Beanstalk Longshot test farm
Menu

Report rp_d549463ccce0ea2f fail

#1866 perf(vk): the recorder takes a whole part and submits it, so the render thread never waits at a part's end

linux+nvidia on jabbaxipil/Blaine,

This report disagrees with its peers.

three-areas field-standing.frametime_max_ms: +172.4% against peers' +9.6%
three-areas kokiri-after-turn.draws_mean: +5.5% against peers' +32.6%
three-areas kokiri-after-turn.frametime_max_ms: -66.9% against peers' +4.0%
three-areas kokiri-standing.frametime_max_ms: +53.3% against peers' +7.7%
three-areas market-after-turn.draws_mean: -64.2% against peers' +31.6%
three-areas market-after-turn.fps_mean: +20.1% against peers' -8.2%
three-areas market-after-turn.frametime_1pct_low_ms: -29.7% against peers' +6.7%
three-areas market-after-turn.frametime_p50_ms: -20.6% against peers' +9.2%
three-areas market-after-turn.frametime_p99_ms: -19.6% against peers' +9.4%
three-areas market-standing.frametime_max_ms: +114.4% against peers' +4.1%
three-areas run_frametime_max_ms: +136.4% against peers' +19.8%

Machine

Commit
f08644e37cd3ce2e6a4801195b6c39d7a2df9f7e
Baseline
89c3d21ea169ed46124e6c97acd90587f660b75a
App
nightly-195-f9b6d598, attempt 1
Game build
e0f3ccfa
os
linux
cpu
AMD Ryzen 7 5800X3D 8-Core Processor
gpu
NVIDIA GeForce RTX 4070
tags
linuxnvidianvidia-proprietarydgpu
driver
615.71.9
ram_gb
64
vram_mb
12,282
gpu_vendor
nvidia
os_version
Ubuntu 26.04.1 LTS (kernel 7.0.0)

Builds

main (baseline)

795.9 s warm

from 89c3d21ea169

lift
98 s
read
4.3 s
finish
14.6 s
compile
660.2 s
shaders
11.6 s

PR

24.5 s warm

Program cache: warm

Scenarios

three-areas

ok
Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count33.529.5−11.9%
log_forbidden_hits00±0%
log_expected_missing00±0%
run_frametime_max_ms1,972.744,663.87+136.4%
field-standing.fps_mean114.83114.66−0.2%
kokiri-standing.fps_mean119.12117.08−1.7%
market-standing.fps_mean87.2887.7+0.5%
field-after-turn.fps_mean127.17129.41+1.8%
field-standing.draws_mean1,186.21,188.25+0.2%
field-standing.hitch_count00.5—
kokiri-after-turn.fps_mean122.16118.99−2.6%
kokiri-standing.draws_mean679.55684.15+0.7%
market-after-turn.fps_mean111.22133.6+20.1%
market-standing.draws_mean2,4752,468.75−0.3%
field-after-turn.draws_mean1,034.351,049.05+1.4%
kokiri-standing.hitch_count00±0%
market-standing.hitch_count01—
field-after-turn.hitch_count01—
field-standing.stutter_count0.52.5+400%
kokiri-after-turn.draws_mean696.15734.1+5.5%
market-after-turn.draws_mean1,394.8499.15−64.2%
kokiri-after-turn.hitch_count10−100%
kokiri-standing.stutter_count28.5+325%
market-after-turn.hitch_count1.51−33.3%
market-standing.stutter_count3.53−14.3%
field-after-turn.stutter_count13+200%
field-standing.frametime_max_ms15.8243.08+172.4%
field-standing.frametime_p50_ms8.528.57+0.6%
field-standing.frametime_p99_ms11.8811.37−4.3%
kokiri-after-turn.stutter_count20−100%
market-after-turn.stutter_count3.54+14.3%
kokiri-standing.frametime_max_ms16.625.44+53.3%
kokiri-standing.frametime_p50_ms8.088.19+1.4%
kokiri-standing.frametime_p99_ms11.5912.36+6.7%
market-standing.frametime_max_ms33.5471.91+114.4%
market-standing.frametime_p50_ms11.0410.89−1.3%
market-standing.frametime_p99_ms17.9816.43−8.6%
field-after-turn.frametime_max_ms42.2749+15.9%
field-after-turn.frametime_p50_ms7.857.57−3.6%
field-after-turn.frametime_p99_ms9.7310.23+5.1%
kokiri-after-turn.frametime_max_ms38.1112.62−66.9%
kokiri-after-turn.frametime_p50_ms8.058.28+2.9%
kokiri-after-turn.frametime_p99_ms10.4210.89+4.5%
market-after-turn.frametime_max_ms70.457.16−18.8%
market-after-turn.frametime_p50_ms9.167.28−20.6%
market-after-turn.frametime_p99_ms13.8711.15−19.6%
field-standing.frametime_1pct_low_ms12.7114.08+10.8%
kokiri-standing.frametime_1pct_low_ms12.7314+10%
market-standing.frametime_1pct_low_ms20.8821.44+2.7%
field-after-turn.frametime_1pct_low_ms12.0815.22+25.9%
kokiri-after-turn.frametime_1pct_low_ms14.2311.67−18%
market-after-turn.frametime_1pct_low_ms23.0916.25−29.7%
screenshot_diff—0.12—

Pictures against main: kokiri diff 0.0924 (main against itself 0.6963), field diff 0.0397 (main against itself 0.0483), market diff 0.1355 (main against itself 0.1968)

Runs (4)
Metricmain 0pr 1pr 2main 3
hangnononono
crashednononono
exit_code0000
run_hitch_count46382121
log_forbidden_hits0000
log_expected_missing0000
run_frametime_max_ms3,725.688,997.64330.09219.79
field-standing.fps_mean110.61112117.31119.05
kokiri-standing.fps_mean117.5115.02119.15120.74
market-standing.fps_mean81.3183.0992.3193.24
field-after-turn.fps_mean135.22131.85126.98119.12
field-standing.draws_mean1,178.11,186.71,189.81,194.3
field-standing.hitch_count0100
kokiri-after-turn.fps_mean124.64118.86119.12119.68
kokiri-standing.draws_mean631.4636.4731.9727.7
market-after-turn.fps_mean88.19133.19134.01134.24
market-standing.draws_mean2,510.12,460.22,477.32,439.9
field-after-turn.draws_mean810.7846.31,251.81,258
kokiri-standing.hitch_count0000
market-standing.hitch_count0110
field-after-turn.hitch_count0020
field-standing.stutter_count1500
kokiri-after-turn.draws_mean688756.4711.8704.3
market-after-turn.draws_mean2,316.8514484.3472.8
kokiri-after-turn.hitch_count0002
kokiri-standing.stutter_count41700
market-after-turn.hitch_count3020
market-standing.stutter_count6331
field-after-turn.stutter_count1151
field-standing.frametime_max_ms18.6872.6113.5512.96
field-standing.frametime_p50_ms8.668.628.528.38
field-standing.frametime_p99_ms13.6212.899.8410.13
kokiri-after-turn.stutter_count0004
market-after-turn.stutter_count5262
kokiri-standing.frametime_max_ms20.5338.4112.4812.67
kokiri-standing.frametime_p50_ms7.877.968.438.28
kokiri-standing.frametime_p99_ms13.5415.299.439.64
market-standing.frametime_max_ms42.8865.3478.4824.21
market-standing.frametime_p50_ms11.5911.2610.5310.49
market-standing.frametime_p99_ms20.8718.5414.3315.09
field-after-turn.frametime_max_ms40.5325.672.4144
field-after-turn.frametime_p50_ms7.337.477.688.38
field-after-turn.frametime_p99_ms9.329.510.9610.15
kokiri-after-turn.frametime_max_ms12.6512.7112.5263.58
kokiri-after-turn.frametime_p50_ms7.818.218.348.28
kokiri-after-turn.frametime_p99_ms10.5310.411.3810.32
market-after-turn.frametime_max_ms92.4140.0374.2948.4
market-after-turn.frametime_p50_ms11.027.387.187.3
market-after-turn.frametime_p99_ms16.9310.5511.7510.81
field-standing.frametime_1pct_low_ms14.6717.5710.5810.75
kokiri-standing.frametime_1pct_low_ms15.0617.7610.2310.39
market-standing.frametime_1pct_low_ms24.6622.8420.0517.1
field-after-turn.frametime_1pct_low_ms11.6811.5618.8812.48
kokiri-after-turn.frametime_1pct_low_ms11.2911.3511.9917.17
market-after-turn.frametime_1pct_low_ms32.2713.0319.4613.92
screenshot_diff—0.140.1—

Rules

ScenarioCellRuleLimitmainPRΔ %Status
three-areascrashFalse00—pass
three-areaskokiri-standing.fps_mean.min_delta_pct10119.12117.08−1.7%pass
three-areaskokiri-after-turn.fps_mean.min_delta_pct10122.16118.99−2.6%pass
three-areasfield-standing.fps_mean.min_delta_pct10114.83114.66−0.2%pass
three-areasfield-after-turn.fps_mean.min_delta_pct10127.17129.41+1.8%pass
three-areasmarket-standing.fps_mean.min_delta_pct1087.2887.7+0.5%pass
three-areasmarket-after-turn.fps_mean.min_delta_pct10111.22133.6+20.1%pass
three-areasscreenshot.max_diff0.050.70.14—passover the limit, but main differs from itself by 0.6963 here: no more than 1.5x that is the scene, not the PR
three-areaskokiri-standing.stutter_count.max_delta_pct2028.5+325%fail
three-areaskokiri-after-turn.stutter_count.max_delta_pct2020−100%pass
three-areasfield-standing.stutter_count.max_delta_pct200.52.5+400%fail
three-areasfield-after-turn.stutter_count.max_delta_pct2013+200%fail
three-areasmarket-standing.stutter_count.max_delta_pct203.53−14.3%pass
three-areasmarket-after-turn.stutter_count.max_delta_pct203.54+14.3%pass
three-areaskokiri-standing.frametime_p99_ms.max_delta_pct1511.5912.36+6.7%pass
three-areaskokiri-after-turn.frametime_p99_ms.max_delta_pct1510.4210.89+4.5%pass
three-areasfield-standing.frametime_p99_ms.max_delta_pct1511.8811.37−4.3%pass
three-areasfield-after-turn.frametime_p99_ms.max_delta_pct159.7310.23+5.1%pass
three-areasmarket-standing.frametime_p99_ms.max_delta_pct1517.9816.43−8.6%pass
three-areasmarket-after-turn.frametime_p99_ms.max_delta_pct1513.8711.15−19.6%pass
three-areaslog_forbidden_hits.max000—pass

Log (PR side, end)

[hos host] vk: write-backs left out (600 submits): 1239, 6077.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 306 dirty pages from their submits' captures, 61467 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): 1325 dispatches, 49570 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6300 runs, 1032.0 MB compared; 177 pages read into the shadows; 372 MB of shadows (114 MB of pages known, in 157 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17300 in captures, 2 only in the collection, 58 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 160 live (2156 MB), 0 retired still imported (0 MB), 41 retired given back (976 MB)
[hos host] vk: queues ordered by their events: 156000 submissions stopped at a wait, 155998 went on once it was reached, 0 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 0 (stopped 0.00 ms at most); waits met 433552, not met 0; captures at a CPU event's wait 16370, 0 of them at a wait met on arrival
[hos Presentation Thread] vi: 25495 frames presented (133.9 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 669, 15.5-17.8 0, 17.8-22 0, 22-30 0, 30-40 0, over 40 1
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: 2 textures retired (2636 left, 4541 MB): budget 0, drawn over 2, dead targets 0
[hos ModuleSystemWorker3] nvn: CommandBufferSignalEvent #310000: event 0x100d6c8008 mode 0 value 309995 (lr 0x80660f34)
[hos host] vk: 3 textures retired (2633 left, 4519 MB): budget 0, drawn over 3, dead targets 0
[hos ModuleSystemWorker5] nvn: 2200000 recordings made, 67884 live copies (18.4 MB), 3848 texture copies; 5 command buffers' segments (4.2 MB)
[hos host] vk: replay reads of guest memory as submitted: 8920000 (4779895 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: write-backs left out (600 submits): 1218, 5974.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 157 dirty pages from their submits' captures, 60478 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): 1308 dispatches, 48708 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6163 runs, 1015.7 MB compared; 188 pages read into the shadows; 372 MB of shadows (114 MB of pages known, in 158 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 16823 in captures, 4 only in the collection, 59 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] nvn: CommandBufferWaitEvent #435000: event 0x100d6c8048 mode 0 value 310828 (now 310809) (lr 0x806661f8)
[hos host] vk: 1 textures retired (2635 left, 4519 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 8940000 (4792765 pages from the captures)
[hos host] vk: a frame: 28 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 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 3.33, present 0.15, draw 1.66, descriptors 0.69, texture lookups 0.05, uploads 0.00, pipelines 0.07, programs 0.27, commands 0.49, dispatch 0.44, memory sync 0.31, queue submit 0.00, slot wait 0.02
[hos host] vk: a frame: mirrors 329 (833 MB): 4.09 MB synced in 839.7 copies, 17.50 MB written back
[hos host] vk: 25560 presents, 137.4 fps; a frame: 483 draws (0 skipped), 81 dispatches, 133 passes, 139 barriers, 6 submits, 0.0 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 37386 pipelines (4479524 ms), 0 programs (0 failed), 2635 textures (4519 MB)
[hos host] vk: host time in GPU resource waits: 0.05 ms a frame (fences 0.01, timelines 0.04, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 307.0 full barriers recorded, 179.4 left out (nothing recorded since the one before), 12.5 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: 31.2; 0x12: 75.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 3.6 ms a frame (the longest submission 2.0 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.2 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 864 of 3000 waited for their write-back, 0.01 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: 8898 kept their virtual core (nothing there to wait for), 3134 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 6710 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.04 over 853 ranges, 2955 MB), 203 pages written, 203 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.015; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.013, bytes copied 0.000; 25 kept (94 MB), 11 spare buffers, 4621 texture pages copied (18 MB)
[hos host] vk: 2 textures retired (2633 left, 4508 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: sparse mirrors: 60, 833 MB with memory of 1352 MB (518 MB spared); blocks given memory later 6161, given back 10365; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 26238; 840 binds (394.8 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 210 taken from the spares, 119 made new; 269 kept as spares, 0 destroyed; 59 spares now, 1012 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1239, 6077.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 197 dirty pages from their submits' captures, 61524 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): 1327 dispatches, 49514 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6272 runs, 1032.3 MB compared; 185 pages read into the shadows; 372 MB of shadows (114 MB of pages known, in 157 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17161 in captures, 3 only in the collection, 59 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 160 live (2156 MB), 0 retired still imported (0 MB), 41 retired given back (976 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.02, writes collected 0.04
[hos host] vk: replay reads of guest memory as submitted: 8960000 (4805556 pages from the captures)
[hos host] vk: 1 textures retired (2633 left, 4505 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 8980000 (4818359 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: the recording streams: 48 of 1 MB, the most one held 265 KB; 0 made on demand (0 MB kept now); c1 stages: at most 105 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 496 sets (ub 5548, sb 2511, cis 42298, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 1239, 6077.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 145 dirty pages from their submits' captures, 61502 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): 1329 dispatches, 49449 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6208 runs, 1030.2 MB compared; 171 pages read into the shadows; 372 MB of shadows (114 MB of pages known, in 157 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17211 in captures, 1 only in the collection, 58 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] vk: capture: a submit 0.10 ms (lock 0.01, scan 0.05 over 822 ranges, 2811 MB), 192 pages written, 192 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.004; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (100 MB), 11 spare buffers, 4330 texture pages copied (16 MB)
[hos host] vk: 1 textures retired (2639 left, 4536 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a frame: 28 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 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 3.39, present 0.13, draw 1.71, descriptors 0.72, texture lookups 0.05, uploads 0.00, pipelines 0.07, programs 0.27, commands 0.52, dispatch 0.44, memory sync 0.33, queue submit 0.00, slot wait 0.02
[hos host] vk: a frame: mirrors 329 (833 MB): 4.08 MB synced in 838.6 copies, 17.49 MB written back
[hos host] vk: 25680 presents, 137.7 fps; a frame: 481 draws (0 skipped), 80 dispatches, 133 passes, 139 barriers, 6 submits, 0.0 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 37386 pipelines (4479524 ms), 0 programs (0 failed), 2639 textures (4536 MB)
[hos host] vk: host time in GPU resource waits: 0.07 ms a frame (fences 0.01, timelines 0.06, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 305.9 full barriers recorded, 178.7 left out (nothing recorded since the one before), 12.4 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: 31.2; 0x12: 75.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 3.6 ms a frame (the longest submission 2.0 ms); 0.1 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.1 (0.3 MB), ASTC or BC 0.0 (0.0 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: 9000000 (4831208 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 7461 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: readback pages refreshed from guest memory: 5139591 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: program picks: 14068232 kept (5288964 by the last picks' textures, a uniform buffer bound since), 2722 new, 5329045 with other textures (of 19400000)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: 1 textures retired (2638 left, 4528 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: render targets drawn over from another start, let go: 7267 so far (28147 MB), 6723 of them (26568 MB) between the passes
[hos ModuleSystemWorker3] vk: readbacks of GPU results: 870 of 3000 waited for their write-back, 0.01 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 ModuleSystemWorker3] vk: invalidates since the last readbacks line: 8646 kept their virtual core (nothing there to wait for), 2874 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: sparse mirrors: 60, 833 MB with memory of 1352 MB (518 MB spared); blocks given memory later 6161, given back 10365; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 26940; 840 binds (394.8 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 210 taken from the spares, 119 made new; 269 kept as spares, 0 destroyed; 59 spares now, 1012 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1218, 5974.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 295 dirty pages from their submits' captures, 60422 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): 1310 dispatches, 48686 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6105 runs, 1014.8 MB compared; 181 pages read into the shadows; 372 MB of shadows (114 MB of pages known, in 157 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 16968 in captures, 5 only in the collection, 59 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 160 live (2156 MB), 0 retired still imported (0 MB), 41 retired given back (976 MB)
[hos host] vk: replay reads of guest memory as submitted: 9020000 (4844009 pages from the captures)
[hos host] vk: 2 textures retired (2637 left, 4517 MB): budget 0, drawn over 2, dead targets 0
[hos sead::MainThread] nvn: CommandBufferWaitEvent #440000: event 0x100d6c8048 mode 0 value 314628 (now 314609) (lr 0x806661f8)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.04 over 816 ranges, 2785 MB), 204 pages written, 204 copied (0.8 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 0.008, bytes copied 0.000; 25 kept (94 MB), 11 spare buffers, 4444 texture pages copied (17 MB)
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #315000: event 0x100d6c801c mode 0 value 314997 (lr 0x802911b8)
[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: 9040000 (4856820 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 1239, 6077.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 279 dirty pages from their submits' captures, 61331 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): 1322 dispatches, 49454 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6116 runs, 1013.9 MB compared; 170 pages read into the shadows; 372 MB of shadows (113 MB of pages known, in 156 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17238 in captures, 0 only in the collection, 58 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] nvn: DrawMeshTasksIndirect: 720000 so far (x1 0x11791f0000 x2 0x2 x3 0x11791f0000 x4 0x654040)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.00, writes collected 0.04
[hos host] vk: memory: working set 21530 MB (peak 21556), private 12292 MB, staging 32 MB in 2 arenas, 37 made so far (16 ms)
[hos host] vk: memory heaps: heap 0 (device): 216 blocks 6906 MB, 16714 allocations 5977 MB; used 7678 of 10602 MB; heap 1 (system): 15 blocks 944 MB, 39 allocations 858 MB; used 3119 of 48166 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): 27 blocks 864 MB, 152 images 388 MB; 80 free ranges, the largest 32 MB
[hos host] vk: image blocks (textures, compacted): 99 blocks 3168 MB, 2477 images 2967 MB; 1333 free ranges, the largest 19 MB
[hos host] vk: render target images: 4107 reused from the pool, 4807 made new, 4650 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 7678 MB used of 10602 MB allowed (928 MB of it free in the blocks); textures 4534 MB (budget 7164 MB), 2639 textures
[hos host] vk: budget: VRAM 6906 of 10602 MB (7678 MB by the driver, 772 MB unaccounted): textures 4534, blocks free 928, mirrors 833, shadows 372, other buffers 142, other images 63, staging 32; RAM 3048 MB (the process's private memory 12292 MB): imports 2156, capture spares 754, captures 86, capture table 34, other buffers 18, others 0
[hos host] vk: textures: decoded ASTC 2444 MB, rendered 1048 MB (39 drawn in one frame, kept: 31775 KB), other 678 MB; used in the last 8 presents 1910 MB
[hos host] vk: decoded ASTC used again: 390 textures (947 MB) copied from ones the game finalized, not decoded; 390 of those retired, 93 listed; 1171 MB of blocks hashed in 126 ms
[hos host] vk:   288 MB: 0x1147c16000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a22aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 6804 presents ago
[hos host] vk:   214 MB: 0x10a5b22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 6804 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x113cba9000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x100030a4200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1433 presents ago
[hos host] vk:   54 MB: 0x10002fb3c00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 6804 presents ago
[hos host] vk:   54 MB: 0x10002d97a00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 73 (19599 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 50x57x1 NVN 0x25: 1 (11 KB); 62x41x1 NVN 0x25: 1 (9 KB); 74x32x1 NVN 0x25: 1 (9 KB); 90x34x1 NVN 0x25: 1 (11 KB); 114x121x1 NVN 0x1: 1 (13 KB); 126x105x1 NVN 0x1: 1 (12 KB); 138x96x1 NVN 0x1: 1 (12 KB); 140x42x1 NVN 0x25: 1 (22 KB); 154x98x1 NVN 0x1: 1 (14 KB); 165x112x1 NVN 0x1: 1 (18 KB); 198x40x1 NVN 0x25: 2 (61 KB); 204x106x1 NVN 0x1: 1 (21 KB); 242x40x1 NVN 0x25: 3 (113 KB); 244x44x1 NVN 0x25: 1 (41 KB); 260x260x1 NVN 0xd: 36 (6334 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 54 waited for it, 1.50 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 68.4 ms; VRAM 7.5 of 10.4 GB
[hos host] vk: a frame: 28 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 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 3.28, present 0.14, draw 1.67, descriptors 0.71, texture lookups 0.05, uploads 0.00, pipelines 0.07, programs 0.27, commands 0.49, dispatch 0.44, memory sync 0.27, queue submit 0.00, slot wait 0.02
[hos host] vk: a frame: mirrors 329 (833 MB): 4.08 MB synced in 839.0 copies, 17.48 MB written back
[hos host] vk: 25800 presents, 140.4 fps; a frame: 481 draws (0 skipped), 80 dispatches, 133 passes, 139 barriers, 6 submits, 0.0 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 37386 pipelines (4479524 ms), 0 programs (0 failed), 2639 textures (4534 MB)
[hos host] vk: host time in GPU resource waits: 0.08 ms a frame (fences 0.01, timelines 0.07, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 305.6 full barriers recorded, 178.4 left out (nothing recorded since the one before), 12.4 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: 31.1; 0x12: 75.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 3.6 ms a frame (the longest submission 2.0 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.2 MB), ASTC or BC 0.0 (0.0 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: the render thread (600 presents, 7.42 ms a frame): its jobs 3.81 ms a frame (submits 3.20, presents 0.16, others 0.46), idle 3.61; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 2.13 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); 66 captures at a CPU signal; the submits' waits read in 0.020 ms a frame; readbacks: 2982 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 3.28 ms after its hand-over (65.46 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 2.92 + 0.20 (the window's copy is in the idle), after it 0.43 + 0.23; 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): 4.05 parts a frame recorded in the signal's job, 55.1% of them took nothing that came with the signal; of the others, 1.82 a frame took pages of the signal's capture into a mirror (27.9 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 5 of 66 frames, its first three in 5
[hos host] vk: frame ahead, what came with the signal: 66 signals with a capture, each of 77.2 pages in 9.1 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: 0.22+0.14, 0.18+0.04, 0.01+0.03, 0.01+0.02, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00
[hos nn.npln.Shared] vcores: 201097 takes, 45480 waits (mean 81 us, longest 17.8 ms, 0 over HOS_VCORE_LOG_MS), 20763 kicks, 20243 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 1 textures retired (2638 left, 4525 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 9060000 (4869672 pages from the captures)
[hos host] vk: texture slots registered again at another size: 1000000 (of 1252790 registrations; slot 18079 1920x1080 -> 256x256)
[hos host] vk: 1 textures retired (2640 left, 4525 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: sparse mirrors: 60, 833 MB with memory of 1352 MB (518 MB spared); blocks given memory later 6161, given back 10365; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 27642; 840 binds (394.8 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 210 taken from the spares, 119 made new; 269 kept as spares, 0 destroyed; 59 spares now, 1012 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1218, 5974.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 93 dirty pages from their submits' captures, 60935 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): 1326 dispatches, 48917 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6200 runs, 1031.2 MB compared; 189 pages read into the shadows; 372 MB of shadows (114 MB of pages known, in 157 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 16787 in captures, 2 only in the collection, 59 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 160 live (2156 MB), 0 retired still imported (0 MB), 41 retired given back (976 MB)
[hos host] vk: 1 textures retired (2639 left, 4517 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 9080000 (4882490 pages from the captures)
[hos ModuleSystemWorker5] vk: readbacks of GPU results: 838 of 3000 waited for their write-back, 0.01 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: 8519 kept their virtual core (nothing there to wait for), 3001 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos Presentation Thread] vk: capture: a submit 0.11 ms (lock 0.00, scan 0.04 over 818 ranges, 2785 MB), 209 pages written, 209 copied (0.8 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 0.008, bytes copied 0.000; 25 kept (94 MB), 10 spare buffers, 4712 texture pages copied (18 MB)
[hos host] vk: 3 textures retired (2636 left, 4499 MB): budget 0, drawn over 3, dead targets 0
[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: write-backs left out (600 submits): 1239, 6077.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 153 dirty pages from their submits' captures, 61094 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): 1314 dispatches, 49244 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6198 runs, 1031.2 MB compared; 187 pages read into the shadows; 372 MB of shadows (113 MB of pages known, in 157 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17169 in captures, 1 only in the collection, 59 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 9100000 (4895300 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: 1 textures retired (2637 left, 4507 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: 3 textures retired (2634 left, 4490 MB): budget 0, drawn over 3, dead targets 0
[hos host] vk: a frame: 28 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 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 3.67, present 0.16, draw 1.88, descriptors 0.78, texture lookups 0.06, uploads 0.00, pipelines 0.08, programs 0.30, commands 0.57, dispatch 0.49, memory sync 0.31, queue submit 0.00, slot wait 0.02
[hos host] vk: a frame: mirrors 329 (833 MB): 4.08 MB synced in 838.8 copies, 17.48 MB written back
[hos host] vk: 25920 presents, 124.6 fps; a frame: 481 draws (0 skipped), 80 dispatches, 133 passes, 139 barriers, 6 submits, 0.0 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 37386 pipelines (4479524 ms), 0 programs (0 failed), 2636 textures (4501 MB)
[hos host] vk: host time in GPU resource waits: 0.05 ms a frame (fences 0.01, timelines 0.04, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 305.3 full barriers recorded, 179.2 left out (nothing recorded since the one before), 12.4 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: 75.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 3.6 ms a frame (the longest submission 2.0 ms); 0.1 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.2 MB), ASTC or BC 0.0 (0.0 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: shadow staging (600 takes): 0 shadow copies from staging, 8955 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos rmbl::VibrationThread] hid: 32768 vibration values, strongest 0.00 since the last count
[hos host] vk: replay reads of guest memory as submitted: 9120000 (4908140 pages from the captures)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #445000: event 0x100d6c8048 mode 0 value 318428 (now 318409) (lr 0x806661f8)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 265 KB; 0 made on demand (0 MB kept now); c1 stages: at most 105 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 496 sets (ub 5548, sb 2511, cis 42298, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 60, 833 MB with memory of 1352 MB (518 MB spared); blocks given memory later 6161, given back 10365; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 28344; 840 binds (394.8 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 210 taken from the spares, 119 made new; 269 kept as spares, 0 destroyed; 59 spares now, 1012 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1239, 6077.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 282 dirty pages from their submits' captures, 61313 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): 1322 dispatches, 49448 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6104 runs, 1013.9 MB compared; 188 pages read into the shadows; 372 MB of shadows (113 MB of pages known, in 155 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17085 in captures, 0 only in the collection, 58 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 160 live (2156 MB), 0 retired still imported (0 MB), 41 retired given back (976 MB)
[hos host] vk: 3 textures retired (2639 left, 4507 MB): budget 0, drawn over 3, dead targets 0
[hos sead::MainThread] vk: capture: a submit 0.11 ms (lock 0.00, scan 0.04 over 819 ranges, 2783 MB), 209 pages written, 209 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.013; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.011, bytes copied 0.000; 25 kept (94 MB), 11 spare buffers, 5436 texture pages copied (21 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 64 sync waits (14 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: readback pages refreshed from guest memory: 5213776 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 9140000 (4920951 pages from the captures)
[hos ModuleSystemWorker2] nvn: 2250000 recordings made, 67859 live copies (18.8 MB), 3848 texture copies; 3 command buffers' segments (4.1 MB)
[hos host] vk: 1 textures retired (2640 left, 4513 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: 1 textures retired (2640 left, 4511 MB): budget 0, drawn over 1, dead targets 0
[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: write-backs left out (600 submits): 1218, 5974.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 341 dirty pages from their submits' captures, 60448 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): 1312 dispatches, 48706 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6194 runs, 1031.2 MB compared; 175 pages read into the shadows; 372 MB of shadows (114 MB of pages known, in 157 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 16888 in captures, 1 only in the collection, 59 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 855 of 3000 waited for their write-back, 0.01 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 ModuleSystemWorker1] vk: invalidates since the last readbacks line: 8676 kept their virtual core (nothing there to wait for), 3100 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: 9160000 (4933785 pages from the captures)
[hos sead::MainThread] nvn: CommandBufferSignalEvent #320000: event 0x100d6c8030 mode 0 value 319999 (lr 0x806612a8)
[hos host] vk: 1 textures retired (2646 left, 4551 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a frame: 28 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 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 3.42, present 0.13, draw 1.75, descriptors 0.73, texture lookups 0.09, uploads 0.00, pipelines 0.08, programs 0.27, commands 0.51, dispatch 0.45, memory sync 0.33, queue submit 0.00, slot wait 0.02
[hos host] vk: a frame: mirrors 329 (833 MB): 4.09 MB synced in 839.0 copies, 17.48 MB written back
[hos host] vk: 26040 presents, 138.1 fps; a frame: 481 draws (0 skipped), 80 dispatches, 133 passes, 139 barriers, 6 submits, 0.0 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 37386 pipelines (4479524 ms), 0 programs (0 failed), 2646 textures (4551 MB)
[hos host] vk: host time in GPU resource waits: 0.08 ms a frame (fences 0.01, timelines 0.07, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 305.4 full barriers recorded, 178.6 left out (nothing recorded since the one before), 12.3 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: 31.2; 0x12: 75.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 3.7 ms a frame (the longest submission 2.0 ms); 0.1 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.1 (0.7 MB), ASTC or BC 0.0 (0.0 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: shadow staging (600 takes): 0 shadow copies from staging, 10403 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.05 over 811 ranges, 2731 MB), 192 pages written, 192 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.004; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (108 MB), 11 spare buffers, 316 texture pages copied (1 MB)
[hos host] vk: sparse mirrors: 60, 833 MB with memory of 1352 MB (518 MB spared); blocks given memory later 6161, given back 10365; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 29052; 840 binds (394.8 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 210 taken from the spares, 119 made new; 269 kept as spares, 0 destroyed; 59 spares now, 1012 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1239, 6077.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 232 dirty pages from their submits' captures, 61446 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): 1328 dispatches, 49486 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6200 runs, 1031.2 MB compared; 198 pages read into the shadows; 372 MB of shadows (114 MB of pages known, in 158 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 16972 in captures, 1 only in the collection, 59 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 160 live (2156 MB), 0 retired still imported (0 MB), 41 retired given back (976 MB)
[hos host] vk: replay reads of guest memory as submitted: 9180000 (4946624 pages from the captures)
[hos host] vk: 1 textures retired (2648 left, 4568 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: program picks: 14173111 kept (5301789 by the last picks' textures, a uniform buffer bound since), 2722 new, 5424166 with other textures (of 19600000)
[hos host] vk: 1 textures retired (2647 left, 4562 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: 1 textures retired (2646 left, 4556 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.02, writes collected 0.04
[hos host] vk: 3 textures retired (2644 left, 4540 MB): budget 0, drawn over 3, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 9200000 (4959450 pages from the captures)
[hos host] vk: 1 textures retired (2644 left, 4542 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: write-backs left out (600 submits): 1218, 5974.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 213 dirty pages from their submits' captures, 60635 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): 1314 dispatches, 48784 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6094 runs, 1013.8 MB compared; 182 pages read into the shadows; 372 MB of shadows (114 MB of pages known, in 157 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 16850 in captures, 2 only in the collection, 58 only by the replay itself
[hos host] vk:   read live, by pool:
[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 sead::MainThread] nvn: CommandBufferWaitEvent #450000: event 0x100d6c8048 mode 0 value 322228 (now 322209) (lr 0x806661f8)
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 863 of 3000 waited for their write-back, 0.01 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 ModuleSystemWorker1] vk: invalidates since the last readbacks line: 8571 kept their virtual core (nothing there to wait for), 2949 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: 9220000 (4972275 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.04 over 808 ranges, 2721 MB), 207 pages written, 207 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.016; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.013, bytes copied 0.000; 24 kept (92 MB), 12 spare buffers, 4924 texture pages copied (19 MB)
[hos host] vk: a frame: 28 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 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 3.65, present 0.14, draw 1.91, descriptors 0.77, texture lookups 0.09, uploads 0.00, pipelines 0.08, programs 0.31, commands 0.60, dispatch 0.49, memory sync 0.34, queue submit 0.00, slot wait 0.02
[hos host] vk: a frame: mirrors 329 (833 MB): 4.08 MB synced in 838.2 copies, 17.48 MB written back
[hos host] vk: 26160 presents, 133.8 fps; a frame: 481 draws (0 skipped), 80 dispatches, 133 passes, 139 barriers, 6 submits, 0.0 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 37386 pipelines (4479524 ms), 0 programs (0 failed), 2651 textures (4598 MB)
[hos host] vk: host time in GPU resource waits: 0.07 ms a frame (fences 0.01, timelines 0.06, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 305.3 full barriers recorded, 178.8 left out (nothing recorded since the one before), 12.4 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: 31.4; 0x12: 75.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 3.7 ms a frame (the longest submission 2.0 ms); 0.1 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.1 (0.7 MB), ASTC or BC 0.0 (0.0 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: shadow staging (600 takes): 0 shadow copies from staging, 9376 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 1 textures retired (2650 left, 4590 MB): budget 0, drawn over 1, dead targets 0
[hos Presentation Thread] vi: 26170 frames presented (134.9 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 672, 15.5-17.8 1, 17.8-22 1, 22-30 1, 30-40 0, over 40 0
[hos ModuleSystemWorker1] tex: the game's texture heap: 165 compaction rounds asked for in 25200 frames without one running
[hos host] vk: 1 textures retired (2650 left, 4590 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: sparse mirrors: 60, 833 MB with memory of 1352 MB (518 MB spared); blocks given memory later 6161, given back 10365; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 29754; 840 binds (394.8 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 210 taken from the spares, 119 made new; 269 kept as spares, 0 destroyed; 59 spares now, 1012 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1239, 6077.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 289 dirty pages from their submits' captures, 61040 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): 1318 dispatches, 49281 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6202 runs, 1031.2 MB compared; 192 pages read into the shadows; 372 MB of shadows (113 MB of pages known, in 157 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17005 in captures, 3 only in the collection, 59 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 160 live (2156 MB), 0 retired still imported (0 MB), 41 retired given back (976 MB)
[hos host] vk: 2 textures retired (2648 left, 4578 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 9240000 (4985085 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.00, writes collected 0.04
[hos host] vk: 4 textures retired (2644 left, 4545 MB): budget 0, drawn over 4, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 9260000 (4997927 pages from the captures)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 265 KB; 0 made on demand (0 MB kept now); c1 stages: at most 105 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 496 sets (ub 5548, sb 2511, cis 42298, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 1239, 6077.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 313 dirty pages from their submits' captures, 61322 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): 1323 dispatches, 49488 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6104 runs, 1013.8 MB compared; 183 pages read into the shadows; 372 MB of shadows (113 MB of pages known, in 157 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17111 in captures, 1 only in the collection, 58 only by the replay itself
[hos host] vk:   read live, by pool:
[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 Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.04 over 808 ranges, 2721 MB), 197 pages written, 197 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.004, bytes copied 0.000; 25 kept (94 MB), 11 spare buffers, 5044 texture pages copied (19 MB)
[hos host] vk: 3 textures retired (2641 left, 4527 MB): budget 0, drawn over 3, dead targets 0
[hos host] vk: readback pages refreshed from guest memory: 5287489 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 9280000 (5010745 pages from the captures)
[hos host] vk: a frame: 28 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 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 3.34, present 0.13, draw 1.76, descriptors 0.74, texture lookups 0.05, uploads 0.00, pipelines 0.07, programs 0.26, commands 0.54, dispatch 0.45, memory sync 0.27, queue submit 0.00, slot wait 0.02
[hos host] vk: a frame: mirrors 329 (833 MB): 4.08 MB synced in 838.6 copies, 17.48 MB written back
[hos host] vk: 26280 presents, 136.9 fps; a frame: 483 draws (0 skipped), 80 dispatches, 133 passes, 139 barriers, 6 submits, 0.0 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 37386 pipelines (4479524 ms), 0 programs (0 failed), 2641 textures (4527 MB)
[hos host] vk: host time in GPU resource waits: 0.11 ms a frame (fences 0.01, timelines 0.10, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 305.6 full barriers recorded, 179.1 left out (nothing recorded since the one before), 12.4 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: 31.2; 0x12: 75.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 3.6 ms a frame (the longest submission 2.0 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.0 (0.0 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: shadow staging (600 takes): 0 shadow copies from staging, 3252 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 2 textures retired (2641 left, 4522 MB): budget 0, drawn over 2, dead targets 0
[hos ModuleSystemWorker4] nvn: CommandBufferSignalEvent #325000: event 0x100d6c803c mode 0 value 324999 (lr 0x80661bfc)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: sparse mirrors: 60, 833 MB with memory of 1352 MB (518 MB spared); blocks given memory later 6161, given back 10365; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 30456; 840 binds (394.8 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 210 taken from the spares, 119 made new; 269 kept as spares, 0 destroyed; 59 spares now, 1012 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1218, 5974.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 327 dirty pages from their submits' captures, 60516 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): 1311 dispatches, 48750 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6209 runs, 1031.3 MB compared; 189 pages read into the shadows; 372 MB of shadows (114 MB of pages known, in 158 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 16794 in captures, 1 only in the collection, 59 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 160 live (2156 MB), 0 retired still imported (0 MB), 41 retired given back (976 MB)
[hos ModuleSystemWorker5] vk: readbacks of GPU results: 853 of 3000 waited for their write-back, 0.01 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: 8744 kept their virtual core (nothing there to wait for), 3032 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: render targets drawn over from another start, let go: 7267 so far (28147 MB), 6723 of them (26568 MB) between the passes
[hos host] vk: 1 textures retired (2642 left, 4528 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 9300000 (5023539 pages from the captures)
[hos sead::MainThread] nvn: DrawMeshTasksIndirect: 740000 so far (x1 0x117a4f2020 x2 0 x3 0x117a4f2000 x4 0xf33080)
[hos host] vk: 1 textures retired (2642 left, 4525 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: queues ordered by their events: 162000 submissions stopped at a wait, 161998 went on once it was reached, 0 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 0 (stopped 0.00 ms at most); waits met 454646, not met 0; captures at a CPU event's wait 16464, 0 of them at a wait met on arrival
[hos host] vk: 1 textures retired (2641 left, 4519 MB): budget 0, drawn over 1, dead targets 0
[hos sead::MainThread] nvn: CommandBufferWaitEvent #455000: event 0x100d6c8048 mode 0 value 326028 (now 326009) (lr 0x806661f8)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.04 over 808 ranges, 2721 MB), 199 pages written, 199 copied (0.8 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 0.005, bytes copied 0.000; 24 kept (86 MB), 12 spare buffers, 5044 texture pages copied (19 MB)
[hos host] vk: write-backs left out (600 submits): 1239, 6077.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 191 dirty pages from their submits' captures, 61392 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): 1329 dispatches, 49447 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6227 runs, 1031.6 MB compared; 198 pages read into the shadows; 372 MB of shadows (114 MB of pages known, in 158 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17065 in captures, 1 only in the collection, 59 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 9320000 (5036384 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: 1 textures retired (2645 left, 4555 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: memory: working set 21543 MB (peak 21556), private 12304 MB, staging 32 MB in 2 arenas, 37 made so far (16 ms)
[hos host] vk: memory heaps: heap 0 (device): 216 blocks 6906 MB, 16723 allocations 6021 MB; used 7678 of 10601 MB; heap 1 (system): 15 blocks 944 MB, 39 allocations 858 MB; used 3119 of 48166 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): 27 blocks 864 MB, 157 images 432 MB; 81 free ranges, the largest 32 MB
[hos host] vk: image blocks (textures, compacted): 99 blocks 3168 MB, 2481 images 2967 MB; 1337 free ranges, the largest 19 MB
[hos host] vk: render target images: 4109 reused from the pool, 4850 made new, 4688 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 7678 MB used of 10601 MB allowed (884 MB of it free in the blocks); textures 4578 MB (budget 7164 MB), 2648 textures
[hos host] vk: budget: VRAM 6906 of 10601 MB (7678 MB by the driver, 772 MB unaccounted): textures 4578, blocks free 884, mirrors 833, shadows 372, other buffers 142, other images 63, staging 32; RAM 3048 MB (the process's private memory 12304 MB): imports 2156, capture spares 762, captures 86, capture table 34, other buffers 10, others 0
[hos host] vk: textures: decoded ASTC 2444 MB, rendered 1090 MB (39 drawn in one frame, kept: 31775 KB), other 678 MB; used in the last 8 presents 1934 MB
[hos host] vk: decoded ASTC used again: 394 textures (950 MB) copied from ones the game finalized, not decoded; 394 of those retired, 105 listed; 1171 MB of blocks hashed in 126 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 7404 presents ago
[hos host] vk:   214 MB: 0x10a22aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 7404 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: 0x10002fb3c00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 7404 presents ago
[hos host] vk:   54 MB: 0x100030a4200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 2033 presents ago
[hos host] vk:   54 MB: 0x10002d97a00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 73 (19599 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 50x57x1 NVN 0x25: 1 (11 KB); 62x41x1 NVN 0x25: 1 (9 KB); 74x32x1 NVN 0x25: 1 (9 KB); 90x34x1 NVN 0x25: 1 (11 KB); 114x121x1 NVN 0x1: 1 (13 KB); 126x105x1 NVN 0x1: 1 (12 KB); 138x96x1 NVN 0x1: 1 (12 KB); 140x42x1 NVN 0x25: 1 (22 KB); 154x98x1 NVN 0x1: 1 (14 KB); 165x112x1 NVN 0x1: 1 (18 KB); 198x40x1 NVN 0x25: 2 (61 KB); 204x106x1 NVN 0x1: 1 (21 KB); 242x40x1 NVN 0x25: 3 (113 KB); 244x44x1 NVN 0x25: 1 (41 KB); 260x260x1 NVN 0xd: 36 (6334 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 62 waited for it, 0.46 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 29.3 ms; VRAM 7.5 of 10.4 GB
[hos host] vk: a frame: 28 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 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 3.33, present 0.14, draw 1.67, descriptors 0.71, texture lookups 0.08, uploads 0.00, pipelines 0.07, programs 0.27, commands 0.49, dispatch 0.44, memory sync 0.29, queue submit 0.00, slot wait 0.02
[hos host] vk: a frame: mirrors 329 (833 MB): 4.09 MB synced in 839.6 copies, 17.49 MB written back
[hos host] vk: 26400 presents, 137.9 fps; a frame: 483 draws (0 skipped), 80 dispatches, 133 passes, 139 barriers, 6 submits, 0.0 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 37386 pipelines (4479524 ms), 0 programs (0 failed), 2648 textures (4578 MB)
[hos host] vk: host time in GPU resource waits: 0.07 ms a frame (fences 0.01, timelines 0.06, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 306.3 full barriers recorded, 179.2 left out (nothing recorded since the one before), 12.4 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: 31.2; 0x12: 75.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 3.6 ms a frame (the longest submission 2.0 ms); 0.1 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.1 (0.8 MB), ASTC or BC 0.0 (0.0 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: the render thread (600 presents, 7.46 ms a frame): its jobs 3.78 ms a frame (submits 3.24, presents 0.15, others 0.39), idle 3.68; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 2.03 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); 68 captures at a CPU signal; the submits' waits read in 0.021 ms a frame; readbacks: 2959 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 3.25 ms after its hand-over (14.90 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 3.01 + 0.20 (the window's copy is in the idle), after it 0.36 + 0.23; 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): 4.06 parts a frame recorded in the signal's job, 54.3% of them took nothing that came with the signal; of the others, 1.85 a frame took pages of the signal's capture into a mirror (30.6 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 3 of 68 frames, its first three in 3
[hos host] vk: frame ahead, what came with the signal: 68 signals with a capture, each of 83.0 pages in 9.6 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: 0.24+0.14, 0.09+0.04, 0.02+0.03, 0.01+0.02, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 4356 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: replay reads of guest memory as submitted: 9340000 (5049203 pages from the captures)
[hos host] vk: sparse mirrors: 60, 833 MB with memory of 1352 MB (518 MB spared); blocks given memory later 6161, given back 10365; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 31158; 840 binds (394.8 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 210 taken from the spares, 119 made new; 269 kept as spares, 0 destroyed; 59 spares now, 1012 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1239, 6077.5 MB; 34 bindings followed, 42 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 213 dirty pages from their submits' captures, 61495 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): 1323 dispatches, 49520 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6154 runs, 1014.7 MB compared; 175 pages read into the shadows; 372 MB of shadows (113 MB of pages known, in 156 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17235 in captures, 1 only in the collection, 58 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 160 live (2156 MB), 0 retired still imported (0 MB), 41 retired given back (976 MB)
[hos host] control: quit asked by a client
[hos host] vk: indexed descriptor arrays (close): 2610 in 2610 of 52174 program stages, 0 of them near a limit
[hos host] vk: indexed descriptor arrays (close), fragment stages: 2610 texture (17-64: 2610); the longest 64
[hos host] vk: page table census (close): no pool outside its windows, 64-72 GB and 508-512 GB (4261 pools in the table now)
[hos host] vk: pipeline cache: play file 16 saved (3122 KB): 9 pipelines, 9 from the driver's cache, 0 compiled; 8 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)