Rhys Perry b318fe47e9 aco: don't byte align global VMEM loads if it might be unsafe ...

Using the byte align path can be unsafe even when 12 byte loads are
supported.

fossil-db (navi21):
Totals from 185 (0.23% of 79395) affected shaders:
Instrs: 391501 -> 391575 (+0.02%); split: -0.03%, +0.05%
CodeSize: 2147336 -> 2147672 (+0.02%); split: -0.03%, +0.05%
Latency: 3762613 -> 3860941 (+2.61%); split: -0.01%, +2.62%
InvThroughput: 871429 -> 888013 (+1.90%); split: -0.08%, +1.98%
VClause: 9712 -> 10210 (+5.13%)
Copies: 53775 -> 53010 (-1.42%); split: -1.46%, +0.04%
VALU: 254009 -> 252146 (-0.73%)
SALU: 56698 -> 56699 (+0.00%); split: -0.00%, +0.00%
VMEM: 18503 -> 19601 (+5.93%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Fixes: 391bf3ea30 ("aco: don't expand smem/mubuf global loads")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31807>

2024-11-06 19:07:16 +00:00

..

amd

aco: don't byte align global VMEM loads if it might be unsafe

2024-11-06 19:07:16 +00:00

android_stub

vulkan/android: Add helper to probe AHB support

2024-05-14 14:53:44 +00:00

asahi

asahi/clc: ingest spir-v

2024-11-01 13:25:37 -07:00

broadcom

broadcom/ci: document flakes seen lately

2024-11-06 14:14:38 +00:00

c11

build: pass licensing information in SPDX form

2024-06-29 12:42:49 -07:00

compiler

nir/lower_clip_disable: handle non-scalar store intrinsics

2024-11-06 15:51:51 +00:00

drm-shim

drm-shim: stub synobj_timeline_wait and query ioctl

2024-07-16 11:17:59 +02:00

egl

egl: error out during setup if the configuration is invalid

2024-10-29 17:25:08 +00:00

etnaviv

etnaviv: isa: Add img_store instruction

2024-10-23 07:58:04 +00:00

freedreno

ir3: merge is_reg_gpr and reg_gpr

2024-11-06 17:00:25 +00:00

gallium

freedreno/ci: document flakes seen lately

2024-11-06 14:14:38 +00:00

gbm

gbm/dri: Fix color format for big endian.

2024-10-25 14:18:24 +00:00

getopt

build: pass licensing information in SPDX form

2024-06-29 12:42:49 -07:00

gfxstream

gfxstream: update Kumquat API

2024-11-05 16:28:56 -08:00

glx

glx: Fix the GLX_EXT_swap_control_tear drawable attributes

2024-10-23 19:55:53 +00:00

gtest

build: pass licensing information in SPDX form

2024-06-29 12:42:49 -07:00

imagination

pvr: Make Vulkan driver depend only on nir headers, not library

2024-10-23 16:47:34 +00:00

imgui

…

intel

anv: move pipe control debug to anv_util.c

2024-11-06 12:20:23 +00:00

loader

loader: Fix typo in __DRI_IMAGE_FORMAT_XBGR16161616 definition

2024-10-25 14:18:24 +00:00

mapi

meson: remove selinux option

2024-10-21 01:14:35 +00:00

mesa

st/mesa: implement key->persample_shading for lowered IO

2024-11-06 15:51:51 +00:00

microsoft

nir: always emit ddx intrinsics

2024-10-17 09:50:19 +00:00

nouveau

nvk/ci: document flakes seen lately

2024-11-06 14:14:38 +00:00

panfrost

pan/bi: Lower nir_texop_txd to TEXC in GRDESC_DER mode followed by sampling TEXC

2024-11-05 11:20:21 +00:00

tool

perfetto: Add Panfrost data sources to system.cfg

2024-08-22 18:33:45 +00:00

util

util/ralloc: add total_size helper

2024-10-30 12:59:10 +00:00

virtio

virtio/drm: Update header

2024-10-17 18:17:18 +00:00

vulkan

vulkan/runtime: fix allocation failure handling

2024-11-06 11:23:27 +00:00

x11

meson: require dri3 modifiers

2024-09-06 17:34:17 +00:00

.clang-format

panfrost: Move ForEachMacros into panfrost

2024-09-12 20:04:58 +00:00

meson.build

mesa: add gfxstream-experimental to -Dvulkan-drivers

2024-09-19 20:06:04 +00:00