feat(vk): bind storage and texel descriptors as piece arrays in the SPIR-V back end #648

Closed

mooki wants to merge 2 commits from mooki/Longshot:mac/seg-shader into main

pull from: mooki/Longshot:mac/seg-shader

merge into: MagicBeanDev:main

MagicBeanDev:main

Conversation 1 Commits 2 Files changed 20 +1840 -60

mooki commented 2026-11-15 02:17:23 +00:00

Member

• Copy link
• Copy Markdown

Type: feat · Risk: low · Tested on: macOS 26.7.1, Apple M4 Max (MoltenVK 1.4.2), game-free at this head; GPU probes on both CPU routes (history) · Needs: Windows, Linux

Why

On a Mac, MoltenVK caps one device allocation at 41,747,087,360 bytes (M4 Max). The guest memory spans the game binds as
storage and texel buffers can be larger, so the macOS port imports such a span as several allocations (segments), and a
descriptor whose range crosses a cut cannot be one buffer. This PR is the shader half of binding it anyway: the SPIR-V back
end can declare such a descriptor as an array of pieces, one per segment, at its own binding, and every access finds its
piece. It also carries the fix for the one pattern of it MoltenVK refused: a segmented program with a compare-exchange
loop (an increment, a decrement, or a float add the GPU has no atomic for) failed pipeline creation with "no matching
function for call". Nothing is drawn with pieces until the draw side (#663) asks for a segmented variant, so on its own
this PR changes no frame on any platform. It is part of the macOS port. See #420.

What changed

The back end learns RuntimeInfo::segmented: a list of descriptors to bind as pieces. With the list empty, which is every
program today, it emits byte for byte what it did. With it, the descriptor is an array of pieces at its binding, and each
load, store, atomic, texel fetch and size query goes through a small helper function that walks the pieces by their lengths
with constant indices. The renderer keeps such a variant apart from the plain program (never cached on disk, never picked for
a plain draw). The second commit writes the compare-exchange loops out inside those helpers, because SPIRV-Cross's MSL cannot
pass a pointer to one piece of a buffer array to a function.

• back end, gpu/vk/eden/shader_recompiler/runtime_info.h: SegEntry {Storage|Table|TexBuf|ImgBuf, desc, pieces} and
  RuntimeInfo::segmented.
• back end, backend/spirv/spirv_emit_context.{h,cpp}: DefineSegmentedDescriptors checks the list against the program
  (refused with LogicError otherwise); DefineSsbos, DefineTextureBuffers and DefineImageBuffers declare a flagged
  descriptor as TypeArray(…, pieces) at the same binding (a table's count * pieces, entry k's piece j at
  k * pieces + j); SegmentedStorageCall, SegmentedTextureBufferCall, SegmentedImageBufferCall and
  SegmentedTextureBufferSize ask for one DontInline helper per (view, op, result type), made after main by
  DefineSegmentedHelpers and named in the calls by PatchSegmentedCalls; DefineGlobalMemoryFunctions sums a segmented
  descriptor's piece lengths for its bound end and branches to the piece (single, native candidate, table written out and in
  the loop).
• back end, emit_spirv_memory.cpp, emit_spirv_atomic.cpp, emit_spirv_image.cpp, emit_spirv_image_atomic.cpp: the
  storage loads, stores and atomics (CAS loops, float adds, the 64-bit and 32x2 fallbacks), texel fetches, image buffer
  reads, writes and atomics dispatch to the helpers for a flagged descriptor; emit_spirv.cpp calls the two new steps,
  which do nothing without helpers.
• back end, the MslArrayPtr fix (second commit): EmitContext::CasWordHere, CasLoopHere and CasStoreHere write
  CasLoop's loop (and write_storage_cas_loop_func's) out in the piece, attempt for attempt, with the same
  CAS_LOOP_MAX_TRIES ceiling; CasLoop records each loop's operation (cas_operations); SegmentedCasLoop and
  WriteStorageByCasLoop use them, and DefineSegmentedHelpers makes the loop's two variables first in the helper.
• renderer, gpu/vk/vk_shader.{h,cpp}: SegSig (sorted entries and a hash) on Program, seg_sig(), seg_variant(),
  program_segmented() (finds or translates the variant of a plain program with plain's questions and answers, and fails
  it loudly if its descriptors are not plain's); program_pick, mesh_program_locked, program_find_plain,
  loaded_variants and portable_variant keep the variants apart; every stage translation sets RuntimeInfo::segmented
  from Program::seg (empty in every plain variant).
• renderer, gpu/vk/vk_progcache.cpp: a segmented variant is never written to the program cache nor listed among the
  loaded programs.
• tests: tests/test_seg_shader.py and tests/seg_shader_ir.cpp (new) run synthetic IR programs through the real back
  end, plain and in pieces, check the segmented modules' structure and capabilities, run every module through spirv-val
  (Vulkan 1.3, uniform buffer standard layout, as CI validates the back end), check that no module passes a pointer into a
  buffer descriptor array to a function, and with --base-tree compare the plain modules byte for byte with an older
  tree's. Seven source-reading checks follow the new field, the atomics' new branch and the new loop:
  test_native_float_atomic_add, test_pipeline_state_portable, test_program_cache_cleanup,
  test_program_cache_one_variant, test_seed_kept_loaded, test_seed_variant_cleanup, test_cas_loop_ceiling.

Measurements

Verdict: passes. The plain SPIR-V is byte for byte main's on both CPU routes, every module validates, and the array
pointer check fails without the fix (3 segmented modules, 54 calls) and passes with it. No speed to measure: nothing is
drawn with pieces until #663, and the plain path's SPIR-V is unchanged.

Hardware: macOS 26.7.1, Apple M4 Max, 64 GB, Apple clang 21 (game-free: no GPU, no game)
Scene: none (an emitter change: its proof is the fixture's modules)
Method: tests/test_seg_shader.py at this head with --base-tree a git archive of main 6c9edffe's back end, on arm64
and with --arch x86_64 (compiled for Intel, run under Rosetta); the same test file against the first commit's back end
(cbcdb34d, the pieces without the fix). Every output is a byte comparison or a hash, so one run per route and side; the
two routes' 29 modules have equal SHA-256.
How to reproduce: python -I -B tests/test_seg_shader.py --headers <SPIRV-Headers/include> --spirv-val spirv-val --base-tree <git archive of main> [--arch x86_64]
Baseline: the middle column is this PR's first commit alone, the pieces before the fix; main has no segmented module to
check.

	Before: main @ 6c9edffe	After, first commit only (no fix): this PR @ cbcdb34d	After: this PR @ 9d48901a
plain modules, arm64 and x86_64	14 emitted (the base tree)	—	14 of 14 byte-identical to main's on both routes
modules passing spirv-val	—	29 of 29	29 of 29, both routes
modules passing a pointer into a buffer array to a function	—	3 of 29 (seg-storage 18 calls, seg-storage-native-fadd 15, seg-storage-narrow 21): exit 1	0 of 29: exit 0
arm E (SegProbe's module) SHA-256	—	01a0f43f…	01a0f43f…, as SegProbe pinned

this PR @ 9d48901a, arm64 (x86_64 the same lines):
array pointers: 29 of 29 modules pass no pointer into a buffer descriptor array to a function
base tree …/base-6c9edffe: 14 of 14 plain modules byte-identical
PASS: segmented descriptors through the real back end; every module validates
this PR's test against cbcdb34d's back end, arm64:
seg-storage-narrow.spv: 21 calls pass a pointer into a buffer array: %105(ssbo0), %132(ssbo0), %162(ssbo0), %191(ssbo0), %219(ssbo0), %240(ssbo0), %74(ssbo0)
seg-storage-native-fadd.spv: 15 calls pass a pointer into a buffer array: %118(ssbo0), %149(ssbo0), %176(ssbo0), %205(ssbo0), %233(ssbo0)
seg-storage.spv: 18 calls pass a pointer into a buffer array: %110(ssbo0), %141(ssbo0), %168(ssbo0), %198(ssbo0), %227(ssbo0), %255(ssbo0)
array pointers: 26 of 29 modules pass no pointer into a buffer descriptor array to a function

History, GPU (MoltenVK 1.4.2, M4 Max, arm64 and x86_64 under Rosetta), and why it holds for this head:

• SegProbe revision 02 ran this emitter's arm E module (a 4-piece storage buffer, read and written through the helpers):
  E.read_g1_layout and E.write_g2_layout PASS on both routes, 0 [mvk-error] (reports/2026-11-14-seg-probe.md
  section 3d). This head emits the same module (SHA-256 01a0f43f7da317c61ab1ee0dab18448d8dcafb798a1fcabdf5d08a0db8b69cc0).
• MslProbe revision 02 ran the fix's modules (reports/2026-11-14-msl-arrayptr.md sections 11 and 12): every pipeline was
  made, and the written-out loops (increment, decrement, float add, f32x2 add/min/max, the narrow profile's 8- and 16-bit
  stores) gave the CPU interpreter's bytes in every piece; ma-cas-seg, ma-cas-narrow-seg, their whole controls and the
  table program mc-table MATCH on both routes. Those five modules were built from the NullShader branch's back end; this
  head's back end emits the same five programs byte for byte (reports/PubWave-files/PubShader/compare-mslprobe.txt), so
  the result is this head's.
• Without the fix, that branch's segmented program NS2 failed pipeline creation on MoltenVK with no matching function for call, &ssbo0_4[0u] given to a device _66 * parameter (NullProbe revision 02; the cause in SPIRV-Cross's source is in
  reports/2026-11-14-msl-arrayptr.md section 1).

Flags and settings

None.

Testing

Tested: on macOS 26.7.1, Apple M4 Max, at 9d48901a, this PR's head before the rebases onto e090cc48 and aa95d35f
(main 6c9edffe plus this PR), game-free:

• tests/test_seg_shader.py --base-tree <main 6c9edffe> on arm64 and x86_64 (Rosetta): exit 0 both, as in Measurements;
  the 29 modules of the two routes are byte-identical, and byte-identical to the macOS port's integration tree after the
  fix merged (7cd720da, both routes: every plain and segmented module and arm E). Without the fix (the first commit's back
  end) the three segmented storage modules are the topic's original bytes (b7928876 on a4802cb5) and fail the array
  pointer check.
• The five MslProbe programs (build_cas_run whole and in 3 pieces, both profiles, and build_table_run, copied from the
  NullShader branch's fixture into a scratch copy of this head's, bit off) through this head's back end: byte-identical to
  MslProbe revision 02's frozen GPU inputs; all validate and pass the array pointer check.
• Renderer files that include vk_shader.h (vk_shader.cpp, vk_progcache.cpp, vk_draw.cpp, vk_predict.cpp,
  vk_replay.cpp, vk_programs.cpp) compiled for arm64 and x86_64 at this head and on main 6c9edffe with main's renderer
  flags (nx2/build.py's vflags plus -arch, and main's non-Windows -fPIC -I host/posix -D_GNU_SOURCE). Every tree got
  #492's two Darwin fixes (environ, RUSAGE_THREAD) and a compile-only prelude declaring posix_fallocate and
  posix_fadvise, which main's renderer calls and macOS lacks (#476 gives them Darwin paths): 12 of 12 objects on each,
  exit 0.
• The seven source-reading checks this PR updates (tools/ls-suite.py named run): 5 pass. test_program_cache_cleanup and
  test_program_cache_one_variant cannot compile their fixture on this Mac, exactly as on main 6c9edffe (main's
  vk_progcache.cpp uses POSIX_FADV_WILLNEED; main's own macOS suite fails both the same way). With #476's two Darwin
  hunks (vk_progcache.cpp, pipeline_cache_file.h) applied to scratch copies of main and this head, both pass on both
  (main 9 tests OK and the same variant for 9 picks; this head the same).
• Logs: reports/PubWave-files/PubShader/proof/p20-*.log, proof/fadv-*.log, compare-p20.txt, compare-mslprobe.txt,
  compile-*.out, named/, merge-tree.txt, stamps.txt.
• After the rebase onto main aa95d35f, at 9b73e829 (9d48901a before the rebases onto e090cc48 and aa95d35f):
  Sweep3's checks of this head (the C/C++ it changes compiled, and its own tests run, on arm64 and x86_64 where they
  apply) gave the same results as before; python -I -B tests/run_all.py on wave B's merge db0e2e16, which carries
  this head: 569/32/18 (passed/failed/skipped) against main aa95d35f's 543/33/17 on this Mac, no new failure
  (reports/2026-11-15-rebase-sweep-3.md).

Not tested: the game at this head: per-PR game runs are not planned, and nothing here runs before #663 asks for a
segmented variant. The GPU at this head: the probes above are history whose modules this head reproduces byte for byte. The
f16x2 compare-exchange loops have no fixture program (the CPU interpreter has no 16-bit float vectors); they share the f32x2
loops' code path. Windows and Linux: not built or run there (11 changed renderer and back-end files differ from main's off
Apple, unifdef -U__APPLE__); its test request covers the build, run_all on Linux and a start.

Must be tested:

• Linux and Windows, any GPU (Linux runs the SPIR-V test): build this PR (head 9b73e829) and main aa95d35f the same way
  (the 11 every-platform files: the back end's emit_spirv.cpp, emit_spirv_atomic.cpp, emit_spirv_image.cpp,
  emit_spirv_image_atomic.cpp, emit_spirv_memory.cpp, spirv_emit_context.{h,cpp}, runtime_info.h;
  vk_shader.{h,cpp}, vk_progcache.cpp); on Linux run python -I -B tests/run_all.py, expecting
  tests/test_seg_shader.py to pass (it needs spirv-val and clang++; Windows skips it); start each build to the
  title, expecting this PR to translate the shader caches again once (the stamp changed) and nothing else to differ from
  main, in the log or on screen (by hand: test request T-WLDY)

Risks

What could break: nothing runs differently until #663 calls program_segmented: every program's
RuntimeInfo::segmented is empty, and with it empty the back end's output is byte for byte main's (checked on 14 programs,
both routes). Program grows by a SegSig (an empty vector and a hash). Once #663 uses it on macOS: a segmented variant is
translated on the drawing thread under shader_mu at its first draw, and one whose descriptors differ from plain's is failed
and logged, never drawn with pieces.

Turning it off: revert the commits (there is no switch: nothing asks for pieces before #663).

Compatibility: every platform's shader caches are translated again once: eden/ and vk_shader.cpp are in the
translator's stamp (nx2/translator_files.py: ebbf160ce9626041 on main, 40b414b30dad80e1 here), so each player's
program cache, prepare record and pipeline cache are dropped at the first start after updating. kProgramCacheFormat is
unchanged here (#649 raises it). Nothing changes in saves, longshot_options.ini or the vk ABI (MKWVK_API_VERSION).

Reviewer notes

• Review cbcdb34d..9d48901a: this PR's two commits, cbcdb34d (the pieces, -x from macos/seg-shader b7928876)
  and 9d48901a (the MslArrayPtr fix, -x from macos/msl-arrayptr bd396010). Read runtime_info.h, then
  DefineSegmentedHelpers and SegmentedStorageCall in spirv_emit_context.cpp, then program_segmented in
  vk_shader.cpp; then CasWordHere/CasLoopHere.
• One PR for two topics (owner decision): the pieces and their compare-exchange fix ship together, so no state of main has
  segmented atomics that MoltenVK refuses.
• How the fix was ported: bd396010 was committed on the NullShader branch, where a storage descriptor bound whole can also
  reach the segmented helpers (StorageThroughHelper). Here every access that reaches a helper is in pieces, so the loops are
  always written out; the bound-whole branch, the guarded helper's cas_loop, and the tests that need NullShader or
  tests/spirv_run.h (the loops' and a table's CPU runs, --changed-segmented, the *.cpurun comparison) stay for #674.
  The cherry-pick's conflicts (emit_spirv_memory.cpp, spirv_emit_context.{h,cpp}, seg_shader_ir.cpp,
  test_seg_shader.py, spirv_run.h absent here) were all that difference.
• Stacking: on main 6c9edffe, no dependency. #649 stacks on it; the draw side (#662, #663, #663) and #674 use it.
• Conflicts with main: none in the cherry-picks. No semantic rework: main changed vk_shader.cpp (#252 make_module's
  Intel gate and driver_faulted_program in translate_into, which program_segmented reaches through translate_into,
  so a program the driver faulted on is not asked again as a segmented variant either), vk_progcache.cpp (#321's paths,
  #252) and test_native_float_atomic_add.py (#191, #444) beside this PR's lines. #370 moved the stamp's file list to
  nx2/translator_files.py and took vk_api.h out of it: the first commit's message still says to land it with a
  vk_api.h change; the stamp changes here either way.
• With the open macOS PRs (#431, #437, #438, #479, #480, #490, #491, #492) and this wave's branches on main 6c9edffe:
  none (git merge-tree).
• Code comments cite reports/2026-11-14-*.md (the port's design notes and the MslArrayPtr report), which are not in this
  repository.

Issues

See #420

Checklist

Everyone

• One topic.
• The title is a Conventional Commit (type(scope): description) of the status line's type.
• Why gives the reason plainly and links the issues that fit (the tracker searched).
• Measurements has a verdict, or one line saying why there is nothing to measure.
• Testing names what was tested, what wasn't, and what must still be tested, where and how. (Test request T-WLDY.)
• Rebased on current main: on aa95d35f (#761, 2026-11-15), head 9b73e829, without a conflict; this PR's own
  commits, old → new: cbcdb34d → 1f9183c2, 9d48901a → 9b73e829; the commit names elsewhere in this description
  are from before the rebase.
• python -I -B tests/run_all.py passes (any failure that also fails on main is named). (On the wave's merge of every macOS PR built with this one, 453eb678: 553 passed, 32 failed, 18 skipped against main 6c9edffe's 531/32/17 on this Mac; the failures are main's own except test_split_ub_flush, which #658's tree at the time broke and its head fixes; test_collect_wide now passes; the new tests pass, test_seg_shader among them, and test_darwin_audio_service, Darwin-only, is skipped by the runner; reports/PubWave-files/waveA/. After the rebase onto main aa95d35f (head 9b73e829): wave B's merge db0e2e16, 569/32/18 against main aa95d35f's 543/33/17, no new failure.)
• No game files, dumps, build output or generated files.
• third_party/ untouched (a new folder with a PROVENANCE.md instead).
• The license is OK: GPL-3.0-or-later, nothing incompatible and nothing decompiled.

When it applies

• AI assistance recorded with Assisted-by: / Co-Authored-By: trailers.
• A game-free test added for the fix, failing on main and passing here. (It fails on this PR's first commit, the
  pieces without the fix; main has no segmented module.)
• A check under tests/ that reads changed source updated in this PR, not weakened.

**Type:** feat · **Risk:** low · **Tested on:** macOS 26.7.1, Apple M4 Max (MoltenVK 1.4.2), game-free at this head; GPU probes on both CPU routes (history) · **Needs:** Windows, Linux ## Why On a Mac, MoltenVK caps one device allocation at 41,747,087,360 bytes (M4 Max). The guest memory spans the game binds as storage and texel buffers can be larger, so the macOS port imports such a span as several allocations (segments), and a descriptor whose range crosses a cut cannot be one buffer. This PR is the shader half of binding it anyway: the SPIR-V back end can declare such a descriptor as an array of pieces, one per segment, at its own binding, and every access finds its piece. It also carries the fix for the one pattern of it MoltenVK refused: a segmented program with a compare-exchange loop (an increment, a decrement, or a float add the GPU has no atomic for) failed pipeline creation with "no matching function for call". Nothing is drawn with pieces until the draw side (#663) asks for a segmented variant, so on its own this PR changes no frame on any platform. It is part of the macOS port. See #420. ## What changed The back end learns `RuntimeInfo::segmented`: a list of descriptors to bind as pieces. With the list empty, which is every program today, it emits byte for byte what it did. With it, the descriptor is an array of pieces at its binding, and each load, store, atomic, texel fetch and size query goes through a small helper function that walks the pieces by their lengths with constant indices. The renderer keeps such a variant apart from the plain program (never cached on disk, never picked for a plain draw). The second commit writes the compare-exchange loops out inside those helpers, because SPIRV-Cross's MSL cannot pass a pointer to one piece of a buffer array to a function. - back end, `gpu/vk/eden/shader_recompiler/runtime_info.h`: `SegEntry {Storage|Table|TexBuf|ImgBuf, desc, pieces}` and `RuntimeInfo::segmented`. - back end, `backend/spirv/spirv_emit_context.{h,cpp}`: `DefineSegmentedDescriptors` checks the list against the program (refused with `LogicError` otherwise); `DefineSsbos`, `DefineTextureBuffers` and `DefineImageBuffers` declare a flagged descriptor as `TypeArray(…, pieces)` at the same binding (a table's `count * pieces`, entry k's piece j at `k * pieces + j`); `SegmentedStorageCall`, `SegmentedTextureBufferCall`, `SegmentedImageBufferCall` and `SegmentedTextureBufferSize` ask for one DontInline helper per (view, op, result type), made after main by `DefineSegmentedHelpers` and named in the calls by `PatchSegmentedCalls`; `DefineGlobalMemoryFunctions` sums a segmented descriptor's piece lengths for its bound end and branches to the piece (single, native candidate, table written out and in the loop). - back end, `emit_spirv_memory.cpp`, `emit_spirv_atomic.cpp`, `emit_spirv_image.cpp`, `emit_spirv_image_atomic.cpp`: the storage loads, stores and atomics (CAS loops, float adds, the 64-bit and 32x2 fallbacks), texel fetches, image buffer reads, writes and atomics dispatch to the helpers for a flagged descriptor; `emit_spirv.cpp` calls the two new steps, which do nothing without helpers. - back end, the MslArrayPtr fix (second commit): `EmitContext::CasWordHere`, `CasLoopHere` and `CasStoreHere` write CasLoop's loop (and `write_storage_cas_loop_func`'s) out in the piece, attempt for attempt, with the same `CAS_LOOP_MAX_TRIES` ceiling; `CasLoop` records each loop's operation (`cas_operations`); `SegmentedCasLoop` and `WriteStorageByCasLoop` use them, and `DefineSegmentedHelpers` makes the loop's two variables first in the helper. - renderer, `gpu/vk/vk_shader.{h,cpp}`: `SegSig` (sorted entries and a hash) on `Program`, `seg_sig()`, `seg_variant()`, `program_segmented()` (finds or translates the variant of a plain program with plain's questions and answers, and fails it loudly if its descriptors are not plain's); `program_pick`, `mesh_program_locked`, `program_find_plain`, `loaded_variants` and `portable_variant` keep the variants apart; every stage translation sets `RuntimeInfo::segmented` from `Program::seg` (empty in every plain variant). - renderer, `gpu/vk/vk_progcache.cpp`: a segmented variant is never written to the program cache nor listed among the loaded programs. - tests: `tests/test_seg_shader.py` and `tests/seg_shader_ir.cpp` (new) run synthetic IR programs through the real back end, plain and in pieces, check the segmented modules' structure and capabilities, run every module through `spirv-val` (Vulkan 1.3, uniform buffer standard layout, as CI validates the back end), check that no module passes a pointer into a buffer descriptor array to a function, and with `--base-tree` compare the plain modules byte for byte with an older tree's. Seven source-reading checks follow the new field, the atomics' new branch and the new loop: `test_native_float_atomic_add`, `test_pipeline_state_portable`, `test_program_cache_cleanup`, `test_program_cache_one_variant`, `test_seed_kept_loaded`, `test_seed_variant_cleanup`, `test_cas_loop_ceiling`. ## Measurements **Verdict:** passes. The plain SPIR-V is byte for byte main's on both CPU routes, every module validates, and the array pointer check fails without the fix (3 segmented modules, 54 calls) and passes with it. No speed to measure: nothing is drawn with pieces until #663, and the plain path's SPIR-V is unchanged. Hardware: macOS 26.7.1, Apple M4 Max, 64 GB, Apple clang 21 (game-free: no GPU, no game) Scene: none (an emitter change: its proof is the fixture's modules) Method: `tests/test_seg_shader.py` at this head with `--base-tree` a `git archive` of main 6c9edffe's back end, on arm64 and with `--arch x86_64` (compiled for Intel, run under Rosetta); the same test file against the first commit's back end (`cbcdb34d`, the pieces without the fix). Every output is a byte comparison or a hash, so one run per route and side; the two routes' 29 modules have equal SHA-256. How to reproduce: `python -I -B tests/test_seg_shader.py --headers <SPIRV-Headers/include> --spirv-val spirv-val --base-tree <git archive of main> [--arch x86_64]` Baseline: the middle column is this PR's first commit alone, the pieces before the fix; main has no segmented module to check. | | Before: main @ 6c9edffe | After, first commit only (no fix): this PR @ cbcdb34d | After: this PR @ 9d48901a | |---|---|---|---| | plain modules, arm64 and x86_64 | 14 emitted (the base tree) | — | 14 of 14 byte-identical to main's on both routes | | modules passing `spirv-val` | — | 29 of 29 | 29 of 29, both routes | | modules passing a pointer into a buffer array to a function | — | 3 of 29 (`seg-storage` 18 calls, `seg-storage-native-fadd` 15, `seg-storage-narrow` 21): exit 1 | 0 of 29: exit 0 | | arm E (SegProbe's module) SHA-256 | — | `01a0f43f…` | `01a0f43f…`, as SegProbe pinned | ``` this PR @ 9d48901a, arm64 (x86_64 the same lines): array pointers: 29 of 29 modules pass no pointer into a buffer descriptor array to a function base tree …/base-6c9edffe: 14 of 14 plain modules byte-identical PASS: segmented descriptors through the real back end; every module validates this PR's test against cbcdb34d's back end, arm64: seg-storage-narrow.spv: 21 calls pass a pointer into a buffer array: %105(ssbo0), %132(ssbo0), %162(ssbo0), %191(ssbo0), %219(ssbo0), %240(ssbo0), %74(ssbo0) seg-storage-native-fadd.spv: 15 calls pass a pointer into a buffer array: %118(ssbo0), %149(ssbo0), %176(ssbo0), %205(ssbo0), %233(ssbo0) seg-storage.spv: 18 calls pass a pointer into a buffer array: %110(ssbo0), %141(ssbo0), %168(ssbo0), %198(ssbo0), %227(ssbo0), %255(ssbo0) array pointers: 26 of 29 modules pass no pointer into a buffer descriptor array to a function ``` History, GPU (MoltenVK 1.4.2, M4 Max, arm64 and x86_64 under Rosetta), and why it holds for this head: - SegProbe revision 02 ran this emitter's arm E module (a 4-piece storage buffer, read and written through the helpers): `E.read_g1_layout` and `E.write_g2_layout` PASS on both routes, 0 `[mvk-error]` (`reports/2026-11-14-seg-probe.md` section 3d). This head emits the same module (SHA-256 `01a0f43f7da317c61ab1ee0dab18448d8dcafb798a1fcabdf5d08a0db8b69cc0`). - MslProbe revision 02 ran the fix's modules (`reports/2026-11-14-msl-arrayptr.md` sections 11 and 12): every pipeline was made, and the written-out loops (increment, decrement, float add, f32x2 add/min/max, the narrow profile's 8- and 16-bit stores) gave the CPU interpreter's bytes in every piece; `ma-cas-seg`, `ma-cas-narrow-seg`, their whole controls and the table program `mc-table` MATCH on both routes. Those five modules were built from the NullShader branch's back end; this head's back end emits the same five programs byte for byte (`reports/PubWave-files/PubShader/compare-mslprobe.txt`), so the result is this head's. - Without the fix, that branch's segmented program NS2 failed pipeline creation on MoltenVK with `no matching function for call`, `&ssbo0_4[0u]` given to a `device _66 *` parameter (NullProbe revision 02; the cause in SPIRV-Cross's source is in `reports/2026-11-14-msl-arrayptr.md` section 1). ### Flags and settings None. ## Testing **Tested:** on macOS 26.7.1, Apple M4 Max, at `9d48901a`, this PR's head before the rebases onto e090cc48 and aa95d35f (main 6c9edffe plus this PR), game-free: - `tests/test_seg_shader.py --base-tree <main 6c9edffe>` on arm64 and x86_64 (Rosetta): exit 0 both, as in Measurements; the 29 modules of the two routes are byte-identical, and byte-identical to the macOS port's integration tree after the fix merged (`7cd720da`, both routes: every plain and segmented module and arm E). Without the fix (the first commit's back end) the three segmented storage modules are the topic's original bytes (`b7928876` on `a4802cb5`) and fail the array pointer check. - The five MslProbe programs (`build_cas_run` whole and in 3 pieces, both profiles, and `build_table_run`, copied from the NullShader branch's fixture into a scratch copy of this head's, bit off) through this head's back end: byte-identical to MslProbe revision 02's frozen GPU inputs; all validate and pass the array pointer check. - Renderer files that include `vk_shader.h` (`vk_shader.cpp`, `vk_progcache.cpp`, `vk_draw.cpp`, `vk_predict.cpp`, `vk_replay.cpp`, `vk_programs.cpp`) compiled for arm64 and x86_64 at this head and on main 6c9edffe with main's renderer flags (`nx2/build.py`'s `vflags` plus `-arch`, and main's non-Windows `-fPIC -I host/posix -D_GNU_SOURCE`). Every tree got #492's two Darwin fixes (`environ`, `RUSAGE_THREAD`) and a compile-only prelude declaring `posix_fallocate` and `posix_fadvise`, which main's renderer calls and macOS lacks (#476 gives them Darwin paths): 12 of 12 objects on each, exit 0. - The seven source-reading checks this PR updates (`tools/ls-suite.py` named run): 5 pass. `test_program_cache_cleanup` and `test_program_cache_one_variant` cannot compile their fixture on this Mac, exactly as on main 6c9edffe (main's `vk_progcache.cpp` uses `POSIX_FADV_WILLNEED`; main's own macOS suite fails both the same way). With #476's two Darwin hunks (`vk_progcache.cpp`, `pipeline_cache_file.h`) applied to scratch copies of main and this head, both pass on both (main 9 tests OK and the same variant for 9 picks; this head the same). - Logs: `reports/PubWave-files/PubShader/proof/p20-*.log`, `proof/fadv-*.log`, `compare-p20.txt`, `compare-mslprobe.txt`, `compile-*.out`, `named/`, `merge-tree.txt`, `stamps.txt`. - **After the rebase onto main aa95d35f**, at `9b73e829` (`9d48901a` before the rebases onto e090cc48 and aa95d35f): Sweep3's checks of this head (the C/C++ it changes compiled, and its own tests run, on arm64 and x86_64 where they apply) gave the same results as before; `python -I -B tests/run_all.py` on wave B's merge `db0e2e16`, which carries this head: 569/32/18 (passed/failed/skipped) against main aa95d35f's 543/33/17 on this Mac, no new failure (`reports/2026-11-15-rebase-sweep-3.md`). **Not tested:** the game at this head: per-PR game runs are not planned, and nothing here runs before #663 asks for a segmented variant. The GPU at this head: the probes above are history whose modules this head reproduces byte for byte. The f16x2 compare-exchange loops have no fixture program (the CPU interpreter has no 16-bit float vectors); they share the f32x2 loops' code path. Windows and Linux: not built or run there (11 changed renderer and back-end files differ from main's off Apple, `unifdef -U__APPLE__`); its test request covers the build, run_all on Linux and a start. **Must be tested:** - Linux and Windows, any GPU (Linux runs the SPIR-V test): build this PR (head 9b73e829) and main aa95d35f the same way (the 11 every-platform files: the back end's `emit_spirv.cpp`, `emit_spirv_atomic.cpp`, `emit_spirv_image.cpp`, `emit_spirv_image_atomic.cpp`, `emit_spirv_memory.cpp`, `spirv_emit_context.{h,cpp}`, `runtime_info.h`; `vk_shader.{h,cpp}`, `vk_progcache.cpp`); on Linux run `python -I -B tests/run_all.py`, expecting `tests/test_seg_shader.py` to pass (it needs `spirv-val` and clang++; Windows skips it); start each build to the title, expecting this PR to translate the shader caches again once (the stamp changed) and nothing else to differ from main, in the log or on screen (by hand: test request T-WLDY) ## Risks **What could break:** nothing runs differently until #663 calls `program_segmented`: every program's `RuntimeInfo::segmented` is empty, and with it empty the back end's output is byte for byte main's (checked on 14 programs, both routes). `Program` grows by a `SegSig` (an empty vector and a hash). Once #663 uses it on macOS: a segmented variant is translated on the drawing thread under `shader_mu` at its first draw, and one whose descriptors differ from plain's is failed and logged, never drawn with pieces. **Turning it off:** revert the commits (there is no switch: nothing asks for pieces before #663). **Compatibility:** every platform's shader caches are translated again once: `eden/` and `vk_shader.cpp` are in the translator's stamp (`nx2/translator_files.py`: `ebbf160ce9626041` on main, `40b414b30dad80e1` here), so each player's program cache, prepare record and pipeline cache are dropped at the first start after updating. `kProgramCacheFormat` is unchanged here (#649 raises it). Nothing changes in saves, `longshot_options.ini` or the vk ABI (`MKWVK_API_VERSION`). ## Reviewer notes - Review `cbcdb34d..9d48901a`: this PR's two commits, `cbcdb34d` (the pieces, `-x` from `macos/seg-shader` `b7928876`) and `9d48901a` (the MslArrayPtr fix, `-x` from `macos/msl-arrayptr` `bd396010`). Read `runtime_info.h`, then `DefineSegmentedHelpers` and `SegmentedStorageCall` in `spirv_emit_context.cpp`, then `program_segmented` in `vk_shader.cpp`; then `CasWordHere`/`CasLoopHere`. - One PR for two topics (owner decision): the pieces and their compare-exchange fix ship together, so no state of `main` has segmented atomics that MoltenVK refuses. - How the fix was ported: `bd396010` was committed on the NullShader branch, where a storage descriptor bound whole can also reach the segmented helpers (`StorageThroughHelper`). Here every access that reaches a helper is in pieces, so the loops are always written out; the bound-whole branch, the guarded helper's `cas_loop`, and the tests that need NullShader or `tests/spirv_run.h` (the loops' and a table's CPU runs, `--changed-segmented`, the `*.cpurun` comparison) stay for #674. The cherry-pick's conflicts (`emit_spirv_memory.cpp`, `spirv_emit_context.{h,cpp}`, `seg_shader_ir.cpp`, `test_seg_shader.py`, `spirv_run.h` absent here) were all that difference. - Stacking: on main 6c9edffe, no dependency. #649 stacks on it; the draw side (#662, #663, #663) and #674 use it. - Conflicts with main: none in the cherry-picks. No semantic rework: main changed `vk_shader.cpp` (#252 `make_module`'s Intel gate and `driver_faulted_program` in `translate_into`, which `program_segmented` reaches through `translate_into`, so a program the driver faulted on is not asked again as a segmented variant either), `vk_progcache.cpp` (#321's paths, #252) and `test_native_float_atomic_add.py` (#191, #444) beside this PR's lines. #370 moved the stamp's file list to `nx2/translator_files.py` and took `vk_api.h` out of it: the first commit's message still says to land it with a `vk_api.h` change; the stamp changes here either way. - With the open macOS PRs (#431, #437, #438, #479, #480, #490, #491, #492) and this wave's branches on main 6c9edffe: none (`git merge-tree`). - Code comments cite `reports/2026-11-14-*.md` (the port's design notes and the MslArrayPtr report), which are not in this repository. ## Issues See #420 ## Checklist ### Everyone - [x] One topic. - [x] The title is a Conventional Commit (`type(scope): description`) of the status line's type. - [x] Why gives the reason plainly and links the issues that fit (the tracker searched). - [x] Measurements has a verdict, or one line saying why there is nothing to measure. - [x] Testing names what was tested, what wasn't, and what must still be tested, where and how. (Test request T-WLDY.) - [x] Rebased on current `main`: on aa95d35f (#761, 2026-11-15), head `9b73e829`, without a conflict; this PR's own commits, old → new: `cbcdb34d` → `1f9183c2`, `9d48901a` → `9b73e829`; the commit names elsewhere in this description are from before the rebase. - [x] `python -I -B tests/run_all.py` passes (any failure that also fails on `main` is named). (On the wave's merge of every macOS PR built with this one, `453eb678`: 553 passed, 32 failed, 18 skipped against main 6c9edffe's 531/32/17 on this Mac; the failures are main's own except `test_split_ub_flush`, which #658's tree at the time broke and its head fixes; `test_collect_wide` now passes; the new tests pass, `test_seg_shader` among them, and `test_darwin_audio_service`, Darwin-only, is skipped by the runner; `reports/PubWave-files/waveA/`. After the rebase onto main aa95d35f (head `9b73e829`): wave B's merge `db0e2e16`, 569/32/18 against main aa95d35f's 543/33/17, no new failure.) - [x] No game files, dumps, build output or generated files. - [x] `third_party/` untouched (a new folder with a `PROVENANCE.md` instead). - [x] The license is OK: GPL-3.0-or-later, nothing incompatible and nothing decompiled. ### When it applies - [x] AI assistance recorded with `Assisted-by:` / `Co-Authored-By:` trailers. - [x] A game-free test added for the fix, failing on `main` and passing here. (It fails on this PR's first commit, the pieces without the fix; `main` has no segmented module.) - [x] A check under `tests/` that reads changed source updated in this PR, not weakened.

mooki added 2 commits 2026-11-15 02:17:23 +00:00

feat(vk): descriptors bound as pieces in the SPIR-V back end and program variants ... cbcdb34df1

A guest span larger than one device allocation (the cap on MoltenVK/M4 Max is 41,747,087,360 bytes) is bound as
pieces: one descriptor per allocation segment, at the descriptor's own binding (reports/2026-11-14-full-range-design.md
sections 4.1-4.4). This is the shader half; the draw side (descriptors, layout, gathers) comes later.

- runtime_info.h: SegEntry {Storage|Table|TexBuf|ImgBuf, desc, pieces} and RuntimeInfo::segmented.
- Back end: a segmented storage buffer is an array of pieces at its binding (a table count * pieces, entry k's piece j
  at k * pieces + j); its loads, stores and atomics call DontInline helpers made after main, which find the piece by
  the pieces' OpArrayLength in order, with constant indices. The global memory fallbacks sum the pieces' lengths for a
  segmented descriptor's bound end and branch to its piece (single, native candidate, table written out and in the
  loop). Texel and image buffers are arrays too, their fetches, reads, writes and atomics through helpers that pick
  the piece by OpImageQuerySize, and a texel buffer's size query sums the pieces. ImageQuery was declared for every
  module already. With RuntimeInfo::segmented empty nothing new is made: the SPIR-V is byte-identical.
- vk_shader: SegSig (entries, hash) on Program, seg_sig and seg_variant; program_segmented finds or translates the
  variant of a plain program with plain's texture questions and answers and checks its descriptors are plain's;
  program_pick, mesh_program_locked, program_find_plain, loaded_variants and portable_variant keep the plain and
  segmented variants apart; every stage's RuntimeInfo::segmented comes from Program::seg.
- vk_progcache: a segmented variant is never written to the program cache nor listed among the loaded programs.
- tests/test_seg_shader.py + tests/seg_shader_ir.cpp: synthetic IR programs through the real back end, plain and
  segmented; structure checks, spirv-val (Vulkan 1.3, uniform buffer standard layout), and with --base-tree a byte
  comparison of the plain modules against an older tree. Five source-reading checks follow the new Program field and
  the atomics' new branch.

The translator stamp changes with this (eden/, vk_shader.cpp, sizeof(Program)): every program cache is rebuilt once.
Land it in the same release as the vk_api.h change that already does so.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit b7928876dab846485f667c1c934d34d25ecef035)

fix(vk): MslArrayPtr: no pointer into a storage buffer array is passed to a function ... 9d48901abf

On MoltenVK a segmented storage program with a compare-exchange loop (an
increment, decrement or float add the device does not have) failed pipeline
creation: NullProbe revision 02's NS2, "no matching function for call",
&ssbo0_4[0u] given to a device _66 * parameter. SPIRV-Cross's MSL (6c09849f,
MoltenVK 1.4.2) passes a storage buffer pointer to a function by value, and
takes the address of the argument's expression: for an element of a buffer
array, which MSL holds as an array of pointers, that is the address of the
pointer (reports/2026-11-14-msl-arrayptr.md).

A descriptor bound as pieces now has its loops written out in the piece,
inside the segmented helper (CasLoopHere, CasStoreHere: CasLoop's loop,
attempt for attempt, its variables made first in the helper), instead of
calling the CasLoop function or write_storage_cas_loop_func with a pointer
into the pieces' array. Every other program is as before byte for byte.

Tests: test_seg_shader.py checks that no module passes a pointer into a
buffer descriptor array to a function. The two checks that cut the back
end's source follow the new loop and parameter.

Ported from macos/msl-arrayptr, where it was committed on NullShader, to the
pieces alone: here every storage access that reaches a segmented helper is
in pieces, so the loops are always written out. Left for the NullShader
change: the bound-whole branch that still calls the function with its
block, the guarded helper's cas_loop, and the tests that need NullShader or
tests/spirv_run.h (the CPU runs of the loops and of a storage buffer table,
--changed-segmented and the *.cpurun comparison).

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit bd396010887afab9520181009db1b56a4cc425e3)

mooki referenced this pull request 2026-11-15 02:18:02 +00:00

fix(vk): give programs that read storage buffer lengths descriptor sets on MoltenVK #644

mooki referenced this pull request 2026-11-15 02:18:07 +00:00

feat(vk): read texel formats MoltenVK cannot view through R8/R16/R32 carrier codecs #649

mooki referenced this pull request 2026-11-15 02:18:11 +00:00

WIP: fix(macos): let MoltenVK's precompile go on behind the title and name slow compiles #652

mooki referenced this pull request 2026-11-15 02:50:25 +00:00

feat(vk): add the inert contract for segmented binding consumers #662

mooki referenced this pull request 2026-11-15 02:50:27 +00:00

feat(vk): bind segment-straddling buffers and texel views as piece arrays, and split transfers at segment cuts, on macOS #663

mooki referenced this pull request 2026-11-15 03:48:58 +00:00

feat(vk): emulate null descriptors in the SPIR-V back end on MoltenVK #674

mooki referenced this pull request 2026-11-15 04:00:40 +00:00

WIP: feat(macos): count NVN's polygon modes and restart-off strip draws on Apple, log-only #677

mooki referenced this pull request 2026-11-15 04:00:45 +00:00

WIP: fix(macos): bind a direct draw's 8-bit indices with their size on MoltenVK #678

mooki referenced this pull request 2026-11-15 04:00:51 +00:00

WIP: fix(macos): count draws rewritten where MoltenVK has no drawIndirectCount, and buffer updates copied from the stream #679

mooki referenced this pull request 2026-11-15 04:00:56 +00:00

WIP: fix(macos): direct draws that read the draw index drawn as one indirect record on MoltenVK #680

mooki referenced this pull request 2026-11-15 04:01:45 +00:00

WIP: feat(macos): RX_VK_LMEM_LOG names each module's private memory and each pipeline's first bind #681

mooki referenced this pull request 2026-11-15 05:23:50 +00:00

WIP: fix(macos): fans keyed apart from the other triangle topologies on MoltenVK #701

mooki referenced this pull request 2026-11-15 05:24:19 +00:00

WIP: feat(vk): log-only frame cost lines, a GPU timeline and the bench's A/B exactness arm, all off by default #704

mooki changed title from WIP: feat(vk): bind storage and texel descriptors as piece arrays in the SPIR-V back end to feat(vk): bind storage and texel descriptors as piece arrays in the SPIR-V back end 2026-11-15 05:25:32 +00:00

mooki force-pushed mac/seg-shader from 9d48901abf to 407a0cfc93 2026-11-15 07:16:33 +00:00 Compare

mooki force-pushed mac/seg-shader from 407a0cfc93 to 9b73e82942 2026-11-15 13:10:01 +00:00 Compare

SocialWare referenced this pull request 2026-11-15 13:27:34 +00:00

Game: Crash in Hyrule field after opening and closing map #809

mooki added the

area/renderer

kind

feature

platform/linux

platform/macos

platform/windows

priority

low

labels 2026-11-15 13:57:02 +00:00

mooki referenced this pull request 2026-11-15 17:39:50 +00:00

WIP: feat(vk): the SPIR-V back end on MoltenVK and guest spans past one device allocation #848

mooki commented 2026-11-15 18:07:51 +00:00

Author

Member

• Copy link
• Copy Markdown

Folded into #848 (feat(vk): the SPIR-V back end on MoltenVK and guest spans past one device allocation)

Folded into #848 (feat(vk): the SPIR-V back end on MoltenVK and guest spans past one device allocation)

mooki closed this pull request 2026-11-15 18:07:52 +00:00

Some checks failed

Hide checks

CI / Python source checks (ubuntu-24.04, part 1/2) (pull_request) Has been skipped

Details

CI / Python source checks (ubuntu-24.04, part 2/2) (pull_request) Has been skipped

Details

CI / SPIR-V fixtures (global narrow load/store, direct surface atomics, F2I immediate) (pull_request) Has been skipped

Details

Third-party parts / The same as upstream (pull_request) Has been skipped

Details

Website / Website on Cloudflare (pull_request) Has been skipped

Details

CI / Conventional Commit title (pull_request) Successful in 4s

Details

CI / Workflow lint (actionlint) (pull_request) Successful in 6s

Details

Guide screenshots / Guide images and local UI source review (pull_request) Failing after 6s

Details

Third-party parts / As their notes say (pull_request) Successful in 11s

Details

PR description / Description follows the template (pull_request_target) Successful in 4s

Details

Website / Website on Cloudflare (pull_request_target) Has been skipped

Details

Pull request closed

Please reopen this pull request to perform a merge.

Sign in to join this conversation.

Reviewers

No reviewers

Labels

Clear labels   

area/accessibility

Barriers for people with disabilities

  

area/audio

Sound and music

  

area/input

Controllers, keyboard, mouse

  

area/install

Installing, updating, building from your game

  

area/menu

In-game menu and settings

  

area/renderer

Vulkan renderer, shaders, effects

  

area/saves

Save files

  

area/upscaling

DLSS, FSR, frame generation

  

beanstalk: approved @ 1697603

Set by Beanstalk for show; the farm's database is what counts

  

beanstalk: approved @ 2c5a62e

Set by Beanstalk for show; the farm's database is what counts

  

beanstalk: approved @ 4fd82b7

Set by Beanstalk for show; the farm's database is what counts

  

beanstalk: approved @ 64d6d09

Set by Beanstalk for show; the farm's database is what counts

  

beanstalk: approved @ 6f91fae

Set by Beanstalk for show; the farm's database is what counts

  

beanstalk: approved @ 70a78f9

Set by Beanstalk for show; the farm's database is what counts

  

beanstalk: approved @ 8d361dc

Set by Beanstalk for show; the farm's database is what counts

  

beanstalk: approved @ 9d8356e

Set by Beanstalk for show; the farm's database is what counts

  

beanstalk: approved @ abf2227

Set by Beanstalk for show; the farm's database is what counts

  

beanstalk: approved @ d82f202

Set by Beanstalk for show; the farm's database is what counts

  

beanstalk: approved @ e5f1e40

Set by Beanstalk for show; the farm's database is what counts

  

beanstalk: approved @ ee8ab5f

Set by Beanstalk for show; the farm's database is what counts

  

ci: approved

A maintainer approved CircleCI for this fork PR (sticky; removed after 14 quiet days)

  

good first issue

A good start for a new contributor

  

gpu/amd

AMD GPUs

  

gpu/igpu

Integrated GPUs

  

gpu/intel

Intel GPUs

  

gpu/nvidia

NVIDIA GPUs

  

help wanted

Extra hands or knowledge wanted

  

kind

bug

Something doesn't work as it should

  

kind

chore

Build, CI, dependencies, tidying

  

kind

docs

Documentation

  

kind

feature

A new feature or an improvement

  

kind

performance

Frame rate, frame time, loading, memory

  

kind

question

A question rather than a change

  

needs measurements

Waiting for before/after measurements

  

platform/linux

Affects Linux (a change for every system carries both Windows and Linux)

  

platform/macos

macOS (Apple silicon or Intel Macs)

  

platform/steam-deck

Steam Deck

  

platform/windows

Affects Windows (a change for every system carries both Windows and Linux)

  

preview

Maintainers: deploy a preview of this PR's website (read its changes first). Removed after the run.

  

priority

critical

Crashes, data loss, or blocks most players

  

priority

high

Important: next in line

  

priority

low

Nice to have

  

regression

Worked in an earlier build, broken in a newer one: highest priority, above other bugs

  

status

blocked

Waits on something else first

  

status

confirmed

Reproduced or agreed on

  

status

duplicate

Already reported or proposed

  

status

invalid

Not a problem in this project

  

status

needs-info

Waiting for more information from the reporter

  

status

wontfix

Will not be worked on

  

testing wanted

Wanted: a test on other hardware (posts a test request in #test-requests)

  

ci: approved

No labels

area/accessibility

area/audio

area/input

area/install

area/menu

area/renderer

area/saves

area/upscaling

beanstalk: approved @ 1697603

beanstalk: approved @ 2c5a62e

beanstalk: approved @ 4fd82b7

beanstalk: approved @ 64d6d09

beanstalk: approved @ 6f91fae

beanstalk: approved @ 70a78f9

beanstalk: approved @ 8d361dc

beanstalk: approved @ 9d8356e

beanstalk: approved @ abf2227

beanstalk: approved @ d82f202

beanstalk: approved @ e5f1e40

beanstalk: approved @ ee8ab5f

ci: approved

good first issue

gpu/amd

gpu/igpu

gpu/intel

gpu/nvidia

help wanted

kind

bug

kind

chore

kind

docs

kind

feature

kind

performance

kind

question

needs measurements

platform/linux

platform/macos

platform/steam-deck

platform/windows

preview

priority

critical

priority

high

priority

low

regression

status

blocked

status

confirmed

status

duplicate

status

invalid

status

needs-info

status

wontfix

testing wanted

ci: approved

Milestone

Clear milestone

No items

No milestone

Projects

Clear projects

No items

No project

Assignees

Clear assignees

No assignees

1 participant

Notifications

Subscribe

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference

MagicBeanDev/Longshot!648

No description provided.