WIP: perf(host): notify message queue waiters on readiness transitions #1598
No reviewers
Labels
No labels
area/accessibility
area/audio
area/input
area/install
area/menu
area/renderer
area/saves
area/tooling
area/upscaling
beanstalk: approved
beanstalk: failed
beanstalk: needs approval
beanstalk: running
beanstalk: tested
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
risk
high
risk
low
risk
medium
status
blocked
status
confirmed
status
duplicate
status
invalid
status
needs-info
status
wontfix
ci: approved
No milestone
No project
No assignees
3 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set
Reference
MagicBeanDev/Longshot!1598
Loading…
Reference in a new issue
No description provided.
Delete branch "HumanBean/Longshot:perf/queue-readiness-transitions"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Type: perf · Risk: medium · Tested on: CachyOS Linux, Ryzen 9 7900X, 32 GiB, GeForce RTX 3090, NVIDIA 620.70.06
TL;DR
Wake message queue waiters when a queue becomes readable or writable. A fill-and-drain burst now needs two readiness broadcasts in the synthetic test.
Why and what changed
Condition-variable and multiwait notifications follow the empty-to-readable and full-to-writable transitions. The regression covers competing consumers, blocking operations and transitions that satisfy a previously false predicate.
Measurements
The worker-only integration arm completed the headless route at 61.58 FPS in Market and 75.12 FPS in Field in its first full-profile run. Other runs varied with system load. Those rates describe that run and do not establish a general FPS improvement.
Testing
The relevant host path ran on Linux through the Market and Field route with player movement verified. The synthetic regression checks wait and wake behavior. A long-running multiwait stress pass and Windows coverage remain outstanding.
Risks and remaining work
Incorrect notification conditions could strand a waiter under contention. Keep this branch separate from the GPU buffer changes so it can be stress-tested independently.
Checklist
876dc360. Measurements were taken before the cold-install packaging change on that revision.Tested in game: Linux, Ryzen 9 7900X, RTX 3090, NVIDIA 620.70.06. The worker-only implementation completed Market and Field movement in the headless game. The marked starting locations and movement were checked.
Possibly related, for context:
6c745d779c299dac6538Corrected author and committer attribution to HumanBean. Verified identical file contents before and after the metadata change.
@Beanstalk
Beanstalk test farm
Commit
299dac6· tier reviewed · approved by HumanBean · dashboardWhat will be tested · AI plan (
gpt-6.1-terra)The plan compares movement and repeated scene loads in Market and Hyrule Field against main, taking about 12 minutes per tester. It measures frame times and load times and compares scene images; crashes, stalled loads, large performance regressions, or visual differences fail. This gameplay check cannot establish that all queue and multiwait contention cases are safe.
kokiri-start, then teleport to Market, move the stick, take the picture market to compare with main, measure market for 20 s (after 5 s warm-up), teleport to Hyrule Field, take the picture field to compare with main, measure field for 20 s (after 5 s warm-up), 3 times: teleport to Market, teleport to Hyrule Field; 1 run on each side.Done (started 06:18 UTC, last report 06:46 UTC). Comment
@Beanstalkto run it again.NVIDIA GeForce RTX 3050 Ti Laptop GPU, 621.24.0; 2: linux+amd+radv+dgpuAMD Radeon RX 7900 XTX (RADV NAVI31), 26.2.7Measurements · any
Verdict: could not be judged on any (2/1 report).
Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu
NVIDIA GeForce RTX 3050 Ti Laptop GPU, 621.24.0AMD Ryzen 7 5800H with Radeon Graphics, 16 GB RAM; 2: linux+amd+radv+dgpuAMD Radeon RX 7900 XTX (RADV NAVI31), 26.2.7AMD Ryzen 7 5800X3D 8-Core Processor, 64 GB RAMScene: market-field, from the repository's save kokiri-start; each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); market-field: 1 game start a side, market 20 s after 5 s warm-up, field 20 s after 5 s warm-up; each number the median of the runs, then of 2 reports
How to reproduce: the farm's plan above (
@Beanstalkon this PR runs it again)Nothing was measured: no metric came back.
Flags and settings
None: the plan varied no setting, switch or build option.
Markdown for the PR description
View command line instructions
Manual merge helper
Use this merge commit message when completing the merge manually.
Checkout
From your project repository, check out a new branch and test the changes.