neccessary, and save them in the
context.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21527>
Emma Anholt [Tue, 21 Feb 2023 19:32:41 +0000 (11:32 -0800)]
anv: Skip BTI RT flush if we're doing an op that doesn't use render targets.
rt_flushes emitted on zink sauer.trace --loop=500 -2.02118% +/- 1.15992% (n=8).
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21508>
Emma Anholt [Tue, 21 Feb 2023 17:43:46 +0000 (09:43 -0800)]
anv: Skip the RT flush when doing depth-only rendering.
The spec citation says it's just for when the RT write message BTI might
point to a different RT, and if we don't have any color attachments then
we won't have one of those at all.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21508>
Mike Blumenkrantz [Mon, 27 Feb 2023 18:52:30 +0000 (13:52 -0500)]
aux/tc: only call tc_parse_draw() when parsing renderpass info
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21533>
Mike Blumenkrantz [Wed, 22 Feb 2023 19:58:44 +0000 (14:58 -0500)]
aux/tc: track whether queries have been terminated in a renderpass
on tilers it's important to know whether a query is ended mid-renderpass
so that the query begin can occur inside/outside of the renderpass
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21533>
Caio Oliveira [Mon, 27 Feb 2023 06:02:49 +0000 (22:02 -0800)]
panfrost: Use NIR scoped barriers instead of memory barriers
Now both GLSL and SPIR-V will produce the scoped barriers, so no
need to handle the old ones.
Control barriers are still present in some cases, so keep that
for now.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3339>
Caio Oliveira [Mon, 27 Feb 2023 05:57:49 +0000 (21:57 -0800)]
pan/midgard: Handle nir_intrinsic_scoped_barrier in Midgard compiler
Behave the same as the existing more specific barrier intrinsics.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3339>
Caio Oliveira [Mon, 27 Feb 2023 05:40:16 +0000 (21:40 -0800)]
pan/compiler: Fix handling of nir_intrinsic_scoped_barrier
Only emit anything if this is a control barrier, i.e. contains an
execution scope. Also change the assertion to look at that scope
instead of the memory scope.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3339>