test(pacing): characterize current cap and present policy #1591
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
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set
Reference
MagicBeanDev/Longshot!1591
Loading…
Reference in a new issue
No description provided.
Delete branch "futondestroyer/Longshot:test/issue-1519-pacing-policy"
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?
Summary
The host, renderer and FPS overlay make pacing decisions from the same settings. The game-free fixture in
tests/test_pacing_policy.pyrecords the current choices so later policy changes can identify which behavior moved. It covers movie caps, V-Sync and present modes, and fixed frame-generation image and slot counts. The refresh rebases the identical fixture onto main89c3d21e, including the presenter rewrite in #1763. See #1519.Testing
Tested: Windows with llvm-mingw at
e62b8e90. All 13 focused checks passed, including the pacing matrix, presenter, swapchain semaphores and CPU acquire fence. The matrix checks 13 host and overlay rows, 192 present-mode combinations plus the empty-list fallback, and 16 fixed frame-generation allocation cases. The Windows strict suite completed in batches with 693 passes, 140 failures reproduced on exact main, and 48 declared platform skips. The Windows queue self-test additionally failed its RAM-filesystem case on both revisions and interrupted both suite invocations. Four intermittent candidate outcomes passed on targeted reruns. Pinned Ruff, actionlint and the Windows Gitleaks release passed. The Guide screenshots checker reports the same 33 review items on main and this head; see #1501.Failures reproduced on main
nx2/test_mesh_first_types.pynx2/test_nup_codec.pynx2/test_nup_read_name.pynx2/test_nup_stand_in.pynx2/test_nup_verify.pynx2/test_texture_init_barrier.pytests/test_antilag.pytests/test_appimage_player_build.pytests/test_async_safe_frees.pytests/test_auto_handheld.pytests/test_bug_report_defaults.pytests/test_build_long_paths.pytests/test_capture_dimensions.pytests/test_capture_read_once.pytests/test_capture_readback_visibility.pytests/test_capture_resources.pytests/test_coopmat_probe_failures.pytests/test_crash_dialog.pytests/test_crt_options.pytests/test_damage_debug_menu.pytests/test_desc_counts.pytests/test_dev_ab.pytests/test_dev_build.pytests/test_dev_migrate.pytests/test_dev_packs.pytests/test_device_policy.pytests/test_discover.pytests/test_display_startup_failure.pytests/test_dlss_feature_size.pytests/test_draw_binds.pytests/test_draw_stream_visibility.pytests/test_dummy_views.pytests/test_dump_readback.pytests/test_early_split_before_draw.pytests/test_failed_compile_keeps_objects.pytests/test_fallback_pipeline_state.pytests/test_fastfp_fma.pytests/test_fma_soft.pytests/test_fpx_sticky.pytests/test_frame_overlap.pytests/test_framegen_resources.pytests/test_framelog_sink.pytests/test_fsr3_release_frees.pytests/test_fsr3fi_hud_layer.pytests/test_fsr3fi_hud_mask.pytests/test_fsr3fi_hudless_colour.pytests/test_fsr4_bind_dispatch.pytests/test_fsr_image_retirement.pytests/test_fsr_uniform_ownership.pytests/test_game_menu_mouse_buttons.pytests/test_guest_imported.pytests/test_high_fps_fixes.pytests/test_hold_gpu.pytests/test_host_guest_writes.pytests/test_hud_layout_row.pytests/test_image_view_retry.pytests/test_import_buffer_binding.pytests/test_import_reach.pytests/test_inline_write_order.pytests/test_insn_fixed_convert.pytests/test_job_queue.pytests/test_libc_hooks.pytests/test_lift_cache.pytests/test_lift_lookups.pytests/test_linux_preflight.pytests/test_linux_pressure.pytests/test_longshot_ctl.pytests/test_lost_wait_said.pytests/test_low_vram_texture_mips.pytests/test_make_game_cache.pytests/test_make_game_ngx.pytests/test_mise_build.pytests/test_missing_nup.pytests/test_mods_list.pytests/test_native_zstd.pytests/test_ngx_sdk.pytests/test_pads_after_menu.pytests/test_pass_barrier.pytests/test_pass_hazard.pytests/test_pcache_header_read.pytests/test_pending_guest_import.pytests/test_pipeline_cache_save.pytests/test_pipeline_key_values.pytests/test_pipeline_play_cache.pytests/test_pipeline_state_portable.pytests/test_pipeline_states.pytests/test_prepare_cache_save.pytests/test_present_queue_default.pytests/test_present_source_crop.pytests/test_profiling_mode.pytests/test_program_answer_domains.pytests/test_prompt_guide_lines.pytests/test_prompt_hand_guides.pytests/test_prompt_picture_tint.pytests/test_prompt_speed.pytests/test_psychov35_fsr3_fg.pytests/test_push_state.pytests/test_readback_early.pytests/test_readback_late.pytests/test_readback_timeout_source.pytests/test_recorder_own_idle.pytests/test_recorder_sync.pytests/test_recording_admission.pytests/test_replay_write_order.pytests/test_same_draw.pytests/test_same_dyn.pytests/test_sampler_admission.pytests/test_sdiv_overflow.pytests/test_setup_tar_diagnostics.pytests/test_slow_frames.pytests/test_song_note_cue.pytests/test_sparse_bind_failure.pytests/test_sparse_bind_ranges.pytests/test_spawn_emit.pytests/test_split_before_draw.pytests/test_split_ub_flush.pytests/test_staging_burst.pytests/test_taau_typed_dispatch.pytests/test_table_track.pytests/test_texel_buffer_ownership.pytests/test_texture_staging_publication.pytests/test_ub_snapshot_default.pytests/test_ub_update_batch.pytests/test_ui_record.pytests/test_ui_texture_upload.pytests/test_upscaler_rows.pytests/test_vbytes.pytests/test_vcore_escape.pytests/test_vecfp.pytests/test_vendor_fixes.pytests/test_vertex_bind.pytests/test_vertex_bind_range.pytests/test_vfpx_elem.pytests/test_vk_occlusion_queries.pytests/test_vulkan_clear_mip.pytests/test_vulkan_copy_span.pytests/test_vulkan_copy_span_gpu_compile.pytests/test_walk.pytests/test_walk_key.pytests/test_wb_diff_observations_gpu.pytests/test_widen_alpha.pyStill to test: None for the fixture. Actual display cadence and VRR engagement are outside the fixture's scope.
Measurements
The change adds a game-free fixture. Performance measurement does not apply.
Risks
Source moves can require updates to the fixture's extraction points. Reverting the test commit removes the fixture.
Checklist
Ready for review
main; the title is a Conventional Commit.mise run testpasses (any failure that also fails onmainis named).main, or Measurements says why there is nothing to measure.third_party/untouched; nothing GPL-incompatible.When it applies
Assisted-by:/Co-Authored-By:trailers.Ready to merge
main, measured again or a comment says why the numbers still hold.WIP: test(pacing): characterize current cap and present policyto test(pacing): characterize current cap and present policy14dd5f4de548f608ca5cRebased onto main
876dc360. The new head is48f608caand the fixture is byte-identical to the reviewed14dd5f4d. All 12 focused Windows checks passed. The full game-free suite recorded 616 passes, 44 failures reproduced on main, and 121 skips. The Guide screenshots failure also occurs on main and is tracked in #1501. Please confirm the new head against the existing review.48f608ca5ce62b8e9092Rebased the unchanged fixture onto main
89c3d21e. The new head ise62b8e90. All 13 focused Windows checks passed, including the current presenter and CPU acquire-fence checks. The description names the strict-suite baseline failures and the Windows queue self-test interruption reproduced on both revisions. The Graphics guide review remains tracked in #1501. Please confirm the new revision pair in review.