This is used when printing the program and to avoid updating register
demand during post-RA liveness analysis.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10315>
p_extract_vector copy-propagation can create byte sels for v2b operands.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10315>
It isn't intended to be accurate after RA, so num_waves can become zero,
breaking the sgpr_limit calculation.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10315>
DCC decompress/FMASK expand are supported on compute queues. Since
the driver doesn't perform fast clears with concurrent images, we
don't need to perform a FCE on compute (we can't anyways).
This fixes a performance regression with Control and
VKD3D_CONFIG=multi_queue.
One more optimization (as discussed with Bas) is to implement FCE
on compute to allow fast clears.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10323>
We can just check whether tex_instr is NULL instead.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10036>
Turns out both kernel v5.10 and v5.11 have the same amdgpu driver
version and only one has modifiers ... In addition the version check
is kinda annoying for backports.
So lets use the cap. Since the cap is technically about ADDFB2 I
tested that this works on rendernodes (and reading the code there
is no distinction from what kind of node this is called).
Fixes: 9a937330ef ("radeonsi: Only set modifier creation function for GFX9+ & with kernel support.")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10337>
If depth bias is enabled but zero values used, they were never
emitted to the command buffer because they are equal to the default
values.
Previously, they were always emitted when the bound DS attachment
changed.
This should fix some sort of Z fighting with Dota2 on all GPUs.
This also fixes a different issue (ie. some occlusion queries failures)
on GFX6 because CLEAR_STATE is not used on that chip.
Fixes: 8a47422d97 ("radv: do not scale the depth bias for D16_UNORM depth surfaces")
Cc: 21.1 mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10310>
Make sure to preserve signed zeroes.
Fixes dEQP-VK.spirv_assembly.instruction.compute.opquantize.flush_to_zero
on GFX6 (Pitcairn). Untested on GFX7.
Fixes: 54a09545ec ("aco: optimize a*0.0")
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/10319>
because gfx103.json is automatically generated and can't be changed
manually. This fixes the file generator without changing the generated
header.
Missing registers must be in registers-manually-defined.json, and
missing fields must be in parse_kernel_headers.py.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10261>
Fixes
dEQP-VK.api.image_clearing.core.clear_color_image.2d.optimal.single_layer.e5b9g9r9_ufloat_pack32_33x128
with RADV_DEBUG=forcecompress on GFX10.3.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Cc: 21.1 <mesa-stable>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10176>
Without image stores, DCC is always decompressed on compute.
Cc: 21.1 <mesa-stable@lists.freedesktop.org>
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10168>
This would be easy to support except that it doesn't support RDNA 2.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10199>
It doesn't support Navi1x and the removal enables this nice code cleanup.
v2: rebase - mareko
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> (v1)
Acked-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10199>
clang generates a warning if there's no explicit break or fall-through
annotation. The latter would be kind of silly in this case, and not
robust against any future changes turning the fall-through invalid.
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10220>
One exception is src/amd/addrlib/, for which -Wimplicit-fallthrough is
explicitly disabled.
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10220>
ACO doesn't create a waitcnt for barriers between texture samples and
image stores because texture samples are supposed to use read-only
memory. It could also schedule the barrier to above the texture sample.
We also have use a larger memory scope to avoid an ACO optimization.
Tested on GFX8 with Sachsa Willems deferred sample. With some DCC
decompressions and the compute path forced.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Cc: 21.1 <mesa-stable>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9496>
Allocating a descriptor set is aligned to 32 bytes, so just like the
other buffer types, bump the descriptor size to 32 bytes when allocating
MUTABLE descriptor types from a pool.
Fixes: 86644b84b9 ("radv: Implement VK_VALVE_mutable_descriptor_type.")
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10132>
Accept non-linear tiling for multi-planar formats on GFX9+, as long
as DCC is disabled. DCC support is possible in theory, but untested
for now.
GFX8 is still restricted to linear tiling because it's not yet clear
how modifiers should be handled on these chips for multi-planar
formats. Each plane may need a different modifier.
Signed-off-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Acked-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10134>
util_format_get_blocksize asserts that the blocksize isn't zero.
However the blocksize will be zero if the format's channel encoding
is unspecified. The channel encoding is only meaningful for the
plain u_format layout, so util_format_get_blocksize can't be used
for formats with another layout. For example, YUV formats don't have
the channel encoding specified.
Use util_format_get_blocksizebits, which just returns zero without
an assertion for formats which don't have a channel encoding.
Signed-off-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10134>
Otherwise, handle_loop_phis() might pass it to handle_live_in() and then
we could have two phis for this variable.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Fixes: 7c64623e94 ("aco/ra: refactor SSA repairing during register allocation")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10236>
In most games I tested we use 32 MiB of cmdbuffers+cmd upload buffers
at most. Especially since we have mutable descriptors it seems
somewhat unlikely anything else will eat it up so be a bit more
aggressive allocating them in VRAM.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10198>
Late export is theoretically better if used with LATE_ALLOC,
but in practice, the early export has an advantage of
lower register usage, therefore more concurrent waves.
The idea of this commit is that "small" shaders benefit from early
primitive export more, due to being able to launch much more waves.
Let's consider a NIR shader "small" when it has only 1 block.
This yields both better performance, and better stats, than always
using late export.
Fossil DB on Sienna:
Totals from 12807 (8.76% of 146265) affected shaders:
VGPRs: 609128 -> 620216 (+1.82%); split: -0.01%, +1.83%
SpillSGPRs: 1458 -> 1538 (+5.49%)
CodeSize: 37028204 -> 37019320 (-0.02%); split: -0.17%, +0.14%
MaxWaves: 282902 -> 278516 (-1.55%)
Instrs: 7163142 -> 7162925 (-0.00%); split: -0.18%, +0.18%
VClause: 169285 -> 169547 (+0.15%); split: -1.15%, +1.30%
SClause: 267373 -> 267151 (-0.08%); split: -0.24%, +0.16%
Copies: 446442 -> 444567 (-0.42%); split: -2.68%, +2.26%
Branches: 156245 -> 156195 (-0.03%); split: -0.30%, +0.26%
PreSGPRs: 434701 -> 447396 (+2.92%)
PreVGPRs: 527783 -> 540527 (+2.41%)
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10106>
The user-set priority of shaders matters very little, but we hope
this might still help speed up VS input loads especially.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10106>
We learned that the gs_alloc_req is not actually when the export
space allocation happens. So it makes no sense to prioritize it.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10106>
Previously, every wave had multiple active lanes read the LDS, and
the data was processed by VALU DPP instructions.
Now, only the first lane reads the LDS in order to avoid bank
conflicts, and the results are processed by SALU.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10155>
No fossil-db changes, probably because all fp16 shaders have at least one
16-bit mov or vec2 somehwere.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10227>
Layered VRS attachments is for later.
The CTS failures are similar to the existing ones, I will investigate
soon.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10187>
