Beanstalk Longshot test farm
Menu

Report rp_521666ea63125b4b fail

#1813 perf(vk): RX_VK_TEX_AHEAD=2 prepares every block-compressed texture the game makes, not only those an earlier load drew

windows+vram8gb on thebasedpisces/agent-8d2999,

This report disagrees with its peers.

teleport-tour#2 run_frametime_max_ms: -50.4% against peers' -17.1%
teleport-tour#2 run_hitch_count: -100.0% against peers' +22.2%
teleport-tour#2 value_textures_retired: +141.0% against peers' +563.5%

Machine

Commit
3af1d156e575f9fa99a99f0de40f7a629fbfb341
Baseline
65488813f084c48b81013006e1b07f5ae3b9ecb4
App
nightly-195-f9b6d598, attempt 2
Game build
e0f3ccfa
os
windows
cpu
AMD Ryzen 5 5600X3D 6-Core Processor
gpu
NVIDIA GeForce RTX 3070
tags
windowsnvidianvidia-proprietarydgpu
driver
621.24.0
ram_gb
32
vram_mb
8,017
gpu_vendor
nvidia
os_version
Windows 10.0.26200

Builds

main (baseline)

0 s warm

from 65488813f084

the base was built already

PR

83.2 s warm

Program cache: warm

Scenarios

teleport-tour RX_VK_TEX_AHEAD=0

ok
Metric (median)mainPRΔ
exit_code00±0%
field.cpu_s22.3622.2−0.7%
load_time_s4.013.9−2.7%
market.cpu_s27.8328.31+1.7%
kakariko.cpu_s27.8427.7−0.5%
load_time_max_s6.085.99−1.5%
run_hitch_count5542−23.6%
field.hitch_count32−33.3%
field.rss_peak_mb13,196.813,346.7+1.1%
field.vram_mean_mb6,2146,150−1%
field.vram_peak_mb6,2146,150−1%
kokiri-first.cpu_s48.6648.67+0%
log_forbidden_hits00±0%
market.hitch_count166−62.5%
market.rss_peak_mb12,327.212,378.1+0.4%
field.stutter_count56+20%
kokiri-return.cpu_s14.6915.77+7.4%
market.vram_mean_mb4,4414,334.9−2.4%
market.vram_peak_mb6,3136,231−1.3%
kakariko.hitch_count43−25%
kakariko.rss_peak_mb12,721.412,888.6+1.3%
log_expected_missing00±0%
market.stutter_count1613−18.8%
run_frametime_max_ms554.15627.23+13.2%
kakariko.vram_mean_mb6,3136,157−2.5%
kakariko.vram_peak_mb6,3136,157−2.5%
field.frametime_max_ms126.73122.93−3%
field.frametime_p50_ms16.7116.74+0.2%
field.frametime_p99_ms46.8640.1−14.4%
kakariko.stutter_count54−20%
value_textures_retired2,0252,148+6.1%
market.frametime_max_ms163.43163.96+0.3%
market.frametime_p50_ms16.7816.64−0.8%
market.frametime_p99_ms124.8468.88−44.8%
kokiri-first.hitch_count45+25%
kokiri-first.rss_peak_mb10,621.910,506.9−1.1%
kakariko.frametime_max_ms77.7463.73−18%
kakariko.frametime_p50_ms17.2816.86−2.4%
kakariko.frametime_p99_ms55.5348.71−12.3%
kokiri-first.vram_mean_mb4,4414,334.9−2.4%
kokiri-first.vram_peak_mb6,3136,231−1.3%
kokiri-return.hitch_count11±0%
kokiri-return.rss_peak_mb13,078.513,355.7+2.1%
kokiri-first.stutter_count127−41.7%
kokiri-return.vram_mean_mb4,4414,334.9−2.4%
kokiri-return.vram_peak_mb6,3136,231−1.3%
field.frametime_1pct_low_ms88.779.54−10.3%
kokiri-return.stutter_count32−33.3%
market.frametime_1pct_low_ms140.83118.07−16.2%
kokiri-first.frametime_max_ms82.64173.79+110.3%
kokiri-first.frametime_p50_ms18.5618.03−2.9%
kokiri-first.frametime_p99_ms55.6671.46+28.4%
kakariko.frametime_1pct_low_ms68.0260.37−11.3%
kokiri-return.frametime_max_ms67.8351.36−24.3%
kokiri-return.frametime_p50_ms16.5916.48−0.6%
kokiri-return.frametime_p99_ms30.6528.64−6.6%
kokiri-first.frametime_1pct_low_ms66.06119.87+81.5%
kokiri-return.frametime_1pct_low_ms49.0839.22−20.1%
screenshot_diff—0.31—

Pictures against main: market diff 0.2351 (main against itself ), kakariko diff 0.3136 (main against itself ), field diff 0.0427 (main against itself ), kokiri-return diff 0.1644 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
field.cpu_s22.3622.2
load_time_s4.013.9
market.cpu_s27.8328.31
kakariko.cpu_s27.8427.7
load_time_max_s6.085.99
run_hitch_count5542
field.hitch_count32
field.rss_peak_mb13,196.813,346.7
field.vram_mean_mb6,2146,150
field.vram_peak_mb6,2146,150
kokiri-first.cpu_s48.6648.67
log_forbidden_hits00
market.hitch_count166
market.rss_peak_mb12,327.212,378.1
field.stutter_count56
kokiri-return.cpu_s14.6915.77
market.vram_mean_mb4,4414,334.9
market.vram_peak_mb6,3136,231
kakariko.hitch_count43
kakariko.rss_peak_mb12,721.412,888.6
log_expected_missing00
market.stutter_count1613
run_frametime_max_ms554.15627.23
kakariko.vram_mean_mb6,3136,157
kakariko.vram_peak_mb6,3136,157
field.frametime_max_ms126.73122.93
field.frametime_p50_ms16.7116.74
field.frametime_p99_ms46.8640.1
kakariko.stutter_count54
value_textures_retired2,0252,148
market.frametime_max_ms163.43163.96
market.frametime_p50_ms16.7816.64
market.frametime_p99_ms124.8468.88
kokiri-first.hitch_count45
kokiri-first.rss_peak_mb10,621.910,506.9
kakariko.frametime_max_ms77.7463.73
kakariko.frametime_p50_ms17.2816.86
kakariko.frametime_p99_ms55.5348.71
kokiri-first.vram_mean_mb4,4414,334.9
kokiri-first.vram_peak_mb6,3136,231
kokiri-return.hitch_count11
kokiri-return.rss_peak_mb13,078.513,355.7
kokiri-first.stutter_count127
kokiri-return.vram_mean_mb4,4414,334.9
kokiri-return.vram_peak_mb6,3136,231
field.frametime_1pct_low_ms88.779.54
kokiri-return.stutter_count32
market.frametime_1pct_low_ms140.83118.07
kokiri-first.frametime_max_ms82.64173.79
kokiri-first.frametime_p50_ms18.5618.03
kokiri-first.frametime_p99_ms55.6671.46
kakariko.frametime_1pct_low_ms68.0260.37
kokiri-return.frametime_max_ms67.8351.36
kokiri-return.frametime_p50_ms16.5916.48
kokiri-return.frametime_p99_ms30.6528.64
kokiri-first.frametime_1pct_low_ms66.06119.87
kokiri-return.frametime_1pct_low_ms49.0839.22
screenshot_diff—0.31
loadsKokiriForest_Child 1.43 sMarket_Child 6.08 sKakarikoVillage_Child 4.91 sHyruleField_Child 3.86 sKokiriForest_Child 3.75 sKokiriForest_Child 1.18 sMarket_Child 5.99 sKakarikoVillage_Child 4.8 sHyruleField_Child 3.93 sKokiriForest_Child 3.59 s

teleport-tour RX_VK_TEX_AHEAD=1

crashed
Metric (median)mainPRΔ
exit_code-1,073,741,571——
load_time_s2.950.04−98.6%
load_time_max_s5.60.04−99.3%
run_hitch_count35+66.7%
kokiri-first.cpu_s26.0825.16−3.5%
log_forbidden_hits00±0%
log_expected_missing00±0%
run_frametime_max_ms78.3279.16+1.1%
value_textures_retired1,4991,163−22.4%
kokiri-first.hitch_count34+33.3%
kokiri-first.rss_peak_mb20,597.821,032.4+2.1%
kokiri-first.vram_mean_mb3,233.45,506+70.3%
kokiri-first.vram_peak_mb5,5165,506−0.2%
kokiri-first.stutter_count66±0%
kokiri-first.frametime_max_ms78.3279.16+1.1%
kokiri-first.frametime_p50_ms19.9116.58−16.7%
kokiri-first.frametime_p99_ms50.8465.47+28.8%
kokiri-first.frametime_1pct_low_ms61.8576.62+23.9%
Runs (2)
Metricmain 0pr 1
hangyesno
crashedyesno
exit_code-1,073,741,571—
load_time_s2.950.04
load_time_max_s5.60.04
run_hitch_count35
kokiri-first.cpu_s26.0825.16
log_forbidden_hits00
log_expected_missing00
run_frametime_max_ms78.3279.16
value_textures_retired1,4991,163
kokiri-first.hitch_count34
kokiri-first.rss_peak_mb20,597.821,032.4
kokiri-first.vram_mean_mb3,233.45,506
kokiri-first.vram_peak_mb5,5165,506
kokiri-first.stutter_count66
kokiri-first.frametime_max_ms78.3279.16
kokiri-first.frametime_p50_ms19.9116.58
kokiri-first.frametime_p99_ms50.8465.47
kokiri-first.frametime_1pct_low_ms61.8576.62
loadsKokiriForest_Child 0.31 sMarket_Child 5.6 sKokiriForest_Child 0.04 s

teleport-tour RX_VK_TEX_AHEAD=2

crashed
Metric (median)mainPRΔ
load_time_s0.040.47+1075%
load_time_max_s0.040.47+1075%
run_hitch_count30−100%
kokiri-first.cpu_s23.2729.31+26%
log_forbidden_hits00±0%
log_expected_missing00±0%
run_frametime_max_ms84.2441.75−50.4%
value_textures_retired1,2322,969+141%
kokiri-first.hitch_count30−100%
kokiri-first.rss_peak_mb20,842.220,665.9−0.8%
kokiri-first.vram_mean_mb4,121.84,256+3.3%
kokiri-first.vram_peak_mb5,6465,296−6.2%
kokiri-first.stutter_count71−85.7%
kokiri-first.frametime_max_ms84.2441.75−50.4%
kokiri-first.frametime_p50_ms16.5216.61+0.6%
kokiri-first.frametime_p99_ms50.5126.64−47.3%
kokiri-first.frametime_1pct_low_ms73.9332.96−55.4%
exit_code—-1,073,741,819—
Runs (2)
Metricmain 0pr 1
hangnoyes
crashednoyes
exit_code—-1,073,741,819
load_time_s0.040.47
load_time_max_s0.040.47
run_hitch_count30
kokiri-first.cpu_s23.2729.31
log_forbidden_hits00
log_expected_missing00
run_frametime_max_ms84.2441.75
value_textures_retired1,2322,969
kokiri-first.hitch_count30
kokiri-first.rss_peak_mb20,842.220,665.9
kokiri-first.vram_mean_mb4,121.84,256
kokiri-first.vram_peak_mb5,6465,296
kokiri-first.stutter_count71
kokiri-first.frametime_max_ms84.2441.75
kokiri-first.frametime_p50_ms16.5216.61
kokiri-first.frametime_p99_ms50.5126.64
kokiri-first.frametime_1pct_low_ms73.9332.96
loadsKokiriForest_Child 0.04 sKokiriForest_Child 0.47 s

Rules

ScenarioCellRuleLimitmainPRΔ %Status
teleport-tourRX_VK_TEX_AHEAD=0crashFalse00—pass
teleport-tourRX_VK_TEX_AHEAD=0screenshot.max_diff0.05—0.31—fail
teleport-tourRX_VK_TEX_AHEAD=0kokiri-first.vram_peak_mb.max_delta_pct156,3136,231−1.3%pass
teleport-tourRX_VK_TEX_AHEAD=0market.vram_peak_mb.max_delta_pct156,3136,231−1.3%pass
teleport-tourRX_VK_TEX_AHEAD=0kakariko.vram_peak_mb.max_delta_pct156,3136,157−2.5%pass
teleport-tourRX_VK_TEX_AHEAD=0field.vram_peak_mb.max_delta_pct156,2146,150−1%pass
teleport-tourRX_VK_TEX_AHEAD=0kokiri-return.vram_peak_mb.max_delta_pct156,3136,231−1.3%pass
teleport-tourRX_VK_TEX_AHEAD=0load_time_max_s.max_delta_pct306.085.99−1.5%pass
teleport-tourRX_VK_TEX_AHEAD=0kokiri-first.frametime_p99_ms.max_delta_pct3055.6671.46+28.4%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
teleport-tourRX_VK_TEX_AHEAD=0market.frametime_p99_ms.max_delta_pct30124.8468.88−44.8%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
teleport-tourRX_VK_TEX_AHEAD=0kakariko.frametime_p99_ms.max_delta_pct3055.5348.71−12.3%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
teleport-tourRX_VK_TEX_AHEAD=0field.frametime_p99_ms.max_delta_pct3046.8640.1−14.4%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
teleport-tourRX_VK_TEX_AHEAD=0kokiri-return.frametime_p99_ms.max_delta_pct3030.6528.64−6.6%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
teleport-tourRX_VK_TEX_AHEAD=0log_forbidden_hits.max000—pass
teleport-tourRX_VK_TEX_AHEAD=1crashFalse10—pass
teleport-tourRX_VK_TEX_AHEAD=1screenshot.max_diff0.05———skippedthe plan has no screenshot probe
teleport-tourRX_VK_TEX_AHEAD=1kokiri-first.vram_peak_mb.max_delta_pct155,5165,506−0.2%pass
teleport-tourRX_VK_TEX_AHEAD=1market.vram_peak_mb.max_delta_pct15———skippedthe metric was not measured
teleport-tourRX_VK_TEX_AHEAD=1kakariko.vram_peak_mb.max_delta_pct15———skippedthe metric was not measured
teleport-tourRX_VK_TEX_AHEAD=1field.vram_peak_mb.max_delta_pct15———skippedthe metric was not measured
teleport-tourRX_VK_TEX_AHEAD=1kokiri-return.vram_peak_mb.max_delta_pct15———skippedthe metric was not measured
teleport-tourRX_VK_TEX_AHEAD=1load_time_max_s.max_delta_pct305.60.04−99.3%pass
teleport-tourRX_VK_TEX_AHEAD=1kokiri-first.frametime_p99_ms.max_delta_pct30———inconclusiveneed two usable starts per side (baseline: 0, PR: 0)
teleport-tourRX_VK_TEX_AHEAD=1market.frametime_p99_ms.max_delta_pct30———inconclusiveneed two usable starts per side (baseline: 0, PR: 0)
teleport-tourRX_VK_TEX_AHEAD=1kakariko.frametime_p99_ms.max_delta_pct30———inconclusiveneed two usable starts per side (baseline: 0, PR: 0)
teleport-tourRX_VK_TEX_AHEAD=1field.frametime_p99_ms.max_delta_pct30———inconclusiveneed two usable starts per side (baseline: 0, PR: 0)
teleport-tourRX_VK_TEX_AHEAD=1kokiri-return.frametime_p99_ms.max_delta_pct30———inconclusiveneed two usable starts per side (baseline: 0, PR: 0)
teleport-tourRX_VK_TEX_AHEAD=1log_forbidden_hits.max000—pass
teleport-tourRX_VK_TEX_AHEAD=2crashFalse01—fail
teleport-tourRX_VK_TEX_AHEAD=2screenshot.max_diff0.05———skippedthe plan has no screenshot probe
teleport-tourRX_VK_TEX_AHEAD=2kokiri-first.vram_peak_mb.max_delta_pct155,6465,296−6.2%pass
teleport-tourRX_VK_TEX_AHEAD=2market.vram_peak_mb.max_delta_pct15———skippedthe metric was not measured
teleport-tourRX_VK_TEX_AHEAD=2kakariko.vram_peak_mb.max_delta_pct15———skippedthe metric was not measured
teleport-tourRX_VK_TEX_AHEAD=2field.vram_peak_mb.max_delta_pct15———skippedthe metric was not measured
teleport-tourRX_VK_TEX_AHEAD=2kokiri-return.vram_peak_mb.max_delta_pct15———skippedthe metric was not measured
teleport-tourRX_VK_TEX_AHEAD=2load_time_max_s.max_delta_pct300.040.47+1075%fail
teleport-tourRX_VK_TEX_AHEAD=2kokiri-first.frametime_p99_ms.max_delta_pct30———inconclusiveneed two usable starts per side (baseline: 0, PR: 0)
teleport-tourRX_VK_TEX_AHEAD=2market.frametime_p99_ms.max_delta_pct30———inconclusiveneed two usable starts per side (baseline: 0, PR: 0)
teleport-tourRX_VK_TEX_AHEAD=2kakariko.frametime_p99_ms.max_delta_pct30———inconclusiveneed two usable starts per side (baseline: 0, PR: 0)
teleport-tourRX_VK_TEX_AHEAD=2field.frametime_p99_ms.max_delta_pct30———inconclusiveneed two usable starts per side (baseline: 0, PR: 0)
teleport-tourRX_VK_TEX_AHEAD=2kokiri-return.frametime_p99_ms.max_delta_pct30———inconclusiveneed two usable starts per side (baseline: 0, PR: 0)
teleport-tourRX_VK_TEX_AHEAD=2log_forbidden_hits.max000—pass

Pictures of the failure

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

teleport-tour-c0-market-main-failed.png
teleport-tour-c0-market-main-failed.png
teleport-tour-c0-market-pr-failed.png
teleport-tour-c0-market-pr-failed.png

Log (PR side, end)

[hos host] vk: mirror syncs' pass (600 submits): 1263 dispatches, 38528 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 9625 runs, 855.2 MB compared; 216 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 19124 in captures, 1 only in the collection, 50 only by the replay itself
[hos host] vk:   read live, by pool:
[hos OptimizeChangeBanc] nvn: 150000 recordings made, 41464 live copies (12.0 MB), 285 texture copies; 3 command buffers' segments (4.2 MB)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 4673 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: a submit's memory work (ms): maps 0.01, capture taken 0.01, writes collected 0.28
[hos host] vk: textures ahead: 3016 prepared so far (2605 MB of images, 566 ms); a draw's first uploads: 0 new to the list, 0 known and not prepared in time, 0 drawn too long after they were made to be kept; made and never drawn before 0, let go for their age 0; 2416 waiting, 0 in the list; the cache had no room at times
[hos sead::MainThread] vk: capture: a submit 0.39 ms (lock 0.06, scan 0.21 over 1355 ranges, 3243 MB), 358 pages written, 358 copied (1.4 MB)
[hos host] vk: a capture taken (ms): all 0.011; GPU wait for late pages 0.001 (1 of 600 takes), pools lock 0.000, retires and stamps 0.006, bytes copied 0.000; 25 kept (200 MB), 11 spare buffers, 2070 texture pages copied (8 MB)
[hos Presentation Thread] vcores: 158469 takes, 65914 waits (mean 127 us, longest 157.9 ms, 6 over HOS_VCORE_LOG_MS), 32864 kicks, 32269 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: replay reads of guest memory as submitted: 140000 (65952 pages from the captures)
[hos host] vk: recorder: 1660438 packets run in 10.0 s (busy 0.72 s, spun idle 1.86 s, slept 6095 times, 1514 of them under 200 us), 0 calls made directly (no room); the replay waited for it 20878 times, 78.94 ms in all, the longest 1.39 ms; vkEndCommandBuffer 6974/36.05 ms (max 1.39), replay_mu unlock 6918/23.97 ms (max 0.16), vkResetDescriptorPool 6927/18.10 ms (max 0.04), vkQueueBindSparse 50/0.76 ms (max 0.21), vmaInvalidateAllocation 4/0.03 ms (max 0.02), vmaDestroyImage 1/0.01 ms (max 0.01)
[hos host] vk: readback pages refreshed from guest memory: 69194 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 197 KB; 0 made on demand (0 MB kept now); c1 stages: at most 76 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 63, 835 MB with memory of 1064 MB (228 MB spared); blocks given memory later 90, given back 1245; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 91 binds (11.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 67 made new; 11 kept as spares, 0 destroyed; 4 spares now, 260 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1150, 4681.6 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 89 dirty pages from their submits' captures, 57420 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1252 dispatches, 38910 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 9584 runs, 869.2 MB compared; 151 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 20216 in captures, 1 only in the collection, 50 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 65 live (1108 MB), 0 retired still imported (0 MB), 11 retired given back (320 MB)
[hos host] vk: late pages' fix: stamps for 0x103cb90000 (pool 0x103c800000; 64 blocks, 4096 KB)
[hos host] vk: replay reads of guest memory as submitted: 160000 (74958 pages from the captures)
[hos host] vk: a frame: 51 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 7.06, present 0.21, draw 3.14, descriptors 1.36, texture lookups 0.11, uploads 0.05, pipelines 0.12, programs 0.49, commands 0.94, dispatch 0.91, memory sync 1.11, queue submit 0.47, slot wait 0.05
[hos host] vk: a frame: mirrors 74 (835 MB): 4.53 MB synced in 772.7 copies, 17.41 MB written back
[hos host] vk: 2160 presents, 60.0 fps; a frame: 630 draws (0 skipped), 166 dispatches, 139 passes, 144 barriers, 6 submits, 0.0 uploads (0.1 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37478 pipelines (5315077 ms), 0 programs (0 failed), 1876 textures (2845 MB)
[hos host] vk: host time in GPU resource waits: 0.16 ms a frame (fences 0.03, 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: 399.4 full barriers recorded, 185.5 left out (nothing recorded since the one before), 18.0 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 8.5 ms a frame (the longest submission 3.3 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.2 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.12 MB unswizzled on the CPU in 0.05 ms
[hos host] vk: replay reads of guest memory as submitted: 180000 (83930 pages from the captures)
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #15000: event 0x100ce5800c mode 0 value 7553 (now 7534) (lr 0x8066179c)
[hos host] vk: write-backs left out (600 submits): 1150, 4681.6 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 55 dirty pages from their submits' captures, 51285 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1206 dispatches, 38443 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 9400 runs, 865.4 MB compared; 106 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 20573 in captures, 0 only in the collection, 50 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1013 of 3000 waited for their write-back, 0.01 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker1] vk: invalidates since the last readbacks line: 3846 kept their virtual core (nothing there to wait for), 2810 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: 1 textures retired (1876 left, 2840 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos Presentation Thread] vk: capture: a submit 0.15 ms (lock 0.02, scan 0.07 over 1352 ranges, 3007 MB), 202 pages written, 202 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.011; GPU wait for late pages 0.004 (1 of 600 takes), pools lock 0.000, retires and stamps 0.005, bytes copied 0.000; 25 kept (174 MB), 11 spare buffers, 2072 texture pages copied (8 MB)
[hos host] vk: 1 textures retired (1892 left, 2917 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 200000 (94403 pages from the captures)
[hos host] vk: 1 textures retired (1901 left, 2931 MB): budget 1, drawn over 0, dead targets 0
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 4966 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: 42 textures retired (1867 left, 2955 MB): budget 42, drawn over 0, dead targets 0
[hos host] vk: sparse mirrors: 63, 835 MB with memory of 1064 MB (228 MB spared); blocks given memory later 90, given back 1245; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 91 binds (11.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 67 made new; 11 kept as spares, 0 destroyed; 4 spares now, 260 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1219, 4962.5 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 15 dirty pages from their submits' captures, 55009 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1303 dispatches, 40748 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6499 runs, 852.0 MB compared; 109 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 21936 in captures, 1 only in the collection, 53 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 65 live (1108 MB), 0 retired still imported (0 MB), 11 retired given back (320 MB)
[hos host] vk: 74 textures retired (1800 left, 2968 MB): budget 74, drawn over 0, dead targets 0
[hos Presentation Thread] vi: 2255 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 33, 15.5-17.8 249, 17.8-22 18, 22-30 1, 30-40 0, over 40 0
[hos host] vk: 82 textures retired (1728 left, 2993 MB): budget 81, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 220000 (105210 pages from the captures)
[hos host] vk: 36 textures retired (1698 left, 2988 MB): budget 28, drawn over 8, dead targets 0
[hos host] vk: 10 textures retired (1688 left, 2936 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: texture #3446 0x10007e8ce00 1036x720 x1 NVN 0x34 the GPU drew, finalized by the game, sampled again at frame 2280 (drawn at 2219)
[hos host] vk: a frame: 44 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 5.81, present 0.27, draw 2.90, descriptors 1.25, texture lookups 0.46, uploads 0.00, pipelines 0.11, programs 0.46, commands 0.88, dispatch 0.69, memory sync 0.36, queue submit 0.41, slot wait 0.04
[hos host] vk: a frame: mirrors 74 (835 MB): 4.05 MB synced in 770.7 copies, 16.51 MB written back
[hos host] vk: 2280 presents, 60.0 fps; a frame: 638 draws (0 skipped), 113 dispatches, 139 passes, 144 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37478 pipelines (5315077 ms), 0 programs (0 failed), 1688 textures (2936 MB)
[hos host] vk: host time in GPU resource waits: 0.09 ms a frame (fences 0.02, timelines 0.07, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 344.7 full barriers recorded, 185.5 left out (nothing recorded since the one before), 17.6 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 45.5; 0x12: 68.8; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.7 ms a frame (the longest submission 3.3 ms); 1.6 textures made a frame: colour targets 0.7 (3.2 MB), depth 0.9 (4.9 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.01 MB unswizzled on the CPU in 0.00 ms
[hos rmbl::VibrationThread] hid: 4096 vibration values, strongest 0.00 since the last count
[hos host] vk: VRAM compaction 1 begins: the textures' blocks hold 1760 MB, 551 MB of it free space
[hos host] vk: write-backs left out (600 submits): 1230, 5002.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 23 dirty pages from their submits' captures, 55725 from earlier ones', 0 from guest memory as it was at the replay; 24 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1296 dispatches, 41511 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 7018 runs, 893.3 MB compared; 108 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 22880 in captures, 0 only in the collection, 53 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: program picks: 235000 kept (67418 by the last picks' textures, a uniform buffer bound since), 710 new, 164289 with other textures (of 400000)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos Presentation Thread] vk: capture: a submit 0.14 ms (lock 0.01, scan 0.07 over 1351 ranges, 2851 MB), 207 pages written, 207 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.005, bytes copied 0.000; 25 kept (188 MB), 10 spare buffers, 2070 texture pages copied (8 MB)
[hos host] vk: replay reads of guest memory as submitted: 240000 (115772 pages from the captures)
[hos host] vk: VRAM compaction 1 over: 158 textures moved (52 MB) in 15 passes, 0 left where they were; 17 blocks (544 MB) given back to the driver, 544 MB in all so far
[hos ModuleSystemWorker2] nvn: DrawMeshTasksIndirect: 20000 so far (x1 0x117832a000 x2 0 x3 0x117832a000 x4 0x654040)
[hos ModuleSystemWorker5] nvn: CommandBufferSignalEvent #10000: event 0x100ce58044 mode 0 value 9999 (lr 0x802742f0)
[hos host] vk: texture #3543 0x100074c0800 384x384 x1 NVN 0x34 the GPU drew, finalized by the game, sampled again at frame 2330 (drawn at 2269)
[hos host] vk: replay reads of guest memory as submitted: 260000 (126366 pages from the captures)
[hos host] vk: sparse mirrors: 63, 835 MB with memory of 1064 MB (228 MB spared); blocks given memory later 90, given back 1245; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 91 binds (11.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 67 made new; 11 kept as spares, 0 destroyed; 4 spares now, 260 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1254, 5110.0 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 27 dirty pages from their submits' captures, 56478 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1314 dispatches, 42397 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 7485 runs, 967.0 MB compared; 116 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 22754 in captures, 0 only in the collection, 55 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 65 live (1108 MB), 0 retired still imported (0 MB), 11 retired given back (320 MB)
[hos host] vk: 7 textures retired (1686 left, 2937 MB): budget 0, drawn over 7, dead targets 0
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1085 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 ModuleSystemWorker2] vk: invalidates since the last readbacks line: 4367 kept their virtual core (nothing there to wait for), 3057 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: 10 textures retired (1676 left, 2871 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 4500 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: 8 textures retired (1669 left, 2822 MB): budget 0, drawn over 8, dead targets 0
[hos host] vk: 10 textures retired (1660 left, 2754 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: handle table: element 19 of program 2fd8c7ca964e5a1d names texture #3690 (format 97) for a depth comparison: bound as nothing
[hos host] vk: handle table: element 19 of program 21c7c8a2d0fe7171 names texture #3690 (format 97) for a depth comparison: bound as nothing
[hos host] vk: handle table: element 19 of program 71a3d4bd81edf4d7 names texture #3690 (format 97) for a depth comparison: bound as nothing
[hos host] vk: handle table: element 19 of program 71a3d4bd81edf4d7 names texture #3690 (format 97) for a depth comparison: bound as nothing
[hos host] vk: handle table: element 19 of program 9f7eab75bd327abd names texture #3690 (format 97) for a depth comparison: bound as nothing
[hos host] vk: handle table: element 8 of program 2fd8c7ca964e5a1d names texture #3690 (format 97) for a depth comparison: bound as nothing
[hos host] vk: handle table: element 8 of program 21c7c8a2d0fe7171 names texture #3690 (format 97) for a depth comparison: bound as nothing
[hos host] vk: handle table: element 8 of program 71a3d4bd81edf4d7 names texture #3690 (format 97) for a depth comparison: bound as nothing
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #20000: event 0x100ce58010 mode 0 value 11354 (now 11335) (lr 0x8066179c)
[hos host] vk: replay reads of guest memory as submitted: 280000 (137106 pages from the captures)
[hos host] vk: 2 textures retired (1657 left, 2734 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: memory: working set 20677 MB (peak 20694), private 27532 MB, staging 96 MB in 6 arenas, 10 made so far (18 ms)
[hos host] vk: memory heaps: heap 0 (device): 170 blocks 4420 MB, 15259 allocations 4236 MB; used 5296 of 7249 MB; heap 1 (system): 26 blocks 772 MB, 38 allocations 661 MB; used 1980 of 15577 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 4 images 12 MB; 1 free ranges, the largest 19 MB
[hos host] vk: image blocks (render targets): 27 blocks 432 MB, 108 images 326 MB; 45 free ranges, the largest 16 MB
[hos host] vk: image blocks (textures, compacted): 38 blocks 1216 MB, 1539 images 1208 MB; 589 free ranges, the largest 5 MB
[hos host] vk: render target images: 149 reused from the pool, 278 made new, 166 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 5296 MB used of 7249 MB allowed (183 MB of it free in the blocks); textures 2743 MB (budget 4676 MB), 1658 textures
[hos host] vk: budget: VRAM 4420 of 7249 MB (5296 MB by the driver, 875 MB unaccounted): textures 2743, mirrors 835, shadows 344, blocks free 183, other buffers 158, staging 96, other images 60; RAM 1819 MB (the process's private memory 27532 MB): imports 1108, capture spares 496, captures 188, capture table 27, others 0
[hos host] vk: textures: decoded ASTC 1182 MB, rendered 912 MB (102 drawn in one frame, kept: 49063 KB), other 380 MB; used in the last 8 presents 1748 MB
[hos host] vk: decoded ASTC used again: 0 textures (0 MB) copied from ones the game finalized, not decoded; 0 of those retired, 3 listed; 849 MB of blocks hashed in 85 ms
[hos host] vk:   288 MB: 0x114737c000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a52b2c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a1a3a600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x1136cec000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x113c30f000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x1136cec000 4096x4096x1 x1 levels 1 NVN 0xa, last used 511 presents ago
[hos host] vk:   47 MB: 0x10bb22a200 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 4 presents ago
[hos host] vk:   47 MB: 0x10af4c2400 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 1 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 2 (177 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 260x260x1 NVN 0xd: 1 (175 KB);
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos Presentation Thread] vk: presents after the CPU's signal (600): 110 waited for it, 7.29 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: readback pages refreshed from guest memory: 140056 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: status: 600 frames, the slowest 344.1 ms; VRAM 5.2 of 7.1 GB
[hos host] vk: a frame: 35 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 5.29, present 0.58, draw 2.91, descriptors 1.28, texture lookups 0.10, uploads 0.00, pipelines 0.11, programs 0.46, commands 0.87, dispatch 0.71, memory sync 0.34, queue submit 0.40, slot wait 0.04
[hos host] vk: a frame: mirrors 74 (835 MB): 4.06 MB synced in 777.4 copies, 17.16 MB written back
[hos host] vk: 2400 presents, 60.0 fps; a frame: 596 draws (0 skipped), 101 dispatches, 140 passes, 146 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37478 pipelines (5315077 ms), 0 programs (0 failed), 1658 textures (2743 MB)
[hos host] vk: host time in GPU resource waits: 0.04 ms a frame (fences 0.02, 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: 332.9 full barriers recorded, 188.1 left out (nothing recorded since the one before), 18.3 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 47.0; 0x12: 68.8; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.3 ms a frame (the longest submission 3.0 ms); 0.1 textures made a frame: colour targets 0.1 (0.2 MB), depth 0.1 (0.7 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: the render thread (600 presents, 18.15 ms a frame): its jobs 8.23 ms a frame (submits 5.89, presents 0.35, others 1.99), idle 9.91; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 4.35 jobs queued at a present); 2 catch-ups (a finish, a texture read back, a write-back not recorded yet: 0), 3.31 ms each; 0 presents held for the CPU's signal, 0.00 ms each (0 let go by RX_VK_HOLD_MS); 114 captures at a CPU signal; the submits' waits read in 0.021 ms a frame; readbacks: 2026 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 7.03 ms after its hand-over (299.41 at most)
[hos host] vk: frame overlap (600 presents): the GPU between recordings, ms a frame, idle + in the next one's uploads: up to a frame's first recording with a draw 8.81 + 0.34 (the window's copy is in the idle), after it 1.23 + 0.29; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.03 ms of GPU time), after its last that reads it 290.03 (0.05 ms)
[hos host] vk: frame ahead (600 presents): 3.25 parts a frame recorded in the signal's job, 78.4% of them took nothing that came with the signal; of the others, 0.70 a frame took pages of the signal's capture into a mirror (52.5 pages a frame) and 0.01 pages read as guest memory had them (1.4 pages a frame); a frame's first part took nothing in 82 of 114 frames, its first three in 81
[hos host] vk: frame ahead, what came with the signal: 114 signals with a capture, each of 87.4 pages in 2.3 pools; between a frame's last submit and that signal 0.60 texture and sampler registrations, 0.54 releases and 0.00 other CPU signals a signal (11 of them with any)
[hos host] vk: frame overlap, by place: idle + uploads before a frame's 2nd to 12th recording with a draw, then all later ones: 0.19+0.06, 0.50+0.09, 0.04+0.06, 0.04+0.02, 0.03+0.03, 0.04+0.01, 0.01+0.00, 0.10+0.01, 0.02+0.00, 0.06+0.00, 0.14+0.01, 0.06+0.00
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 197 KB; 0 made on demand (0 MB kept now); c1 stages: at most 76 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: write-backs left out (600 submits): 1264, 5148.4 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 72 dirty pages from their submits' captures, 57204 from earlier ones', 0 from guest memory as it was at the replay; 24 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1328 dispatches, 42738 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6897 runs, 896.0 MB compared; 114 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 22864 in captures, 0 only in the collection, 54 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vk: capture: a submit 0.13 ms (lock 0.00, scan 0.07 over 1351 ranges, 2851 MB), 207 pages written, 207 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.005, bytes copied 0.000; 25 kept (182 MB), 11 spare buffers, 2071 texture pages copied (8 MB)
[hos sead::MainThread] vcores: 136183 takes, 21009 waits (mean 54 us, longest 5.3 ms, 0 over HOS_VCORE_LOG_MS), 10470 kicks, 10315 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: 1 textures retired (1658 left, 2743 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: mirror of pool 0x104e800000 (8 MB) given back: idle, 6018 submissions since its sync
[hos host] vk: mirror of pool 0x7fe2400000 (4 MB) given back: idle, 6018 submissions since its sync
[hos host] vk: mirror of pool 0x1041c00000 (4 MB) given back: idle, 6017 submissions since its sync
[hos host] vk: mirror of pool 0x1042000000 (4 MB) given back: idle, 6015 submissions since its sync
[hos host] vk: 2 textures retired (1656 left, 2726 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: 1 textures retired (1656 left, 2726 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 300000 (147798 pages from the captures)
[hos sead::MainThread] nvn: 200000 recordings made, 42894 live copies (11.8 MB), 357 texture copies; 5 command buffers' segments (4.3 MB)
[hos host] vk: 1 textures retired (1656 left, 2726 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: sparse mirrors: 59, 823 MB with memory of 1044 MB (220 MB spared); blocks given memory later 90, given back 1245; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 92 binds (11.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 67 made new; 15 kept as spares, 0 destroyed; 8 spares now, 280 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1243, 5057.6 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 21 dirty pages from their submits' captures, 56099 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1308 dispatches, 41851 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6740 runs, 883.7 MB compared; 115 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 22405 in captures, 0 only in the collection, 55 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 65 live (1108 MB), 0 retired still imported (0 MB), 11 retired given back (320 MB)
[hos host] vk: 1 textures retired (1655 left, 2717 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: 1 textures retired (1654 left, 2709 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 320000 (158398 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 4356 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: render targets drawn over from another start, let go: 253 so far (1159 MB), 249 of them (1137 MB) between the passes
[hos host] vk: 1 textures retired (1655 left, 2715 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: 1 textures retired (1654 left, 2706 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos sead::MainThread] nvn: EventSignal (CPU) #5000: event 0x10065c8004 mode 0 value 1 (lr 0x8348c78c)
[hos Presentation Thread] vk: capture: a submit 0.13 ms (lock 0.00, scan 0.07 over 1351 ranges, 2851 MB), 207 pages written, 207 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.005, bytes copied 0.000; 25 kept (174 MB), 11 spare buffers, 2071 texture pages copied (8 MB)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1113 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 ModuleSystemWorker2] vk: invalidates since the last readbacks line: 4057 kept their virtual core (nothing there to wait for), 2855 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: write-backs left out (600 submits): 1264, 5148.4 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 62 dirty pages from their submits' captures, 57012 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1326 dispatches, 42555 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6736 runs, 883.8 MB compared; 115 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 22851 in captures, 1 only in the collection, 54 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 340000 (168988 pages from the captures)
[hos host] vk: 1 textures retired (1653 left, 2698 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a frame: 36 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 5.25, present 0.24, draw 2.86, descriptors 1.23, texture lookups 0.10, uploads 0.00, pipelines 0.11, programs 0.45, commands 0.87, dispatch 0.65, memory sync 0.33, queue submit 0.40, slot wait 0.04
[hos host] vk: a frame: mirrors 74 (823 MB): 4.06 MB synced in 774.4 copies, 16.07 MB written back
[hos host] vk: 2520 presents, 60.0 fps; a frame: 634 draws (0 skipped), 98 dispatches, 141 passes, 146 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37478 pipelines (5315077 ms), 0 programs (0 failed), 1653 textures (2698 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.02, 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: 330.6 full barriers recorded, 188.4 left out (nothing recorded since the one before), 18.6 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 47.0; 0x12: 69.2; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.4 ms a frame (the longest submission 3.0 ms); 0.1 textures made a frame: colour targets 0.1 (0.2 MB), depth 0.1 (0.4 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: 1 textures retired (1652 left, 2689 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: queues ordered by their events: 12000 submissions stopped at a wait, 11999 went on once it was reached, 0 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 0 (stopped 0.00 ms at most); waits met 23813, not met 0; captures at a CPU event's wait 1781, 0 of them at a wait met on arrival
[hos host] vk: 1 textures retired (1652 left, 2689 MB): budget 0, drawn over 1, dead targets 0
[hos Presentation Thread] vi: 2555 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 300, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos host] vk: replay reads of guest memory as submitted: 360000 (179499 pages from the captures)
[hos host] vk: sparse mirrors: 59, 823 MB with memory of 1044 MB (220 MB spared); blocks given memory later 90, given back 1245; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 92 binds (11.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 67 made new; 15 kept as spares, 0 destroyed; 8 spares now, 280 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1243, 5057.6 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 51 dirty pages from their submits' captures, 55921 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1304 dispatches, 41826 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6696 runs, 876.7 MB compared; 117 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 22487 in captures, 1 only in the collection, 55 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 65 live (1108 MB), 0 retired still imported (0 MB), 11 retired given back (320 MB)
[hos host] vk: 1 textures retired (1652 left, 2689 MB): budget 0, drawn over 1, dead targets 0
[hos ModuleSystemWorker3] nvn: CommandBufferSignalEvent #15000: event 0x100ce58010 mode 0 value 15002 (lr 0x80660bcc)
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #25000: event 0x100ce5800c mode 0 value 15153 (now 15134) (lr 0x8066179c)
[hos host] vk: program picks: 348222 kept (104013 by the last picks' textures, a uniform buffer bound since), 712 new, 251065 with other textures (of 600000)
[hos host] vk: replay reads of guest memory as submitted: 380000 (189928 pages from the captures)
[hos host] vk: texture #3696 0x10006bc0600 1453x1453 x1 NVN 0x34 the GPU drew, finalized by the game, sampled again at frame 2598 (drawn at 2537)
[hos host] vk: 1 textures retired (1656 left, 2691 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 4285 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos Presentation Thread] vk: capture: a submit 0.13 ms (lock 0.00, scan 0.07 over 1352 ranges, 2855 MB), 207 pages written, 207 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.005, bytes copied 0.000; 25 kept (188 MB), 11 spare buffers, 2070 texture pages copied (8 MB)
[hos host] vk: write-backs left out (600 submits): 1265, 5149.8 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 80 dirty pages from their submits' captures, 56973 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1327 dispatches, 42576 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6744 runs, 883.8 MB compared; 111 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 22945 in captures, 0 only in the collection, 54 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 400000 (200352 pages from the captures)
[hos host] vk: a frame: 32 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 0.24, draw 3.21, descriptors 1.38, texture lookups 0.09, uploads 0.00, pipelines 0.12, programs 0.51, commands 0.97, dispatch 0.74, memory sync 0.35, queue submit 0.43, slot wait 0.05
[hos host] vk: a frame: mirrors 74 (823 MB): 4.06 MB synced in 774.1 copies, 16.14 MB written back
[hos host] vk: 2640 presents, 60.0 fps; a frame: 622 draws (0 skipped), 98 dispatches, 140 passes, 146 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37478 pipelines (5315077 ms), 0 programs (0 failed), 1657 textures (2702 MB)
[hos host] vk: host time in GPU resource waits: 0.04 ms a frame (fences 0.03, 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: 330.9 full barriers recorded, 188.0 left out (nothing recorded since the one before), 18.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 47.0; 0x12: 68.8; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.4 ms a frame (the longest submission 3.0 ms); 0.1 textures made a frame: colour targets 0.0 (0.2 MB), depth 0.1 (0.5 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: 9 textures retired (1658 left, 2708 MB): budget 0, drawn over 0, dead targets 9
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1100 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 ModuleSystemWorker2] vk: invalidates since the last readbacks line: 4256 kept their virtual core (nothing there to wait for), 2912 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: readback pages refreshed from guest memory: 210589 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 197 KB; 0 made on demand (0 MB kept now); c1 stages: at most 76 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 59, 823 MB with memory of 1044 MB (220 MB spared); blocks given memory later 90, given back 1245; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 92 binds (11.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 67 made new; 15 kept as spares, 0 destroyed; 8 spares now, 280 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1242, 5056.2 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 27 dirty pages from their submits' captures, 56544 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1306 dispatches, 41926 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6900 runs, 900.6 MB compared; 122 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 22622 in captures, 0 only in the collection, 55 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 65 live (1108 MB), 0 retired still imported (0 MB), 11 retired given back (320 MB)
[hos host] vk: 1 textures retired (1657 left, 2699 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 420000 (211010 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos sead::MainThread] vk: capture: a submit 0.13 ms (lock 0.00, scan 0.07 over 1352 ranges, 2855 MB), 208 pages written, 208 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.005, bytes copied 0.000; 25 kept (188 MB), 11 spare buffers, 2070 texture pages copied (8 MB)
[hos BaseProcExecutor0] vcores: 136449 takes, 20651 waits (mean 58 us, longest 4.8 ms, 0 over HOS_VCORE_LOG_MS), 10118 kicks, 10020 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: 1 textures retired (1656 left, 2693 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 4391 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: replay reads of guest memory as submitted: 440000 (221397 pages from the captures)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: write-backs left out (600 submits): 1265, 5149.8 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 29 dirty pages from their submits' captures, 56865 from earlier ones', 0 from guest memory as it was at the replay; 24 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1327 dispatches, 42535 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 7066 runs, 922.9 MB compared; 119 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 22830 in captures, 0 only in the collection, 54 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: 2 textures retired (1655 left, 2685 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: recorder: 3275923 packets run in 10.0 s (busy 1.20 s, spun idle 2.53 s, slept 4843 times, 1407 of them under 200 us), 0 calls made directly (no room); the replay waited for it 20529 times, 103.92 ms in all, the longest 2.56 ms; replay_mu unlock 7205/42.99 ms (max 0.23), vkEndCommandBuffer 6687/41.87 ms (max 2.56), vkResetDescriptorPool 6637/19.06 ms (max 0.15)
[hos host] vk: 1 textures retired (1654 left, 2676 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a frame: 29 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 5.24, present 0.22, draw 2.83, descriptors 1.20, texture lookups 0.08, uploads 0.00, pipelines 0.11, programs 0.48, commands 0.85, dispatch 0.67, memory sync 0.34, queue submit 0.39, slot wait 0.04
[hos host] vk: a frame: mirrors 74 (823 MB): 4.06 MB synced in 774.7 copies, 16.67 MB written back
[hos host] vk: 2760 presents, 60.1 fps; a frame: 609 draws (0 skipped), 98 dispatches, 141 passes, 146 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 37478 pipelines (5315077 ms), 0 programs (0 failed), 1654 textures (2676 MB)
[hos host] vk: host time in GPU resource waits: 0.07 ms a frame (fences 0.02, timelines 0.04, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 330.4 full barriers recorded, 188.2 left out (nothing recorded since the one before), 18.6 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 47.0; 0x12: 69.0; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.4 ms a frame (the longest submission 3.0 ms); 0.2 textures made a frame: colour targets 0.1 (0.5 MB), depth 0.1 (0.4 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: Auto (this PC), measured in play: settled (60 fps of 60, the GPU at 7.4 ms a frame of 16.7): Render resolution 100% kept
[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: 460000 (231761 pages from the captures)
[hos sead::MainThread] nvn: DrawMeshTasksIndirect: 40000 so far (x1 0x117c74b040 x2 0x2 x3 0x117c74b040 x4 0x9b040)
[hos host] vk: sparse mirrors: 59, 823 MB with memory of 1044 MB (220 MB spared); blocks given memory later 90, given back 1245; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 92 binds (11.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 67 made new; 15 kept as spares, 0 destroyed; 8 spares now, 280 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1242, 5056.2 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 9 dirty pages from their submits' captures, 56041 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1303 dispatches, 41764 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6647 runs, 867.8 MB compared; 116 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 22391 in captures, 2 only in the collection, 55 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 65 live (1108 MB), 0 retired still imported (0 MB), 11 retired given back (320 MB)
[hos ModuleSystemWorker3] nvn: CommandBufferWaitEvent #30000: event 0x100ce5801c mode 0 value 18957 (now 18938) (lr 0x8066179c)
[hos host] control: teleport to destination 3 asked for
[hos BaseProcExecutor2] cheats: teleport child to Market_Child at Market_Entrance (boot Dm_SY_BancChangeArrive_Market_02, event Dm_SY_BancChangeGo, starter 1): request sent, result 0x1
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos sead::MainThread] vk: capture: a submit 0.13 ms (lock 0.00, scan 0.07 over 1352 ranges, 2855 MB), 209 pages written, 209 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.005, bytes copied 0.000; 25 kept (174 MB), 11 spare buffers, 2070 texture pages copied (8 MB)
[hos host] vk: replay reads of guest memory as submitted: 480000 (242494 pages from the captures)
[hos sead::MainThread] vk: readbacks of GPU results: 1076 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 sead::MainThread] vk: invalidates since the last readbacks line: 4193 kept their virtual core (nothing there to wait for), 3231 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: write-backs left out (600 submits): 1218, 4961.1 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 17 dirty pages from their submits' captures, 58734 from earlier ones', 0 from guest memory as it was at the replay; 24 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1337 dispatches, 42306 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6262 runs, 831.9 MB compared; 108 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 22668 in captures, 0 only in the collection, 52 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker2] nvn: 250000 recordings made, 43276 live copies (12.2 MB), 397 texture copies; 3 command buffers' segments (4.1 MB)
[hos host] vk: replay reads of guest memory as submitted: 500000 (253876 pages from the captures)
[hos sead::MainThread] nvn: CommandBufferSignalEvent #20000: event 0x100ce58014 mode 0 value 20000 (lr 0x80660bcc)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 4392 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: texture #1660 0x1001f347e00 285x94x1 x1 NVN 0x25 the GPU drew goes with no copy of its texels (finalized, unused for 2593 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: texture #1671 0x1001f2a9000 631x94x1 x1 NVN 0x25 the GPU drew goes with no copy of its texels (finalized, unused for 2593 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: 154 textures retired (1566 left, 3043 MB): budget 153, drawn over 1, dead targets 0; GPU-drawn with no copy kept: 0 live, 2 finalized, 0 MB
[hos Presentation Thread] vi: 2855 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 300, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos host] vk: 145 textures retired (1439 left, 3074 MB): budget 144, drawn over 1, dead targets 0
[hos host] vk: program picks: 465865 kept (147972 by the last picks' textures, a uniform buffer bound since), 731 new, 333403 with other textures (of 800000)
[hos host] vk: slow submit: 17.0 ms: texture lookups 16.8 uploads 16.3
[hos host] vk: 66 textures retired (1407 left, 3072 MB): budget 56, drawn over 10, dead targets 0
[hos sead::MainThread] nvn: GPU waits in 5 s: 18 sync waits (477 ms), 0 polls timed out (0 ms), 2 queue finishes (13 ms)
[hos host] vk: slow submit: 18.1 ms: texture lookups 18.0 uploads 15.7
[hos host] vk: slow upload 16.8 ms: #3994 0x10dbba6000 1024x1024x1 x1 levels 11 NVN 0x7b ASTC (BC3), 912 of 912 KB
[hos host] vk: slow submit: 17.9 ms: texture lookups 17.8 uploads 17.7 queue submit 1.8
[hos host] vk: texture 0x10dbd41e00 683x683x1 x1 levels 10 NVN 0x8d -> vk 137 (ASTC decoded) (1500 textures, 3224 MB)
[hos host] vk: slow submit: 20.9 ms: texture lookups 20.8 uploads 20.5
[hos host] vk: slow submit: 18.4 ms: texture lookups 18.2 uploads 17.9
[hos host] vk: slow submit: 20.4 ms: texture lookups 20.3 uploads 20.0
[hos host] vk: texture #1648 0x1001f3b1c00 240x94x1 x1 NVN 0x25 the GPU drew goes with no copy of its texels (finalized, unused for 2329 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: texture #2797 0x1001ef9aa00 1466x282x1 x1 NVN 0x25 the GPU drew goes with no copy of its texels (finalized, unused for 1298 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: texture #429 0x10000348400 2560x1440x1 x1 NVN 0x38 the GPU drew goes with no copy of its texels (finalized, unused for 1290 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: texture #427 0x10755c1000 1920x1080x1 x1 NVN 0x3d the GPU drew goes with no copy of its texels (finalized, unused for 1290 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: texture #426 0x1074d44000 1920x1080x1 x1 NVN 0x25 the GPU drew goes with no copy of its texels (finalized, unused for 1290 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: texture #425 0x1000021c200 960x540x1 x1 NVN 0xd the GPU drew goes with no copy of its texels (finalized, unused for 1290 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: 261 textures retired (1523 left, 3124 MB): budget 254, drawn over 7, dead targets 0; GPU-drawn with no copy kept: 0 live, 6 finalized, 32 MB
[hos host] vk: a frame: 42 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 7.64, present 0.27, draw 3.05, descriptors 1.31, texture lookups 2.22, uploads 1.52, pipelines 0.11, programs 0.51, commands 0.91, dispatch 0.60, memory sync 0.40, queue submit 0.45, slot wait 0.04
[hos host] vk: a frame: mirrors 74 (823 MB): 3.85 MB synced in 712.2 copies, 14.30 MB written back
[hos host] vk: 2880 presents, 59.2 fps; a frame: 713 draws (0 skipped), 87 dispatches, 124 passes, 131 barriers, 6 submits, 4.0 uploads (1.4 MB), 73 ub updates (51 overlays), GPU-bound wait 0.0 ms (recording slots only); 37478 pipelines (5315077 ms), 0 programs (0 failed), 1555 textures (3145 MB)
[hos host] vk: host time in GPU resource waits: 1.93 ms a frame (fences 1.30, timelines 0.62, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 351.0 full barriers recorded, 171.2 left out (nothing recorded since the one before), 16.5 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 1.8; 0x8: 2.7; 0x9: 0.9; 0x11: 41.9; 0x12: 62.3; 0x13: 1.8; 0x21: 3.6; 0x28: 12.5; 0x92: 2.7; 0x97: 0.9;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 3.1 ms); 4.7 textures made a frame: colour targets 0.2 (0.9 MB), depth 0.6 (5.3 MB), ASTC or BC 3.9 (3.2 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.34 MB unswizzled on the CPU in 0.03 ms
[hos host] vk: slow submit: 15.8 ms: texture lookups 15.7 uploads 15.6
[hos host] vk: slow submit: 17.7 ms: texture lookups 17.6 uploads 17.5
[hos host] vk: slow submit: 17.4 ms: texture lookups 16.8 uploads 16.6
[hos host] vk: slow submit: 18.9 ms: texture lookups 18.8 uploads 18.5
[hos host] vk: texture #3143 0x10003b89600 1036x720x1 x1 NVN 0x1 the GPU drew goes with no copy of its texels (finalized, unused for 989 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: texture #3188 0x10004308e00 2072x1440x1 x1 NVN 0x25 the GPU drew goes with no copy of its texels (finalized, unused for 989 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: texture #3144 0x100039f1200 1036x720x1 x1 NVN 0x1 the GPU drew goes with no copy of its texels (finalized, unused for 989 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: texture #3142 0x10003abd400 1036x720x1 x1 NVN 0x1 the GPU drew goes with no copy of its texels (finalized, unused for 989 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: texture #3136 0x10003c55800 2200x2200x1 x3 NVN 0x32 the GPU drew goes with no copy of its texels (finalized, unused for 989 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: 90 textures retired (1773 left, 3170 MB): budget 81, drawn over 9, dead targets 0; GPU-drawn with no copy kept: 0 live, 5 finalized, 41 MB
[hos host] vk: slow submit: 21.9 ms: texture lookups 21.7 uploads 21.3
[hos host] vk: slow submit: 20.2 ms: texture lookups 20.0 uploads 19.6
[hos host] vk: slow submit: 19.8 ms: texture lookups 19.6 uploads 19.4
[hos host] vk: slow submit: 22.1 ms: texture lookups 22.0 uploads 21.5
[hos host] vk: texture 0x10eb8bba00 512x512x1 x1 levels 10 NVN 0x8c -> vk 137 (ASTC decoded) (2000 textures, 3305 MB)
[hos host] vk: slow submit: 21.7 ms: texture lookups 21.6 uploads 21.1
[hos host] vk: sparse mirrors: 59, 823 MB with memory of 1044 MB (220 MB spared); blocks given memory later 90, given back 1245; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 92 binds (11.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 67 made new; 15 kept as spares, 0 destroyed; 8 spares now, 280 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 713, 2902.6 MB; 32 bindings followed, 48 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 62 dirty pages from their submits' captures, 35660 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 913 dispatches, 25634 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 3907 runs, 514.2 MB compared; 100 pages read into the shadows; 344 MB of shadows; pages the shadows knew that the CPU wrote: 13650 in captures, 3 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 66 live (1124 MB), 0 retired still imported (0 MB), 11 retired given back (320 MB)
[hos Presentation Thread] vk: capture: a submit 0.13 ms (lock 0.01, scan 0.06 over 1223 ranges, 3198 MB), 177 pages written, 177 copied (0.7 MB)
[hos host] vk: slow submit: 20.5 ms: texture lookups 20.4 uploads 20.1
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.006, bytes copied 0.000; 25 kept (170 MB), 10 spare buffers, 2071 texture pages copied (8 MB)
[hos host] control: state play -> loading at frame 2898 (6.4 draws a frame)
[hos host] vk: slow submit: 22.0 ms: texture lookups 21.0 uploads 20.5
[hos host] vk: 145 textures retired (2059 left, 3220 MB): budget 135, drawn over 10, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.07
[hos host] vk: slow submit: 15.1 ms: texture lookups 15.0 uploads 14.9
[hos host] vk: slow submit: 20.5 ms: texture lookups 20.4 uploads 20.1
[hos host] vk: slow submit: 20.5 ms: texture lookups 20.3 uploads 19.8
[hos host] vk: slow submit: 20.3 ms: texture lookups 20.2 uploads 19.9
[hos host] vk: slow submit: 20.9 ms: texture lookups 20.7 uploads 20.3
[hos host] vk: slow submit: 22.0 ms: texture lookups 21.9 uploads 21.5
[hos host] vk: 24 textures retired (2329 left, 3292 MB): budget 14, drawn over 10, dead targets 0
[hos host] vk: slow submit: 22.0 ms: texture lookups 21.8 uploads 21.3
[hos host] vk: texture 0x10f801da00 256x256x1 x1 levels 9 NVN 0x49 -> vk 139 (2500 textures, 3375 MB)
[hos host] vk: slow upload 15.2 ms: #5552 0x10f85abe00 768x768x1 x1 levels 10 NVN 0x7b ASTC (BC3), 528 of 528 KB
[hos host] vk: slow submit: 16.4 ms: texture lookups 16.3 uploads 16.1
[hos host] vk: slow submit: 15.5 ms: texture lookups 15.3 uploads 15.2
[hos host] vk: a 64 MB capture buffer made ahead in 1.8 ms, off the game's thread
[hos host] vk: slow submit: 17.5 ms: texture lookups 16.5 uploads 2.6
[hos host] vk: slow submit: 26.8 ms: memory sync 26.7 [memory sync: maps 0.0 (lock waits 0.0), capture 26.6, collect 0.1]
[hos host] vk: a 32 MB capture buffer made ahead in 0.8 ms, off the game's thread
[hos host] vk: slow submit: 21.7 ms: texture lookups 20.3 uploads 19.8 memory sync 1.2 [memory sync: maps 0.0 (lock waits 0.0), capture 0.2, collect 0.9]
[hos host] vk: 10 textures retired (2715 left, 3352 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: slow submit: 20.2 ms: texture lookups 20.1 uploads 19.7
[hos host] vk: slow submit: 17.0 ms: texture lookups 16.8 uploads 16.7
[hos host] vk: slow submit: 20.3 ms: texture lookups 19.2 uploads 18.8
[hos host] vk: slow submit: 19.9 ms: texture lookups 19.8 uploads 19.4
[hos host] vk: slow submit: 19.9 ms: texture lookups 19.8 uploads 18.3
[hos host] vk: slow submit: 21.3 ms: texture lookups 21.0 uploads 19.9
[hos PreActorModelCompose] mem: commit of 0x10000 bytes at the commit limit made after 50 ms (1 so far)
[hos ModuleSystemWorker1] vcores: ModuleSystemWorker1 (priority 16) waited 50.1 ms for a core: given by rmbl::VibrationThread (priority 15), which held it 0.0 ms and was never asked; the cores it may use now: core 1: it;
[hos PreActorModelCompose] mem: commit of 0x10000 bytes at the commit limit made after 50 ms (2 so far)
[hos host] vk: 8 textures retired (2969 left, 3561 MB): budget 3, drawn over 5, dead targets 0
[hos host] vk: texture 0x1103868000 341x341x1 x1 levels 9 NVN 0x49 -> vk 139 (3000 textures, 3579 MB)
[hos host] CRASH: exception 0xc0000005 (write 0x20, guest 0xfffff00000000020) at nvgpucomp64.dll+0x60aea0
[hos host] CRASH: host backtrace: nvgpucomp64.dll+0x60aea0 nvgpucomp64.dll+0x40f75be nvgpucomp64.dll+0x40ea3bb nvgpucomp64.dll+0x3f6ac00 nvgpucomp64.dll+0x3f6ace7 nvgpucomp64.dll+0x3383508 nvgpucomp64.dll+0x33835d4 nvgpucomp64.dll+0x45f4f46 nvgpucomp64.dll+0x45f3bae nvgpucomp64.dll+0x4608cd1 nvgpucomp64.dll+0x1fd3679 nvgpucomp64.dll+0x1fcf935 nvgpucomp64.dll+0x1fcf0be nvgpucomp64.dll+0x2058f93 nvgpucomp64.dll+0x1fcf82a nvoglv64.dll+0x6bcca nvoglv64.dll+0x6c54f nvoglv64.dll+0xf8b506 nvoglv64.dll+0xf3c9f6 nvoglv64.dll+0xf3e18d