docs(worklist): row 5.6 measured on main on Windows: the commit settles at ~20 GB and does not grow over a second lap #995

Open
d4nkn3ss wants to merge 2 commits from d4nkn3ss/Longshot:docs/worklist-memory-measured into main
Contributor

Type: docs · Risk: low · Tested on: Windows 11, RTX 3080 · Needs: none

Why

Worklist row 5.6 (memory fills up over play time, the top row) still says "Next: retest on the next release". There was no measurement of main on Windows, where most of its reporters play. Two runs today give one: on a 10 GB card with 64 GB of RAM the game's commit settles at ~20 GB and doesn't grow over a second lap. The row should say so, say what the 20 GB is made of, and name the runs it still needs: a 16 GB PC and a 6-8 GB card, which this PC can't stand in for. See #388, #939, #992.

What changed

Row 5.6 gets the measurement and its next step. The two runs are files in docs/tested/.

  • docs: docs/worklist.md row 5.6: "Next: retest on the next release ..." now reads "Measured on main a8a30c22 ...". It gives the split (guest window 7.6 GB, VRAM blocks charged to the commit 7.4-8.1 GB, the rest 4.3 GB), the Nov 14 build's growth, the open issues and PRs on it (#388, #939, #992), and a next step: a memtour run from a 16 GB Windows PC and from a 6-8 GB card, plus the reporters' logs. Score, owner and links are unchanged.
  • docs: row 5.6 also names the crashes at the limit: six reports end with an uncaught C++ exception (0x20474343) after reaching Windows' commit limit (#1029, #1030, #991, #981, #964, #894), shown as "An unexpected error" until #1056.
  • docs: docs/tested/2026-11-15-memtour-windows-d4nkn3ss-main.md and ...-dc390e59.md: the two runs, one row each.

Measurements

Verdict: nothing to measure for this change, which is docs only. The runs it records are the measurement: main @ a8a30c22's commit is 20,355 MB after lap 1 and 20,035 MB after lap 2, while dc390e59's "rest" grew by about 440 MB over lap 2. Method and log lines are on #992 (comment).

Flags and settings

None

Testing

Tested:

  • python -I -B tests/test_tested_configurations.py: the two new files pass its format checks. Its one failure, test_cli_ascii_pipe_emits_the_arrow_containing_source_as_utf8 (CRLF against LF on Windows), fails on main too.
  • python -I -B tests/run_all.py from PowerShell on Windows 11, with llvm-mingw from the Longshot Builder: 596 passed, 18 failed, 42 skipped on this PR @ 936bfc24 (rebased on main @ bbe421f5 since, with the crashes' sentence added: a0b90618; before that on cefce0d1: 9351bdb1, docs only). All 18 fail on main @ a0be15b3 too: test_auto_handheld, test_crash_dialog, test_crt_options, test_fsr_uniform_ownership, test_guest_imported, test_import_reach, test_linux_preflight, test_make_game_cache, test_missing_nup, test_pads_after_menu, test_parallel_init_order, test_pending_guest_import, test_release, test_replay_write_order, test_run_all_changed, test_tested_configurations, test_vk_occlusion_queries and test_walk_key. Three are fixed by #993, test_release by #144 and test_run_all_changed by #954.

Not tested: the website's catalogue build. The files follow docs/tested/README.md and its check.

Must be tested: None. The change is to docs only.

Risks

What could break: nothing: two new files and one table cell.

Turning it off: revert the commit.

Compatibility: Nothing changes.

Issues

See #388, #939

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.
  • Rebased on current main.
  • python -I -B tests/run_all.py passes (any failure that also fails on main is named).
  • No game files, dumps, build output or generated files.
  • third_party/ untouched (a new folder with a PROVENANCE.md instead).
  • The license is OK: GPL-3.0-or-later, nothing incompatible and nothing decompiled.

When it applies

  • AI assistance recorded with Assisted-by: / Co-Authored-By: trailers.
  • A file in docs/tested/ for each new measured run (format in docs/tested-configurations.md).
  • Docs updated.

🤖 Generated with Claude Code

**Type:** docs · **Risk:** low · **Tested on:** Windows 11, RTX 3080 · **Needs:** none ## Why Worklist row 5.6 (memory fills up over play time, the top row) still says "Next: retest on the next release". There was no measurement of main on Windows, where most of its reporters play. Two runs today give one: on a 10 GB card with 64 GB of RAM the game's commit settles at ~20 GB and doesn't grow over a second lap. The row should say so, say what the 20 GB is made of, and name the runs it still needs: a 16 GB PC and a 6-8 GB card, which this PC can't stand in for. See #388, #939, #992. ## What changed Row 5.6 gets the measurement and its next step. The two runs are files in `docs/tested/`. - docs: `docs/worklist.md` row 5.6: "Next: retest on the next release ..." now reads "Measured on main a8a30c22 ...". It gives the split (guest window 7.6 GB, VRAM blocks charged to the commit 7.4-8.1 GB, the rest 4.3 GB), the Nov 14 build's growth, the open issues and PRs on it (#388, #939, #992), and a next step: a memtour run from a 16 GB Windows PC and from a 6-8 GB card, plus the reporters' logs. Score, owner and links are unchanged. - docs: row 5.6 also names the crashes at the limit: six reports end with an uncaught C++ exception (0x20474343) after reaching Windows' commit limit (#1029, #1030, #991, #981, #964, #894), shown as "An unexpected error" until #1056. - docs: `docs/tested/2026-11-15-memtour-windows-d4nkn3ss-main.md` and `...-dc390e59.md`: the two runs, one row each. ## Measurements **Verdict:** nothing to measure for this change, which is docs only. The runs it records are the measurement: main @ a8a30c22's commit is 20,355 MB after lap 1 and 20,035 MB after lap 2, while dc390e59's "rest" grew by about 440 MB over lap 2. Method and log lines are on https://git.magicbean.dev/MagicBeanDev/Longshot/pulls/992#issuecomment-31580. ### Flags and settings None ## Testing **Tested:** - `python -I -B tests/test_tested_configurations.py`: the two new files pass its format checks. Its one failure, `test_cli_ascii_pipe_emits_the_arrow_containing_source_as_utf8` (CRLF against LF on Windows), fails on main too. - `python -I -B tests/run_all.py` from PowerShell on Windows 11, with llvm-mingw from the Longshot Builder: 596 passed, 18 failed, 42 skipped on this PR @ 936bfc24 (rebased on main @ bbe421f5 since, with the crashes' sentence added: a0b90618; before that on cefce0d1: 9351bdb1, docs only). All 18 fail on main @ a0be15b3 too: `test_auto_handheld`, `test_crash_dialog`, `test_crt_options`, `test_fsr_uniform_ownership`, `test_guest_imported`, `test_import_reach`, `test_linux_preflight`, `test_make_game_cache`, `test_missing_nup`, `test_pads_after_menu`, `test_parallel_init_order`, `test_pending_guest_import`, `test_release`, `test_replay_write_order`, `test_run_all_changed`, `test_tested_configurations`, `test_vk_occlusion_queries` and `test_walk_key`. Three are fixed by #993, `test_release` by #144 and `test_run_all_changed` by #954. **Not tested:** the website's catalogue build. The files follow `docs/tested/README.md` and its check. **Must be tested:** None. The change is to docs only. ## Risks **What could break:** nothing: two new files and one table cell. **Turning it off:** revert the commit. **Compatibility:** Nothing changes. ## Issues See #388, #939 ## Checklist ### Everyone - [x] One topic. - [x] The title is a Conventional Commit (`type(scope): description`) of the status line's type. - [x] Why gives the reason plainly and links the issues that fit (the tracker searched). - [x] Measurements has a verdict, or one line saying why there is nothing to measure. - [x] Testing names what was tested, what wasn't, and what must still be tested, where and how. - [x] Rebased on current `main`. - [x] `python -I -B tests/run_all.py` passes (any failure that also fails on `main` is named). - [x] No game files, dumps, build output or generated files. - [x] `third_party/` untouched (a new folder with a `PROVENANCE.md` instead). - [x] The license is OK: GPL-3.0-or-later, nothing incompatible and nothing decompiled. ### When it applies - [x] AI assistance recorded with `Assisted-by:` / `Co-Authored-By:` trailers. - [x] A file in `docs/tested/` for each new measured run (format in `docs/tested-configurations.md`). - [x] Docs updated. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
docs(worklist): row 5.6 measured on main on Windows: the commit settles at ~20 GB and does not grow over a second lap
Some checks are pending
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 / Conventional Commit title (pull_request) Blocked by required conditions
CI / Workflow lint (actionlint) (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 3s
Website / Website on Cloudflare (pull_request_target) Has been skipped
9351bdb116
Two runs of a two-lap memory tour on an RTX 3080 10 GB / 64 GB PC, main
a8a30c22 and the Nov 14 main dc390e59, as docs/tested files, and the row's
next step: a run from a 16 GB Windows PC and a 6-8 GB card.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
d4nkn3ss changed title from WIP: docs(worklist): row 5.6 measured on main on Windows: the commit settles at ~20 GB and does not grow over a second lap to docs(worklist): row 5.6 measured on main on Windows: the commit settles at ~20 GB and does not grow over a second lap 2026-11-16 01:30:25 +00:00
d4nkn3ss force-pushed docs/worklist-memory-measured from 9351bdb116
Some checks are pending
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 / Conventional Commit title (pull_request) Blocked by required conditions
CI / Workflow lint (actionlint) (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 3s
Website / Website on Cloudflare (pull_request_target) Has been skipped
to a0b90618f2
All checks were successful
PR description / Description follows the template (pull_request_target) Successful in 3s
2026-11-16 05:29:32 +00:00
Compare
Author
Contributor

Rebased on main @ bbe421f5 (now a0b90618), and row 5.6 now also names the crashes at the commit limit: six reports with an uncaught C++ exception (0x20474343) after their logs reached Windows' commit limit, which #1056 makes the crash window call out of memory. test_doc_anchors and test_tested_configurations' format checks pass on the two run files as before. The latter's Windows CRLF case still fails, as on main (#493).

Posted by an AI agent (Claude Code) for d4nkn3ss.

Rebased on main @ bbe421f5 (now a0b90618), and row 5.6 now also names the crashes at the commit limit: six reports with an uncaught C++ exception (0x20474343) after their logs reached Windows' commit limit, which #1056 makes the crash window call out of memory. `test_doc_anchors` and `test_tested_configurations`' format checks pass on the two run files as before. The latter's Windows CRLF case still fails, as on main (#493). *Posted by an AI agent (Claude Code) for d4nkn3ss.*
All checks were successful
PR description / Description follows the template (pull_request_target) Successful in 3s
This pull request doesn't have enough approvals yet. 0 of 1 approvals granted.
Some workflows are waiting to be reviewed.
This branch is out-of-date with the base branch
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 'docs/worklist-memory-measured:d4nkn3ss-docs/worklist-memory-measured'
git switch 'd4nkn3ss-docs/worklist-memory-measured'
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!995
No description provided.