WIP: fix(macos): leave MoltenVK's compute pipelines to the precompile and bound its giant compiles #664

Closed

mooki wants to merge 19 commits from mooki/Longshot:mac/warm-handoff into main

pull from: mooki/Longshot:mac/warm-handoff

merge into: MagicBeanDev:main

MagicBeanDev:main

Conversation 1 Commits 19 Files changed 51 +4021 -125

mooki commented 2026-11-15 02:55:02 +00:00

Member

• Copy link
• Copy Markdown

Type: fix · Risk: medium · Tested on: macOS 26.7.1, Apple M4 Max (MoltenVK 1.4.2), arm64 and x86_64 under Rosetta · Needs: macOS, Apple silicon (a start on the rebased stack); Windows and Linux, any GPU (build and start)

Why

On a Mac starting without a program cache, the title did not come: the shader preparation compiled every compute program's
Metal pipeline itself, before the precompile that can hand slow compiles to the game, and Metal's compiler service took
minutes over the largest of them. In start 14, 13 of the preparation's 16 workers were each held for 205-274 s by one compute
compile of 122,000-209,000 SPIR-V words; the preparation reached 278 of 38,809 programs in 430 s and the precompile never
began. Once those compiles are the precompile's, the same giants could still take every one of its workers at once, so nothing
small finished and nothing told the start to go on. This PR leaves the compute pipelines to the precompile on MoltenVK and lets
only a few giant compiles run at a time, always leaving a worker for the small ones. It is part of the macOS port. See #420.

How to reproduce

Seen on: macOS 26.7.1, Apple M4 Max, MoltenVK 1.4.2 (main does not run on macOS yet: seen on the macOS port's integration
builds, which carry the open macOS PRs)
Settings and switches: defaults (start mode), a start whose program cache is translated anew (a first start, or a changed
translator)
Steps: start the game and wait for the title.
Expected: the preparation translates the programs in a minute or two, then the precompile builds what it can and hands the
slowest compiles to the game.
Actual: the preparation sits for minutes inside the compute warm's Metal compiles (start 14: 278 of 38,809 programs in 430 s;
start 16: 1,550 s, 1,381 compute pipelines compiled there), and the precompile with its hand-off comes only after it
(reports/2026-11-14-mac-startup-14.md, -16.md; analysis in reports/2026-11-14-warm-handoff.md).

What changed

Under compiler_service (MoltenVK only, #652) the preparation still translates and keeps every compute program but compiles
none of their pipelines: the precompile takes every kept one on a first start (its order puts them last, smallest first) or the
first dispatch compiles it, as any program this start did not prepare. In the precompile, at most RX_VK_GIANT_COMPILES (4)
compiles of more than RX_VK_GIANT_WORDS (100,000) SPIR-V words are in flight at once, and never as many as the workers of a
phase (4 of 16 before the game, 3 of the 4 behind it); one more is deferred, named once, and built as soon as one ends, never
left out. Every other driver's preparation and precompile are unchanged.

• gpu/vk/vk_draw.cpp prepare_run: if (warm_cs && compiler_service_on()) clears the compute warm before its .warming
  marker and says where the compiles went; the stage-library warm (RX_VK_GPL), the state list and the prepared record are as
  before.
• gpu/vk/precompile_handoff.h (no Vulkan): claim_below (a claim that never passes the end, so a worker waiting for a
  deferred giant does not count as one that ran out), giant_compiles/giant_words (the two switches), Giants (next
  returns Run, Wait or None; end; limit_for(cap, workers) = at most one fewer than the workers, at least 1; phase for the
  share behind the game).
• gpu/vk/vk_draw.cpp precompile_run, under pre->hand only: each item's SPIR-V words, pre->giants, behind_worker
  through Giants::next with a deferral line once per giant and the count in behind_done's line; phase(share) at the
  hand-off. The classic worker (every other driver) is untouched.
• docs/development.md: the RX_VK_COMPILER_SERVICE row (the preparation's compute pipelines) and a row for
  RX_VK_GIANT_COMPILES/RX_VK_GIANT_WORDS.
• tests: tests/test_warm_handoff.py (new: prepare_run's warm decision and warm_compute cut from the source and run with
  stand-ins, six cases; source checks for the chain that keeps each variant for the precompile);
  tests/precompile_handoff_test.cpp and tests/test_precompile_handoff.py (the gate before the game and behind it, one worker
  always on small items, the wiring).

Measurements

Verdict: helps: with this change a start that translates the programs anew spent 118.3 s in the preparation with 0 compute
pipelines compiled there (run #18), where it spent 1,550.4 s compiling 1,381 of them before (run #16); the precompile then
handed off after 670 s (#18) and 626 s (#19) with at most 4 giants in flight before the game (1,411 and 1,415 deferral lines),
and the title came 4.5-5 s later.

Baseline: main does not run on macOS yet, so both columns are the macOS port's integration builds, arm64, MoltenVK 1.4.2,
start mode, each translating the program cache anew. Before: 4e472987 (run #16), without this change. After: 1cbb594a (run
#18, Integration 25, which carries this topic). One run each; the runs are the port's start runs, not an A/B on one build.

Hardware: macOS 26.7.1, Apple M4 Max (40-core GPU), MoltenVK 1.4.2, 16-core CPU, 64 GB
Scene: a start to the title, windowless, defaults
Method: the game log's vk: prepared …, vk: precompile: … and deferral lines, one run each
How to reproduce: a start of a macOS build from a seeded Metal cache with a program cache that must be translated again

	Before: integration @ 4e472987 (run #16)	After: integration @ 1cbb594a (run #18)
preparation	1,550.4 s, 1,381 compute pipelines compiled in it	118.3 s, 0 compute pipelines compiled in it
precompile	reached 5,806 of 42,234 by the end of its 1,800 s hold	670.3 s, then the hand-off; 1,411 giants deferred while 4 ran
title	not in the hold	4.6 s after the hand-off; menu-ready

#16: vk: prepared 38809 shader programs (43678 new, 70 failed, 1381 compute pipelines; 1381 compute variants, 140 programs with texture variants, 0 at the limit) in 1550.4 s
#18: vk: preparation: the compute programs are translated and kept; their pipelines are compiled by the precompile, which hands what is still compiling to the game (RX_VK_COMPILER_SERVICE: Metal's compiler service)
#18: vk: prepared 38809 shader programs (46597 new, 70 failed, 0 compute pipelines; 9707 compute variants, 775 programs with texture variants, 59 at the limit) in 118.3 s
#18: vk: precompile: in this order for Metal's compiler service (RX_VK_COMPILER_SERVICE): 30329 noted pipelines, then 2198 mesh fallback programs and 9707 compute programs by their SPIR-V size, and last 0 an earlier run found slow (…); at most 4 compiles of more than 100000 SPIR-V words in flight at once (RX_VK_GIANT_COMPILES=4, never all 16 workers), 1442 such items here (RX_VK_GIANT_WORDS)
#18: vk: precompile: 42134 of 42234 done (42234 handed out), every item still running has run for longer than the slow time: the game starts, the rest go on behind it on 4 workers, every one of them still built (RX_VK_PRECOMPILE_STALL_S, RX_VK_SLOW_COMPILE_S); at most 3 giant compiles at once there

Run reports: reports/2026-11-14-mac-startup-14.md, -16.md, -18.md, -19.md.

Flags and settings

Two new switches for the giant bound, on with the compiler service; the RX_VK_COMPILER_SERVICE row gains the preparation's
part. Nothing changes on other drivers.

Flag	What it is	Before this PR	After this PR	Effect	Helps / doesn't help / untested	Recommendation
RX_VK_GIANT_COMPILES	how many of the precompile's giant compiles may be in flight at once (never all of a phase's workers); 0: no bound	(no switch)	4, with RX_VK_COMPILER_SERVICE	the small items finish while the giants compile; the hand-off can come	helps: MoltenVK (runs #18, #19); untested: other values	as it ships
RX_VK_GIANT_WORDS	the SPIR-V size above which a compile counts as a giant	(no switch)	100000	which compiles the bound holds	helps: MoltenVK; untested: other values	as it ships
RX_VK_COMPILER_SERVICE	(from #652) now also: the preparation compiles no compute pipeline	on for MoltenVK only	on for MoltenVK only	the preparation ends in about 2 minutes on a new program cache	helps: MoltenVK	as it ships; =0 only to compare

Testing

Tested: on macOS 26.7.1, Apple M4 Max, at this head (main 6c9edffe plus #650, #651, #652 and this PR):

• tests/test_warm_handoff.py and tests/test_precompile_handoff.py, each --arch arm64 and --arch x86_64 (the fixtures run
  under Rosetta): pass on both arches (reports/PubWave-files/PubLate1/wh-proof-02/). On this branch's base (#652 7ef7cff9)
  with the test files copied in (sha256 in wh-test-files-02.sha256): both fail on both arches (the warm compiled 1 pipeline
  under compiler_service, wrote the .warming marker and said nothing; the gate fixture does not build: no claim_below,
  Giants, limit_for) (wh-proof-02/wh-base-warm-*, wh-proof-01/wh-base-handoff-*). Before the adaptation commit
  (e10c0ad5), test_warm_handoff fails on both arches: its fixture lacks main's warm stand-ins (wh-proof-01/wh-warm-*).
• Every check that reads a renderer file changed by #650, #651, #652 or this PR (298 checks), through tools/ls-suite.py: 279
  pass, 19 fail, the same 19 by name as at #652's head (16 also fail on main 6c9edffe on macOS; test_precompile_compute_guard,
  test_shader_module_guard and test_present_aspect_vulkan are skipped on macOS in a full run and fail only because named)
  (wh-related-02/, wh-related-02-compare.txt).
• gpu/vk/vk_draw.cpp compiled whole for arm64 and x86_64 with main's renderer flags (with #476's and #492's macOS compile pieces
  in a scratch tree): both objects made, exit 0 (reports/PubWave-files/PubLate1/compile/warm-handoff/).
• Off Apple (unifdef -U__APPLE__ of each changed file, base against head): vk_draw.cpp is the same; precompile_handoff.h
  differs (its new pieces are compiled on every platform, used only under compiler_service)
  (reports/PubWave-files/PubLate1/offapple-warm-handoff.txt).
• In the game (integration builds carrying these commits, before the rebase): runs #18 and #19, above.

Not tested: the game at this head (main does not run on macOS without the other macOS PRs); Windows and Linux (no change in
behaviour there: the new code runs only under compiler_service, off on their drivers; the header's new pieces are compiled
there): not built on Windows or Linux; its test request covers it.

Must be tested:

• macOS, Apple silicon, on the rebased macOS stack: a start that translates the program cache anew, expecting the vk: preparation: the compute programs are translated and kept … line, 0 compute pipelines in vk: prepared …, deferral lines
  in the precompile and the hand-off (by hand: the parent's GPU slot).
• Windows and Linux, any GPU: build this PR (head 20367871, it carries #650, #651 and #652) and main 6c9edffe the same way and
  start each to the title, expecting no in this order for Metal's compiler service, deferred or the compute programs are translated and kept line (precompile_handoff.h's new code is compiled there and not reached), and every other line to match
  main's (by hand: test request T-XXXX).

Risks

What could break: on MoltenVK a compute program the precompile does not reach before the hand-off is compiled behind the
game or at its first dispatch, which leaves that dispatch out until it is ready (as with play's own builds); with the bound, the
precompile may wait on four giants while small items are left, so the hand-off comes on the stall rule. Other drivers: one if
that compiler_service gates in prepare_run, and the header's new code, unused there.

Turning it off: RX_VK_GIANT_COMPILES=0 (no bound); RX_VK_COMPILER_SERVICE=0 (the warm and the precompile as on every
other driver); or revert the PR.

Compatibility: Nothing changes for players on Windows and Linux: no shader stamp, cache format or MKWVK_API_VERSION
change. On macOS the preparation writes no .warming marker for compute pipelines any more.

Reviewer notes

• Review abfc05c8..20367871 (5 commits; in git's notation 7ef7cff9..20367871): the warm left to the precompile, the gate,
  the docs rows, the per-phase bound, and the test adaptation (20367871).
• Depends on #652 (compiler_service, precompile_handoff.h, behind_worker, create_watched), and through it on #650 and
  #651. The branch is #652's head 7ef7cff9 plus this PR's commits.
• The topic's four commits were cherry-picked with -x from macos/warm-handoff (84c32d68 f1e028c8 8aaa2f33 acd3a71a), made
  on the integration branch. Conflicts in the pick:
  • tests/test_precompile_handoff.py behind_worker's needles: #652's adaptation needles (#252's quick builds:
    precompile_compute_quick, optimize_owed) and this PR's giant needles, both kept.
  • docs/development.md (8aaa2f33, acd3a71a): the RX_VK_COMPILER_SERVICE and RX_VK_SLOW_COMPILE_S rows take main's
    cache/longshot_pipelines.bin.slow (#321) with this PR's text.
• The semantic adaptation is its own commit, 20367871: test_warm_handoff's fixture gets stand-ins for main's warm (#252's
  IntelDriverGate and guarded_compute_held, #190's and #650's create_cached, #652's create_watched/compiles_watched), and
  its dispatch check reads main's p (the integration branch's plain is #663's rename).
• Not carried: the integration branch's two fixture fixes for this test, 4253b972 (#include "crash_markers.h", needed once
  #428 puts crash_markers::begin in prepare_run) and 1cbb594a (a comment-tolerant seed_compute check, needed once the
  SPIR-V profile switch, a later macOS PR, adds to that line). Neither is needed on this base. Whichever of this PR and #428 lands
  second adds that include to test_warm_handoff.py's fixture.
• Sibling macOS PRs: #665 and #666 build on this branch. git merge-tree of this head: no conflict with
  #423, #427, #431, #438, #476, #479, #480, #490, #491, #492; gpu/vk/vk_main.cpp with #428, #498 and #653, and
  vk_main.cpp/vk_progcache.cpp/vk_shader.h/vk_vendor.h with #646: the same conflicts as #652's own head
  (reports/PubWave-files/PubLate1/merge-check-*.txt), none from this PR's lines.
• Conflicts with main: none.

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 requests not open yet.)
• Rebased on current main (main e090cc48).
• python -I -B tests/run_all.py passes (any failure that also fails on main is named). (On the wave's merge of the macOS PRs this one is built with, bfe5c752 (metal-cache2 + giant-mem-gate, which carry this PR): 544 passed, 32 failed, 17 skipped against main 6c9edffe's 531/32/17 on this Mac; the 32 failures are main's own, nothing that passes on main fails, 13 new tests pass; reports/PubWave-files/waveC1/.)
• 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.
• Switches documented (docs/development.md; the README for a menu row) and shipped at their recommended value.
• A game-free test added for the fix, failing on main and passing here.
• A check under tests/ that reads changed source updated in this PR, not weakened.
• Docs updated.

**Type:** fix · **Risk:** medium · **Tested on:** macOS 26.7.1, Apple M4 Max (MoltenVK 1.4.2), arm64 and x86_64 under Rosetta · **Needs:** macOS, Apple silicon (a start on the rebased stack); Windows and Linux, any GPU (build and start) ## Why On a Mac starting without a program cache, the title did not come: the shader preparation compiled every compute program's Metal pipeline itself, before the precompile that can hand slow compiles to the game, and Metal's compiler service took minutes over the largest of them. In start 14, 13 of the preparation's 16 workers were each held for 205-274 s by one compute compile of 122,000-209,000 SPIR-V words; the preparation reached 278 of 38,809 programs in 430 s and the precompile never began. Once those compiles are the precompile's, the same giants could still take every one of its workers at once, so nothing small finished and nothing told the start to go on. This PR leaves the compute pipelines to the precompile on MoltenVK and lets only a few giant compiles run at a time, always leaving a worker for the small ones. It is part of the macOS port. See #420. ## How to reproduce **Seen on:** macOS 26.7.1, Apple M4 Max, MoltenVK 1.4.2 (main does not run on macOS yet: seen on the macOS port's integration builds, which carry the open macOS PRs) **Settings and switches:** defaults (start mode), a start whose program cache is translated anew (a first start, or a changed translator) **Steps:** start the game and wait for the title. **Expected:** the preparation translates the programs in a minute or two, then the precompile builds what it can and hands the slowest compiles to the game. **Actual:** the preparation sits for minutes inside the compute warm's Metal compiles (start 14: 278 of 38,809 programs in 430 s; start 16: 1,550 s, 1,381 compute pipelines compiled there), and the precompile with its hand-off comes only after it (`reports/2026-11-14-mac-startup-14.md`, `-16.md`; analysis in `reports/2026-11-14-warm-handoff.md`). ## What changed Under `compiler_service` (MoltenVK only, #652) the preparation still translates and keeps every compute program but compiles none of their pipelines: the precompile takes every kept one on a first start (its order puts them last, smallest first) or the first dispatch compiles it, as any program this start did not prepare. In the precompile, at most `RX_VK_GIANT_COMPILES` (4) compiles of more than `RX_VK_GIANT_WORDS` (100,000) SPIR-V words are in flight at once, and never as many as the workers of a phase (4 of 16 before the game, 3 of the 4 behind it); one more is deferred, named once, and built as soon as one ends, never left out. Every other driver's preparation and precompile are unchanged. - `gpu/vk/vk_draw.cpp` `prepare_run`: `if (warm_cs && compiler_service_on())` clears the compute warm before its `.warming` marker and says where the compiles went; the stage-library warm (`RX_VK_GPL`), the state list and the prepared record are as before. - `gpu/vk/precompile_handoff.h` (no Vulkan): `claim_below` (a claim that never passes the end, so a worker waiting for a deferred giant does not count as one that ran out), `giant_compiles`/`giant_words` (the two switches), `Giants` (`next` returns Run, Wait or None; `end`; `limit_for(cap, workers)` = at most one fewer than the workers, at least 1; `phase` for the share behind the game). - `gpu/vk/vk_draw.cpp` `precompile_run`, under `pre->hand` only: each item's SPIR-V words, `pre->giants`, `behind_worker` through `Giants::next` with a deferral line once per giant and the count in `behind_done`'s line; `phase(share)` at the hand-off. The classic worker (every other driver) is untouched. - `docs/development.md`: the `RX_VK_COMPILER_SERVICE` row (the preparation's compute pipelines) and a row for `RX_VK_GIANT_COMPILES`/`RX_VK_GIANT_WORDS`. - tests: `tests/test_warm_handoff.py` (new: `prepare_run`'s warm decision and `warm_compute` cut from the source and run with stand-ins, six cases; source checks for the chain that keeps each variant for the precompile); `tests/precompile_handoff_test.cpp` and `tests/test_precompile_handoff.py` (the gate before the game and behind it, one worker always on small items, the wiring). ## Measurements **Verdict:** helps: with this change a start that translates the programs anew spent 118.3 s in the preparation with 0 compute pipelines compiled there (run #18), where it spent 1,550.4 s compiling 1,381 of them before (run #16); the precompile then handed off after 670 s (#18) and 626 s (#19) with at most 4 giants in flight before the game (1,411 and 1,415 deferral lines), and the title came 4.5-5 s later. Baseline: main does not run on macOS yet, so both columns are the macOS port's integration builds, arm64, MoltenVK 1.4.2, start mode, each translating the program cache anew. Before: 4e472987 (run #16), without this change. After: 1cbb594a (run #18, Integration 25, which carries this topic). One run each; the runs are the port's start runs, not an A/B on one build. Hardware: macOS 26.7.1, Apple M4 Max (40-core GPU), MoltenVK 1.4.2, 16-core CPU, 64 GB Scene: a start to the title, windowless, defaults Method: the game log's `vk: prepared …`, `vk: precompile: …` and deferral lines, one run each How to reproduce: a start of a macOS build from a seeded Metal cache with a program cache that must be translated again | | Before: integration @ 4e472987 (run #16) | After: integration @ 1cbb594a (run #18) | |---|---|---| | preparation | 1,550.4 s, 1,381 compute pipelines compiled in it | 118.3 s, 0 compute pipelines compiled in it | | precompile | reached 5,806 of 42,234 by the end of its 1,800 s hold | 670.3 s, then the hand-off; 1,411 giants deferred while 4 ran | | title | not in the hold | 4.6 s after the hand-off; menu-ready | ``` #16: vk: prepared 38809 shader programs (43678 new, 70 failed, 1381 compute pipelines; 1381 compute variants, 140 programs with texture variants, 0 at the limit) in 1550.4 s #18: vk: preparation: the compute programs are translated and kept; their pipelines are compiled by the precompile, which hands what is still compiling to the game (RX_VK_COMPILER_SERVICE: Metal's compiler service) #18: vk: prepared 38809 shader programs (46597 new, 70 failed, 0 compute pipelines; 9707 compute variants, 775 programs with texture variants, 59 at the limit) in 118.3 s #18: vk: precompile: in this order for Metal's compiler service (RX_VK_COMPILER_SERVICE): 30329 noted pipelines, then 2198 mesh fallback programs and 9707 compute programs by their SPIR-V size, and last 0 an earlier run found slow (…); at most 4 compiles of more than 100000 SPIR-V words in flight at once (RX_VK_GIANT_COMPILES=4, never all 16 workers), 1442 such items here (RX_VK_GIANT_WORDS) #18: vk: precompile: 42134 of 42234 done (42234 handed out), every item still running has run for longer than the slow time: the game starts, the rest go on behind it on 4 workers, every one of them still built (RX_VK_PRECOMPILE_STALL_S, RX_VK_SLOW_COMPILE_S); at most 3 giant compiles at once there ``` Run reports: `reports/2026-11-14-mac-startup-14.md`, `-16.md`, `-18.md`, `-19.md`. ### Flags and settings Two new switches for the giant bound, on with the compiler service; the `RX_VK_COMPILER_SERVICE` row gains the preparation's part. Nothing changes on other drivers. | Flag | What it is | Before this PR | After this PR | Effect | Helps / doesn't help / untested | Recommendation | |---|---|---|---|---|---|---| | `RX_VK_GIANT_COMPILES` | how many of the precompile's giant compiles may be in flight at once (never all of a phase's workers); `0`: no bound | (no switch) | 4, with `RX_VK_COMPILER_SERVICE` | the small items finish while the giants compile; the hand-off can come | helps: MoltenVK (runs #18, #19); untested: other values | as it ships | | `RX_VK_GIANT_WORDS` | the SPIR-V size above which a compile counts as a giant | (no switch) | 100000 | which compiles the bound holds | helps: MoltenVK; untested: other values | as it ships | | `RX_VK_COMPILER_SERVICE` | (from #652) now also: the preparation compiles no compute pipeline | on for MoltenVK only | on for MoltenVK only | the preparation ends in about 2 minutes on a new program cache | helps: MoltenVK | as it ships; `=0` only to compare | ## Testing **Tested:** on macOS 26.7.1, Apple M4 Max, at this head (main 6c9edffe plus #650, #651, #652 and this PR): - `tests/test_warm_handoff.py` and `tests/test_precompile_handoff.py`, each `--arch arm64` and `--arch x86_64` (the fixtures run under Rosetta): pass on both arches (`reports/PubWave-files/PubLate1/wh-proof-02/`). On this branch's base (#652 `7ef7cff9`) with the test files copied in (sha256 in `wh-test-files-02.sha256`): both fail on both arches (the warm compiled 1 pipeline under compiler_service, wrote the `.warming` marker and said nothing; the gate fixture does not build: no `claim_below`, `Giants`, `limit_for`) (`wh-proof-02/wh-base-warm-*`, `wh-proof-01/wh-base-handoff-*`). Before the adaptation commit (`e10c0ad5`), `test_warm_handoff` fails on both arches: its fixture lacks main's warm stand-ins (`wh-proof-01/wh-warm-*`). - Every check that reads a renderer file changed by #650, #651, #652 or this PR (298 checks), through `tools/ls-suite.py`: 279 pass, 19 fail, the same 19 by name as at #652's head (16 also fail on main 6c9edffe on macOS; `test_precompile_compute_guard`, `test_shader_module_guard` and `test_present_aspect_vulkan` are skipped on macOS in a full run and fail only because named) (`wh-related-02/`, `wh-related-02-compare.txt`). - `gpu/vk/vk_draw.cpp` compiled whole for arm64 and x86_64 with main's renderer flags (with #476's and #492's macOS compile pieces in a scratch tree): both objects made, exit 0 (`reports/PubWave-files/PubLate1/compile/warm-handoff/`). - Off Apple (`unifdef -U__APPLE__` of each changed file, base against head): `vk_draw.cpp` is the same; `precompile_handoff.h` differs (its new pieces are compiled on every platform, used only under `compiler_service`) (`reports/PubWave-files/PubLate1/offapple-warm-handoff.txt`). - In the game (integration builds carrying these commits, before the rebase): runs #18 and #19, above. **Not tested:** the game at this head (main does not run on macOS without the other macOS PRs); Windows and Linux (no change in behaviour there: the new code runs only under `compiler_service`, off on their drivers; the header's new pieces are compiled there): not built on Windows or Linux; its test request covers it. **Must be tested:** - macOS, Apple silicon, on the rebased macOS stack: a start that translates the program cache anew, expecting the `vk: preparation: the compute programs are translated and kept …` line, `0 compute pipelines` in `vk: prepared …`, deferral lines in the precompile and the hand-off (by hand: the parent's GPU slot). - Windows and Linux, any GPU: build this PR (head `20367871`, it carries #650, #651 and #652) and main 6c9edffe the same way and start each to the title, expecting no `in this order for Metal's compiler service`, `deferred` or `the compute programs are translated and kept` line (`precompile_handoff.h`'s new code is compiled there and not reached), and every other line to match main's (by hand: test request T-XXXX). ## Risks **What could break:** on MoltenVK a compute program the precompile does not reach before the hand-off is compiled behind the game or at its first dispatch, which leaves that dispatch out until it is ready (as with play's own builds); with the bound, the precompile may wait on four giants while small items are left, so the hand-off comes on the stall rule. Other drivers: one `if` that `compiler_service` gates in `prepare_run`, and the header's new code, unused there. **Turning it off:** `RX_VK_GIANT_COMPILES=0` (no bound); `RX_VK_COMPILER_SERVICE=0` (the warm and the precompile as on every other driver); or revert the PR. **Compatibility:** Nothing changes for players on Windows and Linux: no shader stamp, cache format or `MKWVK_API_VERSION` change. On macOS the preparation writes no `.warming` marker for compute pipelines any more. ## Reviewer notes - Review `abfc05c8..20367871` (5 commits; in git's notation `7ef7cff9..20367871`): the warm left to the precompile, the gate, the docs rows, the per-phase bound, and the test adaptation (`20367871`). - Depends on #652 (`compiler_service`, `precompile_handoff.h`, `behind_worker`, `create_watched`), and through it on #650 and #651. The branch is #652's head `7ef7cff9` plus this PR's commits. - The topic's four commits were cherry-picked with `-x` from `macos/warm-handoff` (`84c32d68 f1e028c8 8aaa2f33 acd3a71a`), made on the integration branch. Conflicts in the pick: - `tests/test_precompile_handoff.py` `behind_worker`'s needles: #652's adaptation needles (#252's quick builds: `precompile_compute_quick`, `optimize_owed`) and this PR's giant needles, both kept. - `docs/development.md` (`8aaa2f33`, `acd3a71a`): the `RX_VK_COMPILER_SERVICE` and `RX_VK_SLOW_COMPILE_S` rows take main's `cache/longshot_pipelines.bin.slow` (#321) with this PR's text. - The semantic adaptation is its own commit, `20367871`: `test_warm_handoff`'s fixture gets stand-ins for main's warm (#252's `IntelDriverGate` and `guarded_compute_held`, #190's and #650's `create_cached`, #652's `create_watched`/`compiles_watched`), and its dispatch check reads main's `p` (the integration branch's `plain` is #663's rename). - Not carried: the integration branch's two fixture fixes for this test, `4253b972` (`#include "crash_markers.h"`, needed once #428 puts `crash_markers::begin` in `prepare_run`) and `1cbb594a` (a comment-tolerant `seed_compute` check, needed once the SPIR-V profile switch, a later macOS PR, adds to that line). Neither is needed on this base. Whichever of this PR and #428 lands second adds that include to `test_warm_handoff.py`'s fixture. - Sibling macOS PRs: #665 and #666 build on this branch. `git merge-tree` of this head: no conflict with #423, #427, #431, #438, #476, #479, #480, #490, #491, #492; `gpu/vk/vk_main.cpp` with #428, #498 and #653, and `vk_main.cpp`/`vk_progcache.cpp`/`vk_shader.h`/`vk_vendor.h` with #646: the same conflicts as #652's own head (`reports/PubWave-files/PubLate1/merge-check-*.txt`), none from this PR's lines. - Conflicts with main: none. ## 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. - [ ] Testing names what was tested, what wasn't, and what must still be tested, where and how. (Test requests not open yet.) - [x] Rebased on current `main` (main e090cc48). - [x] `python -I -B tests/run_all.py` passes (any failure that also fails on `main` is named). (On the wave's merge of the macOS PRs this one is built with, `bfe5c752` (metal-cache2 + giant-mem-gate, which carry this PR): 544 passed, 32 failed, 17 skipped against main 6c9edffe's 531/32/17 on this Mac; the 32 failures are main's own, nothing that passes on main fails, 13 new tests pass; `reports/PubWave-files/waveC1/`.) - [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] Switches documented (docs/development.md; the README for a menu row) and shipped at their recommended value. - [x] A game-free test added for the fix, failing on `main` and passing here. - [x] A check under `tests/` that reads changed source updated in this PR, not weakened. - [x] Docs updated.

mooki added 19 commits 2026-11-15 02:55:02 +00:00

fix(macos): the shader preparation runs on every usable CPU on macOS ... 1843f6f809

prepare_run's first part (every shader translated, and each compute
variant compiled for the pipeline cache) ran on half the CPUs
everywhere: prepare_cpus gives half_cpus unless the Windows process heap
is the segment heap. Half exists for the NT heap's lock that an
in-process driver compiler queued on, and for whole-PC freezes when
those compilers ran on every CPU at High priority. Neither holds on
macOS: MoltenVK converts SPIR-V to MSL in a few ms and hands every
compile to Metal's compiler service (MTLCompilerService, a launchd
service at nice 0), so the workers mostly wait on it. The Mac's first
start logged "8 workers for the shader programs (16 CPUs usable; half
the CPUs)".

prepare_workers::count takes an `every` argument (default false): each
usable CPU gets a worker, at least half_cpus and 1, at most kMaxWorkers.
kEveryCpu is true where __APPLE__ is defined, and prepare_cpus passes
it; the preparation's log line says "every one: macOS". In a synthetic
MoltenVK 1.4.2 benchmark on an M4 Max (16 CPUs, macOS 26.7.1) 16 threads
compiled 56-61 % more compute and 35-39 % more graphics pipelines a
second than 8, but only with a pipeline cache per thread; with the one
shared cache LS uses today the rate is flat from 4 to 16 threads. The
next commit gives each worker its own cache.

tests/test_prepare_every_cpu.py compiles the real prepare_workers.h and
prepare_cpus() cut from vk_progcache.cpp: on this compiler's target
prepare_cpus() gives 16 of 16 usable CPUs on macOS (8 elsewhere); count
with and without `every`; prepare_cpus with kEveryCpu forced either way.
On the base it fails: prepare_cpus() gives 8 of 16 on macOS.

Windows and Linux: unchanged (kEveryCpu false; the existing
test_prepare_workers and test_prepare_cpus cases are untouched).

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit 3573e736e15078554543e4740c5e2d53c719fa89)

fix(macos): a pipeline cache per preparation worker on MoltenVK ... 55749103ac

MoltenVK 1.4.2 holds one mutex per VkPipelineCache across the SPIR-V to
MSL conversion and Metal's library compile (MVKPipeline.mm
getShaderLibrary, _shaderCacheLock). Every worker of the start's
preparation compiled with the one start cache, pcache, so they compiled
one pipeline at a time: the Mac's first start (mac-arm64-a5f2d6f9-01,
M4 Max) logged strict pairs of Metal compiles every ~195 ms, 5.05
compute variants a second, an estimated 33 minutes for the compute
programs alone. A synthetic benchmark on that Mac (PrepSpeed, MoltenVK
1.4.2, macOS 26.7.1): one shared cache stays at 10.5 compute and 9.6
graphics pipelines a second from 4 to 16 threads; a cache per thread,
merged after, gives 31.5 / 29.7 at 8 threads and 49-51 / 40-41 at 16.

gpu/vk/worker_caches.h: a Set of empty caches, one per worker of a
phase, that a worker enters with Scope; create() asks the shared cache
first under VK_PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT
and compiles what it lacks with the worker's own; merge_into() merges
them into the start cache with one vkMergePipelineCaches in slot order
(MoltenVK retains each Metal library, no compile). The caches are not
externally synchronized, so a shared slot stays correct.

vk_draw.cpp:
- create_cached: before play, a worker in a phase with a set goes
  through worker_caches::create (pcache asked first, so a saved file's
  or an earlier phase's pipelines are found as before); every use of
  pcache holds pcache_mu shared, and worker_caches_merge holds it alone
  (vkMergePipelineCaches's dstCache is externally synchronized:
  vk.xml externsync without INTERNALLY_SYNCHRONIZED_MERGE). A save's
  reads of pcache's data hold it shared too.
- prepare_run (the compute warm), precompile_run, precompile_warm (both
  passes) and precompile_at_create make a set under worker_caches_on,
  enter it in their workers and merge it right after the join; the
  precompile's stragglers left running are merged as far as they got.
- precompile_run, with the caches on, sorts the notes by program
  variant and hands a variant's notes to one worker in runs of up to
  32 (worker_caches::runs/claim), so its shader libraries are compiled
  once, not by every worker that met one of its notes at the same time.
- In play, with the caches on, a compile goes into a new cache of its
  own merged into the generation's at once under the generation's lock
  (made_alone), which the asks of the generation hold shared: the asks
  keep their order and find what any thread compiled.

vk_vendor.h: worker_caches, on for VK_DRIVER_ID_MOLTENVK only, and only
with pipelineCreationCacheControl; RX_VK_WORKER_CACHES forces it either
way. Elsewhere nothing measured says a driver's cache is locked across
its compiles, and the caches would only add merges. The pipelines built,
the persisted cache's format and its stamp are the same either way.

tests/test_worker_caches.py runs vk_draw.cpp's own create_cached,
play_cache_make, worker_caches_on and worker_caches_merge and the real
header against a made-up driver whose caches lock across a compile:
four workers compile at once (without a set they cannot); pcache asked
first, nothing compiled with it; ask-only callers never compiled; caches
not externally synchronized; one merge in slot order whatever order the
slots were taken in, with no ask of pcache beside it; play's compiles
merged into the generation at once; runs and claims. The four-workers
part runs against the base's create_cached and fails there (1 of 4
compiles met the others). tests/test_pipeline_play_cache.py gets an
"alone" case, and both play-cache and cache-save harnesses the new
stand-ins (pcache_mu, worker_caches_on).

Windows and Linux: the same code paths as before (no set, one shared
cache), plus an uncontended shared lock around pcache's uses.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit 2174e33f14ccce8c1635565e734773a33bc32455)

fix(macos): the worker caches make no Vulkan call from a header ... b5f4f1febd

tests/test_recorder_sync.py holds every header to no recorder boundary
call: vk_work.h, included last by each renderer .cpp, turns
vkDestroyPipelineCache into a drain of RX_VK_RECORDER's packets and then
the call, and a header's calls would be wrapped or not depending on what
included it. worker_caches.h called vkCreatePipelineCache,
vkDestroyPipelineCache and vkMergePipelineCaches itself, from vk_draw.cpp
before vk_work.h.

worker_caches::Set and made_alone now take a Calls struct with the
three operations; vk_draw.cpp defines worker_cache_calls after
vk_work.h, so a set's cache destruction goes through the recorder's
boundary like every other vkDestroy* in the file. Behaviour is
otherwise the same: the same caches, made, merged and destroyed in the
same order.

tests/test_worker_caches.py and tests/pipeline_play_cache_test.cpp take
worker_cache_calls from the source and the model respectively.
test_recorder_sync now fails only as on the base (vk_vma.cpp's
vmaDestroyAllocator), and the four tests that touch the caches pass.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit aa8ecba546465f1ce65ac872ecf21d232b65cef0)

fix(macos): MoltenVK starts with an empty pipeline cache and writes none ... f59051783e

MoltenVK 1.4.2 keeps MSL, not GPU code, in a VkPipelineCache, and
vkCreatePipelineCache with saved data compiles every stored library one
after another on the calling thread (MVKPipeline.mm readData).
pcache_init hands it all of mkw_vk_pipelines.bin, and play_files_merge
every play file, at every start. The PrepFast GPU check on the M4 Max,
with Metal's per-user cache blocked as in the runner, ran three later
starts of the same 64 pipelines (80 libraries):
- reading the one file: 5.7 s of serial compiles, 6.6 s in all;
- 16 shards read in parallel: 1.4 s;
- nothing read, pipelines compiled from SPIR-V with the worker caches:
  1.4 s.
At the game's scale the read is tens of minutes per start.

vk_vendor.h gets pcache_no_load, on for VK_DRIVER_ID_MOLTENVK only and
logged in the vendor fixes' line with worker_caches. It is the device's
default for RX_VK_PCACHE_LOAD: pcache_load_off reads the switch first
(any value, as before) and falls back to the fix, so
RX_VK_PCACHE_LOAD=1 reads the files on MoltenVK as everywhere else. With
the fix on and the switch unset or off, pcache_files_unused holds:
- pcache_init reads neither the main file nor the play files and says
  so in the log; the cache lives in memory for the run;
- nothing is saved (pcache_no_save, as RX_VK_PCACHE_LOAD=0 already
  did), and play_cache_save writes no play file and keeps the
  generation in memory: unread play files would pile up, since they
  are compacted only where they are read;
- the stamp is left alone (pcache_drop_stale returns as for any
  start that does not read the files);
- precompile_run drops RX_VK_LOAD=create's longshot_pipelines.bin.built
  marker at each start, as it does when the files were dropped: no
  cache a later start reads holds the notes, so the warm builds them
  into the in-memory cache again before loads at create ask for them.

Saving does not continue on MoltenVK because nothing reads the file or
its stamp there:
- the prepared record (prepared_for) has its own file and stamp;
- --prepare-shaders' saved cache is useful only to a start that reads
  it;
- pcache_saved_now serves graphics pipeline libraries, which MoltenVK
  does not have;
- a start with RX_VK_PCACHE_LOAD=1 makes, saves and stamps its own.
Every other driver is unchanged, RX_VK_PCACHE_LOAD=0 there included
(main file unread and unsaved, play files still merged and written).

tests/test_vendor_fixes.py checks the two cache fixes on its 16 devices
and on MoltenVK: on for MoltenVK alone, RX_VK_WORKER_CACHES forcing
the first, vendor_fixes_env leaving pcache_no_load to the files'
switch. It also checks the log line and pcache_load_off's order.
tests/test_pcache_stamp.py runs pcache_load_off, pcache_files_unused,
pcache_drop_stale and pcache_init, cut from the source, in eight starts:
other drivers unset, 0, empty and 1 as before, and MoltenVK unset, 0,
off and 1. test_pipeline_play_cache gets an "unused" case. All three
fail on aa8ecba5. tests/test_warm_bounded.py's source check of the
marker's removal takes the new condition.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit 61bfbf1ae7c0966a9436ab0c556a9ee547bb3b5f)

fix(vk, macos): the renderer's threads start with a 16 MiB stack, not macOS's 512 KiB ... f0a62f7ba0

macOS starts a std::thread with 512 KiB of stack, where Linux starts one
with 8 MiB and Windows with longshot.exe's 8 MiB. On macOS MoltenVK
converts each shader with SPIRV-Cross on the thread that creates the
pipeline, recursing once per block of a program's control flow, and a
prepare_run worker overran its 512 KiB there (CFG::walk_from under
vkCreateComputePipelines from seed_compute, start #6, 2026-11-14).

gpu/vk/thread_stack.h: thread_stack::thread is std::thread itself off
macOS; on macOS a std::thread-like thread (construction with decayed
arguments, joinable, join, detach, native_handle, moves) started by
pthread_create with thread_stack::kBytes (16 MiB) of stack. Every
thread the renderer starts goes through it: the preparation's pools
(prepare_run, state_list_prepare, warm_libraries, precompile_at_create,
precompile_run, predict_run, warm_bounded::run), the job workers and
the serial builder, the replay, parallel-replay, presenter and
frame-generation threads, the program cache's loaders and the rest.
mkwvk_init logs the size once per start on macOS.

tests/test_thread_stack.py: no file under gpu/vk starts a thread any
other way; each kind of thread runs a recursion 2 MiB deep through the
helper, and on macOS each overflows as a std::thread (caught, no crash
report). Seven source checks follow the new spelling.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit 48b49054b1c64851b3e8c974e64dc0b757b2ced3)

test(macos): one overflow message from thread_stack_test, however many threads overflow at once ... ea4081d942

The first thread whose stack overflows says so and ends the process;
the others wait in their handlers, so the message is not interleaved.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit 62d0991e0cfb5402202f8f14e9eb799d5c20d262)

test: the checks that compile renderer code with its threads see thread_stack.h ... 7707c63433

Eight checks cut renderer functions that start threads (or copy
vk_jobs.cpp and vk_recorder.h) into fixtures of their own; they now
build with gpu/vk/thread_stack.h beside them: test_capture_copy_helpers,
test_gpu_hang_watch, test_jobs_failure, test_jobs_wait_runs_queued,
test_pipeline_cache_save, test_prepare_stall, test_present_source_crop
and test_recorder_sync.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit fd46ff431f9f62b1125582a4a3434a8cc92c551e)

fix(macos): the worker caches follow main's fetch lock (#190) and Intel's gate (#252) ... 078c9a7a54

Main's #190 (f41a5c59) put a fetch lock (PcacheFetch, PcacheUse) on the pipeline cache's save and compile path, the
very path where the worker caches add the lock their merges need, and #252 (57f4eaec) put Intel's compiler gate
(IntelDriverGate) around the compiles. The worker caches now sit on top of both:

- pcache_mu, the merge lock, is taken only where the worker caches are on (pcache_use_held: MoltenVK by default).
  Elsewhere create_cached, the save's fetch and pcache_saved_now take exactly #190's locks, as on main: no new lock
  on any other driver. A play generation's lock is likewise taken only where play's compiles are merged into it
  (g->alone).
- The worker caches are never on with the fetch lock (worker_caches_on: not with pcache_fetch_lock_on, Intel's
  Windows driver or RX_VK_PCACHE_FETCH_LOCK=1): a merge reads its caches' data as a fetch does, which that lock keeps
  apart from the builds there, and the gate lets one compile at a time into Intel's compiler there anyway.
- The order of the locks: Intel's gate, then pcache_mu, then the fetch lock. The save's fetch and pcache_saved_now
  take pcache_mu before PcacheFetch; a merge takes pcache_mu alone.
- prepare_run's compute warm keeps #252's gate around its compile and, with the caches, goes through create_cached
  under it (cached_compute would take the gate a second time).
- tests: test_worker_caches checks the gating (with the caches off a pipeline is made while pcache_mu or the
  generation's lock is held alone elsewhere; with them on it waits for pcache_mu), worker_caches_on against the
  fetch lock, the warm's gate and pcache_mu ahead of PcacheFetch in the save. test_pipeline_play_cache's route case
  checks that no lock of theirs is taken off MoltenVK; pipeline_play_cache_test.cpp and pipeline_cache_save_test.cpp
  get pcache_use_held stand-ins. test_pcache_stamp's load fixture follows #321's cache_path and #190's PcacheFetch in
  pcache_init, and play_cache_save's new parameters.
- docs: #321's cache/longshot_vk_pipelines.bin in the RX_VK_WORKER_CACHES row and in vk_vendor.h, and a row for
  RX_VK_PCACHE_LOAD, whose default this PR changes on MoltenVK.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>

merge(macos): the renderer threads' 16 MiB stacks (mac/thread-stack 7707c634) under the precompile hand-off ... f9f166f7cb

The precompile hand-off starts its workers and its compile watcher through thread_stack::thread. One conflict file,
gpu/vk/vk_draw.cpp (4 blocks: the lines P31 had to resolve against main, the other way round): the include block keeps prepare_workers.h,
worker_caches.h and thread_stack.h; prepare_run's, precompile_at_create's and precompile_run's thread lines take
thread_stack::thread and keep the worker caches' merges beside them.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>

fix(vk): MoltenVK: the precompile goes on behind the game; slow compiles named; a failed compile tried once more ... 89606310bb

On the Mac's start 12 every precompile worker was held by a Metal compile of minutes while 37,600 of
42,234 items were still to hand out, so RX_VK_PRECOMPILE_STALL_S could not apply (it waits until every
item is handed out) and the title never came. With the new vendor fix compiler_service (MoltenVK only,
RX_VK_COMPILER_SERVICE):
- the precompile hands the rest to the game once nothing has finished for the stall time, or once all
  that is left is compiles running past RX_VK_SLOW_COMPILE_S; a quarter of the workers build the rest
  behind the game, every item still built, each straggler named when it ends;
- its order: noted pipelines first, then fallbacks and computes by SPIR-V words, what an earlier run
  found slow (longshot_pipelines.bin.slow) last;
- a compile over the slow time (10 s) is named with key, variant, kind, stage and SPIR-V words while it
  runs and when it ends, and written to the slow list;
- a pipeline whose compile failed is compiled once more with a new pipeline cache (MoltenVK keeps a
  failed library in the cache it was compiled with), and said made or not made.
Every other driver: the classic worker, wait and log line as they were.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit 77d4971cbdb6356d9064a44cdee923f4bde354f9)

test(vk): test_precompile_handoff: a tree without the change says each missing piece ... 306ffd28ca

Assisted-by: Claude Opus 5.5

Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit f9a179c08d3f9e102729ec8fe5fe923336f581b8)

docs: RX_VK_PRECOMPILE_STALL_S, RX_VK_COMPILER_SERVICE and RX_VK_SLOW_COMPILE_S ... 44f210178c

Assisted-by: Claude Opus 5.5

Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit 194a230dd959706797203501edad3b5a2331118a)

fix(vk): name a graphics compile at its precompile callers, not inside build_graphics ... ab2bf44680

test_fallback_pipeline_state compiles build_graphics cut from the source; the noted pipelines of behind_worker and precompile_warm name what they build instead.

Assisted-by: Claude Opus 5.5

Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit 528784dc5c0c293da1b692f6fd71da070a940d8b)

fix(vk): the precompile hand-off on main's Intel gate and quick builds (#252), fetch lock (#190) and game folder (#321) ... 7ef7cff977

The hand-off was written on the macOS integration branch, before main's #190 (f41a5c59), #252 (57f4eaec) and #321
(76699aad), and before the worker caches followed them (this branch's base). What it needs on main's code:

- cached_graphics and cached_compute keep #252's gate and main's `return create_cached(guarded_*_held, ci, pl);`, and go
  through create_watched only where compiles are watched (compiles_watched: compiler_service, MoltenVK's, or
  RX_VK_SLOW_COMPILE_S set). Every other driver makes its pipelines exactly as on main, and main's own source checks
  (test_intel_compiler_gate, test_pcache_fetch_lock) read the lines they read there.
- prepare_run's compute warm: with the worker caches and watched compiles through create_watched, under the gate it holds;
  otherwise P22's create_cached or main's guarded_compute_held, as they are.
- behind_worker, the precompile's worker under compiler_service, builds a compute pipeline as the classic worker now
  does (#252): on Intel's one-at-a-time builds (serial_builds_on) without the optimizer first, the optimized build owed
  to its first dispatch (precompile_compute_quick, optimize_owed). On MoltenVK, where compiler_service is on, that never
  applies unless RX_VK_COMPUTE_SERIAL=1; it keeps the two workers' builds the same.
- test_precompile_handoff's source checks follow main's spellings: the watched and plain lines in cached_* after the gate,
  compiles_watched, the warm, the dispatch's inline build named as main spells it (p, not the integration branch's plain),
  and behind_worker's quick build.
- docs: the slow list is cache/longshot_pipelines.bin.slow since #321 moved the notes into the game folder's cache/.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>

fix(vk): under compiler_service the preparation leaves the compute pipelines to the precompile ... abfc05c8ea

On the Mac's cold start 14, prepare_run's compute warm kept 13 of its 16 workers on compute compiles of
122k-209k SPIR-V words for 205-274 s each. The preparation reached 278 of 38,809 programs in 430 s, and the
precompile, whose hand-off lets the title start while such compiles go on behind it, never began.

With compiler_service (MoltenVK only) the warm compiles nothing. Each compute variant is still translated and
kept by seed_compute (program_cache_save), and the state list and the prepared record follow as before. The
precompile compiles the pipelines: every kept compute program on a first start (computes_plan), in its order
and with its hand-off; one it does not take is compiled at its first dispatch. Every other driver's warm is
unchanged.

tests/test_warm_handoff.py runs prepare_run's own decision and warm_compute against stand-ins on both arches,
and pins the source chain that keeps each variant for the precompile.

Assisted-by: Claude Opus 5.5

Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit 84c32d68c536468b61af24230ea8c283c5f091ef)

fix(vk): at most four giant compiles at once in the precompile under compiler_service ... bd313bffae

Start 14 had 13 compute compiles of 122k-209k SPIR-V words in flight at once while the other programs went
at 0.1-0.3 a second, and Metal's compiler service aborted twice (LLVM bad_alloc).

precompile_handoff.h Giants: at most RX_VK_GIANT_COMPILES (4) compiles of more than RX_VK_GIANT_WORDS
(100,000) SPIR-V words in flight at once among behind_worker's workers, before the game and behind it. A
giant claimed while the bound is reached is deferred, never dropped, and named once with a count; its
worker takes the next item, and the deferred ones are taken first once a giant ends. A worker with nothing
else waits for them and leaves only when none is deferred. behind_worker claims with claim_below, which
never passes the end, and a worker adds its one "ran out" to next as it leaves, so the wait's
next >= total + threads still means every worker has left. The classic worker (every other driver) is
unchanged.

The precompile_handoff fixture runs the gate with blocking giants: at most 2 of 6 in flight, the small
items going on meanwhile, each deferral named once with its count, every item built once, before the game
and behind it with the hand-off.

Assisted-by: Claude Opus 5.5

Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit f1e028c8db05ef31383fa44c05fa156984c0f6c9)

docs: RX_VK_GIANT_COMPILES and RX_VK_GIANT_WORDS; the preparation's compute pipelines under compiler_service ... 1725c432f4

Assisted-by: Claude Opus 5.5

Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit 8aaa2f33d51b6b6f8db335cac69639137833537e)

fix(vk): giant compiles never hold every worker of a phase ... e10c0ad542

The bound on giant compiles in flight is now min(RX_VK_GIANT_COMPILES, workers in the phase - 1), and at
least 1. Before the hand-off the phase is the precompile's workers (16 on an M4 Max: 4); at the hand-off
precompile_run sets it to the share behind the game (4 workers: 3), so one worker there always stays on the
smaller items. Giants already in flight over the new bound finish; no more start until they are under it.
RX_VK_GIANT_COMPILES=0 is still no bound. The order line, the hand-off line and each deferral line give the
bound in use; docs/development.md says so.

The precompile_handoff fixture gains a hand-off by the stall time with three giants (4 workers: 3) and a
stuck small item holding every worker and most of the list still to hand out: behind the game, on a share
of 2, one giant at a time and the other worker on the small items, every item built once. Plus limit_for
and phase.

Assisted-by: Claude Opus 5.5

Co-Authored-By: Claude Opus 5.5 <[email protected]>
(cherry picked from commit acd3a71a2717b1e1bfc83c1952932fdcf1afd9de)

test(vk): test_warm_handoff follows main's warm and dispatch (#252, #190) ... 20367871b5

The fixture cuts prepare_run's warm out of vk_draw.cpp. On main the warm builds
under Intel's compiler gate (#252 IntelDriverGate) through guarded_compute_held,
with the worker caches through create_cached (#190, PrepFast) or create_watched
where compiles are watched (PrecompileStall's compiles_watched). The fixture now
has stand-ins for those; without worker caches, as there, the warm still goes to
guarded_compute_held into pcache and is counted the same way.

The source check for a dispatch's build of an unmade compute pipeline reads
main's name for the dispatch's program, p (the topic's tree had a later
rename to plain).

Assisted-by: Claude Opus 5.5

Co-Authored-By: Claude Opus 5.5 <[email protected]>

mooki referenced this pull request 2026-11-15 02:55:22 +00:00

WIP: fix(macos): hold MoltenVK's giant compiles behind the game until its first frame, at a lower QoS class #665

mooki referenced this pull request 2026-11-15 02:55:23 +00:00

WIP: fix(macos): put a dispatch's compute pipeline ahead of MoltenVK's giant compiles and retry it alone, not fail it for good #666

mooki referenced this pull request 2026-11-15 02:55:25 +00:00

WIP: feat(macos): RX_VK_METAL_ARCHIVE_DIR keeps MoltenVK's pipelines in Metal binary archives #667

mooki referenced this pull request 2026-11-15 03:11:18 +00:00

WIP: fix(macos): start MoltenVK's giant compiles behind the game only while memory allows, and say a lost device as one #669

mooki force-pushed mac/warm-handoff from 20367871b5 to 658b854052 2026-11-15 07:25:37 +00:00 Compare

mooki referenced this pull request 2026-11-15 11:56:02 +00:00

WIP: fix(vk): heavy compiles run alone under Metal's compiler service; the first giant behind the game no longer waits for 16 GB #784

mooki referenced this pull request 2026-11-15 11:56:08 +00:00

WIP: fix(macos): KosmicKrisp's compiles go through the compiler service's precompile, as MoltenVK's do #785

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 force-pushed mac/warm-handoff from 658b854052 to 1f7065181d 2026-11-15 13:54:53 +00:00 Compare

mooki added the

area/renderer

kind

bug

platform/macos

priority

high

labels 2026-11-15 13:57:35 +00:00

mooki referenced this pull request 2026-11-15 17:40:00 +00:00

WIP: fix(macos): MoltenVK's precompile and pipelines on Metal #849

mooki commented 2026-11-15 18:08:31 +00:00

Author

Member

• Copy link
• Copy Markdown

Folded into #849 (fix(macos): MoltenVK's precompile and pipelines on Metal)

Folded into #849 (fix(macos): MoltenVK's precompile and pipelines on Metal)

mooki closed this pull request 2026-11-15 18:08:32 +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

PR description / Description follows the template (pull_request_target) Successful in 6s

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 12s

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 @ 8015c9a

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 @ 8015c9a

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!664

No description provided.