fix(build): an update keeps the headers its host and renderer objects were keyed by #2199

Open
dazemc wants to merge 2 commits from dazemc/Longshot:fix/object-reuse-cleanup into main
Contributor

Summary

An update that changes any header under host/ or gpu/ compiles every one of the Vulkan back end's own files and the host's again, even the ones that read nothing that changed, and the shared object store cannot stop it. The build in #1906's log says it in two lines: object store …: 0 objects from it, then 561 files.

How to see it on main: build a game (any Linux or Windows build: the launcher always sets LONGSHOT_OBJ_STORE). Build it again from the same work folder with the same sources: the second build's log says object store …: 0 objects from it for the host's and the back end's own objects, and its per-file lines show them compiled again. A developer's build does not show it: scripts/devtools/packs.py:509 and scripts/devtools/build.py:457 set LONGSHOT_KEEP_BUILD_FILES=1.

Why: every back-end and host object is named by the headers its clang depfile lists (renderer_deps, #37 and #673). A player's build ends in keep_translation, which deletes the build folder and the depfiles with it:

  • nx2/make_game.py:2483 — shutil.rmtree(os.path.join(work, "sa", "build"), ignore_errors=True).
  • nx2/build.py:555-556 — with no depfile, renderer_deps returns the whole-folders' hash instead: if not picked: return blob.
  • nx2/build.py:1649 — after each compile, rekey_depfiles caches the object under its depfile key: key = object_key(src, fl, renderer_deps(blob, roots, obj) + extras) then cache_put(obj, key).

So the store and the folder's cache hold <depfile key>.o while the next update asks for <blob key>.o: object_store.fetch links nothing, the file is compiled again, and rekey_depfiles then re-keys it to a key the store already had. It happens on every update that changes a header under host/ or gpu/, which most nightlies do. #1602 counted the same update for its own measurement: 105 objects (65 host .c, 40 renderer .cpp), 239 compiler-seconds, 37 s of wall time on 16 threads.

What this changes: the headers a compile read now stay beside the object, in obj.deps, as the root's place and the path below that root — path-free, so the same key comes back from another builder folder (the AppImage's mount is new at every start, a launcher's tools folder at every update). They are read as they are now, so an edited header still compiles the file, and a deleted one changes the key rather than being taken for the one that was. renderer_deps falls back to the list when the depfile is gone or names nothing under the roots; rekey_depfiles writes it; keep_translation keeps it (a few hundred small files) and nothing else of the build folder.

Compatibility: the digest is made by one function for both routes, so the lists reproduce the depfile's digest exactly — checked against the 170 depfiles of a real work folder, 170 of 170 the same. No key of the objects those depfiles name moves, so the update that ships this does not compile them again for the new digest's sake (it compiles the 105 once for the missing lists, below).

Split out: the plan's flags line — the one that said -O2 in #1906's log because the object store's loop rebound fl — is #2217: review asked for the two bugs as two changes. This PR is the object-key fix only; no object key depends on that line, and the measurements below are unaffected by it.

Read first: nx2/build.py (renderer_deps, deps_digest, written_deps, rekey_depfiles), nx2/make_game.py (keep_translation), then tests/test_object_reuse_cleanup.py.

See #1906, #1907, #2108.

Testing

Tested: game-free only. In the game: N/A — build tooling only, no game code path changes; the evidence is the suite below and the object-store measurement in Measurements.

  • tests/test_object_reuse_cleanup.py (new, 11 checks, no compiler, no game): the key rekey_depfiles stores an object under is the one a later plan asks for; the clean-up does not change it; it is the same from another builder folder; an edited or deleted header changes it; a header of another root with the same name does not; an object taken from the store keeps its list; a work folder with neither a depfile nor a list keeps the blob; the store's loop does not rebind fl. On main the key check fails ('3d52f65a…' != '279d918c…', "the clean-up changed an object's key") and the plan-line check fails; on this branch all 11 pass.
  • The existing checks around this code pass unchanged: test_renderer_depfiles (11), test_host_depfiles (6), test_kept_translation (7), test_header_ext_deps (8), test_object_store, test_object_cache_prune, test_build_hash_cache (11), test_dev_update (25), test_make_game_cache (48), test_build_cleanup_scope (11), test_doc_anchors, test_lift_cache.
  • mise run test (the whole suite) on the pre-split head 613b775a, same command and environment as the baseline (prrefresh-logs/baseline-a2dc5385.log, a clean a2dc5385 checkout, pinned Vulkan/SPIR-V headers, --strict --jobs 3, 8 GB scope): 912 passed, 23 failed, 18 skipped in 619 s. The failing names are exactly the baseline's 23 (diff empty): nx2/test_import_windows.py and the 22 tests/ names of main's set, tests/test_sim_turf_pools.py among them. The one extra pass over the baseline's 911 is this PR's new check. The suite also caught two checks that cut build.py's Vulkan jobs out and exec them (test_shader_stamp_cache, test_dlss_sdk_cache, NameError on the new helpers); they are updated here with their assertions unchanged.
  • The head 2b65ab36 has its own full suite, same command, environment and scope as the baseline: 912 passed, 23 failed, 18 skipped in 442 s (--jobs 5, 12 GiB), the same 23 names as the baseline (diff empty, no new name). It ran from a worktree at that commit, not borrowed from the pre-split run.
  • The measurements below, taken as two plans of one work folder with the same toolchain, environment and sources.

Still to test:

  • Windows or Linux with a builder: build a nightly from before this PR, then update it to one with it from the same work folder, expecting object store …: N objects from it to name the host's and the back end's own objects and their compile count to fall to the files that really changed (by hand; it needs a builder and a game, not a GPU test).

Measurements

Verdict: with the same sources, flags and translation, a plan with the lists takes 92 objects from the object store; the same plan without them — a player's post-clean-up state, which main plans the same way — takes 0. Those are the host and back end objects that used to be keyed by the whole folders' hash. That cost is the one #1602 measured; this removes its cause, and the end-to-end saving on a real update is the first item of Still to test. Nothing in the plan gets slower.

Setup: Linux, Ryzen 9 5950X (16 cores, 32 threads, 16 GB), one work folder copied from the shared base at a053f9e1 (the same sources, toolchain and store), planned twice — once as a player's clean-up leaves it (sa/build deleted: no depfile, no list) and once with the obj.deps lists this PR writes (the same work folder with the build folder emptied of everything else). The two runs' object keys: fingerprints are identical: game-code headers 5e2c6c3841, flags 9244793c00, hot flags a0a8ecd2c6, lifted code c1d9965f69. Nothing was compiled: the plans stop after the object store's line. Measured at 868c7df7. The head 2b65ab36 differs from it by the split-out commit (removed, now #2217) and the two stale-check updates — none of it is in an object key; the suite on the head is 912/23/18, the baseline's 23 names.

Before, no obj.deps (a player's post-clean-up state): this PR @ 868c7df7 After, the lists kept: this PR @ 868c7df7
objects the store serves to a player's post-clean-up work folder (of 163 depfile-keyed objects) 0 92

Baseline: both columns are the same work folder, copied from the shared base at a053f9e1 (the same sources, toolchain and object store), planned by this PR's build.py at 868c7df7; the only difference between them is the *.deps lists, which the Before column does not have. That state is a player's post-clean-up work folder, and main plans it the same way: no depfile and no list, so renderer_deps gives the whole folders' hash, the keys are the same, and the store answers 0 — which is the line #1906 reports. A plan on a main checkout itself was not run for this table.

The 92 are what the store answered; the other 71 of the 163 were already in the folder's own cache, and the game's chunks are served from it in both runs (their key does not depend on depfiles). The 7 of the 170 depfiles that name no header under the watched folders were left out of the lists, as before.

Risks

  • What could break: an object reused when a header it read changed. The list names the headers and their digest is computed from their contents at plan time, so an edit compiles the file; a deleted header counts as gone and changes the key; an include added to a source changes the source, which is in the key. The digest function is the depfile route's own, and 170 of 170 existing digests are identical, so nothing that worked before changes meaning.
  • The first update that ships this still compiles the ~105 objects once: the build that wrote its work folder ran without the lists. Every update after it reuses them.
  • What this does not fix: the lifted game chunks. Their key is the translation and split_giants.flag's hash of split_giants.py, not a depfile, so an update that changes either still compiles all of them (#1727's key, by design). This PR is about the host's and the back end's own objects; a log that still says 0 objects from it for the chunks in a nightly that changed the translator is expected.
  • What changes on disk: sa/build/obj/*.deps, a few hundred files of a few KB (about 0.2 MB) in the work folder a player's build keeps. LONGSHOT_KEEP_BUILD_FILES=1 and Free up space behave as before; a build folder that is deleted whole (a work folder nobody built in, sa gone) keeps today's behaviour, the whole folders' hash.
  • Turning it off: no switch; a revert is enough, and a reverted build simply compiles those objects again.
  • Nothing a player notices in the game: no save, setting, cache or shader format changes.

Checklist

Ready for review

  • One topic, on current main; the title is a Conventional Commit. (The second commit is the log line for this same reuse; see Summary.)
  • Tested on my own PC, in the game for a change the game runs, as Testing says. (In the game: N/A — build tooling only, no game code path changes; the evidence is the suite and the object-store measurement.)
  • mise run test passes (any failure that also fails on main is named): 912 passed, 23 failed, 18 skipped on this head (2b65ab36, --jobs 5, 12 GiB, pinned headers); the 23 names are exactly a2dc5385's baseline set, no new name.
  • Measured against main, or Measurements says why there is nothing to measure.
  • No game files, build output or hand-edited generated files; third_party/ untouched; nothing GPL-incompatible.

When it applies

  • A fix: a game-free test that fails on main and passes here.
  • A check under tests/ that reads changed source is updated, not weakened.
  • Docs updated.
  • AI assistance recorded with Assisted-by: / Co-Authored-By: trailers.

Ready to merge

  • Everything in Still to test has run, its results in Tested.
  • Approved in review, every comment answered.
  • CI passes on the head; on current main, measured again or a comment says why the numbers still hold.
## Summary An update that changes any header under `host/` or `gpu/` compiles every one of the Vulkan back end's own files and the host's again, even the ones that read nothing that changed, and the shared object store cannot stop it. The build in #1906's log says it in two lines: `object store …: 0 objects from it`, then `561 files`. **How to see it on main:** build a game (any Linux or Windows build: the launcher always sets `LONGSHOT_OBJ_STORE`). Build it again from the same work folder with the same sources: the second build's log says `object store …: 0 objects from it` for the host's and the back end's own objects, and its per-file lines show them compiled again. A developer's build does not show it: `scripts/devtools/packs.py:509` and `scripts/devtools/build.py:457` set `LONGSHOT_KEEP_BUILD_FILES=1`. **Why:** every back-end and host object is named by the headers its clang depfile lists (`renderer_deps`, #37 and #673). A player's build ends in `keep_translation`, which deletes the build folder and the depfiles with it: - `nx2/make_game.py:2483` — `shutil.rmtree(os.path.join(work, "sa", "build"), ignore_errors=True)`. - `nx2/build.py:555-556` — with no depfile, `renderer_deps` returns the whole-folders' hash instead: `if not picked: return blob`. - `nx2/build.py:1649` — after each compile, `rekey_depfiles` caches the object under its **depfile** key: `key = object_key(src, fl, renderer_deps(blob, roots, obj) + extras)` then `cache_put(obj, key)`. So the store and the folder's cache hold `<depfile key>.o` while the next update asks for `<blob key>.o`: `object_store.fetch` links nothing, the file is compiled again, and `rekey_depfiles` then re-keys it to a key the store already had. It happens on every update that changes a header under `host/` or `gpu/`, which most nightlies do. #1602 counted the same update for its own measurement: **105 objects (65 host `.c`, 40 renderer `.cpp`), 239 compiler-seconds, 37 s of wall time on 16 threads**. **What this changes:** the headers a compile read now stay beside the object, in `obj.deps`, as *the root's place and the path below that root* — path-free, so the same key comes back from another builder folder (the AppImage's mount is new at every start, a launcher's tools folder at every update). They are read **as they are now**, so an edited header still compiles the file, and a deleted one changes the key rather than being taken for the one that was. `renderer_deps` falls back to the list when the depfile is gone or names nothing under the roots; `rekey_depfiles` writes it; `keep_translation` keeps it (a few hundred small files) and nothing else of the build folder. **Compatibility:** the digest is made by one function for both routes, so the lists reproduce the depfile's digest exactly — checked against the 170 depfiles of a real work folder, **170 of 170 the same**. No key of the objects those depfiles name moves, so the update that ships this does not compile them again for the new digest's sake (it compiles the 105 once for the missing lists, below). **Split out:** the plan's flags line — the one that said `-O2` in #1906's log because the object store's loop rebound `fl` — is **#2217**: review asked for the two bugs as two changes. This PR is the object-key fix only; no object key depends on that line, and the measurements below are unaffected by it. Read first: `nx2/build.py` (`renderer_deps`, `deps_digest`, `written_deps`, `rekey_depfiles`), `nx2/make_game.py` (`keep_translation`), then `tests/test_object_reuse_cleanup.py`. See #1906, #1907, #2108. ## Testing **Tested:** game-free only. **In the game: N/A** — build tooling only, no game code path changes; the evidence is the suite below and the object-store measurement in Measurements. - `tests/test_object_reuse_cleanup.py` (new, 11 checks, no compiler, no game): the key `rekey_depfiles` stores an object under is the one a later plan asks for; the clean-up does not change it; it is the same from another builder folder; an edited or deleted header changes it; a header of another root with the same name does not; an object taken from the store keeps its list; a work folder with neither a depfile nor a list keeps the blob; the store's loop does not rebind `fl`. On main the key check fails (`'3d52f65a…' != '279d918c…'`, "the clean-up changed an object's key") and the plan-line check fails; on this branch all 11 pass. - The existing checks around this code pass unchanged: `test_renderer_depfiles` (11), `test_host_depfiles` (6), `test_kept_translation` (7), `test_header_ext_deps` (8), `test_object_store`, `test_object_cache_prune`, `test_build_hash_cache` (11), `test_dev_update` (25), `test_make_game_cache` (48), `test_build_cleanup_scope` (11), `test_doc_anchors`, `test_lift_cache`. - `mise run test` (the whole suite) on the pre-split head `613b775a`, same command and environment as the baseline (`prrefresh-logs/baseline-a2dc5385.log`, a clean `a2dc5385` checkout, pinned Vulkan/SPIR-V headers, `--strict --jobs 3`, 8 GB scope): **912 passed, 23 failed, 18 skipped** in 619 s. The failing names are **exactly the baseline's 23** (diff empty): `nx2/test_import_windows.py` and the 22 `tests/` names of main's set, `tests/test_sim_turf_pools.py` among them. The one extra pass over the baseline's 911 is this PR's new check. The suite also caught two checks that cut `build.py`'s Vulkan jobs out and exec them (`test_shader_stamp_cache`, `test_dlss_sdk_cache`, `NameError` on the new helpers); they are updated here with their assertions unchanged. - The head `2b65ab36` has its own full suite, same command, environment and scope as the baseline: **912 passed, 23 failed, 18 skipped** in 442 s (`--jobs 5`, 12 GiB), **the same 23 names** as the baseline (diff empty, no new name). It ran from a worktree at that commit, not borrowed from the pre-split run. - The measurements below, taken as two plans of one work folder with the same toolchain, environment and sources. **Still to test:** - Windows or Linux with a builder: build a nightly from before this PR, then update it to one with it from the same work folder, expecting `object store …: N objects from it` to name the host's and the back end's own objects and their compile count to fall to the files that really changed (by hand; it needs a builder and a game, not a GPU test). ## Measurements **Verdict:** with the same sources, flags and translation, a plan with the lists takes **92 objects from the object store**; the same plan without them — a player's post-clean-up state, which main plans the same way — takes **0**. Those are the host and back end objects that used to be keyed by the whole folders' hash. That cost is the one #1602 measured; this removes its cause, and the end-to-end saving on a real update is the first item of Still to test. Nothing in the plan gets slower. Setup: Linux, Ryzen 9 5950X (16 cores, 32 threads, 16 GB), one work folder copied from the shared base at a053f9e1 (the same sources, toolchain and store), planned twice — once as a player's clean-up leaves it (`sa/build` deleted: no depfile, no list) and once with the `obj.deps` lists this PR writes (the same work folder with the build folder emptied of everything else). The two runs' `object keys:` fingerprints are identical: game-code headers `5e2c6c3841`, flags `9244793c00`, hot flags `a0a8ecd2c6`, lifted code `c1d9965f69`. Nothing was compiled: the plans stop after the object store's line. Measured at 868c7df7. The head `2b65ab36` differs from it by the split-out commit (removed, now #2217) and the two stale-check updates — none of it is in an object key; the suite on the head is 912/23/18, the baseline's 23 names. | | Before, no `obj.deps` (a player's post-clean-up state): this PR @ 868c7df7 | After, the lists kept: this PR @ 868c7df7 | |---|---|---| | objects the store serves to a player's post-clean-up work folder (of 163 depfile-keyed objects) | 0 | 92 | **Baseline:** both columns are the same work folder, copied from the shared base at `a053f9e1` (the same sources, toolchain and object store), planned by this PR's `build.py` at `868c7df7`; the only difference between them is the `*.deps` lists, which the Before column does not have. That state is a player's post-clean-up work folder, and main plans it the same way: no depfile and no list, so `renderer_deps` gives the whole folders' hash, the keys are the same, and the store answers 0 — which is the line #1906 reports. A plan on a main checkout itself was not run for this table. The 92 are what the store answered; the other 71 of the 163 were already in the folder's own cache, and the game's chunks are served from it in both runs (their key does not depend on depfiles). The 7 of the 170 depfiles that name no header under the watched folders were left out of the lists, as before. ## Risks - **What could break:** an object reused when a header it read changed. The list names the headers and their digest is computed from their contents at plan time, so an edit compiles the file; a deleted header counts as gone and changes the key; an include added to a source changes the source, which is in the key. The digest function is the depfile route's own, and 170 of 170 existing digests are identical, so nothing that worked before changes meaning. - **The first update that ships this** still compiles the ~105 objects once: the build that wrote its work folder ran without the lists. Every update after it reuses them. - **What this does not fix:** the lifted game chunks. Their key is the translation and `split_giants.flag`'s hash of `split_giants.py`, not a depfile, so an update that changes either still compiles all of them (#1727's key, by design). This PR is about the host's and the back end's own objects; a log that still says `0 objects from it` for the chunks in a nightly that changed the translator is expected. - **What changes on disk:** `sa/build/obj/*.deps`, a few hundred files of a few KB (about 0.2 MB) in the work folder a player's build keeps. `LONGSHOT_KEEP_BUILD_FILES=1` and **Free up space** behave as before; a build folder that is deleted whole (a work folder nobody built in, `sa` gone) keeps today's behaviour, the whole folders' hash. - **Turning it off:** no switch; a revert is enough, and a reverted build simply compiles those objects again. - Nothing a player notices in the game: no save, setting, cache or shader format changes. ## Checklist ### Ready for review - [x] One topic, on current `main`; the title is a Conventional Commit. (The second commit is the log line for this same reuse; see Summary.) - [x] Tested on my own PC, in the game for a change the game runs, as Testing says. (In the game: N/A — build tooling only, no game code path changes; the evidence is the suite and the object-store measurement.) - [x] `mise run test` passes (any failure that also fails on `main` is named): **912 passed, 23 failed, 18 skipped** on this head (`2b65ab36`, `--jobs 5`, 12 GiB, pinned headers); the 23 names are exactly `a2dc5385`'s baseline set, no new name. - [x] Measured against `main`, or Measurements says why there is nothing to measure. - [x] No game files, build output or hand-edited generated files; `third_party/` untouched; nothing GPL-incompatible. ### When it applies - [x] A fix: a game-free test that fails on `main` and passes here. - [x] A check under `tests/` that reads changed source is updated, not weakened. - [x] Docs updated. - [x] AI assistance recorded with `Assisted-by:` / `Co-Authored-By:` trailers. ### Ready to merge - [ ] Everything in Still to test has run, its results in Tested. - [ ] Approved in review, every comment answered. - [ ] CI passes on the head; on current `main`, measured again or a comment says why the numbers still hold.
A player's build ends with make_game.py's clean-up, which lets the build folder go: `sa/build`, and with it every depfile
clang wrote. The next update names each back-end and host object by `renderer_deps`, which, with no depfile, falls back to the
whole folders' hash (`vdeps`, `host_blob`). That is not the key the object store holds: `rekey_depfiles` cached each object
under its depfile's key after the compile. So every update compiled the host's and the back end's own objects again, then
re-keyed them to a key the store already had, and the store could not stop it.

The size of it is measured in #1602, which counted the same update: 105 objects (65 host `.c`, 40 renderer `.cpp`), 239
compiler-seconds, 37 s of wall time on 16 threads, on any update that changes a header under `host/` or `gpu/` - which every
nightly does. A build with `LONGSHOT_KEEP_BUILD_FILES=1` (every development build) keeps the depfiles and never sees it.

The headers a compile read now stay beside the object (`obj.deps`, `DEPS_STAMP`): the root's place and the path below it, so
the digest is the same from another builder folder (the AppImage's mount is new at every start, a launcher's tools folder at
every update). `renderer_deps` reads those headers as they are now, so an edited or deleted one still compiles the file, and
`keep_translation` keeps the lists through the clean-up.

- nx2/build.py: `header_entries`, `deps_digest`, `written_deps`; `renderer_deps` falls back to the list; `rekey_depfiles`
  writes it (and takes it away when a depfile names nothing under the roots).
- nx2/make_game.py: `keep_translation` keeps `sa/build/obj/*.deps` and nothing else of the build folder, as before.
- tests/test_object_reuse_cleanup.py: an object's stored key is the one a later plan asks for, after the clean-up and from
  another builder folder; an edited or deleted header changes it; an object taken from the store keeps its list; a work folder
  with neither keeps the blob.
- docs: what an update keeps, and the route in "What builds again".

Fixes the cause behind #1906/#1907/#2108's "object store: 0 objects from it" for the host and the back end (the game's chunks
are a separate key, moved by the lift and the splitter).

Assisted-by: opencode (DeepSeek V4.1 Flash)
`fl`, the game's flags, was rebound by the object store's loop in `main()` (`for _, src, obj, fl, deps in jobs:`: Python's loop
variables outlive the loop). Every player's build has `LONGSHOT_OBJ_STORE` set, so that loop always ran, and the line after it
reported the last job's flags instead: on Linux the posix object's `-O2 -D_GNU_SOURCE`, and in #1906's log a bare `-O2` where
the game's lifted chunks are at `-O1`. The `object keys:` fingerprints above it are printed before the loop and were never
affected; this is the line players paste when an update compiles everything.

The loop keys each object by the job's own flags, as before: only the name it left behind changed.

tests/test_object_reuse_cleanup.py checks that the store's loop does not rebind `fl` (it fails on main with the old loop).

Assisted-by: opencode (DeepSeek V4.1 Flash)
test(build): the two checks that cut build.py's Vulkan jobs know the depfile list's helpers
Some checks failed
Website / Website on Cloudflare (pull_request_target) Has been skipped
PR description / Description follows the template (pull_request_target) Successful in 4s
CI / Conventional Commit title (pull_request) Has been cancelled
CI / Python source checks (ubuntu-24.04, part 1/2) (pull_request) Has been cancelled
CI / Python source checks (ubuntu-24.04, part 2/2) (pull_request) Has been cancelled
CI / SPIR-V fixtures (global narrow load/store, direct surface atomics, F2I immediate) (pull_request) Has been cancelled
CI / Workflow lint (actionlint) (pull_request) Has been cancelled
CI / Website build and checks (pull_request) Has been cancelled
Guide screenshots / Guide images and local UI source review (pull_request) Has been cancelled
Third-party parts / As their notes say (pull_request) Has been cancelled
Third-party parts / The same as upstream (pull_request) Has been cancelled
Website / Website on Cloudflare (pull_request) Has been cancelled
613b775a1a
`renderer_deps` now asks `header_entries`, `written_deps` and the stamp beside the object, so the two checks that cut a
part of `nx2/build.py` out and exec it (`tests/test_shader_stamp_cache.py`, `tests/test_dlss_sdk_cache.py`) must cut those
too: without them the cut branch stops at `NameError: name 'header_entries' is not defined`. Their own assertions are
unchanged; only the list of what they take from the source grew, as it did for `renderer_deps` and its helpers.

Found by the full suite on the branch (910 passed, 25 failed: the 23 of main's set at a2dc5385 plus these two).

Assisted-by: opencode (DeepSeek V4.1 Flash)
Author
Contributor

Tested in game: N/A — build tooling only; no game code path changes. Evidence: the suite and the object-store measurement in the body.

Tested in game: N/A — build tooling only; no game code path changes. Evidence: the suite and the object-store measurement in the body.
Author
Contributor

Body: the suite's numbers on the head (912 passed, 23 failed, 18 skipped, the baseline's 23 names exactly, no new name), and the measurement's commit named (868c7df7; the head's third commit is two checks only). No code change since the suite.

Body: the suite's numbers on the head (912 passed, 23 failed, 18 skipped, the baseline's 23 names exactly, no new name), and the measurement's commit named (868c7df7; the head's third commit is two checks only). No code change since the suite.
Author
Contributor

Body only, no code change: the Measurements table's Before column relabelled. It is not a main plan — it is the same work folder (from the shared base a053f9e1) planned by this PR's build.py at 868c7df7 with the *.deps lists absent, i.e. a player's post-clean-up state, which main plans to the same keys (no depfile, no list, the folders' hash — the 0 line #1906 reports). Added the Baseline line the template asks for when the before side is not main; a main plan itself was not run. Suite box untouched (full suite on the head: 912 passed, 23 failed, 18 skipped, the baseline's 23 names exactly).

Body only, no code change: the Measurements table's Before column relabelled. It is not a main plan — it is the same work folder (from the shared base a053f9e1) planned by this PR's build.py at 868c7df7 with the *.deps lists absent, i.e. a player's post-clean-up state, which main plans to the same keys (no depfile, no list, the folders' hash — the 0 line #1906 reports). Added the Baseline line the template asks for when the before side is not main; a main plan itself was not run. Suite box untouched (full suite on the head: 912 passed, 23 failed, 18 skipped, the baseline's 23 names exactly).
dazemc force-pushed fix/object-reuse-cleanup from 613b775a1a
Some checks failed
Website / Website on Cloudflare (pull_request_target) Has been skipped
PR description / Description follows the template (pull_request_target) Successful in 4s
CI / Conventional Commit title (pull_request) Has been cancelled
CI / Python source checks (ubuntu-24.04, part 1/2) (pull_request) Has been cancelled
CI / Python source checks (ubuntu-24.04, part 2/2) (pull_request) Has been cancelled
CI / SPIR-V fixtures (global narrow load/store, direct surface atomics, F2I immediate) (pull_request) Has been cancelled
CI / Workflow lint (actionlint) (pull_request) Has been cancelled
CI / Website build and checks (pull_request) Has been cancelled
Guide screenshots / Guide images and local UI source review (pull_request) Has been cancelled
Third-party parts / As their notes say (pull_request) Has been cancelled
Third-party parts / The same as upstream (pull_request) Has been cancelled
Website / Website on Cloudflare (pull_request) Has been cancelled
to 2b65ab3667
Some checks are pending
CI / Conventional Commit title (pull_request) Blocked by required conditions
CI / Python source checks (ubuntu-24.04, part 1/2) (pull_request) Blocked by required conditions
CI / Python source checks (ubuntu-24.04, part 2/2) (pull_request) Blocked by required conditions
CI / SPIR-V fixtures (global narrow load/store, direct surface atomics, F2I immediate) (pull_request) Blocked by required conditions
CI / Workflow lint (actionlint) (pull_request) Blocked by required conditions
CI / Website build and checks (pull_request) Blocked by required conditions
Guide screenshots / Guide images and local UI source review (pull_request) Blocked by required conditions
Third-party parts / As their notes say (pull_request) Blocked by required conditions
Third-party parts / The same as upstream (pull_request) Blocked by required conditions
Website / Website on Cloudflare (pull_request) Blocked by required conditions
PR description / Description follows the template (pull_request_target) Successful in 6s
2026-11-22 05:46:09 +00:00
Compare
Author
Contributor

Split, as review asked. The plan's flags line (the object-store loop rebinding fl) is now #2217; this PR is the object-key fix only. The branch was rewritten: head 2b65ab36 = the pre-split head 613b775a minus that commit only — git diff 613b775a 2b65ab36 is two hunks, nx2/build.py (a loop variable's name) and tests/test_object_reuse_cleanup.py (-9, the moved check). The suite evidence (912 passed, 23 failed, 18 skipped, the baseline's 23 names exactly) was taken on 613b775a and is NOT claimed for 2b65ab36; the diff above is why the code it exercised is the same. Description updated to the new head.

Split, as review asked. The plan's flags line (the object-store loop rebinding fl) is now #2217; this PR is the object-key fix only. The branch was rewritten: head 2b65ab36 = the pre-split head 613b775a minus that commit only — git diff 613b775a 2b65ab36 is two hunks, nx2/build.py (a loop variable's name) and tests/test_object_reuse_cleanup.py (-9, the moved check). The suite evidence (912 passed, 23 failed, 18 skipped, the baseline's 23 names exactly) was taken on 613b775a and is NOT claimed for 2b65ab36; the diff above is why the code it exercised is the same. Description updated to the new head.
dazemc changed title from fix(build): an update keeps the headers its host and renderer objects were keyed by to WIP: fix(build): an update keeps the headers its host and renderer objects were keyed by 2026-11-22 05:50:29 +00:00
Author
Contributor

WIP again until its own head has a full suite, as #2176/#2200 were. Untick: the suite ran on 613b775a; the head 2b65ab36 differs from it by the split-out commit only (two hunks), and the re-run is queued for after the compile arms (--jobs 5, 12 GiB, pinned headers). Expected: 23, the baseline's names. Body keeps the disclosure, the tick is gone.

WIP again until its own head has a full suite, as #2176/#2200 were. Untick: the suite ran on 613b775a; the head 2b65ab36 differs from it by the split-out commit only (two hunks), and the re-run is queued for after the compile arms (--jobs 5, 12 GiB, pinned headers). Expected: 23, the baseline's names. Body keeps the disclosure, the tick is gone.
Author
Contributor

Suite on this head now: 912 passed, 23 failed, 18 skipped in 442 s (--jobs 5, 12 GiB, pinned headers, worktree at 2b65ab36). The 23 names are exactly the a2dc5385 baseline set, no new name. Box ticked with those numbers.

Suite on this head now: 912 passed, 23 failed, 18 skipped in 442 s (--jobs 5, 12 GiB, pinned headers, worktree at 2b65ab36). The 23 names are exactly the a2dc5385 baseline set, no new name. Box ticked with those numbers.
dazemc changed title from WIP: fix(build): an update keeps the headers its host and renderer objects were keyed by to fix(build): an update keeps the headers its host and renderer objects were keyed by 2026-11-22 08:13:23 +00:00
Some checks are pending
CI / Conventional Commit title (pull_request) Blocked by required conditions
CI / Python source checks (ubuntu-24.04, part 1/2) (pull_request) Blocked by required conditions
CI / Python source checks (ubuntu-24.04, part 2/2) (pull_request) Blocked by required conditions
CI / SPIR-V fixtures (global narrow load/store, direct surface atomics, F2I immediate) (pull_request) Blocked by required conditions
CI / Workflow lint (actionlint) (pull_request) Blocked by required conditions
CI / Website build and checks (pull_request) Blocked by required conditions
Guide screenshots / Guide images and local UI source review (pull_request) Blocked by required conditions
Third-party parts / As their notes say (pull_request) Blocked by required conditions
Third-party parts / The same as upstream (pull_request) Blocked by required conditions
Website / Website on Cloudflare (pull_request) Blocked by required conditions
PR description / Description follows the template (pull_request_target) Successful in 6s
This pull request doesn't have enough approvals yet. 0 of 1 approvals granted.
Some workflows are waiting to be reviewed.
You are not authorized to merge this pull request.
View command line instructions

Checkout

From your project repository, check out a new branch and test the changes.
git fetch -u 'fix/object-reuse-cleanup:dazemc-fix/object-reuse-cleanup'
git switch 'dazemc-fix/object-reuse-cleanup'
Sign in to join this conversation.
No reviewers
No milestone
No project
No assignees
1 participant
Notifications
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!2199
No description provided.