Beanstalk Longshot test farm
Menu

Report rp_83d7bca060d28266 fail

#1902 WIP: perf(host): the box-against-view test and the occlusion grid's build on host code, bit for bit (HOS_NATIVE_VISIBILITY, on by default)

linux on Floppy/agent-06957c,

This report disagrees with its peers.

visibility-route field-standing.frametime_max_ms: -28.0% against peers' +21.6%
visibility-route market-standing.frametime_max_ms: +45.8% against peers' +5.6%
visibility-route market-uncapped.frametime_max_ms: +28.8% against peers' +310.2%

Machine

Commit
5f0c55d21895b5773d2a61c948342bb7e171e27c
Baseline
a104182b4037960ba99a46e2c97f766cd604e036
App
nightly-217-14cc1aa3, attempt 6
Game build
e0f3ccfa
os
linux
cpu
AMD Ryzen 7 7800X3D 8-Core Processor
gpu
NVIDIA GeForce RTX 5080
tags
linuxnvidianvidia-proprietarydgpu
driver
620.73.4
ram_gb
32
vram_mb
16,303
gpu_vendor
nvidia
os_version
CachyOS (kernel 7.3.5)

Builds

main (baseline)

925.8 s warm

from a104182b4037

lift
76.5 s
read
2.6 s
finish
11.2 s
compile
778.7 s
shaders
51.5 s

PR

271.7 s warm

lift
75.3 s
read
0.4 s
finish
13.6 s
compile
182.1 s

Program cache: warm

Timed builds

Main and the PR each built whole in a fresh folder, from the same starting caches, alternating; shader preparation not timed. lift: the lifted C differs

CacheSide#SecondsStages (s)Lifted C
liftmain 1126.5 lift 78.6 · read 2 · finish 10.9 · compile 34.8 d15bb9ed8108
liftPR 1273.9 lift 78.9 · read 1.2 · finish 10.9 · compile 182.8 1c6c6a53dd3a
liftPR 2270.2 lift 75.7 · read 1 · finish 10.9 · compile 182.6 1c6c6a53dd3a
liftmain 2122 lift 75.5 · read 1 · finish 10.7 · compile 34.8 d15bb9ed8108

Scenarios

visibility-route HOS_NATIVE_VISIBILITY=0 (PR only)

ok
Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count4643−6.5%
value_activation00±0%
field-turned.cpu_s26.4528.7+8.5%
market-turned.cpu_s34.835.36+1.6%
field-standing.cpu_s47.3847.26−0.3%
field-uncapped.cpu_s110.65114.76+3.7%
run_frametime_max_ms26,382.5126,269.87−0.4%
market-standing.cpu_s122.85112.25−8.6%
market-uncapped.cpu_s140.64141.3+0.5%
field-turned.hitch_count00±0%
field-turned.rss_peak_mb12,529.312,438.3−0.7%
market-turned.hitch_count00±0%
market-turned.rss_peak_mb11,915.612,049+1.1%
field-standing.hitch_count00±0%
field-standing.rss_peak_mb12,496.712,200−2.4%
field-turned.stutter_count00±0%
field-uncapped.hitch_count00±0%
field-uncapped.rss_peak_mb12,321.212,249.1−0.6%
market-standing.hitch_count01—
market-standing.rss_peak_mb11,970.812,072+0.8%
market-turned.stutter_count00±0%
market-uncapped.hitch_count00±0%
market-uncapped.rss_peak_mb12,325.312,432.1+0.9%
field-standing.stutter_count00±0%
field-uncapped.stutter_count10−100%
field-turned.frametime_max_ms24.623.97−2.5%
field-turned.frametime_p50_ms16.6616.68+0.1%
field-turned.frametime_p99_ms18.5318.02−2.7%
market-standing.stutter_count22±0%
market-uncapped.stutter_count03—
market-turned.frametime_max_ms19.1420.29+6%
market-turned.frametime_p50_ms16.6416.64+0%
market-turned.frametime_p99_ms18.4518.52+0.4%
field-standing.frametime_max_ms28.2820.37−28%
field-standing.frametime_p50_ms16.6616.69+0.2%
field-standing.frametime_p99_ms18.4718.43−0.2%
field-uncapped.frametime_max_ms13.59.46−30%
field-uncapped.frametime_p50_ms5.635.84+3.7%
field-uncapped.frametime_p99_ms6.657+5.3%
market-standing.frametime_max_ms40.6959.32+45.8%
market-standing.frametime_p50_ms16.716.66−0.2%
market-standing.frametime_p99_ms18.5218.82+1.6%
market-uncapped.frametime_max_ms14.1418.21+28.8%
market-uncapped.frametime_p50_ms7.87.82+0.3%
market-uncapped.frametime_p99_ms9.729.95+2.3%
field-turned.frametime_1pct_low_ms19.4319.16−1.4%
market-turned.frametime_1pct_low_ms18.8719.01+0.7%
field-standing.frametime_1pct_low_ms20.219.41−4%
field-uncapped.frametime_1pct_low_ms7.177.46+4.1%
market-standing.frametime_1pct_low_ms22.0823.98+8.6%
market-uncapped.frametime_1pct_low_ms10.6811.74+9.9%
screenshot_diff—0.59—

Pictures against main: market-front diff 0.1529 (main against itself ), market-turned diff 0.5873 (main against itself ), field-front diff 0.0249 (main against itself ), field-turned diff 0.1995 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
run_hitch_count4643
value_activation00
field-turned.cpu_s26.4528.7
market-turned.cpu_s34.835.36
field-standing.cpu_s47.3847.26
field-uncapped.cpu_s110.65114.76
run_frametime_max_ms26,382.5126,269.87
market-standing.cpu_s122.85112.25
market-uncapped.cpu_s140.64141.3
field-turned.hitch_count00
field-turned.rss_peak_mb12,529.312,438.3
market-turned.hitch_count00
market-turned.rss_peak_mb11,915.612,049
field-standing.hitch_count00
field-standing.rss_peak_mb12,496.712,200
field-turned.stutter_count00
field-uncapped.hitch_count00
field-uncapped.rss_peak_mb12,321.212,249.1
market-standing.hitch_count01
market-standing.rss_peak_mb11,970.812,072
market-turned.stutter_count00
market-uncapped.hitch_count00
market-uncapped.rss_peak_mb12,325.312,432.1
field-standing.stutter_count00
field-uncapped.stutter_count10
field-turned.frametime_max_ms24.623.97
field-turned.frametime_p50_ms16.6616.68
field-turned.frametime_p99_ms18.5318.02
market-standing.stutter_count22
market-uncapped.stutter_count03
market-turned.frametime_max_ms19.1420.29
market-turned.frametime_p50_ms16.6416.64
market-turned.frametime_p99_ms18.4518.52
field-standing.frametime_max_ms28.2820.37
field-standing.frametime_p50_ms16.6616.69
field-standing.frametime_p99_ms18.4718.43
field-uncapped.frametime_max_ms13.59.46
field-uncapped.frametime_p50_ms5.635.84
field-uncapped.frametime_p99_ms6.657
market-standing.frametime_max_ms40.6959.32
market-standing.frametime_p50_ms16.716.66
market-standing.frametime_p99_ms18.5218.82
market-uncapped.frametime_max_ms14.1418.21
market-uncapped.frametime_p50_ms7.87.82
market-uncapped.frametime_p99_ms9.729.95
field-turned.frametime_1pct_low_ms19.4319.16
market-turned.frametime_1pct_low_ms18.8719.01
field-standing.frametime_1pct_low_ms20.219.41
field-uncapped.frametime_1pct_low_ms7.177.46
market-standing.frametime_1pct_low_ms22.0823.98
market-uncapped.frametime_1pct_low_ms10.6811.74
screenshot_diff—0.59

visibility-route HOS_NATIVE_VISIBILITY=1 (PR only)

ok
Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count4841−14.6%
value_activation01—
field-turned.cpu_s26.526.52+0.1%
market-turned.cpu_s35.7435.11−1.8%
field-standing.cpu_s43.6544.83+2.7%
field-uncapped.cpu_s109109.76+0.7%
run_frametime_max_ms26,033.5826,342.57+1.2%
market-standing.cpu_s98.91120.7+22%
market-uncapped.cpu_s141.29140.68−0.4%
field-turned.hitch_count00±0%
field-turned.rss_peak_mb12,423.512,373.9−0.4%
market-turned.hitch_count00±0%
market-turned.rss_peak_mb12,210.611,946.9−2.2%
field-standing.hitch_count00±0%
field-standing.rss_peak_mb12,376.612,230.5−1.2%
field-turned.stutter_count00±0%
field-uncapped.hitch_count00±0%
field-uncapped.rss_peak_mb12,222.312,216.6−0%
market-standing.hitch_count21−50%
market-standing.rss_peak_mb12,127.911,983.2−1.2%
market-turned.stutter_count10−100%
market-uncapped.hitch_count00±0%
market-uncapped.rss_peak_mb12,39212,307.6−0.7%
field-standing.stutter_count00±0%
field-uncapped.stutter_count00±0%
field-turned.frametime_max_ms20.4318.9−7.5%
field-turned.frametime_p50_ms16.6716.65−0.2%
field-turned.frametime_p99_ms18.1618−0.9%
market-standing.stutter_count31−66.7%
market-uncapped.stutter_count04—
market-turned.frametime_max_ms39.2419.33−50.7%
market-turned.frametime_p50_ms16.6316.62−0%
market-turned.frametime_p99_ms18.5718.52−0.2%
field-standing.frametime_max_ms21.3326.69+25.1%
field-standing.frametime_p50_ms16.6616.65−0%
field-standing.frametime_p99_ms18.1418.12−0.1%
field-uncapped.frametime_max_ms7.949.92+24.9%
field-uncapped.frametime_p50_ms5.815.79−0.4%
field-uncapped.frametime_p99_ms6.846.88+0.7%
market-standing.frametime_max_ms105.0571.66−31.8%
market-standing.frametime_p50_ms16.6816.64−0.2%
market-standing.frametime_p99_ms19.0518.81−1.3%
market-uncapped.frametime_max_ms13.6225.63+88.1%
market-uncapped.frametime_p50_ms7.87.78−0.3%
market-uncapped.frametime_p99_ms9.6710.35+7%
field-turned.frametime_1pct_low_ms19.1418.4−3.9%
market-turned.frametime_1pct_low_ms21.1218.83−10.9%
field-standing.frametime_1pct_low_ms19.2519.37+0.6%
field-uncapped.frametime_1pct_low_ms7.127.14+0.2%
market-standing.frametime_1pct_low_ms30.3622.62−25.5%
market-uncapped.frametime_1pct_low_ms10.4712.51+19.5%
screenshot_diff—0.5—

Pictures against main: market-front diff 0.1221 (main against itself ), market-turned diff 0.4989 (main against itself ), field-front diff 0.0674 (main against itself ), field-turned diff 0.1524 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
run_hitch_count4841
value_activation01
field-turned.cpu_s26.526.52
market-turned.cpu_s35.7435.11
field-standing.cpu_s43.6544.83
field-uncapped.cpu_s109109.76
run_frametime_max_ms26,033.5826,342.57
market-standing.cpu_s98.91120.7
market-uncapped.cpu_s141.29140.68
field-turned.hitch_count00
field-turned.rss_peak_mb12,423.512,373.9
market-turned.hitch_count00
market-turned.rss_peak_mb12,210.611,946.9
field-standing.hitch_count00
field-standing.rss_peak_mb12,376.612,230.5
field-turned.stutter_count00
field-uncapped.hitch_count00
field-uncapped.rss_peak_mb12,222.312,216.6
market-standing.hitch_count21
market-standing.rss_peak_mb12,127.911,983.2
market-turned.stutter_count10
market-uncapped.hitch_count00
market-uncapped.rss_peak_mb12,39212,307.6
field-standing.stutter_count00
field-uncapped.stutter_count00
field-turned.frametime_max_ms20.4318.9
field-turned.frametime_p50_ms16.6716.65
field-turned.frametime_p99_ms18.1618
market-standing.stutter_count31
market-uncapped.stutter_count04
market-turned.frametime_max_ms39.2419.33
market-turned.frametime_p50_ms16.6316.62
market-turned.frametime_p99_ms18.5718.52
field-standing.frametime_max_ms21.3326.69
field-standing.frametime_p50_ms16.6616.65
field-standing.frametime_p99_ms18.1418.12
field-uncapped.frametime_max_ms7.949.92
field-uncapped.frametime_p50_ms5.815.79
field-uncapped.frametime_p99_ms6.846.88
market-standing.frametime_max_ms105.0571.66
market-standing.frametime_p50_ms16.6816.64
market-standing.frametime_p99_ms19.0518.81
market-uncapped.frametime_max_ms13.6225.63
market-uncapped.frametime_p50_ms7.87.78
market-uncapped.frametime_p99_ms9.6710.35
field-turned.frametime_1pct_low_ms19.1418.4
market-turned.frametime_1pct_low_ms21.1218.83
field-standing.frametime_1pct_low_ms19.2519.37
field-uncapped.frametime_1pct_low_ms7.127.14
market-standing.frametime_1pct_low_ms30.3622.62
market-uncapped.frametime_1pct_low_ms10.4712.51
screenshot_diff—0.5

Rules

ScenarioCellRuleLimitmainPRΔ %Status
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)crashFalse00—pass
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)market-standing.cpu_s.max_delta_pct10122.85112.25−8.6%pass
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)market-turned.cpu_s.max_delta_pct1034.835.36+1.6%pass
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)field-standing.cpu_s.max_delta_pct1047.3847.26−0.3%pass
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)field-turned.cpu_s.max_delta_pct1026.4528.7+8.5%pass
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)field-uncapped.cpu_s.max_delta_pct10110.65114.76+3.7%pass
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)market-uncapped.cpu_s.max_delta_pct10140.64141.3+0.5%pass
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)screenshot.max_diff0.02—0.59—fail
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)market-standing.frametime_p99_ms.max_delta_pct1518.5218.82+1.6%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)market-turned.frametime_p99_ms.max_delta_pct1518.4518.52+0.4%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)field-standing.frametime_p99_ms.max_delta_pct1518.4718.43−0.2%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)field-turned.frametime_p99_ms.max_delta_pct1518.5318.02−2.7%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)field-uncapped.frametime_p99_ms.max_delta_pct156.657+5.3%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=0 (PR only)market-uncapped.frametime_p99_ms.max_delta_pct159.729.95+2.3%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)crashFalse00—pass
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)market-standing.cpu_s.max_delta_pct1098.91120.7+22%fail
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)market-turned.cpu_s.max_delta_pct1035.7435.11−1.8%pass
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)field-standing.cpu_s.max_delta_pct1043.6544.83+2.7%pass
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)field-turned.cpu_s.max_delta_pct1026.526.52+0.1%pass
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)field-uncapped.cpu_s.max_delta_pct10109109.76+0.7%pass
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)market-uncapped.cpu_s.max_delta_pct10141.29140.68−0.4%pass
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)screenshot.max_diff0.02—0.5—fail
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)market-standing.frametime_p99_ms.max_delta_pct1519.0518.81−1.3%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)market-turned.frametime_p99_ms.max_delta_pct1518.5718.52−0.2%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)field-standing.frametime_p99_ms.max_delta_pct1518.1418.12−0.1%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)field-turned.frametime_p99_ms.max_delta_pct1518.1618−0.9%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)field-uncapped.frametime_p99_ms.max_delta_pct156.846.88+0.7%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)market-uncapped.frametime_p99_ms.max_delta_pct159.6710.35+7%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)value_activation: HOS_NATIVE_VISIBILITY=1 vs HOS_NATIVE_VISIBILITY=0 (PR)10001—passthe PR's own cells: HOS_NATIVE_VISIBILITY=1 against HOS_NATIVE_VISIBILITY=0 (HOS_NATIVE_VISIBILITY=0 (PR only)); the reference cell's value is 0
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)market-standing.cpu_s: HOS_NATIVE_VISIBILITY=1 vs HOS_NATIVE_VISIBILITY=0 (PR)5112.25120.7+7.5%failthe PR's own cells: HOS_NATIVE_VISIBILITY=1 against HOS_NATIVE_VISIBILITY=0 (HOS_NATIVE_VISIBILITY=0 (PR only))
visibility-routeHOS_NATIVE_VISIBILITY=1 (PR only)field-standing.cpu_s: HOS_NATIVE_VISIBILITY=1 vs HOS_NATIVE_VISIBILITY=0 (PR)547.2644.83−5.1%passthe PR's own cells: HOS_NATIVE_VISIBILITY=1 against HOS_NATIVE_VISIBILITY=0 (HOS_NATIVE_VISIBILITY=0 (PR only))

Pictures of the failure

The last frame of a step that got stuck, or main's and the PR's picture where they differed more than the plan allows; shrunk, and kept a day: until .

visibility-route-c0-market-front-main-failed.png
visibility-route-c0-market-front-main-failed.png
visibility-route-c0-market-front-pr-failed.png
visibility-route-c0-market-front-pr-failed.png
visibility-route-c0-field-front-main-failed.png
visibility-route-c0-field-front-main-failed.png

Log (PR side, end)

[hos sead::MainThread] nvn: CommandBufferWaitEvent #440000: event 0x100d6c8030 mode 0 value 327130 (now 327111) (lr 0x80664f90)
[hos host] vk: replay reads of guest memory as submitted: 11940000 (6909361 pages from the captures)
[hos host] vk: program picks: 17762151 kept (5905245 by the last picks' textures, a uniform buffer bound since), 4280 new, 6833568 with other textures (of 24600000)
[hos host] vk: replay reads of guest memory as submitted: 11960000 (6922478 pages from the captures)
[hos ModuleSystemWorker1] nvn: DrawMeshTasksIndirect: 780000 so far (x1 0x117a4f2000 x2 0x2 x3 0x117a4f2000 x4 0xf33080)
[hos host] vk: write-backs left out (600 submits): 798, 3914.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5340 dirty pages from their submits' captures, 33714 from earlier ones', 0 from guest memory as it was at the replay; 24 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1043 dispatches, 27344 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5744 runs, 693.5 MB compared; 49 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13601 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 11980000 (6935573 pages from the captures)
[hos host] vk: a frame: 381 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 9154 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.80, present 0.12, draw 3.69, descriptors 1.41, texture lookups 0.05, uploads 0.00, pipelines 0.19, programs 0.56, commands 1.27, dispatch 0.37, memory sync 0.22, queue submit 0.00, slot wait 0.01
[hos host] vk: a frame: mirrors 513 (997 MB): 4.10 MB synced in 732.0 copies, 18.25 MB written back
[hos host] vk: 20640 presents, 126.5 fps; a frame: 2432 draws (0 skipped), 119 dispatches, 123 passes, 140 barriers, 6 submits, 0.1 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 2660 pipelines (296419 ms), 1660 programs (0 failed), 4822 textures (7585 MB)
[hos host] vk: host time in GPU resource waits: 1.09 ms a frame (fences 0.01, timelines 1.08, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 382.1 full barriers recorded, 213.7 left out (nothing recorded since the one before), 15.0 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 31.7; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 34.0 from 24.0 batches at the game's barriers (38.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 left as they are (no mesh shaders); scans ended at a barrier 131.2, an op 1918.0, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 4: 2.0; 5: 20.0; 48: 6.0;
[hos host] vk: GPU time 5.0 ms a frame (the longest submission 1.6 ms); 0.1 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.1 (0.1 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.01 MB unswizzled on the CPU in 0.00 ms
[hos sead::MainThread] vk: readbacks of GPU results: 981 of 3000 waited for their write-back, 0.09 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos sead::MainThread] vk: invalidates since the last readbacks line: 8714 kept their virtual core (nothing there to wait for), 3062 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: replay reads of guest memory as submitted: 12000000 (6948730 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 12020000 (6961827 pages from the captures)
[hos host] vk: sparse mirrors: 89, 997 MB with memory of 1604 MB (606 MB spared); blocks given memory later 10279, given back 14878; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 22134; 961 binds (122.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 331 taken from the spares, 182 made new; 424 kept as spares, 0 destroyed; 93 spares now, 1364 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5399 dirty pages from their submits' captures, 34154 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1074 dispatches, 27484 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5573 runs, 674.6 MB compared; 50 pages read into the shadows; 380 MB of shadows (115 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 14035 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 176 live (2256 MB), 0 retired still imported (0 MB), 137 retired given back (1884 MB)
[hos sead::MainThread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 763 ranges, 2850 MB), 223 pages written, 223 copied (0.9 MB)
[hos host] vk: a capture taken (ms): all 0.010; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (130 MB), 11 spare buffers, 7556 texture pages copied (29 MB)
[hos host] vk: replay reads of guest memory as submitted: 12040000 (6974973 pages from the captures)
[hos Presentation Thread] vi: 20677 frames presented (126.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 630, 15.5-17.8 0, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos host] vk: program picks: 17913217 kept (5961024 by the last picks' textures, a uniform buffer bound since), 4280 new, 6882502 with other textures (of 24800000)
[hos host] vk: replay reads of guest memory as submitted: 12060000 (6988101 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 5632466 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 256 KB; 0 made on demand (0 MB kept now); c1 stages: at most 113 a recording; 0 descriptor pools made when a recording's own ran out (0 slots' pools made whole); a recording's pool took at most 470 sets (ub 5314, sb 2337, cis 38913, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 798, 3914.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5101 dirty pages from their submits' captures, 33955 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1035 dispatches, 27366 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5775 runs, 694.0 MB compared; 41 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13744 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 12080000 (7001143 pages from the captures)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: replay reads of guest memory as submitted: 12100000 (7014258 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 4019 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: replay reads of guest memory as submitted: 12120000 (7027375 pages from the captures)
[hos host] vk: sparse mirrors: 89, 997 MB with memory of 1604 MB (606 MB spared); blocks given memory later 10279, given back 14878; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 22590; 961 binds (122.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 331 taken from the spares, 182 made new; 424 kept as spares, 0 destroyed; 93 spares now, 1364 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 798, 3914.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 4748 dirty pages from their submits' captures, 34883 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1058 dispatches, 27711 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5805 runs, 694.2 MB compared; 54 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 14010 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 176 live (2256 MB), 0 retired still imported (0 MB), 137 retired given back (1884 MB)
[hos ModuleSystemWorker3] nvn: CommandBufferSignalEvent #330000: event 0x100d6c8030 mode 0 value 329999 (lr 0x806612a8)
[hos host] vk: replay reads of guest memory as submitted: 12140000 (7040476 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 762 ranges, 2838 MB), 220 pages written, 220 copied (0.9 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 24 kept (130 MB), 12 spare buffers, 7556 texture pages copied (29 MB)
[hos host] vk: a frame: 380 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 9154 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.77, present 0.15, draw 3.74, descriptors 1.45, texture lookups 0.06, uploads 0.00, pipelines 0.19, programs 0.57, commands 1.28, dispatch 0.37, memory sync 0.22, queue submit 0.00, slot wait 0.01
[hos host] vk: a frame: mirrors 513 (997 MB): 4.09 MB synced in 730.4 copies, 18.26 MB written back
[hos host] vk: 20760 presents, 126.1 fps; a frame: 2445 draws (0 skipped), 119 dispatches, 123 passes, 140 barriers, 6 submits, 0.2 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 2660 pipelines (296419 ms), 1660 programs (0 failed), 4831 textures (7587 MB)
[hos host] vk: host time in GPU resource waits: 0.91 ms a frame (fences 0.01, timelines 0.91, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 381.7 full barriers recorded, 213.8 left out (nothing recorded since the one before), 14.7 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 31.7; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 34.0 from 24.0 batches at the game's barriers (38.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 left as they are (no mesh shaders); scans ended at a barrier 131.2, an op 1944.9, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 4: 2.0; 5: 20.0; 48: 6.0;
[hos host] vk: GPU time 5.0 ms a frame (the longest submission 1.6 ms); 0.2 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.2 (0.1 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.01 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: program picks: 18064395 kept (6016364 by the last picks' textures, a uniform buffer bound since), 4280 new, 6931324 with other textures (of 25000000)
[hos host] vk: replay reads of guest memory as submitted: 12160000 (7053636 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5184 dirty pages from their submits' captures, 33804 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1058 dispatches, 27190 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5615 runs, 675.4 MB compared; 48 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13664 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 12180000 (7066782 pages from the captures)
[hos ModuleSystemWorker5] vk: readbacks of GPU results: 981 of 3000 waited for their write-back, 0.07 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker5] vk: invalidates since the last readbacks line: 8086 kept their virtual core (nothing there to wait for), 2922 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #445000: event 0x100d6c8030 mode 0 value 330930 (now 330911) (lr 0x806614d8)
[hos host] vk: replay reads of guest memory as submitted: 12200000 (7079874 pages from the captures)
[hos ModuleSystemWorker1] nvn: 2400000 recordings made, 67277 live copies (24.0 MB), 1702 texture copies; 4 command buffers' segments (4.2 MB)
[hos host] vk: replay reads of guest memory as submitted: 12220000 (7092937 pages from the captures)
[hos host] vk: sparse mirrors: 89, 997 MB with memory of 1604 MB (606 MB spared); blocks given memory later 10279, given back 14878; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 23040; 961 binds (122.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 331 taken from the spares, 182 made new; 424 kept as spares, 0 destroyed; 93 spares now, 1364 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 798, 3914.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5823 dirty pages from their submits' captures, 34178 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1058 dispatches, 27869 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5791 runs, 693.1 MB compared; 47 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 14161 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 176 live (2256 MB), 0 retired still imported (0 MB), 137 retired given back (1884 MB)
[hos host] vk: replay reads of guest memory as submitted: 12240000 (7106067 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 762 ranges, 2818 MB), 224 pages written, 224 copied (0.9 MB)
[hos host] vk: a capture taken (ms): all 0.009; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 24 kept (142 MB), 12 spare buffers, 7972 texture pages copied (31 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 591 sync waits (310 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: program picks: 18215384 kept (6072113 by the last picks' textures, a uniform buffer bound since), 4280 new, 6980335 with other textures (of 25200000)
[hos host] vk: replay reads of guest memory as submitted: 12260000 (7119207 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5700 dirty pages from their submits' captures, 33346 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1039 dispatches, 27198 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5661 runs, 676.0 MB compared; 40 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13866 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 12280000 (7132389 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 12300000 (7145476 pages from the captures)
[hos host] vk: a frame: 378 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 9154 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.79, present 0.13, draw 3.71, descriptors 1.42, texture lookups 0.05, uploads 0.01, pipelines 0.19, programs 0.56, commands 1.28, dispatch 0.37, memory sync 0.23, queue submit 0.00, slot wait 0.01
[hos host] vk: a frame: mirrors 513 (997 MB): 4.11 MB synced in 734.0 copies, 18.27 MB written back
[hos host] vk: 20880 presents, 126.8 fps; a frame: 2458 draws (0 skipped), 120 dispatches, 123 passes, 140 barriers, 6 submits, 0.1 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 2660 pipelines (296419 ms), 1660 programs (0 failed), 4836 textures (7588 MB)
[hos host] vk: host time in GPU resource waits: 1.06 ms a frame (fences 0.01, timelines 1.06, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 383.3 full barriers recorded, 214.0 left out (nothing recorded since the one before), 14.7 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 31.7; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 34.1 from 24.0 batches at the game's barriers (38.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 left as they are (no mesh shaders); scans ended at a barrier 131.2, an op 1974.0, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 4: 2.0; 5: 20.0; 48: 6.0;
[hos host] vk: GPU time 5.0 ms a frame (the longest submission 1.6 ms); 0.1 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.1 (0.1 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.01 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: readback pages refreshed from guest memory: 5705566 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 256 KB; 0 made on demand (0 MB kept now); c1 stages: at most 113 a recording; 0 descriptor pools made when a recording's own ran out (0 slots' pools made whole); a recording's pool took at most 470 sets (ub 5314, sb 2337, cis 38913, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 89, 997 MB with memory of 1604 MB (606 MB spared); blocks given memory later 10279, given back 14878; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 23484; 961 binds (122.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 331 taken from the spares, 182 made new; 424 kept as spares, 0 destroyed; 93 spares now, 1364 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 4997 dirty pages from their submits' captures, 33801 from earlier ones', 0 from guest memory as it was at the replay; 24 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1031 dispatches, 27096 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5674 runs, 677.2 MB compared; 53 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13673 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 176 live (2256 MB), 0 retired still imported (0 MB), 137 retired given back (1884 MB)
[hos host] vk: replay reads of guest memory as submitted: 12320000 (7158622 pages from the captures)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: replay reads of guest memory as submitted: 12340000 (7171722 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 7321 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: replay reads of guest memory as submitted: 12360000 (7184868 pages from the captures)
[hos host] vk: program picks: 18366348 kept (6127898 by the last picks' textures, a uniform buffer bound since), 4280 new, 7029371 with other textures (of 25400000)
[hos sead::MainThread] vk: readbacks of GPU results: 984 of 3000 waited for their write-back, 0.08 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos sead::MainThread] vk: invalidates since the last readbacks line: 8256 kept their virtual core (nothing there to wait for), 3008 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos sead::MainThread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 761 ranges, 2822 MB), 225 pages written, 225 copied (0.9 MB)
[hos host] vk: a capture taken (ms): all 0.009; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (136 MB), 11 spare buffers, 7972 texture pages copied (31 MB)
[hos host] vk: write-backs left out (600 submits): 798, 3914.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5724 dirty pages from their submits' captures, 34138 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1085 dispatches, 27757 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5646 runs, 675.7 MB compared; 45 pages read into the shadows; 380 MB of shadows (115 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 14165 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 12380000 (7198008 pages from the captures)
[hos host] vk: late pages' fix (600 recordings with one): of 617 captures with late pages 617 fixed (85.8 pages, 86.8 fixes a capture), waited as before: 0 no pool, 0 no stamps or shadow yet, 0 stamps too new, 0 whole copies, 0 in place or unknown, 0 a page's earlier fix not recorded yet, 0 other; 1 read on the CPU first, 0 with no table, 0 pages whose shadow changed first; stamps 608 blocks (38912 KB); of the reads on the CPU, 0 waited for an earlier part's fix, 0 copied the page again instead (that part not waited for); the paths: reads on the CPU with the fix done 0, of an earlier recording 1, not recorded yet 0, of this part 0; pages fixed from stamps 51754, from snapshots 1196; the shadows: 231985 pages synced from captures, 229773 made known by it, 2212 left unknown because the replay or the collection wrote them since
[hos host] vk: replay reads of guest memory as submitted: 12400000 (7211148 pages from the captures)
[hos host] vk: sparse mirrors: 89, 997 MB with memory of 1604 MB (606 MB spared); blocks given memory later 10279, given back 14878; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 23934; 961 binds (122.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 331 taken from the spares, 182 made new; 424 kept as spares, 0 destroyed; 93 spares now, 1364 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5001 dirty pages from their submits' captures, 34133 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1037 dispatches, 27365 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5772 runs, 693.6 MB compared; 46 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13802 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 176 live (2256 MB), 0 retired still imported (0 MB), 137 retired given back (1884 MB)
[hos host] vk: replay reads of guest memory as submitted: 12420000 (7224236 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 12440000 (7237334 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: replay reads of guest memory as submitted: 12460000 (7250503 pages from the captures)
[hos ModuleSystemWorker3] nvn: CommandBufferWaitEvent #450000: event 0x100d6c8030 mode 0 value 334730 (now 334711) (lr 0x806614d8)
[hos host] vk: program picks: 18517621 kept (6183719 by the last picks' textures, a uniform buffer bound since), 4280 new, 7078098 with other textures (of 25600000)
[hos host] vk: memory: working set 12268 MB (peak 12949), private 3204 MB, staging 48 MB in 3 arenas, 40 made so far (9 ms)
[hos host] vk: memory heaps: heap 0 (device): 305 blocks 10063 MB, 21596 allocations 9223 MB; used 10174 of 14152 MB; heap 1 (system): 19 blocks 864 MB, 39 allocations 808 MB; used 3133 of 23769 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 5 images 16 MB; 1 free ranges, the largest 15 MB
[hos host] vk: image blocks (render targets): 18 blocks 576 MB, 117 images 263 MB; 52 free ranges, the largest 31 MB
[hos host] vk: image blocks (textures, compacted): 183 blocks 5856 MB, 4712 images 5710 MB; 2503 free ranges, the largest 21 MB
[hos host] vk: render target images: 1394 reused from the pool, 1578 made new, 1452 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 10174 MB used of 14152 MB allowed (839 MB of it free in the blocks); textures 7589 MB (budget 10830 MB), 4848 textures
[hos host] vk: budget: VRAM 10063 of 14152 MB (10174 MB by the driver, 111 MB unaccounted): textures 7589, mirrors 997, blocks free 839, shadows 380, other buffers 144, other images 63, staging 48; RAM 3098 MB (the process's private memory 3204 MB): imports 2256, capture spares 664, captures 134, capture table 34, other buffers 10, others 0
[hos host] vk: textures: decoded ASTC 4613 MB, rendered 1070 MB (53 drawn in one frame, kept: 38279 KB), other 1377 MB; used in the last 8 presents 2169 MB
[hos host] vk: decoded ASTC used again: 1613 textures (3164 MB) copied from ones the game finalized, not decoded; 1613 of those retired, 942 listed; 1536 MB of blocks hashed in 100 ms
[hos host] vk:   288 MB: 0x1147c16000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a5b22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 5970 presents ago
[hos host] vk:   214 MB: 0x10a22aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 5970 presents ago
[hos host] vk:   64 MB: 0x113cba9000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10002ff0200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 15673 presents ago
[hos host] vk:   54 MB: 0x100030a4200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10002ce3a00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 5970 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 49 (10865 KB): 242x40x1 NVN 0x25: 4 (151 KB); 260x260x1 NVN 0xd: 34 (5982 KB); 306x104x1 NVN 0x1: 4 (124 KB); 592x49x1 NVN 0x25: 3 (339 KB); 656x113x1 NVN 0x1: 3 (217 KB); 960x540x1 NVN 0x29: 1 (4050 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 541 waited for it, 0.49 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5242 dirty pages from their submits' captures, 33679 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1037 dispatches, 27168 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5637 runs, 675.6 MB compared; 47 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13674 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: status: 600 frames, the slowest 10.7 ms; VRAM 9.9 of 13.8 GB
[hos host] vk: program census (play): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 0 program records loaded (0.0 ms), 2359 new graphics pipelines, pipelines from the driver's cache 735, from the start's cache 0, compiled 1769; 203825 draws and dispatches left out, 1515 waits (8398.5 ms)
[hos host] vk: program loads at create (play): 0 queued, 0 loaded (median 0.00 ms, 99th percentile 0.00 ms, worst 0.00 ms, 0 ms and 0.0 MB read in all); 0 programs (0 of them compute variants without their code, 0 graphics programs no note names), 0 graphics pipelines (0 not in a cache), 0 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 0 pipelines made at create taken by draws; 8915 create calls with nothing kept for their stages, 0 before the index; units 8915 (0 ready, 0 loading), 14607 programs alive; 0 noted graphics pipeline probes deferred to first use
[hos host] vk: program loads at create (play): 0 shader modules let go once their programs' noted pipelines were built, 0 programs' modules made again for a new pipeline state (RX_VK_DROP_LOADED_MODULES=0 keeps them)
[hos host] vk: module identifiers (play): 0 known, 0 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 378 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 9154 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.83, present 0.16, draw 3.81, descriptors 1.48, texture lookups 0.06, uploads 0.01, pipelines 0.19, programs 0.58, commands 1.30, dispatch 0.37, memory sync 0.22, queue submit 0.00, slot wait 0.01
[hos host] vk: a frame: mirrors 513 (997 MB): 4.11 MB synced in 733.7 copies, 18.26 MB written back
[hos host] vk: 21000 presents, 125.6 fps; a frame: 2461 draws (0 skipped), 120 dispatches, 123 passes, 140 barriers, 6 submits, 0.2 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 2660 pipelines (296419 ms), 1660 programs (0 failed), 4848 textures (7589 MB)
[hos host] vk: host time in GPU resource waits: 0.94 ms a frame (fences 0.01, timelines 0.93, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 382.9 full barriers recorded, 213.8 left out (nothing recorded since the one before), 14.7 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 31.7; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 34.0 from 24.0 batches at the game's barriers (38.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 left as they are (no mesh shaders); scans ended at a barrier 131.2, an op 1982.4, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 4: 2.0; 5: 20.0; 48: 6.0;
[hos host] vk: GPU time 5.0 ms a frame (the longest submission 1.8 ms); 0.2 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.2 (0.3 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.01 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: the render thread (600 presents, 7.93 ms a frame): its jobs 6.06 ms a frame (submits 0.85, presents 0.14, others 5.06), idle 1.87; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 1.10 jobs queued at a present); 0 catch-ups (a finish, a texture read back, a write-back not recorded yet: 0), 0.00 ms each; 0 presents held for the CPU's signal, 0.00 ms each (0 let go by RX_VK_HOLD_MS); 559 captures at a CPU signal; the submits' waits read in 0.049 ms a frame; readbacks: 2893 next to a texture's storage (no catch-up), 0 in one, 0 read exactly (after the work handed over: among the catch-ups); a present made 5.51 ms after its hand-over (8.22 at most)
[hos host] vk: frame overlap (600 presents): the GPU between recordings, ms a frame, idle + in the next one's uploads: up to a frame's first recording with a draw 0.38 + 0.26 (the window's copy is in the idle), after it 1.59 + 0.68; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.02 ms of GPU time), after its last that reads it 1.00 (0.02 ms)
[hos host] vk: frame ahead (600 presents): 9.10 parts a frame recorded in the signal's job, 66.9% of them took nothing that came with the signal; of the others, 3.01 a frame took pages of the signal's capture into a mirror (154.9 pages a frame) and 0.00 pages read as guest memory had them (0.0 pages a frame); a frame's first part took nothing in 2 of 559 frames, its first three in 2
[hos host] vk: frame ahead, what came with the signal: 559 signals with a capture, each of 218.8 pages in 12.6 pools; between a frame's last submit and that signal 0.00 texture and sampler registrations, 0.00 releases and 0.00 other CPU signals a signal (0 of them with any)
[hos host] vk: frame overlap, by place: idle + uploads before a frame's 2nd to 12th recording with a draw, then all later ones: 0.70+0.26, 0.03+0.05, 0.25+0.10, 0.47+0.11, 0.10+0.05, 0.01+0.03, 0.01+0.03, 0.01+0.03, 0.01+0.02, 0.00+0.00, 0.00+0.00, 0.00+0.00
[hos ModuleSystemWorker4] nvn: CommandBufferSignalEvent #335000: event 0x100d6c803c mode 0 value 334999 (lr 0x80661bfc)
[hos host] vk: replay reads of guest memory as submitted: 12480000 (7263688 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 761 ranges, 2814 MB), 218 pages written, 218 copied (0.9 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (138 MB), 11 spare buffers, 7972 texture pages copied (31 MB)
[hos Presentation Thread] vi: real-time step on: ratio 0.535, game's frame rate 60
[hos host] vk: replay reads of guest memory as submitted: 12500000 (7276822 pages from the captures)
[hos host] vk: sparse mirrors: 89, 997 MB with memory of 1604 MB (606 MB spared); blocks given memory later 10279, given back 14878; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 24384; 961 binds (122.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 331 taken from the spares, 182 made new; 424 kept as spares, 0 destroyed; 93 spares now, 1364 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 798, 3914.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 4326 dirty pages from their submits' captures, 35246 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1069 dispatches, 27731 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5772 runs, 693.6 MB compared; 68 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13919 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 176 live (2256 MB), 0 retired still imported (0 MB), 137 retired given back (1884 MB)
[hos host] vk: replay reads of guest memory as submitted: 12520000 (7290078 pages from the captures)
[hos host] vk: render targets drawn over from another start, let go: 2268 so far (6609 MB), 2036 of them (5836 MB) between the passes
[hos host] vk: replay reads of guest memory as submitted: 12540000 (7303243 pages from the captures)
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 980 of 3000 waited for their write-back, 0.07 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker4] vk: invalidates since the last readbacks line: 8241 kept their virtual core (nothing there to wait for), 3023 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: readback pages refreshed from guest memory: 5779092 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: replay reads of guest memory as submitted: 12560000 (7316391 pages from the captures)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 256 KB; 0 made on demand (0 MB kept now); c1 stages: at most 113 a recording; 0 descriptor pools made when a recording's own ran out (0 slots' pools made whole); a recording's pool took at most 470 sets (ub 5314, sb 2337, cis 38913, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5203 dirty pages from their submits' captures, 34231 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1048 dispatches, 27477 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5606 runs, 675.2 MB compared; 47 pages read into the shadows; 380 MB of shadows (115 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 14025 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: program picks: 18668882 kept (6239612 by the last picks' textures, a uniform buffer bound since), 4280 new, 7126837 with other textures (of 25800000)
[hos host] vk: replay reads of guest memory as submitted: 12580000 (7329629 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 2562 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 760 ranges, 2798 MB), 222 pages written, 222 copied (0.9 MB)
[hos host] vk: a capture taken (ms): all 0.008; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 24 kept (132 MB), 12 spare buffers, 8094 texture pages copied (31 MB)
[hos host] vk: replay reads of guest memory as submitted: 12600000 (7342754 pages from the captures)
[hos host] vk: sparse mirrors: 89, 997 MB with memory of 1604 MB (606 MB spared); blocks given memory later 10279, given back 14878; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 24834; 961 binds (122.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 331 taken from the spares, 182 made new; 424 kept as spares, 0 destroyed; 93 spares now, 1364 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 798, 3914.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5177 dirty pages from their submits' captures, 33987 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1045 dispatches, 27390 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5751 runs, 693.2 MB compared; 48 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13644 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 176 live (2256 MB), 0 retired still imported (0 MB), 137 retired given back (1884 MB)
[hos host] vk: replay reads of guest memory as submitted: 12620000 (7355922 pages from the captures)
[hos host] vk: a frame: 378 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 9154 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.82, present 0.15, draw 3.81, descriptors 1.47, texture lookups 0.06, uploads 0.01, pipelines 0.19, programs 0.57, commands 1.31, dispatch 0.38, memory sync 0.22, queue submit 0.00, slot wait 0.01
[hos host] vk: a frame: mirrors 513 (997 MB): 4.09 MB synced in 731.1 copies, 18.25 MB written back
[hos host] vk: 21120 presents, 126.2 fps; a frame: 2462 draws (0 skipped), 119 dispatches, 123 passes, 140 barriers, 6 submits, 0.4 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 2660 pipelines (296419 ms), 1660 programs (0 failed), 4866 textures (7594 MB)
[hos host] vk: host time in GPU resource waits: 0.96 ms a frame (fences 0.01, timelines 0.95, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 382.4 full barriers recorded, 213.8 left out (nothing recorded since the one before), 14.7 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 31.7; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 34.0 from 24.0 batches at the game's barriers (38.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 left as they are (no mesh shaders); scans ended at a barrier 131.2, an op 1979.0, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 4: 2.0; 5: 20.0; 48: 6.0;
[hos host] vk: GPU time 5.0 ms a frame (the longest submission 1.6 ms); 0.3 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.3 (0.2 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.04 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: replay reads of guest memory as submitted: 12640000 (7369023 pages from the captures)
[hos host] vk: texture slots registered again at another size: 1200000 (of 1426489 registrations; slot 17820 120x67 -> 1920x1080)
[hos host] vk: replay reads of guest memory as submitted: 12660000 (7382193 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5146 dirty pages from their submits' captures, 33696 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1044 dispatches, 27138 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5614 runs, 675.1 MB compared; 46 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13750 in captures, 0 only in the collection, 36 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker4] nvn: DrawMeshTasksIndirect: 800000 so far (x1 0x117daa3060 x2 0 x3 0x117daa3040 x4 0x9b040)
[hos BaseProcExecutor1] vcores: 171507 takes, 39346 waits (mean 72 us, longest 11.6 ms, 0 over HOS_VCORE_LOG_MS), 16970 kicks, 17336 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: program picks: 18819720 kept (6295367 by the last picks' textures, a uniform buffer bound since), 4280 new, 7175999 with other textures (of 26000000)
[hos host] vk: replay reads of guest memory as submitted: 12680000 (7395383 pages from the captures)
[hos ModuleSystemWorker4] nvn: 2450000 recordings made, 67223 live copies (23.6 MB), 1702 texture copies; 5 command buffers' segments (4.3 MB)
[hos host] vk: replay reads of guest memory as submitted: 12700000 (7408527 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 763 ranges, 2804 MB), 231 pages written, 231 copied (0.9 MB)
[hos host] vk: a capture taken (ms): all 0.009; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.004, bytes copied 0.000; 24 kept (140 MB), 12 spare buffers, 6659 texture pages copied (26 MB)
[hos host] vk: sparse mirrors: 89, 997 MB with memory of 1604 MB (606 MB spared); blocks given memory later 10279, given back 14878; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 25284; 961 binds (122.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 331 taken from the spares, 182 made new; 424 kept as spares, 0 destroyed; 93 spares now, 1364 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 798, 3914.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5439 dirty pages from their submits' captures, 34416 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1062 dispatches, 27841 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5612 runs, 675.1 MB compared; 45 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 14024 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 176 live (2256 MB), 0 retired still imported (0 MB), 137 retired given back (1884 MB)
[hos host] vk: replay reads of guest memory as submitted: 12720000 (7421679 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #455000: event 0x100d6c8030 mode 0 value 338530 (now 338511) (lr 0x806614d8)
[hos sead::MainThread] vk: readbacks of GPU results: 958 of 3000 waited for their write-back, 0.08 ms each; 0 looked at the stopped queues first, 3 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos sead::MainThread] vk: invalidates since the last readbacks line: 8422 kept their virtual core (nothing there to wait for), 3098 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: replay reads of guest memory as submitted: 12740000 (7434939 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 12760000 (7448087 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5388 dirty pages from their submits' captures, 33474 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1037 dispatches, 27050 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5685 runs, 692.6 MB compared; 47 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13643 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: queues ordered by their events: 144000 submissions stopped at a wait, 143979 went on once it was reached, 18 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 2 (stopped 50.06 ms at most); waits met 455671, not met 0; captures at a CPU event's wait 9694, 0 of them at a wait met on arrival
[hos host] vk: program picks: 18970872 kept (6351413 by the last picks' textures, a uniform buffer bound since), 4280 new, 7224847 with other textures (of 26200000)
[hos host] vk: replay reads of guest memory as submitted: 12780000 (7461274 pages from the captures)
[hos host] vk: a frame: 378 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 9154 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.89, present 0.13, draw 3.79, descriptors 1.47, texture lookups 0.05, uploads 0.01, pipelines 0.19, programs 0.57, commands 1.30, dispatch 0.37, memory sync 0.23, queue submit 0.00, slot wait 0.01
[hos host] vk: a frame: mirrors 514 (998 MB): 4.11 MB synced in 732.1 copies, 18.24 MB written back
[hos host] vk: 21240 presents, 124.7 fps; a frame: 2465 draws (0 skipped), 119 dispatches, 123 passes, 140 barriers, 6 submits, 0.1 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 2660 pipelines (296419 ms), 1660 programs (0 failed), 4870 textures (7594 MB)
[hos host] vk: host time in GPU resource waits: 1.08 ms a frame (fences 0.01, timelines 1.08, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 381.8 full barriers recorded, 213.9 left out (nothing recorded since the one before), 14.7 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 31.8; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 34.1 from 24.0 batches at the game's barriers (38.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 left as they are (no mesh shaders); scans ended at a barrier 131.3, an op 1980.2, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 4: 2.0; 5: 20.0; 48: 6.0;
[hos host] vk: GPU time 5.0 ms a frame (the longest submission 1.7 ms); 0.1 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.1 (0.1 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: replay reads of guest memory as submitted: 12800000 (7474498 pages from the captures)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: readback pages refreshed from guest memory: 5852197 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 256 KB; 0 made on demand (0 MB kept now); c1 stages: at most 113 a recording; 0 descriptor pools made when a recording's own ran out (0 slots' pools made whole); a recording's pool took at most 470 sets (ub 5314, sb 2337, cis 38913, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 90, 998 MB with memory of 1608 MB (609 MB spared); blocks given memory later 10279, given back 14878; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 25728; 962 binds (122.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 332 taken from the spares, 182 made new; 424 kept as spares, 0 destroyed; 92 spares now, 1360 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5755 dirty pages from their submits' captures, 33176 from earlier ones', 0 from guest memory as it was at the replay; 24 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1036 dispatches, 27116 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5605 runs, 675.1 MB compared; 41 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13728 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 176 live (2256 MB), 0 retired still imported (0 MB), 137 retired given back (1884 MB)
[hos host] vk: replay reads of guest memory as submitted: 12820000 (7487672 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 764 ranges, 2806 MB), 222 pages written, 222 copied (0.9 MB)
[hos host] vk: a capture taken (ms): all 0.009; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 24 kept (144 MB), 12 spare buffers, 6659 texture pages copied (26 MB)
[hos sead::MainThread] nvn: CommandBufferSignalEvent #340000: event 0x100d6c8048 mode 0 value 339999 (lr 0x80666598)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 10094 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: replay reads of guest memory as submitted: 12840000 (7500801 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5289 dirty pages from their submits' captures, 34306 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1078 dispatches, 27611 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5598 runs, 674.9 MB compared; 52 pages read into the shadows; 380 MB of shadows (115 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 14110 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 12860000 (7513882 pages from the captures)
[hos Presentation Thread] vi: 21307 frames presented (125.8 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 630, 15.5-17.8 0, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos host] vk: program picks: 19122126 kept (6407363 by the last picks' textures, a uniform buffer bound since), 4280 new, 7273593 with other textures (of 26400000)
[hos host] vk: replay reads of guest memory as submitted: 12880000 (7527108 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 12900000 (7540289 pages from the captures)
[hos host] vk: sparse mirrors: 90, 998 MB with memory of 1608 MB (609 MB spared); blocks given memory later 10279, given back 14878; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 26178; 962 binds (122.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 332 taken from the spares, 182 made new; 424 kept as spares, 0 destroyed; 92 spares now, 1360 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 798, 3914.3 MB; 31 bindings followed, 43 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5703 dirty pages from their submits' captures, 33475 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1043 dispatches, 27364 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5746 runs, 693.2 MB compared; 42 pages read into the shadows; 380 MB of shadows (116 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13651 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 176 live (2256 MB), 0 retired still imported (0 MB), 137 retired given back (1884 MB)
[hos host] control: quit asked by a client
[hos host] vk: program census (close): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 0 program records loaded (0.0 ms), 2359 new graphics pipelines, pipelines from the driver's cache 735, from the start's cache 0, compiled 1769; 203825 draws and dispatches left out, 1515 waits (8398.5 ms)
[hos host] vk: program loads at create (close): 0 queued, 0 loaded (median 0.00 ms, 99th percentile 0.00 ms, worst 0.00 ms, 0 ms and 0.0 MB read in all); 0 programs (0 of them compute variants without their code, 0 graphics programs no note names), 0 graphics pipelines (0 not in a cache), 0 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 0 pipelines made at create taken by draws; 8915 create calls with nothing kept for their stages, 0 before the index; units 8915 (0 ready, 0 loading), 14607 programs alive; 0 noted graphics pipeline probes deferred to first use
[hos host] vk: program loads at create (close): 0 shader modules let go once their programs' noted pipelines were built, 0 programs' modules made again for a new pipeline state (RX_VK_DROP_LOADED_MODULES=0 keeps them)
[hos host] vk: module identifiers (close): 0 known, 0 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: indexed descriptor arrays (close): 352 in 352 of 3245 program stages, 0 of them near a limit
[hos host] vk: indexed descriptor arrays (close), fragment stages: 352 texture (17-64: 352); the longest 64
[hos host] vk: page table census (close): no pool outside its windows, 64-72 GB and 508-512 GB (4253 pools in the table now)
[hos host] vk: pipeline cache: play file 17 saved (2276 KB): 4 pipelines, 0 from the driver's cache, 4 compiled; 0 more from the start's cache
[hos host] control: exiting
[hos host] stop: close work done (a file the game was writing waited for, at most 3 s)