Beanstalk Longshot test farm
Menu

Report rp_1af8f166b08fd0e5 fail

#518 WIP: fix(vk): cap the render-target image pool and count it in the texture budget

windows+vram8gb on AstraLol/agent-81a0b2,

Machine

Commit
dee0fff09c854122db742e8ff0d9d65b91b3cc5c
Baseline
dbad7d1260e405642086823816950c0775486af3
App
nightly-126-05954779, attempt 1
Game build
e0f3ccfa
os
windows
cpu
AMD Ryzen 9 5900HX with Radeon Graphics
gpu
NVIDIA GeForce RTX 3070 Laptop GPU
tags
windowsnvidianvidia-proprietarydgpu
driver
621.24.0
ram_gb
32
vram_mb
8,018
gpu_vendor
nvidia
os_version
Windows 10.0.26300

Builds

main (baseline)

0 s warm

from dbad7d1260e4

the base was built already

PR

32.5 s warm

Program cache: warm

Scenarios

field-market-sweep

ok
Metric (median)mainPRΔ
exit_code00±0%
field.cpu_s169.28150.89−10.9%
load_time_s7.016.5−7.3%
market.cpu_s199.44188.84−5.3%
return.cpu_s87.7782.62−5.9%
load_time_max_s9.39.01−3.1%
field.hitch_count00±0%
field.rss_peak_mb12,235.712,170.9−0.5%
field.vram_mean_mb6,294.76,239.3−0.9%
field.vram_peak_mb6,5086,474−0.5%
log_forbidden_hits00±0%
market.hitch_count00±0%
market.rss_peak_mb14,004.213,866.5−1%
return.hitch_count01—
return.rss_peak_mb14,167.813,710.9−3.2%
field.stutter_count00±0%
market.vram_mean_mb6,7266,494−3.4%
market.vram_peak_mb6,7906,494−4.4%
return.vram_mean_mb6,3616,104−4%
return.vram_peak_mb6,3616,104−4%
log_expected_missing00±0%
market.stutter_count00±0%
return.stutter_count03—
field.frametime_max_ms28.1124.53−12.7%
field.frametime_p50_ms16.5116.61+0.6%
field.frametime_p99_ms24.8120.87−15.9%
market.frametime_max_ms43.6542.86−1.8%
market.frametime_p50_ms23.722.03−7.1%
market.frametime_p99_ms36.6733.91−7.5%
return.frametime_max_ms30.9157.44+85.8%
return.frametime_p50_ms16.3716.89+3.2%
return.frametime_p99_ms23.2326.95+16%
field.frametime_1pct_low_ms25.9421.7−16.4%
market.frametime_1pct_low_ms39.6136.6−7.6%
return.frametime_1pct_low_ms25.4832.93+29.2%
screenshot_diff—0.75—

Pictures against main: field diff 0.2959, market diff 0.7477, field-return diff 0

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
field.cpu_s169.28150.89
load_time_s7.016.5
market.cpu_s199.44188.84
return.cpu_s87.7782.62
load_time_max_s9.39.01
field.hitch_count00
field.rss_peak_mb12,235.712,170.9
field.vram_mean_mb6,294.76,239.3
field.vram_peak_mb6,5086,474
log_forbidden_hits00
market.hitch_count00
market.rss_peak_mb14,004.213,866.5
return.hitch_count01
return.rss_peak_mb14,167.813,710.9
field.stutter_count00
market.vram_mean_mb6,7266,494
market.vram_peak_mb6,7906,494
return.vram_mean_mb6,3616,104
return.vram_peak_mb6,3616,104
log_expected_missing00
market.stutter_count00
return.stutter_count03
field.frametime_max_ms28.1124.53
field.frametime_p50_ms16.5116.61
field.frametime_p99_ms24.8120.87
market.frametime_max_ms43.6542.86
market.frametime_p50_ms23.722.03
market.frametime_p99_ms36.6733.91
return.frametime_max_ms30.9157.44
return.frametime_p50_ms16.3716.89
return.frametime_p99_ms23.2326.95
field.frametime_1pct_low_ms25.9421.7
market.frametime_1pct_low_ms39.6136.6
return.frametime_1pct_low_ms25.4832.93
screenshot_diff—0.75
loadsHyruleField_Child 6.31 sMarket_Child 9.3 sHyruleField_Child 5.43 sHyruleField_Child 5.4 sMarket_Child 9.01 sHyruleField_Child 5.09 s

Rules

ScenarioCellRuleLimitmainPRΔ %Status
field-market-sweepcrashFalse00—pass
field-market-sweepscreenshot.max_diff0.05—0.75—fail
field-market-sweepfield.vram_peak_mb.max_delta_pct106,5086,474−0.5%pass
field-market-sweepmarket.vram_peak_mb.max_delta_pct106,7906,494−4.4%pass
field-market-sweepreturn.vram_peak_mb.max_delta_pct106,3616,104−4%pass
field-market-sweepload_time_max_s.max_delta_pct209.39.01−3.1%pass
field-market-sweepfield.frametime_p99_ms.max_delta_pct1524.8120.87−15.9%pass
field-market-sweepmarket.frametime_p99_ms.max_delta_pct1536.6733.91−7.5%pass
field-market-sweepreturn.frametime_p99_ms.max_delta_pct1523.2326.95+16%fail
field-market-sweeplog_forbidden_hits.max000—pass

Log (PR side, end)

[hos host] vk: a frame: 502.2 full barriers recorded, 187.0 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 16.5 ms a frame (the longest submission 12.4 ms); 0.2 textures made a frame
[hos host] vk: a frame: 0.18 MB unswizzled on the CPU in 0.06 ms
[hos host] vk: sparse mirrors: 113, 1203 MB with memory of 1728 MB (524 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1298 dirty pages from their submits' captures, 52045 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: write-back diff (600 submits): 5494 runs, 2046.9 MB compared; 47 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16365 in captures, 0 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos host] vk: mirror of pool 0x7fe2c00000 (4 MB) given back: idle, 6043 submissions since its sync
[hos host] vk: mirror of pool 0x7fad000000 (4 MB) given back: idle, 6002 submissions since its sync
[hos host] vk: replay reads of guest memory as submitted: 3560000 (1888127 pages from the captures)
[hos host] vk: mirror of pool 0x7f73400000 (4 MB) given back: idle, 6041 submissions since its sync
[hos host] vk: mirror of pool 0x7f73800000 (20 MB) given back: idle, 6041 submissions since its sync
[hos host] vk: mirror of pool 0x7fac800000 (8 MB) given back: idle, 6040 submissions since its sync
[hos host] vk: mirror of pool 0x1044c00000 (4 MB) given back: idle, 6031 submissions since its sync
[hos host] vk: mirror of pool 0x7f74c00000 (24 MB) given back: idle, 6029 submissions since its sync
[hos host] vk: mirror of pool 0x1043000000 (4 MB) given back: idle, 6027 submissions since its sync
[hos host] vk: mirror of pool 0x1043800000 (4 MB) given back: idle, 6027 submissions since its sync
[hos host] vk: mirror of pool 0x1043c00000 (4 MB) given back: idle, 6027 submissions since its sync
[hos host] vk: mirror of pool 0x7f78000000 (4 MB) given back: idle, 6027 submissions since its sync
[hos host] vk: mirror of pool 0x1041c00000 (4 MB) given back: idle, 6027 submissions since its sync
[hos host] vk: mirror of pool 0x7faa000000 (4 MB) given back: idle, 6027 submissions since its sync
[hos host] vk: mirror of pool 0x7fac000000 (8 MB) given back: idle, 6027 submissions since its sync
[hos host] vk: mirror of pool 0x1042800000 (4 MB) given back: idle, 6018 submissions since its sync
[hos host] vk: 195 mirrors given back (1426 MB in all); 1153 MB of mirrors now, 164 MB of them idle and kept for what the GPU wrote there alone
[hos host] vk: mirror of pool 0x7fd3c00000 (8 MB) given back: idle, 6030 submissions since its sync
[hos host] vk: mirror of pool 0x7f86400000 (12 MB) given back: idle, 6062 submissions since its sync
[hos host] vk: mirror of pool 0x7f77800000 (8 MB) given back: idle, 6048 submissions since its sync
[hos host] vk: mirror of pool 0x7fd7400000 (4 MB) given back: idle, 6048 submissions since its sync
[hos host] vk: mirror of pool 0x7fe0c00000 (4 MB) given back: idle, 6048 submissions since its sync
[hos host] vk: mirror syncs (600 submits): 1351 dirty pages from their submits' captures, 51393 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: write-back diff (600 submits): 5392 runs, 2045.1 MB compared; 73 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 15953 in captures, 148 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 72027 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 212 mirrors given back (1523 MB in all); 1057 MB of mirrors now, 164 MB of them idle and kept for what the GPU wrote there alone
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1364 dirty pages from their submits' captures, 53138 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: write-back diff (600 submits): 5446 runs, 2109.5 MB compared; 56 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16594 in captures, 88 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos host] vk: readback pages refreshed from guest memory: 3149437 pages so far (0 held: may be written ahead in their submission)
[hos sead::MainThread] vk: capture: a submit 0.26 ms (lock 0.01, scan 0.05 over 538 ranges, 4407 MB), 540 pages written, 540 copied (2.1 MB)
[hos host] vk: a capture taken (ms): all 0.091; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.074, bytes copied 0.000; 25 kept (202 MB), 6 spare buffers, 14395 texture pages copied (56 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.10, writes collected 0.06
[hos host] vk: replay reads of guest memory as submitted: 3580000 (1894772 pages from the captures)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1482 of 3000 waited for their write-back, 0.19 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 605 KB; 0 made on demand (0 MB kept now); c1 stages: at most 176 a recording
[hos host] vk: mirror syncs (600 submits): 1309 dirty pages from their submits' captures, 51575 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: write-back diff (600 submits): 5480 runs, 2047.0 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16177 in captures, 0 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #130000: event 0x100e640040 mode 0 value 94846 (now 94827) (lr 0x80697644)
[hos BaseProcExecutor1] vcores: 317507 takes, 23874 waits (mean 28 us, longest 4.9 ms), 13051 kicks, 13071 yields, 0 escapes in 5 s
[hos ModuleSystemWorker5] nvn: CommandBufferSignalEvent #95000: event 0x100e640044 mode 0 value 94999 (lr 0x802742f0)
[hos host] vk: a frame: 131 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 8.06, present 0.17, draw 3.52, descriptors 1.28, texture lookups 0.10, uploads 0.03, pipelines 0.18, programs 0.59, commands 1.19, dispatch 0.72, memory sync 1.04, queue submit 0.64, slot wait 0.06
[hos host] vk: a frame: mirrors 293 (1057 MB): 6.88 MB synced in 1143.6 copies, 68.17 MB written back
[hos host] vk: 6840 presents, 46.9 fps; a frame: 1262 draws (0 skipped), 156 dispatches, 151 passes, 130 barriers, 6 submits, 0.1 uploads (0.1 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 43134 pipelines (8834084 ms), 0 programs (0 failed), 725 textures (1832 MB)
[hos host] vk: host time in GPU resource waits: 7.20 ms a frame (fences 0.04, timelines 7.16, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 499.6 full barriers recorded, 186.3 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 15.9 ms a frame (the longest submission 7.9 ms); 0.1 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: replay reads of guest memory as submitted: 3600000 (1901398 pages from the captures)
[hos host] vk: mirror syncs (600 submits): 1325 dirty pages from their submits' captures, 51468 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: write-back diff (600 submits): 5381 runs, 2044.9 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16219 in captures, 1 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos host] vk: program picks: 6538639 kept (2682416 by the last picks' textures, a uniform buffer bound since), 2822 new, 2258538 with other textures (of 8800000)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 107972 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos sead::MainThread] nvn: GPU waits in 5 s: 231 sync waits (1129 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos sead::MainThread] vk: capture: a submit 0.26 ms (lock 0.02, scan 0.05 over 540 ranges, 4377 MB), 581 pages written, 581 copied (2.3 MB)
[hos Presentation Thread] vi: 6880 frames presented (47.1 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 1, 15.5-17.8 5, 17.8-22 189, 22-30 39, 30-40 1, over 40 1
[hos host] vk: a capture taken (ms): all 0.118; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.100, bytes copied 0.000; 25 kept (210 MB), 11 spare buffers, 22515 texture pages copied (87 MB)
[hos host] vk: mirror syncs (600 submits): 1308 dirty pages from their submits' captures, 51844 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: write-back diff (600 submits): 5380 runs, 2046.2 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16195 in captures, 0 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.13, writes collected 0.06
[hos host] vk: replay reads of guest memory as submitted: 3620000 (1908184 pages from the captures)
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1319 dirty pages from their submits' captures, 51602 from earlier ones', 2 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: write-back diff (600 submits): 5371 runs, 2044.9 MB compared; 41 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16197 in captures, 1 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool: 0x7f7b800000 (12 MB, captured, state 2): 2;
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos host] vk: mirror syncs (600 submits): 1309 dirty pages from their submits' captures, 53287 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: write-back diff (600 submits): 5382 runs, 2044.9 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16728 in captures, 0 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: readback pages refreshed from guest memory: 3216168 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 61734 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos rmbl::VibrationThread] hid: 28672 vibration values, strongest 0.00 since the last count
[hos host] vk: replay reads of guest memory as submitted: 3640000 (1914938 pages from the captures)
[hos host] vk: a frame: 131 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 7.58, present 0.16, draw 3.37, descriptors 1.31, texture lookups 0.08, uploads 0.01, pipelines 0.17, programs 0.60, commands 1.01, dispatch 0.71, memory sync 0.99, queue submit 0.64, slot wait 0.06
[hos host] vk: a frame: mirrors 293 (1057 MB): 6.90 MB synced in 1144.0 copies, 68.16 MB written back
[hos host] vk: 6960 presents, 54.6 fps; a frame: 1263 draws (0 skipped), 155 dispatches, 151 passes, 130 barriers, 6 submits, 0.1 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 43134 pipelines (8834084 ms), 0 programs (0 failed), 737 textures (1841 MB)
[hos host] vk: host time in GPU resource waits: 4.60 ms a frame (fences 0.04, timelines 4.56, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 497.4 full barriers recorded, 186.2 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 13.4 ms a frame (the longest submission 11.2 ms); 0.1 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos Presentation Thread] vk: capture: a submit 0.24 ms (lock 0.01, scan 0.05 over 539 ranges, 4363 MB), 519 pages written, 519 copied (2.0 MB)
[hos host] vk: a capture taken (ms): all 0.081; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.063, bytes copied 0.000; 25 kept (198 MB), 9 spare buffers, 8778 texture pages copied (34 MB)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 605 KB; 0 made on demand (0 MB kept now); c1 stages: at most 176 a recording
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 992 dirty pages from their submits' captures, 51958 from earlier ones', 2 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: write-back diff (600 submits): 5546 runs, 2113.0 MB compared; 43 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16156 in captures, 21 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool: 0x7f7b800000 (12 MB, captured, state 2): 2;
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos sead::MainThread] vk: readbacks of GPU results: 1543 of 3000 waited for their write-back, 0.20 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.09, writes collected 0.05
[hos host] vk: program picks: 6688458 kept (2750204 by the last picks' textures, a uniform buffer bound since), 2822 new, 2308719 with other textures (of 9000000)
[hos host] vk: mirror syncs (600 submits): 825 dirty pages from their submits' captures, 53626 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: write-back diff (600 submits): 5541 runs, 2111.6 MB compared; 43 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16743 in captures, 20 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 3660000 (1921712 pages from the captures)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #135000: event 0x100e640040 mode 0 value 98646 (now 98627) (lr 0x80697644)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 75721 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1045 dirty pages from their submits' captures, 52142 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: write-back diff (600 submits): 5383 runs, 2046.2 MB compared; 41 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16174 in captures, 13 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos ModuleSystemWorker5] nvn: DrawMeshTasksIndirect: 240000 so far (x1 0x117991e000 x2 0x1 x3 0x117991e000 x4 0x4040)
[hos sead::MainThread] vk: capture: a submit 0.26 ms (lock 0.01, scan 0.05 over 539 ranges, 4347 MB), 585 pages written, 585 copied (2.3 MB)
[hos host] vk: a capture taken (ms): all 0.111; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.093, bytes copied 0.000; 25 kept (178 MB), 11 spare buffers, 19569 texture pages copied (76 MB)
[hos ModuleSystemWorker5] nvn: 1050000 recordings made, 55907 live copies (17.7 MB), 1635 texture copies; 4 command buffers' segments (4.2 MB)
[hos host] vk: replay reads of guest memory as submitted: 3680000 (1928511 pages from the captures)
[hos host] vk: mirror syncs (600 submits): 1163 dirty pages from their submits' captures, 53192 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: write-back diff (600 submits): 5374 runs, 2044.9 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16735 in captures, 4 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: render targets drawn over from another start, let go: 2381 so far (8396 MB)
[hos host] vk: a frame: 131 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 7.66, present 0.16, draw 3.35, descriptors 1.30, texture lookups 0.08, uploads 0.01, pipelines 0.18, programs 0.59, commands 1.01, dispatch 0.73, memory sync 1.09, queue submit 0.62, slot wait 0.06
[hos host] vk: a frame: mirrors 293 (1057 MB): 6.88 MB synced in 1146.0 copies, 68.16 MB written back
[hos host] vk: 7080 presents, 60.0 fps; a frame: 1262 draws (0 skipped), 155 dispatches, 151 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 43134 pipelines (8834084 ms), 0 programs (0 failed), 741 textures (1846 MB)
[hos host] vk: host time in GPU resource waits: 2.16 ms a frame (fences 0.04, timelines 2.12, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 496.3 full barriers recorded, 186.0 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 11.3 ms a frame (the longest submission 4.2 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: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1020 dirty pages from their submits' captures, 52099 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: write-back diff (600 submits): 5546 runs, 2111.6 MB compared; 41 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16164 in captures, 15 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.13, writes collected 0.06
[hos host] vk: readback pages refreshed from guest memory: 3283638 pages so far (0 held: may be written ahead in their submission)
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #100000: event 0x100e640010 mode 0 value 100002 (lr 0x80660bcc)
[hos host] vk: replay reads of guest memory as submitted: 3700000 (1935293 pages from the captures)
[hos BaseProcExecutor0] vcores: 314860 takes, 28802 waits (mean 26 us, longest 2.7 ms), 15711 kicks, 15701 yields, 0 escapes in 5 s
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 115390 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 605 KB; 0 made on demand (0 MB kept now); c1 stages: at most 176 a recording
[hos host] vk: mirror syncs (600 submits): 1052 dirty pages from their submits' captures, 51815 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: write-back diff (600 submits): 5376 runs, 2046.2 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16171 in captures, 7 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: recorder: 4500528 packets run in 10.0 s (busy 1.29 s, spun idle 2.74 s, slept 4132 times, 1089 of them under 200 us), 0 calls made directly (no room); the replay waited for it 27129 times, 96.52 ms in all, the longest 1.32 ms; vkEndCommandBuffer 10369/59.03 ms (max 1.32), replay_mu unlock 6392/22.49 ms (max 0.10), vkResetDescriptorPool 10368/15.01 ms (max 0.08)
[hos sead::MainThread] vk: capture: a submit 0.27 ms (lock 0.01, scan 0.05 over 537 ranges, 4311 MB), 627 pages written, 627 copied (2.4 MB)
[hos host] vk: a capture taken (ms): all 0.128; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.110, bytes copied 0.000; 25 kept (208 MB), 5 spare buffers, 30594 texture pages copied (119 MB)
[hos host] vk: program picks: 6838201 kept (2818185 by the last picks' textures, a uniform buffer bound since), 2822 new, 2358976 with other textures (of 9200000)
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1188 dirty pages from their submits' captures, 53251 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: write-back diff (600 submits): 5479 runs, 2110.2 MB compared; 43 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16741 in captures, 9 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 235 sync waits (330 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1546 of 3000 waited for their write-back, 0.17 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs)
[hos host] vk: replay reads of guest memory as submitted: 3720000 (1942088 pages from the captures)
[hos Presentation Thread] vi: 7179 frames presented (59.7 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 44, 15.5-17.8 215, 17.8-22 37, 22-30 3, 30-40 0, over 40 0
[hos host] vk: mirror syncs (600 submits): 1254 dirty pages from their submits' captures, 51925 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: write-back diff (600 submits): 5439 runs, 2046.3 MB compared; 40 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16180 in captures, 6 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.13, writes collected 0.05
[hos host] vk: memory: working set 13439 MB (peak 13926), private 20627 MB, staging 256 MB in 4 arenas
[hos host] vk: memory heaps: heap 0 (device): 107 blocks 5505 MB, 17804 allocations 3882 MB; used 6104 of 7250 MB; heap 1 (system): 14 blocks 928 MB, 39 allocations 872 MB; used 3337 of 15319 MB;
[hos host] vk: image blocks: 1 blocks 128 MB, 5 images 16 MB; 1 free ranges, the largest 111 MB
[hos host] vk: image blocks (render targets): 9 blocks 1152 MB, 151 images 525 MB; 59 free ranges, the largest 101 MB
[hos host] vk: image blocks (textures, compacted): 9 blocks 1152 MB, 593 images 512 MB; 179 free ranges, the largest 128 MB
[hos host] vk: render target images: 554 reused from the pool, 2007 made new, 638 destroyed from it, 1217 refused by it; it holds 0 (0 MB)
[hos host] vk: VRAM: 6104 MB used of 7250 MB allowed (1623 MB of it free in the blocks); textures 1851 MB (budget 4677 MB), 747 textures
[hos host] vk: textures: decoded ASTC 599 MB, rendered 852 MB (79 drawn in one frame, kept: 48166 KB), other 195 MB; used in the last 8 presents 1224 MB
[hos host] vk:   288 MB: 0x1148bea000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 957 presents ago
[hos host] vk:   214 MB: 0x10a322a600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a6aa2c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x11384dc000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x113db7d000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   37 MB: 0x10b7654e00 1800x1380x1 x8 levels 1 NVN 0x6, last used 1 presents ago
[hos host] vk:   32 MB: 0x115abeb000 4096x4096x1 x1 levels 1 NVN 0x32 rendered, last used 957 presents ago
[hos host] vk:   21 MB: 0x10005b30000 2400x2400x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 40 (5407 KB): 242x40x1 NVN 0x25: 4 (151 KB); 260x260x1 NVN 0xd: 26 (4574 KB); 306x104x1 NVN 0x1: 4 (124 KB); 592x49x1 NVN 0x25: 3 (339 KB); 656x113x1 NVN 0x1: 3 (217 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 545 waited for it, 2.80 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 57.1 ms; VRAM 6.0 of 7.1 GB
[hos host] vk: program census (play): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 96 program records loaded (27.0 ms), 5 new graphics pipelines, pipelines from the driver's cache 23, from the start's cache 14, compiled 87; 4276 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (play): 5882 queued, 5882 loaded (median 4.97 ms, 99th percentile 28.66 ms, worst 103.11 ms, 40061 ms and 1260.5 MB read in all); 8051 programs (1388 of them compute variants without their code, 0 graphics programs no note names), 10058 graphics pipelines (0 not in a cache), 184 compute pipelines (499 not in a cache); a first use ran 26 itself and waited for 3 (112.7 ms); 2095 pipelines made at create taken by draws; 3060 create calls with nothing kept for their stages, 0 before the index; units 8942 (5882 ready, 0 loading), 15523 programs alive
[hos host] vk: module identifiers (play): 756 known, 756 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 131 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 7.63, present 0.16, draw 3.28, descriptors 1.26, texture lookups 0.08, uploads 0.01, pipelines 0.18, programs 0.58, commands 0.99, dispatch 0.72, memory sync 1.08, queue submit 0.62, slot wait 0.06
[hos host] vk: a frame: mirrors 293 (1057 MB): 6.89 MB synced in 1144.2 copies, 68.16 MB written back
[hos host] vk: 7200 presents, 59.9 fps; a frame: 1263 draws (0 skipped), 155 dispatches, 151 passes, 130 barriers, 6 submits, 0.1 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 43134 pipelines (8834084 ms), 0 programs (0 failed), 747 textures (1851 MB)
[hos host] vk: host time in GPU resource waits: 2.37 ms a frame (fences 0.04, timelines 2.33, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 497.2 full barriers recorded, 186.1 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 11.4 ms a frame (the longest submission 4.2 ms); 0.1 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: queues ordered by their events: 46018 submissions stopped at a wait, 46004 went on once it was reached, 14 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 1 (stopped 50.27 ms at most); waits met 139265, not met 0; captures at a CPU event's wait 3371, 0 of them at a wait met on arrival
[hos host] vk: the render thread's whole frames (600 presents): 0 jobs went on before their present after 1000 ms, 0 times a game's thread waited for the work handed over, 0 times for a fence of the frame in the making
[hos host] vk: the render thread (600 presents, 19.02 ms a frame): its jobs 8.85 ms a frame (submits 2.55, presents 0.32, others 5.98), idle 10.17; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 124.99 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); 547 captures at a CPU signal; the submits' waits read in 0.028 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 8.89 ms after its hand-over (46.18 at most)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 101499 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 950 dirty pages from their submits' captures, 51783 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: write-back diff (600 submits): 5381 runs, 2046.2 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16198 in captures, 10 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos host] vk: replay reads of guest memory as submitted: 3740000 (1948988 pages from the captures)
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #140000: event 0x100e640040 mode 0 value 102446 (now 102427) (lr 0x80697644)
[hos sead::MainThread] vk: capture: a submit 0.25 ms (lock 0.01, scan 0.05 over 537 ranges, 4305 MB), 557 pages written, 557 copied (2.2 MB)
[hos host] vk: a capture taken (ms): all 0.105; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.088, bytes copied 0.000; 25 kept (192 MB), 11 spare buffers, 13921 texture pages copied (54 MB)
[hos host] vk: mirror syncs (600 submits): 1087 dirty pages from their submits' captures, 53599 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: write-back diff (600 submits): 5544 runs, 2111.6 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16734 in captures, 16 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: readback pages refreshed from guest memory: 3350180 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 3760000 (1955612 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 605 KB; 0 made on demand (0 MB kept now); c1 stages: at most 176 a recording
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1088 dirty pages from their submits' captures, 51641 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: write-back diff (600 submits): 5383 runs, 2046.2 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16195 in captures, 18 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos host] vk: program picks: 6988259 kept (2886200 by the last picks' textures, a uniform buffer bound since), 2822 new, 2408918 with other textures (of 9400000)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.12, writes collected 0.05
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 94737 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: mirror syncs (600 submits): 1092 dirty pages from their submits' captures, 53471 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: write-back diff (600 submits): 5546 runs, 2111.7 MB compared; 43 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16753 in captures, 13 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vk: capture: a submit 0.27 ms (lock 0.01, scan 0.05 over 537 ranges, 4267 MB), 577 pages written, 577 copied (2.3 MB)
[hos host] vk: a capture taken (ms): all 0.100; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.082, bytes copied 0.000; 25 kept (134 MB), 10 spare buffers, 16418 texture pages copied (64 MB)
[hos host] vk: a frame: 131 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 7.84, present 0.18, draw 3.45, descriptors 1.33, texture lookups 0.07, uploads 0.00, pipelines 0.18, programs 0.60, commands 1.05, dispatch 0.74, memory sync 1.07, queue submit 0.65, slot wait 0.06
[hos host] vk: a frame: mirrors 293 (1057 MB): 6.88 MB synced in 1143.1 copies, 68.16 MB written back
[hos host] vk: 7320 presents, 60.0 fps; a frame: 1261 draws (0 skipped), 155 dispatches, 150 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 43134 pipelines (8834084 ms), 0 programs (0 failed), 749 textures (1851 MB)
[hos host] vk: host time in GPU resource waits: 2.13 ms a frame (fences 0.04, timelines 2.09, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 495.2 full barriers recorded, 185.9 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 11.3 ms a frame (the longest submission 4.2 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: 3780000 (1962263 pages from the captures)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1546 of 3000 waited for their write-back, 0.16 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs)
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1102 dirty pages from their submits' captures, 53492 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: write-back diff (600 submits): 5545 runs, 2113.0 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16774 in captures, 14 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos ModuleSystemWorker2] nvn: CommandBufferSignalEvent #105000: event 0x100e640000 mode 0 value 104995 (lr 0x80291440)
[hos host] vk: mirror syncs (600 submits): 884 dirty pages from their submits' captures, 51788 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: write-back diff (600 submits): 5387 runs, 2046.2 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16186 in captures, 10 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 3800000 (1968921 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 88155 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.12, writes collected 0.06
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1079 dirty pages from their submits' captures, 53713 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: write-back diff (600 submits): 5443 runs, 2109.5 MB compared; 43 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16714 in captures, 12 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos host] vk: readback pages refreshed from guest memory: 3417630 pages so far (0 held: may be written ahead in their submission)
[hos sead::MainThread] vk: capture: a submit 0.26 ms (lock 0.01, scan 0.05 over 536 ranges, 4253 MB), 576 pages written, 576 copied (2.3 MB)
[hos host] vk: a capture taken (ms): all 0.105; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.087, bytes copied 0.000; 25 kept (204 MB), 6 spare buffers, 17925 texture pages copied (70 MB)
[hos tera::FileResourceMgr] vcores: 314800 takes, 30796 waits (mean 26 us, longest 6.3 ms), 16648 kicks, 16616 yields, 0 escapes in 5 s
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #145000: event 0x100e640040 mode 0 value 106246 (now 106227) (lr 0x80697644)
[hos host] vk: replay reads of guest memory as submitted: 3820000 (1975536 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 605 KB; 0 made on demand (0 MB kept now); c1 stages: at most 176 a recording
[hos host] vk: mirror syncs (600 submits): 869 dirty pages from their submits' captures, 51682 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: write-back diff (600 submits): 5489 runs, 2048.3 MB compared; 41 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16199 in captures, 10 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker4] nvn: DrawMeshTasksIndirect: 260000 so far (x1 0x117991e000 x2 0x1 x3 0x117991e000 x4 0x4040)
[hos host] vk: a frame: 131 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 7.60, present 0.16, draw 3.43, descriptors 1.32, texture lookups 0.07, uploads 0.00, pipelines 0.18, programs 0.61, commands 1.04, dispatch 0.73, memory sync 0.95, queue submit 0.62, slot wait 0.06
[hos host] vk: a frame: mirrors 293 (1057 MB): 6.87 MB synced in 1143.6 copies, 68.16 MB written back
[hos host] vk: 7440 presents, 60.1 fps; a frame: 1261 draws (0 skipped), 155 dispatches, 150 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 43134 pipelines (8834084 ms), 0 programs (0 failed), 751 textures (1853 MB)
[hos host] vk: host time in GPU resource waits: 1.93 ms a frame (fences 0.04, timelines 1.89, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 495.7 full barriers recorded, 185.9 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 11.3 ms a frame (the longest submission 4.2 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: program picks: 7138926 kept (2954198 by the last picks' textures, a uniform buffer bound since), 2822 new, 2458251 with other textures (of 9600000)
[hos host] vk: queues ordered by their events: 48000 submissions stopped at a wait, 47984 went on once it was reached, 14 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 1 (stopped 50.27 ms at most); waits met 145655, not met 0; captures at a CPU event's wait 3562, 0 of them at a wait met on arrival
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 628 dirty pages from their submits' captures, 54062 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: write-back diff (600 submits): 5553 runs, 2113.0 MB compared; 43 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16723 in captures, 29 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 198 sync waits (168 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 51887 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos Presentation Thread] vi: 7480 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 29, 15.5-17.8 241, 17.8-22 31, 22-30 0, 30-40 0, over 40 0
[hos host] vk: replay reads of guest memory as submitted: 3840000 (1982193 pages from the captures)
[hos host] vk: mirror syncs (600 submits): 1068 dirty pages from their submits' captures, 53562 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: write-back diff (600 submits): 5544 runs, 2111.6 MB compared; 43 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16754 in captures, 15 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.10, writes collected 0.06
[hos sead::MainThread] vk: capture: a submit 0.26 ms (lock 0.01, scan 0.05 over 542 ranges, 4219 MB), 567 pages written, 567 copied (2.2 MB)
[hos host] vk: a capture taken (ms): all 0.090; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.073, bytes copied 0.000; 25 kept (206 MB), 6 spare buffers, 14097 texture pages copied (55 MB)
[hos sead::MainThread] nvn: EventSignal (CPU) #15000: event 0x1007db0004 mode 0 value 1 (lr 0x8348c78c)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1548 of 3000 waited for their write-back, 0.17 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs)
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1132 dirty pages from their submits' captures, 51709 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: write-back diff (600 submits): 5379 runs, 2046.3 MB compared; 41 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16197 in captures, 9 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos ModuleSystemWorker2] nvn: 1100000 recordings made, 55906 live copies (17.4 MB), 1635 texture copies; 4 command buffers' segments (4.2 MB)
[hos host] vk: replay reads of guest memory as submitted: 3860000 (1989157 pages from the captures)
[hos host] vk: mirror syncs (600 submits): 1053 dirty pages from their submits' captures, 53888 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: write-back diff (600 submits): 5542 runs, 2111.6 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16741 in captures, 9 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: readback pages refreshed from guest memory: 3484389 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: a frame: 131 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 7.58, present 0.16, draw 3.30, descriptors 1.28, texture lookups 0.09, uploads 0.02, pipelines 0.18, programs 0.58, commands 0.99, dispatch 0.72, memory sync 0.97, queue submit 0.62, slot wait 0.06
[hos host] vk: a frame: mirrors 293 (1057 MB): 6.89 MB synced in 1144.3 copies, 68.16 MB written back
[hos host] vk: 7560 presents, 60.1 fps; a frame: 1262 draws (0 skipped), 155 dispatches, 151 passes, 130 barriers, 6 submits, 0.1 uploads (0.1 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 43134 pipelines (8834084 ms), 0 programs (0 failed), 765 textures (1863 MB)
[hos host] vk: host time in GPU resource waits: 2.15 ms a frame (fences 0.04, timelines 2.12, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 495.4 full barriers recorded, 185.8 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 11.4 ms a frame (the longest submission 4.2 ms); 0.1 textures made a frame
[hos host] vk: a frame: 0.05 MB unswizzled on the CPU in 0.01 ms
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 95894 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 605 KB; 0 made on demand (0 MB kept now); c1 stages: at most 176 a recording
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 896 dirty pages from their submits' captures, 53505 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: write-back diff (600 submits): 5553 runs, 2113.0 MB compared; 43 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16748 in captures, 17 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos Presentation Thread] vk: capture: a submit 0.25 ms (lock 0.02, scan 0.05 over 540 ranges, 4153 MB), 552 pages written, 552 copied (2.2 MB)
[hos host] vk: a capture taken (ms): all 0.090; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.072, bytes copied 0.000; 25 kept (226 MB), 10 spare buffers, 7993 texture pages copied (31 MB)
[hos host] vk: replay reads of guest memory as submitted: 3880000 (1995951 pages from the captures)
[hos host] vk: program picks: 7288736 kept (3022185 by the last picks' textures, a uniform buffer bound since), 2822 new, 2508441 with other textures (of 9800000)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.10, writes collected 0.05
[hos host] vk: mirror syncs (600 submits): 726 dirty pages from their submits' captures, 54020 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: write-back diff (600 submits): 5557 runs, 2113.0 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16744 in captures, 25 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker2] nvn: CommandBufferSignalEvent #110000: event 0x100e640020 mode 0 value 110000 (lr 0x80664f14)
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #150000: event 0x100e640040 mode 0 value 110046 (now 110027) (lr 0x80697644)
[hos host] vk: replay reads of guest memory as submitted: 3900000 (2002718 pages from the captures)
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1025 dirty pages from their submits' captures, 53282 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: write-back diff (600 submits): 5549 runs, 2113.0 MB compared; 43 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16691 in captures, 19 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 63685 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: mirror syncs (600 submits): 963 dirty pages from their submits' captures, 52046 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: write-back diff (600 submits): 5388 runs, 2046.2 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16147 in captures, 13 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] vk: capture: a submit 0.25 ms (lock 0.01, scan 0.05 over 534 ranges, 4069 MB), 559 pages written, 559 copied (2.2 MB)
[hos host] vk: a capture taken (ms): all 0.083; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.066, bytes copied 0.000; 25 kept (208 MB), 5 spare buffers, 13753 texture pages copied (53 MB)
[hos host] vk: render targets drawn over from another start, let go: 2381 so far (8396 MB)
[hos host] vk: a frame: 131 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 7.44, present 0.16, draw 3.38, descriptors 1.31, texture lookups 0.07, uploads 0.00, pipelines 0.18, programs 0.60, commands 1.02, dispatch 0.72, memory sync 0.88, queue submit 0.61, slot wait 0.06
[hos host] vk: a frame: mirrors 293 (1057 MB): 6.88 MB synced in 1145.2 copies, 68.16 MB written back
[hos host] vk: 7680 presents, 60.0 fps; a frame: 1262 draws (0 skipped), 155 dispatches, 151 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 43134 pipelines (8834084 ms), 0 programs (0 failed), 766 textures (1863 MB)
[hos host] vk: host time in GPU resource waits: 1.85 ms a frame (fences 0.04, timelines 1.81, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 495.2 full barriers recorded, 185.8 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 11.4 ms a frame (the longest submission 4.2 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos sead::MainThread] vk: readbacks of GPU results: 1546 of 3000 waited for their write-back, 0.15 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs)
[hos host] vk: replay reads of guest memory as submitted: 3920000 (2009463 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.09, writes collected 0.05
[hos host] vk: sparse mirrors: 81, 1057 MB with memory of 1444 MB (386 MB spared); blocks given memory later 9070, given back 20398; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 402; 760 binds (205.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 927 dirty pages from their submits' captures, 53762 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: write-back diff (600 submits): 5542 runs, 2111.6 MB compared; 42 pages read into the shadows; 404 MB of shadows; pages the shadows knew that the CPU wrote: 16750 in captures, 16 only in the collection, 30 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 163 live (2304 MB), 0 retired still imported (0 MB), 47 retired given back (1472 MB)
[hos host] vk: readback pages refreshed from guest memory: 3551260 pages so far (0 held: may be written ahead in their submission)
[hos AudioFrameworkThread] vcores: 309593 takes, 30104 waits (mean 25 us, longest 2.5 ms), 16217 kicks, 16104 yields, 0 escapes in 5 s
[hos host] control: quit asked by a client
[hos host] vk: program census (close): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 96 program records loaded (27.0 ms), 5 new graphics pipelines, pipelines from the driver's cache 23, from the start's cache 14, compiled 87; 4276 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (close): 5882 queued, 5882 loaded (median 4.97 ms, 99th percentile 28.66 ms, worst 103.11 ms, 40061 ms and 1260.5 MB read in all); 8051 programs (1388 of them compute variants without their code, 0 graphics programs no note names), 10058 graphics pipelines (0 not in a cache), 184 compute pipelines (499 not in a cache); a first use ran 26 itself and waited for 3 (112.7 ms); 2095 pipelines made at create taken by draws; 3060 create calls with nothing kept for their stages, 0 before the index; units 8942 (5882 ready, 0 loading), 15523 programs alive
[hos host] vk: module identifiers (close): 756 known, 756 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: pipeline cache: play file 6 saved (2404 KB): 4 pipelines, 3 from the driver's cache, 1 compiled; 3 more from the start's cache
[hos host] control: exiting
[hos host] stop: close work done (the game's saves are not waited for)