Beanstalk Longshot test farm
Menu

Report rp_7c22275ad3927e77 pass

#1488 perf(vk): let go of the shader modules of ready programs after the start on Windows too, on request (#1474)

linux on SweatyEggs69/agent-cf0be7,

This report disagrees with its peers.

module-release-tour run_frametime_max_ms: -60.3% against peers' +21.4%
module-release-tour#1 market.hitch_count: +100.0% against peers' -50.0%
module-release-tour#1 market.stutter_count: +100.0% against peers' -33.3%
module-release-tour#1 run_frametime_max_ms: -70.8% against peers' +2.3%

Machine

Commit
ce69221b74cfa84b683fd147eefae9c4645c32a4
Baseline
068fb8ebe232d455ef48e817702b03ccc62d8520
App
nightly-191-38dff5fc, attempt 1
Game build
e0f3ccfa
os
linux
cpu
11th Gen Intel(R) Core(TM) i9-11900KB @ 3.30GHz
gpu
AMD Radeon RX 9070 XT (RADV GFX1201)
tags
linuxamdradvdgpu
driver
26.2.6
ram_gb
32
vram_mb
16,304
gpu_vendor
amd
os_version
<host> (kernel 7.3.4)

Builds

main (baseline)

0 s warm

from 068fb8ebe232

the base was built already

PR

37.8 s warm

Program cache: warm

Scenarios

module-release-tour RX_VK_DROP_MODULES=0

ok
Metric (median)mainPRΔ
exit_code00±0%
field.cpu_s50.8250.92+0.2%
load_time_s4.564.31−5.5%
market.cpu_s84.9486.88+2.3%
village.cpu_s71.6171.1−0.7%
load_time_max_s5.255.02−4.4%
run_hitch_count3027−10%
field.hitch_count00±0%
field.rss_peak_mb12,137.911,948.4−1.6%
log_forbidden_hits00±0%
market.hitch_count02—
market.rss_peak_mb13,490.113,333.7−1.2%
field.stutter_count00±0%
village.hitch_count00±0%
village.rss_peak_mb13,07313,161.9+0.7%
log_expected_missing00±0%
market.stutter_count03—
run_frametime_max_ms528.18209.44−60.3%
village.stutter_count00±0%
field.frametime_max_ms24.5221.76−11.3%
field.frametime_p50_ms16.6816.66−0.2%
field.frametime_p99_ms17.919.04+6.4%
value_modules_released00±0%
value_private_after_mb00±0%
market.frametime_max_ms29.0873.99+154.4%
market.frametime_p50_ms16.6816.65−0.2%
market.frametime_p99_ms19.9521.48+7.7%
value_private_before_mb00±0%
village.frametime_max_ms22.4823.72+5.5%
village.frametime_p50_ms16.5816.61+0.2%
village.frametime_p99_ms2119.46−7.3%
field.frametime_1pct_low_ms19.1520.09+4.9%
market.frametime_1pct_low_ms21.7933.53+53.9%
village.frametime_1pct_low_ms21.4820.35−5.2%
screenshot_diff—0—

Pictures against main: field diff 0 (main against itself ), market diff 0 (main against itself ), village diff 0 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
field.cpu_s50.8250.92
load_time_s4.564.31
market.cpu_s84.9486.88
village.cpu_s71.6171.1
load_time_max_s5.255.02
run_hitch_count3027
field.hitch_count00
field.rss_peak_mb12,137.911,948.4
log_forbidden_hits00
market.hitch_count02
market.rss_peak_mb13,490.113,333.7
field.stutter_count00
village.hitch_count00
village.rss_peak_mb13,07313,161.9
log_expected_missing00
market.stutter_count03
run_frametime_max_ms528.18209.44
village.stutter_count00
field.frametime_max_ms24.5221.76
field.frametime_p50_ms16.6816.66
field.frametime_p99_ms17.919.04
value_modules_released00
value_private_after_mb00
market.frametime_max_ms29.0873.99
market.frametime_p50_ms16.6816.65
market.frametime_p99_ms19.9521.48
value_private_before_mb00
village.frametime_max_ms22.4823.72
village.frametime_p50_ms16.5816.61
village.frametime_p99_ms2119.46
field.frametime_1pct_low_ms19.1520.09
market.frametime_1pct_low_ms21.7933.53
village.frametime_1pct_low_ms21.4820.35
screenshot_diff—0
loadsHyruleField_Child 3.85 sMarket_Child 5.25 sKakarikoVillage_Child 4.57 sHyruleField_Child 3.66 sMarket_Child 5.02 sKakarikoVillage_Child 4.24 s

module-release-tour RX_VK_DROP_MODULES=1

ok
Metric (median)mainPRΔ
exit_code00±0%
field.cpu_s51.2651.07−0.4%
load_time_s4.244.22−0.5%
market.cpu_s86.0884.64−1.7%
village.cpu_s71.370.78−0.7%
load_time_max_s4.945+1.2%
run_hitch_count2928−3.4%
field.hitch_count00±0%
field.rss_peak_mb7,671.37,637.3−0.4%
log_forbidden_hits00±0%
market.hitch_count12+100%
market.rss_peak_mb9,173.39,066.5−1.2%
field.stutter_count00±0%
village.hitch_count00±0%
village.rss_peak_mb8,927.58,923.4−0%
log_expected_missing00±0%
market.stutter_count12+100%
run_frametime_max_ms1,073.34313.35−70.8%
village.stutter_count00±0%
field.frametime_max_ms22.2123.95+7.8%
field.frametime_p50_ms16.6716.7+0.2%
field.frametime_p99_ms17.9918.12+0.7%
value_modules_released62,36062,360±0%
value_private_after_mb01,684—
market.frametime_max_ms71.3664.15−10.1%
market.frametime_p50_ms16.6816.67−0%
market.frametime_p99_ms20.0219.97−0.3%
value_private_before_mb07,399—
village.frametime_max_ms28.722.5−21.6%
village.frametime_p50_ms16.6216.65+0.2%
village.frametime_p99_ms20.0919.23−4.3%
field.frametime_1pct_low_ms18.6718.9+1.2%
market.frametime_1pct_low_ms25.4828.56+12.1%
village.frametime_1pct_low_ms21.5720.08−6.9%
screenshot_diff—0—

Pictures against main: field diff 0 (main against itself ), market diff 0 (main against itself ), village diff 0 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
field.cpu_s51.2651.07
load_time_s4.244.22
market.cpu_s86.0884.64
village.cpu_s71.370.78
load_time_max_s4.945
run_hitch_count2928
field.hitch_count00
field.rss_peak_mb7,671.37,637.3
log_forbidden_hits00
market.hitch_count12
market.rss_peak_mb9,173.39,066.5
field.stutter_count00
village.hitch_count00
village.rss_peak_mb8,927.58,923.4
log_expected_missing00
market.stutter_count12
run_frametime_max_ms1,073.34313.35
village.stutter_count00
field.frametime_max_ms22.2123.95
field.frametime_p50_ms16.6716.7
field.frametime_p99_ms17.9918.12
value_modules_released62,36062,360
value_private_after_mb01,684
market.frametime_max_ms71.3664.15
market.frametime_p50_ms16.6816.67
market.frametime_p99_ms20.0219.97
value_private_before_mb07,399
village.frametime_max_ms28.722.5
village.frametime_p50_ms16.6216.65
village.frametime_p99_ms20.0919.23
field.frametime_1pct_low_ms18.6718.9
market.frametime_1pct_low_ms25.4828.56
village.frametime_1pct_low_ms21.5720.08
screenshot_diff—0
loadsHyruleField_Child 3.65 sMarket_Child 4.94 sKakarikoVillage_Child 4.14 sHyruleField_Child 3.78 sMarket_Child 5 sKakarikoVillage_Child 3.88 s

Rules

ScenarioCellRuleLimitmainPRΔ %Status
module-release-tourRX_VK_DROP_MODULES=0crashFalse00—pass
module-release-tourRX_VK_DROP_MODULES=0screenshot.max_diff0.15—0—pass
module-release-tourRX_VK_DROP_MODULES=0field.rss_peak_mb.max_delta_pct1512,137.911,948.4−1.6%pass
module-release-tourRX_VK_DROP_MODULES=0market.rss_peak_mb.max_delta_pct1513,490.113,333.7−1.2%pass
module-release-tourRX_VK_DROP_MODULES=0village.rss_peak_mb.max_delta_pct1513,07313,161.9+0.7%pass
module-release-tourRX_VK_DROP_MODULES=0load_time_max_s.max_delta_pct305.255.02−4.4%pass
module-release-tourRX_VK_DROP_MODULES=0field.frametime_p99_ms.max_delta_pct1517.919.04+6.4%pass
module-release-tourRX_VK_DROP_MODULES=0market.frametime_p99_ms.max_delta_pct1519.9521.48+7.7%pass
module-release-tourRX_VK_DROP_MODULES=0village.frametime_p99_ms.max_delta_pct152119.46−7.3%pass
module-release-tourRX_VK_DROP_MODULES=0log_forbidden_hits.max000—pass
module-release-tourRX_VK_DROP_MODULES=1crashFalse00—pass
module-release-tourRX_VK_DROP_MODULES=1screenshot.max_diff0.15—0—pass
module-release-tourRX_VK_DROP_MODULES=1field.rss_peak_mb.max_delta_pct157,671.37,637.3−0.4%pass
module-release-tourRX_VK_DROP_MODULES=1market.rss_peak_mb.max_delta_pct159,173.39,066.5−1.2%pass
module-release-tourRX_VK_DROP_MODULES=1village.rss_peak_mb.max_delta_pct158,927.58,923.4±0%pass
module-release-tourRX_VK_DROP_MODULES=1load_time_max_s.max_delta_pct304.945+1.2%pass
module-release-tourRX_VK_DROP_MODULES=1field.frametime_p99_ms.max_delta_pct1517.9918.12+0.7%pass
module-release-tourRX_VK_DROP_MODULES=1market.frametime_p99_ms.max_delta_pct1520.0219.97−0.3%pass
module-release-tourRX_VK_DROP_MODULES=1village.frametime_p99_ms.max_delta_pct1520.0919.23−4.3%pass
module-release-tourRX_VK_DROP_MODULES=1log_forbidden_hits.max000—pass
module-release-tourRX_VK_DROP_MODULES=1field.frametime_p99_ms: RX_VK_DROP_MODULES=1 vs RX_VK_DROP_MODULES=0 (PR)1519.0418.12−4.8%passthe PR's own cells: RX_VK_DROP_MODULES=1 against RX_VK_DROP_MODULES=0 (RX_VK_DROP_MODULES=0)
module-release-tourRX_VK_DROP_MODULES=1market.frametime_p99_ms: RX_VK_DROP_MODULES=1 vs RX_VK_DROP_MODULES=0 (PR)1521.4819.97−7%passthe PR's own cells: RX_VK_DROP_MODULES=1 against RX_VK_DROP_MODULES=0 (RX_VK_DROP_MODULES=0)
module-release-tourRX_VK_DROP_MODULES=1village.frametime_p99_ms: RX_VK_DROP_MODULES=1 vs RX_VK_DROP_MODULES=0 (PR)1519.4619.23−1.2%passthe PR's own cells: RX_VK_DROP_MODULES=1 against RX_VK_DROP_MODULES=0 (RX_VK_DROP_MODULES=0)
module-release-tourRX_VK_DROP_MODULES=1run_hitch_count: RX_VK_DROP_MODULES=1 vs RX_VK_DROP_MODULES=0 (PR)502728+3.7%passthe PR's own cells: RX_VK_DROP_MODULES=1 against RX_VK_DROP_MODULES=0 (RX_VK_DROP_MODULES=0)

Log (PR side, end)

[hos host] vk: program picks: 5807048 kept (2253253 by the last picks' textures, a uniform buffer bound since), 20112 new, 1972839 with other textures (of 7800000)
[hos ModuleSystemWorker5] vk: readbacks of GPU results: 1456 of 3000 waited for their write-back, 0.26 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: 4435 kept their virtual core (nothing there to wait for), 2989 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: 3620000 (2054234 pages from the captures)
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 660, 2883.2 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 839 dirty pages from their submits' captures, 43554 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): 791 dispatches, 33250 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6546 runs, 613.9 MB compared; 68 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14856 in captures, 1 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos host] vk: replay reads of guest memory as submitted: 3640000 (2064791 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 3660000 (2075307 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.11 ms (lock 0.00, scan 0.06 over 682 ranges, 3969 MB), 246 pages written, 246 copied (1.0 MB)
[hos host] vk: a capture taken (ms): all 0.023; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.014, bytes copied 0.000; 25 kept (102 MB), 9 spare buffers, 7484 texture pages copied (29 MB)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 347 KB; 0 made on demand (0 MB kept now); c1 stages: at most 212 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: write-backs left out (600 submits): 659, 2881.5 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 951 dirty pages from their submits' captures, 43925 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): 798 dispatches, 33596 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6526 runs, 615.6 MB compared; 84 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14880 in captures, 0 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 19043 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[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: render targets drawn over from another start, let go: 865 so far (3829 MB), 802 of them (3586 MB) between the passes
[hos host] vk: readback pages refreshed from guest memory: 1595188 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 3680000 (2085831 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.03, writes collected 0.05
[hos sead::MainThread] nvn: CommandBufferWaitEvent #110000: event 0x100d6c8048 mode 0 value 79218 (now 79199) (lr 0x806661f8)
[hos Presentation Thread] vi: 6215 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 18, 15.5-17.8 268, 17.8-22 14, 22-30 0, 30-40 0, over 40 0
[hos host] vk: replay reads of guest memory as submitted: 3700000 (2096395 pages from the captures)
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 641, 2797.6 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 508 dirty pages from their submits' captures, 42823 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): 780 dispatches, 32580 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6375 runs, 596.5 MB compared; 67 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14465 in captures, 2 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos host] vk: program picks: 5955354 kept (2313055 by the last picks' textures, a uniform buffer bound since), 20112 new, 2024533 with other textures (of 8000000)
[hos host] vk: replay reads of guest memory as submitted: 3720000 (2106889 pages from the captures)
[hos host] vk: a frame: 193 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 83 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 9.06, present 0.20, draw 5.41, descriptors 2.13, texture lookups 0.08, uploads 0.00, pipelines 0.29, programs 0.98, commands 1.61, dispatch 0.72, memory sync 0.45, queue submit 1.08, slot wait 0.03
[hos host] vk: a frame: mirrors 227 (962 MB): 5.30 MB synced in 1015.2 copies, 18.64 MB written back
[hos host] vk: 6240 presents, 60.0 fps; a frame: 2154 draws (0 skipped), 168 dispatches, 187 passes, 146 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37379 pipelines (1845 ms), 0 programs (10 failed), 4521 textures (8362 MB)
[hos host] vk: host time in GPU resource waits: 2.59 ms a frame (fences 0.00, timelines 2.59, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 662.3 full barriers recorded, 267.2 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 42.0; 0x12: 73.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 10.1 ms a frame (the longest submission 4.1 ms); 0.0 textures made a frame
[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: 3740000 (2117456 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 659, 2881.5 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 406 dirty pages from their submits' captures, 44466 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): 798 dispatches, 33613 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6553 runs, 615.9 MB compared; 72 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14919 in captures, 2 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker3] nvn: CommandBufferSignalEvent #80000: event 0x100d6c8004 mode 0 value 79999 (lr 0x80291504)
[hos host] vk: replay reads of guest memory as submitted: 3760000 (2128580 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.11 ms (lock 0.00, scan 0.06 over 682 ranges, 3969 MB), 226 pages written, 226 copied (0.9 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 0.008, bytes copied 0.000; 25 kept (110 MB), 8 spare buffers, 2903 texture pages copied (11 MB)
[hos host] vk: replay reads of guest memory as submitted: 3780000 (2139262 pages from the captures)
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 641, 2797.6 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 767 dirty pages from their submits' captures, 43372 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): 791 dispatches, 32986 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6394 runs, 596.0 MB compared; 77 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14973 in captures, 1 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 13497 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[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 0.02, writes collected 0.05
[hos host] vk: replay reads of guest memory as submitted: 3800000 (2149927 pages from the captures)
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 1446 of 3000 waited for their write-back, 0.27 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: 4696 kept their virtual core (nothing there to wait for), 2984 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: write-backs left out (600 submits): 660, 2883.2 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 601 dirty pages from their submits' captures, 44157 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): 798 dispatches, 33576 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6587 runs, 614.7 MB compared; 73 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14491 in captures, 2 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: program picks: 6103361 kept (2372776 by the last picks' textures, a uniform buffer bound since), 20112 new, 2076526 with other textures (of 8200000)
[hos host] vk: replay reads of guest memory as submitted: 3820000 (2160594 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 3840000 (2171266 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 347 KB; 0 made on demand (0 MB kept now); c1 stages: at most 212 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 659, 2881.5 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 561 dirty pages from their submits' captures, 43508 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): 781 dispatches, 33121 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6580 runs, 615.2 MB compared; 74 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14975 in captures, 0 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos host] vk: replay reads of guest memory as submitted: 3860000 (2182003 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 1665082 pages so far (0 held: may be written ahead in their submission)
[hos ModuleSystemWorker2] nvn: 900000 recordings made, 58249 live copies (20.6 MB), 638 texture copies; 6 command buffers' segments (4.3 MB)
[hos host] vk: a frame: 192 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 83 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 9.04, present 0.20, draw 5.41, descriptors 2.09, texture lookups 0.08, uploads 0.00, pipelines 0.29, programs 1.00, commands 1.63, dispatch 0.72, memory sync 0.47, queue submit 1.06, slot wait 0.03
[hos host] vk: a frame: mirrors 227 (962 MB): 5.31 MB synced in 1015.3 copies, 18.63 MB written back
[hos host] vk: 6360 presents, 60.0 fps; a frame: 2153 draws (0 skipped), 169 dispatches, 186 passes, 146 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37379 pipelines (1845 ms), 0 programs (10 failed), 4521 textures (8362 MB)
[hos host] vk: host time in GPU resource waits: 2.53 ms a frame (fences 0.00, timelines 2.53, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 663.2 full barriers recorded, 267.0 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 42.0; 0x12: 73.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 10.1 ms a frame (the longest submission 4.1 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos nn.npln.Callback] vcores: 105603 takes, 16135 waits (mean 44 us, longest 4.7 ms, 0 over HOS_VCORE_LOG_MS), 7214 kicks, 7183 yields, 0 starved grants, 0 escapes in 5 s
[hos Presentation Thread] vk: capture: a submit 0.12 ms (lock 0.00, scan 0.06 over 682 ranges, 3969 MB), 251 pages written, 251 copied (1.0 MB)
[hos host] vk: a capture taken (ms): all 0.021; 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 (122 MB), 8 spare buffers, 8561 texture pages copied (33 MB)
[hos host] vk: replay reads of guest memory as submitted: 3880000 (2192618 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 641, 2797.6 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 440 dirty pages from their submits' captures, 43617 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): 789 dispatches, 32917 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6402 runs, 595.9 MB compared; 74 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14926 in captures, 2 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 3900000 (2203344 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 13358 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[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 0.02, writes collected 0.05
[hos host] vk: replay reads of guest memory as submitted: 3920000 (2214062 pages from the captures)
[hos host] vk: program picks: 6251293 kept (2432423 by the last picks' textures, a uniform buffer bound since), 20112 new, 2128594 with other textures (of 8400000)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #115000: event 0x100d6c8048 mode 0 value 83018 (now 82999) (lr 0x806661f8)
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 659, 2881.5 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 1072 dirty pages from their submits' captures, 43118 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): 779 dispatches, 33081 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6561 runs, 616.4 MB compared; 72 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14487 in captures, 2 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos host] vk: replay reads of guest memory as submitted: 3940000 (2224740 pages from the captures)
[hos sead::MainThread] nvn: GPU waits in 5 s: 72 sync waits (85 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: replay reads of guest memory as submitted: 3960000 (2235433 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 641, 2797.6 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 429 dirty pages from their submits' captures, 43512 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): 791 dispatches, 32878 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6373 runs, 596.6 MB compared; 75 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14800 in captures, 1 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 3980000 (2246127 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.11 ms (lock 0.00, scan 0.06 over 682 ranges, 3969 MB), 227 pages written, 227 copied (0.9 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 0.007, bytes copied 0.000; 25 kept (114 MB), 10 spare buffers, 8179 texture pages copied (31 MB)
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1446 of 3000 waited for their write-back, 0.27 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: 4537 kept their virtual core (nothing there to wait for), 3143 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: 4000000 (2256816 pages from the captures)
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 660, 2883.2 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 667 dirty pages from their submits' captures, 43288 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): 789 dispatches, 33011 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6557 runs, 615.1 MB compared; 84 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14361 in captures, 1 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos host] vk: a frame: 191 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 83 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 9.04, present 0.19, draw 5.43, descriptors 2.11, texture lookups 0.09, uploads 0.00, pipelines 0.29, programs 1.00, commands 1.63, dispatch 0.72, memory sync 0.48, queue submit 1.08, slot wait 0.03
[hos host] vk: a frame: mirrors 227 (962 MB): 5.29 MB synced in 1012.3 copies, 18.66 MB written back
[hos host] vk: 6480 presents, 59.9 fps; a frame: 2148 draws (0 skipped), 168 dispatches, 187 passes, 146 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37379 pipelines (1845 ms), 0 programs (10 failed), 4523 textures (8362 MB)
[hos host] vk: host time in GPU resource waits: 2.66 ms a frame (fences 0.00, timelines 2.66, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 662.9 full barriers recorded, 267.3 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 42.0; 0x12: 73.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 10.1 ms a frame (the longest submission 4.1 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos ModuleSystemWorker1] nvn: DrawMeshTasksIndirect: 200000 so far (x1 0x117d4a3040 x2 0 x3 0x117d4a3040 x4 0x44b040)
[hos host] vk: replay reads of guest memory as submitted: 4020000 (2267525 pages from the captures)
[hos host] vk: program picks: 6399262 kept (2492339 by the last picks' textures, a uniform buffer bound since), 20112 new, 2180625 with other textures (of 8600000)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.03, writes collected 0.05
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 17422 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[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: 4040000 (2278201 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 347 KB; 0 made on demand (0 MB kept now); c1 stages: at most 212 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: write-backs left out (600 submits): 659, 2881.5 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 678 dirty pages from their submits' captures, 43874 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): 791 dispatches, 33372 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6541 runs, 616.1 MB compared; 68 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14699 in captures, 1 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker5] nvn: CommandBufferSignalEvent #85000: event 0x100d6c8014 mode 0 value 85000 (lr 0x80660bcc)
[hos Presentation Thread] vi: 6516 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 17, 15.5-17.8 267, 17.8-22 17, 22-30 0, 30-40 0, over 40 0
[hos host] vk: readback pages refreshed from guest memory: 1735200 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 4060000 (2288911 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 4080000 (2299642 pages from the captures)
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 641, 2797.6 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 366 dirty pages from their submits' captures, 43281 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): 791 dispatches, 32678 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6367 runs, 595.4 MB compared; 73 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14856 in captures, 2 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos Presentation Thread] vk: capture: a submit 0.11 ms (lock 0.00, scan 0.06 over 682 ranges, 3965 MB), 248 pages written, 248 copied (1.0 MB)
[hos host] vk: a capture taken (ms): all 0.023; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.015, bytes copied 0.000; 25 kept (130 MB), 10 spare buffers, 7758 texture pages copied (30 MB)
[hos host] vk: replay reads of guest memory as submitted: 4100000 (2310533 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 4120000 (2321130 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 660, 2883.2 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 811 dirty pages from their submits' captures, 43614 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): 791 dispatches, 33260 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6538 runs, 615.6 MB compared; 62 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14413 in captures, 1 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: program picks: 6547223 kept (2552200 by the last picks' textures, a uniform buffer bound since), 20112 new, 2232664 with other textures (of 8800000)
[hos host] vk: replay reads of guest memory as submitted: 4140000 (2331781 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.02, writes collected 0.05
[hos host] vk: memory: working set 8918 MB (peak 9061), private 2321 MB, staging 256 MB in 16 arenas, 37 made so far (1 ms)
[hos host] vk: memory heaps: heap 0 (system): 16 blocks 960 MB, 39 allocations 880 MB; used 3308 of 15604 MB; heap 1 (device): 328 blocks 10787 MB, 20216 allocations 10140 MB; used 11565 of 15545 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, 127 images 327 MB; 72 free ranges, the largest 32 MB
[hos host] vk: image blocks (textures, compacted): 209 blocks 6688 MB, 4380 images 6630 MB; 1000 free ranges, the largest 19 MB
[hos host] vk: render target images: 510 reused from the pool, 601 made new, 468 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 11565 MB used of 15545 MB allowed (646 MB of it free in the blocks); textures 8362 MB (budget 11000 MB), 4523 textures
[hos host] vk: budget: VRAM 10787 of 15545 MB (11565 MB by the driver, 777 MB unaccounted): textures 8362, mirrors 962, blocks free 646, shadows 300, staging 256, other buffers 227, other images 31; RAM 3247 MB (the process's private memory 2321 MB): imports 2332, capture spares 752, captures 110, capture table 35, other buffers 18, others 0
[hos host] vk: textures: decoded ASTC 5760 MB, rendered 1092 MB (106 drawn in one frame, kept: 51568 KB), other 1259 MB; used in the last 8 presents 2376 MB
[hos host] vk: decoded ASTC used again: 65 textures (158 MB) copied from ones the game finalized, not decoded; 65 of those retired, 573 listed; 1104 MB of blocks hashed in 164 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 1 presents ago
[hos host] vk:   214 MB: 0x10a22aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, 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:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0xa, last used 4888 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: 0x10005192200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10005552400 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 3048 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 13 (2113 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 260x260x1 NVN 0xd: 12 (2111 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 126 waited for it, 1.13 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 20.2 ms; VRAM 11.3 of 15.2 GB
[hos host] vk: a frame: 191 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 83 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 9.00, present 0.20, draw 5.34, descriptors 2.08, texture lookups 0.08, uploads 0.00, pipelines 0.28, programs 0.98, commands 1.60, dispatch 0.72, memory sync 0.45, queue submit 1.08, slot wait 0.03
[hos host] vk: a frame: mirrors 227 (962 MB): 5.29 MB synced in 1012.4 copies, 18.63 MB written back
[hos host] vk: 6600 presents, 60.1 fps; a frame: 2149 draws (0 skipped), 168 dispatches, 187 passes, 146 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37379 pipelines (1845 ms), 0 programs (10 failed), 4523 textures (8362 MB)
[hos host] vk: host time in GPU resource waits: 2.45 ms a frame (fences 0.00, timelines 2.45, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 661.9 full barriers recorded, 267.3 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 42.0; 0x12: 73.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 10.1 ms a frame (the longest submission 4.1 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: the render thread (600 presents, 16.67 ms a frame): its jobs 9.93 ms a frame (submits 4.54, presents 0.21, others 5.19), idle 6.73; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 4.03 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); 363 captures at a CPU signal; the submits' waits read in 0.099 ms a frame; readbacks: 3000 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 8.96 ms after its hand-over (11.52 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 3.26 + 0.46 (the window's copy is in the idle), after it 2.35 + 0.52; after a frame's last recording that reads or writes the game's memory come 1.01 recordings (0.02 ms of GPU time), after its last that reads it 1.01 (0.02 ms)
[hos host] vk: frame ahead (600 presents): 11.83 parts a frame recorded in the signal's job, 89.8% of them took nothing that came with the signal; of the others, 1.20 a frame took pages of the signal's capture into a mirror (28.9 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 79 of 363 frames, its first three in 79
[hos host] vk: frame ahead, what came with the signal: 363 signals with a capture, each of 40.0 pages in 4.4 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: 0.47+0.04, 0.05+0.02, 0.16+0.06, 0.42+0.21, 0.13+0.04, 0.38+0.03, 0.21+0.02, 0.06+0.02, 0.24+0.02, 0.08+0.02, 0.09+0.02, 0.08+0.01
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 640, 2795.9 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 972 dirty pages from their submits' captures, 43208 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): 793 dispatches, 32918 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6361 runs, 596.5 MB compared; 63 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14923 in captures, 2 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #120000: event 0x100d6c8048 mode 0 value 86818 (now 86799) (lr 0x806661f8)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 14498 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[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: 4160000 (2342371 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 4180000 (2352991 pages from the captures)
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1443 of 3000 waited for their write-back, 0.26 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: 4465 kept their virtual core (nothing there to wait for), 2959 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.06 over 683 ranges, 3967 MB), 224 pages written, 224 copied (0.9 MB)
[hos host] vk: a capture taken (ms): all 0.014; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.007, bytes copied 0.000; 25 kept (118 MB), 10 spare buffers, 2112 texture pages copied (8 MB)
[hos host] vk: write-backs left out (600 submits): 660, 2883.2 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 901 dirty pages from their submits' captures, 43160 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): 783 dispatches, 33046 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6537 runs, 613.8 MB compared; 69 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14331 in captures, 0 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 4200000 (2363603 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 4220000 (2374181 pages from the captures)
[hos ModuleSystemWorker3] vcores: 104756 takes, 16763 waits (mean 44 us, longest 6.2 ms, 0 over HOS_VCORE_LOG_MS), 7503 kicks, 7497 yields, 0 starved grants, 0 escapes in 5 s
[hos Presentation Thread] nvn: QueueSubmitCommands #40000: queue 0x101004133020 (its #6674), 1 handles (lr 0x80590358)
[hos host] vk: program picks: 6695161 kept (2611946 by the last picks' textures, a uniform buffer bound since), 20112 new, 2284726 with other textures (of 9000000)
[hos host] vk: recorder: 8153451 packets run in 10.0 s (busy 1.17 s, spun idle 4.48 s, slept 4161 times, 1739 of them under 200 us), 0 calls made directly (no room); the replay waited for it 29688 times, 53.53 ms in all, the longest 0.11 ms; vkEndCommandBuffer 11028/25.11 ms (max 0.11), vkResetDescriptorPool 10984/17.84 ms (max 0.09), replay_mu unlock 7560/10.51 ms (max 0.11), vkQueueSubmit2 66/0.04 ms (max 0.00), vmaInvalidateAllocation 50/0.03 ms (max 0.00)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 347 KB; 0 made on demand (0 MB kept now); c1 stages: at most 212 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 660, 2883.2 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 569 dirty pages from their submits' captures, 44113 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): 817 dispatches, 33481 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6539 runs, 613.9 MB compared; 84 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14778 in captures, 0 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos host] vk: replay reads of guest memory as submitted: 4240000 (2384797 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 1805290 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 4260000 (2395444 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.02, writes collected 0.05
[hos host] vk: write-backs left out (600 submits): 659, 2881.5 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 554 dirty pages from their submits' captures, 44063 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): 798 dispatches, 33447 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6527 runs, 613.8 MB compared; 72 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14790 in captures, 1 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 4280000 (2406037 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 11311 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[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 frame: 191 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 83 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 9.05, present 0.22, draw 5.44, descriptors 2.15, texture lookups 0.13, uploads 0.02, pipelines 0.29, programs 0.99, commands 1.61, dispatch 0.72, memory sync 0.44, queue submit 1.08, slot wait 0.03
[hos host] vk: a frame: mirrors 227 (962 MB): 5.29 MB synced in 1012.1 copies, 18.60 MB written back
[hos host] vk: 6720 presents, 60.0 fps; a frame: 2146 draws (0 skipped), 168 dispatches, 186 passes, 146 barriers, 6 submits, 0.5 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37379 pipelines (1845 ms), 0 programs (10 failed), 4552 textures (8367 MB)
[hos host] vk: host time in GPU resource waits: 2.36 ms a frame (fences 0.00, timelines 2.36, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 662.0 full barriers recorded, 266.9 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 42.0; 0x12: 73.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 10.1 ms a frame (the longest submission 4.1 ms); 0.5 textures made a frame
[hos host] vk: a frame: 0.04 MB unswizzled on the CPU in 0.00 ms
[hos sead::MainThread] vk: capture: a submit 0.11 ms (lock 0.00, scan 0.06 over 685 ranges, 3977 MB), 232 pages written, 232 copied (0.9 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 0.008, bytes copied 0.000; 25 kept (136 MB), 11 spare buffers, 4738 texture pages copied (18 MB)
[hos host] vk: replay reads of guest memory as submitted: 4300000 (2416636 pages from the captures)
[hos sead::MainThread] nvn: GPU waits in 5 s: 70 sync waits (78 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 641, 2797.6 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 678 dirty pages from their submits' captures, 43407 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): 797 dispatches, 32935 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6364 runs, 595.4 MB compared; 68 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14889 in captures, 1 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos ModuleSystemWorker4] nvn: 950000 recordings made, 58233 live copies (20.4 MB), 638 texture copies; 5 command buffers' segments (4.2 MB)
[hos host] vk: replay reads of guest memory as submitted: 4320000 (2427311 pages from the captures)
[hos host] vk: program picks: 6843096 kept (2671733 by the last picks' textures, a uniform buffer bound since), 20114 new, 2336789 with other textures (of 9200000)
[hos host] vk: replay reads of guest memory as submitted: 4340000 (2437874 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 659, 2881.5 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 1474 dirty pages from their submits' captures, 42835 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): 780 dispatches, 33093 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6339 runs, 597.1 MB compared; 63 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14481 in captures, 0 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker2] nvn: CommandBufferSignalEvent #90000: event 0x100d6c8020 mode 0 value 90000 (lr 0x80664f14)
[hos host] vk: replay reads of guest memory as submitted: 4360000 (2448491 pages from the captures)
[hos host] vk: render targets drawn over from another start, let go: 865 so far (3829 MB), 802 of them (3586 MB) between the passes
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 1454 of 3000 waited for their write-back, 0.25 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: 4284 kept their virtual core (nothing there to wait for), 2884 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 0.02, writes collected 0.05
[hos host] vk: replay reads of guest memory as submitted: 4380000 (2459147 pages from the captures)
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 641, 2797.6 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 659 dirty pages from their submits' captures, 42754 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): 777 dispatches, 32531 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6543 runs, 613.8 MB compared; 70 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14430 in captures, 0 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #125000: event 0x100d6c8048 mode 0 value 90618 (now 90599) (lr 0x806661f8)
[hos Presentation Thread] vi: 6816 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 16, 15.5-17.8 271, 17.8-22 13, 22-30 0, 30-40 0, over 40 0
[hos Presentation Thread] vk: capture: a submit 0.11 ms (lock 0.00, scan 0.06 over 685 ranges, 3989 MB), 225 pages written, 225 copied (0.9 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 0.006, bytes copied 0.000; 24 kept (118 MB), 11 spare buffers, 4881 texture pages copied (19 MB)
[hos host] vk: replay reads of guest memory as submitted: 4400000 (2469665 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 7938 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[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: 4420000 (2480290 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 347 KB; 0 made on demand (0 MB kept now); c1 stages: at most 212 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: write-backs left out (600 submits): 660, 2883.2 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 520 dirty pages from their submits' captures, 44113 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): 797 dispatches, 33519 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6533 runs, 613.9 MB compared; 80 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14822 in captures, 1 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a frame: 191 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 83 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 9.13, present 0.21, draw 5.55, descriptors 2.23, texture lookups 0.16, uploads 0.06, pipelines 0.29, programs 1.01, commands 1.63, dispatch 0.71, memory sync 0.44, queue submit 1.07, slot wait 0.03
[hos host] vk: a frame: mirrors 227 (962 MB): 5.30 MB synced in 1013.9 copies, 18.62 MB written back
[hos host] vk: 6840 presents, 60.0 fps; a frame: 2148 draws (0 skipped), 168 dispatches, 187 passes, 146 barriers, 6 submits, 0.8 uploads (0.1 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37379 pipelines (1845 ms), 0 programs (10 failed), 4596 textures (8376 MB)
[hos host] vk: host time in GPU resource waits: 2.58 ms a frame (fences 0.00, timelines 2.58, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 663.5 full barriers recorded, 267.1 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 42.0; 0x12: 73.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 10.0 ms a frame (the longest submission 4.1 ms); 0.7 textures made a frame
[hos host] vk: a frame: 0.07 MB unswizzled on the CPU in 0.01 ms
[hos host] vk: readback pages refreshed from guest memory: 1875714 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: program picks: 6990984 kept (2731505 by the last picks' textures, a uniform buffer bound since), 20114 new, 2388901 with other textures (of 9400000)
[hos host] vk: replay reads of guest memory as submitted: 4440000 (2491070 pages from the captures)
[hos host] vk: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 640, 2795.9 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 363 dirty pages from their submits' captures, 43799 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): 792 dispatches, 32953 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6368 runs, 595.9 MB compared; 68 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14841 in captures, 1 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos host] vk: replay reads of guest memory as submitted: 4460000 (2501789 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 4480000 (2512496 pages from the captures)
[hos host] vk: queues ordered by their events: 42000 submissions stopped at a wait, 41853 went on once it was reached, 144 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 11 (stopped 50.15 ms at most); waits met 126858, not met 0; captures at a CPU event's wait 2323, 0 of them at a wait met on arrival
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.02, writes collected 0.06
[hos host] vk: write-backs left out (600 submits): 660, 2883.2 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 363 dirty pages from their submits' captures, 43908 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): 794 dispatches, 33282 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6546 runs, 613.9 MB compared; 79 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14332 in captures, 0 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 4500000 (2523219 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.11 ms (lock 0.00, scan 0.06 over 685 ranges, 3989 MB), 224 pages written, 224 copied (0.9 MB)
[hos host] vk: a capture taken (ms): all 0.012; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.006, bytes copied 0.000; 25 kept (118 MB), 11 spare buffers, 5026 texture pages copied (19 MB)
[hos ModuleSystemWorker5] nvn: DrawMeshTasks: 220000 so far (x1 0 x2 0x10 x3 0x191008eb x4 0x2c0)
[hos rmbl::VibrationThread] hid: 20480 vibration values, strongest 0.00 since the last count
[hos host] vk: replay reads of guest memory as submitted: 4520000 (2533930 pages from the captures)
[hos ModuleSystemWorker4] barrier: 174874 looks: 43802 found the masks moved already, 54399 waited until they moved, 76673 for the whole 1 ms (a wait 728 us on average); 64306 wakes
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 5209 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[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: sparse mirrors: 88, 962 MB with memory of 1556 MB (593 MB spared); blocks given memory later 7522, given back 13552; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 9414; 561 binds (195.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 93 taken from the spares, 134 made new; 139 kept as spares, 0 destroyed; 46 spares now, 1240 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 660, 2883.2 MB; 31 bindings followed, 47 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 593 dirty pages from their submits' captures, 43635 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): 790 dispatches, 33175 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6562 runs, 616.2 MB compared; 70 pages read into the shadows; 300 MB of shadows; pages the shadows knew that the CPU wrote: 14793 in captures, 1 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 155 live (2332 MB), 0 retired still imported (0 MB), 50 retired given back (1220 MB)
[hos host] vk: program picks: 7138953 kept (2791299 by the last picks' textures, a uniform buffer bound since), 20114 new, 2440932 with other textures (of 9600000)
[hos host] vk: replay reads of guest memory as submitted: 4540000 (2544557 pages from the captures)
[hos ModuleSystemWorker5] vk: readbacks of GPU results: 1458 of 3000 waited for their write-back, 0.26 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: 4702 kept their virtual core (nothing there to wait for), 2978 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] control: quit asked by a client
[hos host] vk: indexed descriptor arrays (close): 2631 in 2631 of 55732 program stages, 0 of them near a limit
[hos host] vk: indexed descriptor arrays (close), fragment stages: 2631 texture (17-64: 2631); the longest 64
[hos host] vk: page table census (close): no pool outside its windows, 64-72 GB and 508-512 GB (4295 pools in the table now)
[hos host] control: exiting
[hos host] stop: close work done (a file the game was writing waited for, at most 3 s)