Beanstalk Longshot test farm
Menu

Report rp_618a87b57035785a fail

#391 perf(vk): six small costs of a draw and of a submission's parts removed

any on lick/agent-b0d3c1,

This report disagrees with its peers.

forest-and-field field.frametime_max_ms: +30.9% against peers' -11.6%
forest-and-field forest.stutter_count: -100.0% against peers' +0.0%

Machine

Commit
14a057b0b6e46c708b35c8debaa88efd7b6ec482
Baseline
615b249893f18e488cfa79d80be5c4ecc40dc257
App
nightly-146-23c5052c, attempt 1
Game build
e0f3ccfa
os
linux
cpu
AMD Ryzen 9 3900X 12-Core Processor
gpu
NVIDIA GeForce RTX 3080 Ti
tags
linuxnvidianvidia-proprietarydgpu
driver
620.70.6
ram_gb
48
vram_mb
12,288
gpu_vendor
nvidia
os_version
Arch Linux (kernel 7.2.14)

Builds

main (baseline)

1,442.4 s warm

from 615b249893f1

lift
94.2 s
read
3.7 s
finish
20.9 s
compile
841.9 s
shaders
474.5 s

PR

76 s warm

Program cache: warm

Scenarios

forest-and-field

ok
Metric (median)mainPRΔ
exit_code00±0%
field.fps_mean76.2874.33−2.5%
forest.fps_mean82.4583+0.7%
field.draws_mean1,188.951,189.85+0.1%
field.hitch_count00±0%
forest.draws_mean735.4729.25−0.8%
forest.hitch_count00±0%
log_forbidden_hits00±0%
field.stutter_count01.5—
forest.stutter_count0.50−100%
log_expected_missing00±0%
field.frametime_max_ms22.229.05+30.9%
field.frametime_p50_ms13.0613.38+2.5%
field.frametime_p99_ms15.8317.09+8%
forest.frametime_max_ms27.621.05−23.7%
forest.frametime_p50_ms12.0711.99−0.6%
forest.frametime_p99_ms15.0414.75−1.9%
field.frametime_1pct_low_ms16.8118.76+11.6%
forest.frametime_1pct_low_ms16.415.6−4.9%
screenshot_diff—0.15—

Pictures against main: forest diff 0.1867 (main against itself ), field diff 0.051 (main against itself )

Runs (4)
Metricmain 0pr 1pr 2main 3
hangnononono
crashednononono
exit_code0000
field.fps_mean78.2877.0671.6174.28
forest.fps_mean84.9983.1482.8779.91
field.draws_mean1,190.21,186.81,192.91,187.7
field.hitch_count0000
forest.draws_mean728733725.5742.8
forest.hitch_count0000
log_forbidden_hits0000
field.stutter_count0120
forest.stutter_count1000
log_expected_missing0000
field.frametime_max_ms20.4127.3330.7623.98
field.frametime_p50_ms12.7112.9413.8113.4
field.frametime_p99_ms15.4215.7718.4216.24
forest.frametime_max_ms31.4620.621.523.75
forest.frametime_p50_ms11.711211.9812.43
forest.frametime_p99_ms14.5114.5814.9215.57
field.frametime_1pct_low_ms16.0516.820.7217.58
forest.frametime_1pct_low_ms15.7615.4715.7217.04
screenshot_diff—0.190.1—

Rules

ScenarioCellRuleLimitmainPRΔ %Status
forest-and-fieldcrashFalse00—pass
forest-and-fieldforest.fps_mean.min_delta_pct1582.4583+0.7%pass
forest-and-fieldfield.fps_mean.min_delta_pct1576.2874.33−2.5%pass
forest-and-fieldscreenshot.max_diff0.15—0.19—fail
forest-and-fieldforest.frametime_p99_ms.max_delta_pct2015.0414.75−1.9%pass
forest-and-fieldfield.frametime_p99_ms.max_delta_pct2015.8317.09+8%pass
forest-and-fieldlog_forbidden_hits.max000—pass

Log (PR side, end)

[hos host] vk: mirror syncs (600 submits): 389 dirty pages from their submits' captures, 57098 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): 5566 runs, 2232.9 MB compared; 50 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17854 in captures, 5 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos Presentation Thread] vk: capture: a submit 0.16 ms (lock 0.02, scan 0.04 over 595 ranges, 2158 MB), 305 pages written, 305 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (126 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos BaseProcExecutor5] vcores: 171197 takes, 21579 waits (mean 65 us, longest 6.8 ms), 10355 kicks, 10231 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 674 KB; 0 made on demand (0 MB kept now); c1 stages: at most 164 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: mirror syncs (600 submits): 436 dirty pages from their submits' captures, 57343 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): 5425 runs, 2166.4 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17907 in captures, 6 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 1367 of 3000 waited for their write-back, 0.01 ms each; 0 looked at the stopped queues first, 1 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 host] vk: recorder: 5666410 packets run in 10.0 s (busy 1.83 s, spun idle 2.70 s, slept 3909 times, 965 of them under 200 us), 0 calls made directly (no room); the replay waited for it 33961 times, 176.28 ms in all, the longest 0.87 ms; vkEndCommandBuffer 12772/93.37 ms (max 0.87), vkResetDescriptorPool 12772/42.62 ms (max 0.50), replay_mu unlock 8415/40.27 ms (max 0.47), vkQueueBindSparse 2/0.02 ms (max 0.01)
[hos host] vk: replay reads of guest memory as submitted: 4800000 (2065264 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: sparse mirrors: 72, 977 MB with memory of 1288 MB (310 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 518 binds (71.6 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 418 dirty pages from their submits' captures, 55591 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): 5563 runs, 2232.9 MB compared; 48 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17353 in captures, 6 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos host] vk: memory: working set 18674 MB (peak 18688), private 16509 MB, staging 64 MB in 1 arenas, 15 made so far (12 ms)
[hos host] vk: memory heaps: heap 0 (device): 111 blocks 6121 MB, 17444 allocations 5095 MB; used 7148 of 8815 MB; heap 1 (system): 14 blocks 784 MB, 39 allocations 732 MB; used 2599 of 36053 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): 7 blocks 896 MB, 145 images 403 MB; 35 free ranges, the largest 112 MB
[hos host] vk: image blocks (textures, compacted): 22 blocks 2816 MB, 1505 images 2577 MB; 636 free ranges, the largest 86 MB
[hos host] vk: render target images: 244 reused from the pool, 568 made new, 423 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 7148 MB used of 8815 MB allowed (1026 MB of it free in the blocks); textures 3507 MB (budget 6804 MB), 1652 textures
[hos host] vk: textures: decoded ASTC 2335 MB, rendered 454 MB (92 drawn in one frame, kept: 43901 KB), other 468 MB; used in the last 8 presents 1302 MB
[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: 0x113cba9000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   37 MB: 0x110a282000 1800x1380x1 x8 levels 1 NVN 0x6, last used 1 presents ago
[hos host] vk:   21 MB: 0x10004760000 2400x2400x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   16 MB: 0x110dc60600 4096x4096x1 x1 levels 1 NVN 0x4b, last used 4754 presents ago
[hos host] vk:   16 MB: 0x1113285c00 4096x4096x1 x1 levels 1 NVN 0x4b, last used 4754 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 33 (5806 KB): 260x260x1 NVN 0xd: 33 (5806 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 55 waited for it, 0.18 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 30.8 ms; VRAM 7.0 of 8.6 GB
[hos host] vk: a frame: 124 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 6.56, present 0.11, draw 3.45, descriptors 1.20, texture lookups 0.07, uploads 0.00, pipelines 0.17, programs 0.57, commands 1.20, dispatch 0.73, memory sync 0.25, queue submit 0.23, slot wait 0.02
[hos host] vk: a frame: mirrors 245 (977 MB): 6.83 MB synced in 1138.8 copies, 67.67 MB written back
[hos host] vk: 20400 presents, 67.6 fps; a frame: 1184 draws (0 skipped), 145 dispatches, 147 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 32545 pipelines (5548781 ms), 0 programs (0 failed), 1652 textures (3507 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.01, timelines 0.02, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 475.1 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 8.0 ms a frame (the longest submission 3.0 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'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, 14.45 ms a frame): its jobs 6.65 ms a frame (submits 5.92, presents 0.11, others 0.62), idle 7.80; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 111.00 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); 56 captures at a CPU signal; the submits' waits read in 0.027 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 6.69 ms after its hand-over (14.31 at most)
[hos host] vk: readback pages refreshed from guest memory: 4233654 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: mirror syncs (600 submits): 371 dirty pages from their submits' captures, 57240 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): 5567 runs, 2231.9 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17850 in captures, 6 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] vk: capture: a submit 0.16 ms (lock 0.02, scan 0.04 over 595 ranges, 2158 MB), 312 pages written, 312 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.008; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.004, bytes copied 0.000; 25 kept (134 MB), 6 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 4820000 (2072690 pages from the captures)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #320000: event 0x100d6c8048 mode 0 value 224636 (now 224617) (lr 0x806661f8)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: sparse mirrors: 72, 977 MB with memory of 1288 MB (310 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 518 binds (71.6 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 385 dirty pages from their submits' captures, 56241 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): 5419 runs, 2166.3 MB compared; 48 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17336 in captures, 7 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #225000: event 0x100d6c8010 mode 0 value 225002 (lr 0x80660bcc)
[hos host] vk: program picks: 8563550 kept (3748759 by the last picks' textures, a uniform buffer bound since), 1847 new, 3034602 with other textures (of 11600000)
[hos host] vk: mirror syncs (600 submits): 419 dirty pages from their submits' captures, 56617 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): 5577 runs, 2232.9 MB compared; 50 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17927 in captures, 4 only in the collection, 33 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 0.01, writes collected 0.04
[hos ModuleSystemWorker1] nvn: 1550000 recordings made, 41906 live copies (13.5 MB), 581 texture copies; 7 command buffers' segments (4.4 MB)
[hos host] vk: replay reads of guest memory as submitted: 4840000 (2080119 pages from the captures)
[hos host] vk: render targets drawn over from another start, let go: 658 so far (1552 MB)
[hos host] vk: a frame: 124 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 6.68, present 0.11, draw 3.57, descriptors 1.27, texture lookups 0.07, uploads 0.00, pipelines 0.17, programs 0.57, commands 1.23, dispatch 0.74, memory sync 0.25, queue submit 0.26, slot wait 0.02
[hos host] vk: a frame: mirrors 245 (977 MB): 6.84 MB synced in 1138.5 copies, 67.67 MB written back
[hos host] vk: 20520 presents, 65.2 fps; a frame: 1184 draws (0 skipped), 145 dispatches, 147 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 32545 pipelines (5548781 ms), 0 programs (0 failed), 1652 textures (3507 MB)
[hos host] vk: host time in GPU resource waits: 0.02 ms a frame (fences 0.01, timelines 0.01, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 475.3 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 8.1 ms a frame (the longest submission 3.3 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 recording streams: 48 of 2 MB, the most one held 674 KB; 0 made on demand (0 MB kept now); c1 stages: at most 164 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 72, 977 MB with memory of 1288 MB (310 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 518 binds (71.6 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 354 dirty pages from their submits' captures, 57585 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): 5571 runs, 2231.9 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17875 in captures, 7 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos sead::MainThread] vk: capture: a submit 0.16 ms (lock 0.03, scan 0.04 over 595 ranges, 2158 MB), 309 pages written, 309 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (132 MB), 6 spare buffers, 0 texture pages copied (0 MB)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1379 of 3000 waited for their write-back, 0.01 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos host] vk: replay reads of guest memory as submitted: 4860000 (2087820 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: mirror syncs (600 submits): 414 dirty pages from their submits' captures, 57348 from earlier ones', 24 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): 5562 runs, 2232.8 MB compared; 50 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17911 in captures, 5 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool: 0x1065800000 (8 MB, captured, state 2): 24;
[hos host] vk: readback pages refreshed from guest memory: 4300741 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: sparse mirrors: 72, 977 MB with memory of 1288 MB (310 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 518 binds (71.6 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 407 dirty pages from their submits' captures, 56150 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): 5430 runs, 2166.5 MB compared; 48 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17359 in captures, 7 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: mirror of pool 0x7f98c00000 (12 MB) given back: idle, 6011 submissions since its sync
[hos host] vk: mirror of pool 0x7fb1800000 (16 MB) given back: idle, 6011 submissions since its sync
[hos host] vk: replay reads of guest memory as submitted: 4880000 (2094939 pages from the captures)
[hos host] vk: program picks: 8715000 kept (3815783 by the last picks' textures, a uniform buffer bound since), 1847 new, 3083152 with other textures (of 11800000)
[hos host] vk: mirror syncs (600 submits): 427 dirty pages from their submits' captures, 57581 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): 5568 runs, 2231.9 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17882 in captures, 7 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vk: capture: a submit 0.14 ms (lock 0.02, scan 0.04 over 595 ranges, 2158 MB), 309 pages written, 309 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.005; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (104 MB), 9 spare buffers, 0 texture pages copied (0 MB)
[hos ModuleSystemWorker4] nvn: DrawMeshTasksIndirect: 540000 so far (x1 0x117db40010 x2 0 x3 0x117db40000 x4 0x44b040)
[hos host] vk: a frame: 124 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 5.90, present 0.09, draw 3.23, descriptors 1.14, texture lookups 0.05, uploads 0.00, pipelines 0.15, programs 0.50, commands 1.16, dispatch 0.64, memory sync 0.22, queue submit 0.22, slot wait 0.02
[hos host] vk: a frame: mirrors 245 (954 MB): 6.86 MB synced in 1139.8 copies, 67.67 MB written back
[hos host] vk: 20640 presents, 72.6 fps; a frame: 1186 draws (0 skipped), 145 dispatches, 148 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 32545 pipelines (5548781 ms), 0 programs (0 failed), 1652 textures (3507 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.01, timelines 0.02, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 476.0 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 7.8 ms a frame (the longest submission 3.2 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] nvn: CommandBufferWaitEvent #325000: event 0x100d6c8048 mode 0 value 228436 (now 228417) (lr 0x806661f8)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: sparse mirrors: 70, 954 MB with memory of 1260 MB (305 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 518 binds (71.6 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 426 dirty pages from their submits' captures, 57576 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): 5573 runs, 2232.9 MB compared; 50 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17932 in captures, 5 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos host] vk: replay reads of guest memory as submitted: 4900000 (2102144 pages from the captures)
[hos Presentation Thread] vi: 20673 frames presented (68.9 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 276, 15.5-17.8 57, 17.8-22 12, 22-30 0, 30-40 0, over 40 0
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 674 KB; 0 made on demand (0 MB kept now); c1 stages: at most 164 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: mirror syncs (600 submits): 445 dirty pages from their submits' captures, 56803 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): 5421 runs, 2166.3 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17375 in captures, 9 only in the collection, 33 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 0.00, writes collected 0.03
[hos ModuleSystemWorker3] vcores: 173226 takes, 22089 waits (mean 71 us, longest 7.4 ms), 10504 kicks, 10348 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: replay reads of guest memory as submitted: 4920000 (2109254 pages from the captures)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1397 of 3000 waited for their write-back, 0.00 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 host] vk: sparse mirrors: 70, 954 MB with memory of 1260 MB (305 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 518 binds (71.6 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 412 dirty pages from their submits' captures, 56653 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): 5580 runs, 2232.9 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17874 in captures, 6 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos ModuleSystemWorker5] nvn: CommandBufferSignalEvent #230000: event 0x100d6c8008 mode 0 value 229997 (lr 0x80660f34)
[hos sead::MainThread] vk: capture: a submit 0.14 ms (lock 0.02, scan 0.04 over 595 ranges, 2158 MB), 312 pages written, 312 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.004; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (120 MB), 6 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: readback pages refreshed from guest memory: 4367823 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: mirror syncs (600 submits): 377 dirty pages from their submits' captures, 57187 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): 5415 runs, 2165.3 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17922 in captures, 8 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a frame: 124 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 5.62, present 0.09, draw 3.08, descriptors 1.09, texture lookups 0.05, uploads 0.00, pipelines 0.15, programs 0.44, commands 1.14, dispatch 0.60, memory sync 0.20, queue submit 0.21, slot wait 0.02
[hos host] vk: a frame: mirrors 248 (982 MB): 7.12 MB synced in 1139.6 copies, 67.67 MB written back
[hos host] vk: 20760 presents, 72.6 fps; a frame: 1191 draws (0 skipped), 145 dispatches, 149 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 32545 pipelines (5548781 ms), 0 programs (0 failed), 1652 textures (3507 MB)
[hos host] vk: host time in GPU resource waits: 0.14 ms a frame (fences 0.01, timelines 0.13, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 477.3 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 7.8 ms a frame (the longest submission 3.0 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: 4940000 (2116466 pages from the captures)
[hos host] vk: program picks: 8865656 kept (3882637 by the last picks' textures, a uniform buffer bound since), 1847 new, 3132496 with other textures (of 12000000)
[hos host] vk: sparse mirrors: 73, 982 MB with memory of 1296 MB (313 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 521 binds (71.9 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 494 dirty pages from their submits' captures, 56564 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): 5571 runs, 2232.9 MB compared; 48 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17371 in captures, 5 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.00, writes collected 0.03
[hos host] vk: mirror syncs (600 submits): 395 dirty pages from their submits' captures, 57865 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): 5569 runs, 2231.9 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17874 in captures, 6 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: 4960000 (2123580 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.14 ms (lock 0.02, scan 0.04 over 595 ranges, 2158 MB), 310 pages written, 310 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.005; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (98 MB), 8 spare buffers, 0 texture pages copied (0 MB)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #330000: event 0x100d6c8048 mode 0 value 232236 (now 232217) (lr 0x806661f8)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 674 KB; 0 made on demand (0 MB kept now); c1 stages: at most 164 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 73, 982 MB with memory of 1296 MB (313 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 521 binds (71.9 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 392 dirty pages from their submits' captures, 57557 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): 5569 runs, 2232.9 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17895 in captures, 5 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos host] vk: replay reads of guest memory as submitted: 4980000 (2131117 pages from the captures)
[hos host] vk: a frame: 124 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 5.88, present 0.09, draw 3.16, descriptors 1.10, texture lookups 0.05, uploads 0.00, pipelines 0.15, programs 0.49, commands 1.14, dispatch 0.64, memory sync 0.21, queue submit 0.23, slot wait 0.02
[hos host] vk: a frame: mirrors 248 (982 MB): 6.87 MB synced in 1138.7 copies, 67.67 MB written back
[hos host] vk: 20880 presents, 73.7 fps; a frame: 1185 draws (0 skipped), 145 dispatches, 148 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 32545 pipelines (5548781 ms), 0 programs (0 failed), 1652 textures (3507 MB)
[hos host] vk: host time in GPU resource waits: 0.01 ms a frame (fences 0.01, timelines 0.00, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 475.9 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 7.8 ms a frame (the longest submission 2.8 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: mirror syncs (600 submits): 430 dirty pages from their submits' captures, 57308 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): 5496 runs, 2231.1 MB compared; 50 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17905 in captures, 3 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1398 of 3000 waited for their write-back, 0.01 ms each; 0 looked at the stopped queues first, 1 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 host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.00, writes collected 0.03
[hos host] vk: readback pages refreshed from guest memory: 4434776 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: sparse mirrors: 73, 982 MB with memory of 1296 MB (313 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 521 binds (71.9 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 416 dirty pages from their submits' captures, 55975 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): 5505 runs, 2168.2 MB compared; 50 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17354 in captures, 3 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos Presentation Thread] vk: capture: a submit 0.14 ms (lock 0.02, scan 0.04 over 594 ranges, 2156 MB), 313 pages written, 313 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.004; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (176 MB), 9 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 5000000 (2138521 pages from the captures)
[hos host] vk: program picks: 9016900 kept (3949680 by the last picks' textures, a uniform buffer bound since), 1847 new, 3181252 with other textures (of 12200000)
[hos rmbl::VibrationThread] hid: 36864 vibration values, strongest 0.00 since the last count
[hos host] vk: mirror syncs (600 submits): 397 dirty pages from their submits' captures, 57401 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): 5418 runs, 2165.3 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17903 in captures, 7 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker3] nvn: 1600000 recordings made, 41911 live copies (14.7 MB), 581 texture copies; 4 command buffers' segments (4.2 MB)
[hos host] vk: sparse mirrors: 73, 982 MB with memory of 1296 MB (313 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 521 binds (71.9 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 372 dirty pages from their submits' captures, 54689 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): 5424 runs, 2165.2 MB compared; 47 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 16787 in captures, 8 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos ModuleSystemWorker5] nvn: CommandBufferSignalEvent #235000: event 0x100d6c8020 mode 0 value 235000 (lr 0x80664f14)
[hos host] vk: replay reads of guest memory as submitted: 5020000 (2145937 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.00, writes collected 0.03
[hos host] vk: queues ordered by their events: 114000 submissions stopped at a wait, 113984 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 50.05 ms at most); waits met 333827, not met 0; captures at a CPU event's wait 3409, 0 of them at a wait met on arrival
[hos host] vk: memory: working set 18676 MB (peak 18688), private 16512 MB, staging 64 MB in 1 arenas, 15 made so far (12 ms)
[hos host] vk: memory heaps: heap 0 (device): 111 blocks 6121 MB, 17525 allocations 5100 MB; used 7148 of 8801 MB; heap 1 (system): 14 blocks 784 MB, 39 allocations 732 MB; used 2599 of 36053 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): 7 blocks 896 MB, 145 images 403 MB; 35 free ranges, the largest 112 MB
[hos host] vk: image blocks (textures, compacted): 22 blocks 2816 MB, 1505 images 2577 MB; 636 free ranges, the largest 86 MB
[hos host] vk: render target images: 244 reused from the pool, 568 made new, 423 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 7148 MB used of 8801 MB allowed (1021 MB of it free in the blocks); textures 3507 MB (budget 6804 MB), 1652 textures
[hos host] vk: textures: decoded ASTC 2335 MB, rendered 454 MB (92 drawn in one frame, kept: 43901 KB), other 468 MB; used in the last 8 presents 1329 MB
[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: 0x113cba9000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   37 MB: 0x110a282000 1800x1380x1 x8 levels 1 NVN 0x6, last used 1 presents ago
[hos host] vk:   21 MB: 0x10004760000 2400x2400x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   16 MB: 0x110dc60600 4096x4096x1 x1 levels 1 NVN 0x4b, last used 5354 presents ago
[hos host] vk:   16 MB: 0x1113285c00 4096x4096x1 x1 levels 1 NVN 0x4b, last used 5354 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 33 (5806 KB): 260x260x1 NVN 0xd: 33 (5806 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 55 waited for it, 0.12 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 26.1 ms; VRAM 7.0 of 8.6 GB
[hos host] vk: a frame: 124 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 6.17, present 0.10, draw 3.30, descriptors 1.17, texture lookups 0.05, uploads 0.00, pipelines 0.16, programs 0.50, commands 1.18, dispatch 0.66, memory sync 0.21, queue submit 0.22, slot wait 0.02
[hos host] vk: a frame: mirrors 248 (982 MB): 6.88 MB synced in 1141.4 copies, 67.67 MB written back
[hos host] vk: 21000 presents, 72.0 fps; a frame: 1218 draws (0 skipped), 145 dispatches, 147 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 32545 pipelines (5548781 ms), 0 programs (0 failed), 1652 textures (3507 MB)
[hos host] vk: host time in GPU resource waits: 0.02 ms a frame (fences 0.01, timelines 0.01, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 476.7 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 8.0 ms a frame (the longest submission 3.1 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'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, 14.07 ms a frame): its jobs 6.44 ms a frame (submits 5.68, presents 0.10, others 0.66), idle 7.63; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 111.00 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); 61 captures at a CPU signal; the submits' waits read in 0.027 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 6.48 ms after its hand-over (12.21 at most)
[hos Presentation Thread] vi: real-time step on: ratio 0.839, game's frame rate 60
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 674 KB; 0 made on demand (0 MB kept now); c1 stages: at most 164 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: mirror syncs (600 submits): 428 dirty pages from their submits' captures, 55755 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): 5430 runs, 2165.3 MB compared; 47 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17388 in captures, 6 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vk: capture: a submit 0.14 ms (lock 0.02, scan 0.04 over 594 ranges, 2156 MB), 311 pages written, 311 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.004; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (118 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos ModuleSystemWorker4] nvn: DrawMeshTasksIndirect: 560000 so far (x1 0x117db40020 x2 0 x3 0x117db40000 x4 0x44b040)
[hos Presentation Thread] vi: 21038 frames presented (73.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 353, 15.5-17.8 11, 17.8-22 0, 22-30 1, 30-40 0, over 40 0
[hos sead::MainThread] nvn: CommandBufferWaitEvent #335000: event 0x100d6c8048 mode 0 value 236036 (now 236017) (lr 0x806661f8)
[hos host] vk: replay reads of guest memory as submitted: 5040000 (2153366 pages from the captures)
[hos host] vk: sparse mirrors: 73, 982 MB with memory of 1296 MB (313 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 521 binds (71.9 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 382 dirty pages from their submits' captures, 57769 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): 5568 runs, 2231.9 MB compared; 52 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17908 in captures, 5 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos ModuleSystemWorker5] vk: readbacks of GPU results: 1411 of 3000 waited for their write-back, 0.00 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 rmbl::VibrationThread] vcores: 180315 takes, 23057 waits (mean 64 us, longest 6.2 ms), 10951 kicks, 10786 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: readback pages refreshed from guest memory: 4502264 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: mirror syncs (600 submits): 439 dirty pages from their submits' captures, 55837 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): 5272 runs, 2098.7 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17348 in captures, 8 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: program picks: 9167884 kept (4018424 by the last picks' textures, a uniform buffer bound since), 1847 new, 3230268 with other textures (of 12400000)
[hos host] vk: recorder: 5837665 packets run in 10.0 s (busy 1.84 s, spun idle 2.77 s, slept 3946 times, 1014 of them under 200 us), 0 calls made directly (no room); the replay waited for it 34884 times, 171.34 ms in all, the longest 0.51 ms; vkEndCommandBuffer 13134/90.19 ms (max 0.51), vkResetDescriptorPool 13134/41.26 ms (max 0.18), replay_mu unlock 8613/39.89 ms (max 0.44), vkQueueBindSparse 3/0.00 ms (max 0.00)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: replay reads of guest memory as submitted: 5060000 (2160779 pages from the captures)
[hos host] vk: render targets drawn over from another start, let go: 658 so far (1552 MB)
[hos host] vk: sparse mirrors: 73, 982 MB with memory of 1296 MB (313 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 521 binds (71.9 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 443 dirty pages from their submits' captures, 54192 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): 5425 runs, 2165.2 MB compared; 47 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 16803 in captures, 5 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos host] vk: a frame: 123 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 6.77, present 0.09, draw 3.58, descriptors 1.25, texture lookups 0.06, uploads 0.00, pipelines 0.17, programs 0.54, commands 1.29, dispatch 0.73, memory sync 0.24, queue submit 0.24, slot wait 0.02
[hos host] vk: a frame: mirrors 248 (982 MB): 6.88 MB synced in 1142.9 copies, 67.67 MB written back
[hos host] vk: 21120 presents, 71.4 fps; a frame: 1221 draws (0 skipped), 145 dispatches, 146 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 32545 pipelines (5548781 ms), 0 programs (0 failed), 1652 textures (3507 MB)
[hos host] vk: host time in GPU resource waits: 0.02 ms a frame (fences 0.01, timelines 0.01, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 476.1 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 7.9 ms a frame (the longest submission 3.2 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos Presentation Thread] vk: capture: a submit 0.15 ms (lock 0.02, scan 0.04 over 594 ranges, 2156 MB), 311 pages written, 311 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.005; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (190 MB), 9 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: mirror syncs (600 submits): 384 dirty pages from their submits' captures, 56165 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): 5423 runs, 2165.3 MB compared; 51 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17366 in captures, 7 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: 5080000 (2168409 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 674 KB; 0 made on demand (0 MB kept now); c1 stages: at most 164 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 73, 982 MB with memory of 1296 MB (313 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 521 binds (71.9 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 453 dirty pages from their submits' captures, 56905 from earlier ones', 24 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): 5421 runs, 2165.3 MB compared; 56 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17911 in captures, 5 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool: 0x1065800000 (8 MB, captured, state 2): 24;
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: mirror syncs (600 submits): 488 dirty pages from their submits' captures, 57282 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): 5573 runs, 2231.9 MB compared; 50 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17364 in captures, 4 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: 5100000 (2175632 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.14 ms (lock 0.02, scan 0.04 over 594 ranges, 2156 MB), 314 pages written, 314 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.005; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (120 MB), 7 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: readback pages refreshed from guest memory: 4569032 pages so far (0 held: may be written ahead in their submission)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1423 of 3000 waited for their write-back, 0.01 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos host] vk: a frame: 124 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 6.29, present 0.12, draw 3.49, descriptors 1.23, texture lookups 0.06, uploads 0.00, pipelines 0.16, programs 0.53, commands 1.24, dispatch 0.68, memory sync 0.22, queue submit 0.24, slot wait 0.02
[hos host] vk: a frame: mirrors 248 (982 MB): 6.90 MB synced in 1142.5 copies, 67.67 MB written back
[hos host] vk: 21240 presents, 72.6 fps; a frame: 1201 draws (0 skipped), 145 dispatches, 147 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 32545 pipelines (5548781 ms), 0 programs (0 failed), 1652 textures (3507 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.01, timelines 0.02, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 476.2 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 8.0 ms a frame (the longest submission 3.2 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: 73, 982 MB with memory of 1296 MB (313 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 521 binds (71.9 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 373 dirty pages from their submits' captures, 57490 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): 5419 runs, 2166.3 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17887 in captures, 7 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos host] vk: program picks: 9319205 kept (4086278 by the last picks' textures, a uniform buffer bound since), 1847 new, 3278947 with other textures (of 12600000)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #340000: event 0x100d6c8048 mode 0 value 239836 (now 239817) (lr 0x806661f8)
[hos ModuleSystemWorker5] nvn: CommandBufferSignalEvent #240000: event 0x100d6c8028 mode 0 value 239999 (lr 0x80667410)
[hos host] vk: replay reads of guest memory as submitted: 5120000 (2182783 pages from the captures)
[hos host] vk: mirror syncs (600 submits): 458 dirty pages from their submits' captures, 56177 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): 5568 runs, 2232.9 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17379 in captures, 6 only in the collection, 33 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 0.01, writes collected 0.03
[hos host] vk: sparse mirrors: 73, 982 MB with memory of 1296 MB (313 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 521 binds (71.9 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 438 dirty pages from their submits' captures, 57695 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): 5572 runs, 2232.9 MB compared; 50 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17923 in captures, 7 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos Presentation Thread] vk: capture: a submit 0.14 ms (lock 0.02, scan 0.04 over 594 ranges, 2156 MB), 309 pages written, 309 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.005; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (98 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 5140000 (2189942 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 674 KB; 0 made on demand (0 MB kept now); c1 stages: at most 164 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: mirror syncs (600 submits): 441 dirty pages from their submits' captures, 57432 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): 5556 runs, 2231.9 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17889 in captures, 7 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a frame: 124 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 6.30, present 0.09, draw 3.43, descriptors 1.19, texture lookups 0.06, uploads 0.00, pipelines 0.16, programs 0.54, commands 1.23, dispatch 0.70, memory sync 0.22, queue submit 0.23, slot wait 0.02
[hos host] vk: a frame: mirrors 248 (982 MB): 6.89 MB synced in 1140.9 copies, 67.67 MB written back
[hos host] vk: 21360 presents, 73.4 fps; a frame: 1183 draws (0 skipped), 145 dispatches, 146 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 32545 pipelines (5548781 ms), 0 programs (0 failed), 1652 textures (3507 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.01, timelines 0.02, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 474.4 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 7.7 ms a frame (the longest submission 2.7 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: 73, 982 MB with memory of 1296 MB (313 MB spared); blocks given memory later 4633, given back 6727; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 521 binds (71.9 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 410 dirty pages from their submits' captures, 57050 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): 5424 runs, 2166.3 MB compared; 49 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17902 in captures, 5 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (1796 MB), 0 retired still imported (0 MB), 28 retired given back (780 MB)
[hos host] vk: replay reads of guest memory as submitted: 5160000 (2197160 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 4635895 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.00, writes collected 0.03
[hos host] vk: program picks: 9470500 kept (4153266 by the last picks' textures, a uniform buffer bound since), 1847 new, 3327652 with other textures (of 12800000)
[hos Presentation Thread] vi: 21401 frames presented (72.4 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 353, 15.5-17.8 9, 17.8-22 1, 22-30 0, 30-40 0, over 40 0
[hos host] vk: mirror syncs (600 submits): 441 dirty pages from their submits' captures, 56520 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): 5572 runs, 2232.9 MB compared; 48 pages read into the shadows; 372 MB of shadows; pages the shadows knew that the CPU wrote: 17376 in captures, 4 only in the collection, 33 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1399 of 3000 waited for their write-back, 0.00 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] nvn: DrawMeshTasksIndirect: 580000 so far (x1 0x1178986000 x2 0x1 x3 0x1178986000 x4 0x4040)
[hos Presentation Thread] vk: capture: a submit 0.14 ms (lock 0.02, scan 0.04 over 594 ranges, 2156 MB), 312 pages written, 312 copied (1.2 MB)
[hos host] vk: a capture taken (ms): all 0.005; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (160 MB), 8 spare buffers, 0 texture pages copied (0 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 1 sync waits (2 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: present 21422 written to <data>/jobs/jb_8f2d3a29699284d9/runs/forest-and-field-g0-02-pr/shots/forest-and-field-c0-pr-02-field.png (the control bridge)
[hos host] control: quit asked by a client
[hos host] vk: page table census (close): no pool outside its windows, 64-72 GB and 508-512 GB (4305 pools in the table now)
[hos host] control: exiting
[hos host] stop: close work done (the game's saves are not waited for)