Beanstalk Longshot test farm
Menu

Report rp_96c20d3527be1a22 fail

#1442 fix(vk): frame generation's fixed multipliers on the game's timeline under V-Sync On (FIFO) too

linux+amd on Grimlock456/agent-4835b5,

This report disagrees with its peers.

market-framegen market.fps_mean: -5.6% against peers' +65.3%
market-framegen market.frametime_1pct_low_ms: -0.3% against peers' -97.5%
market-framegen market.frametime_max_ms: -0.6% against peers' -97.3%
market-framegen market.frametime_p99_ms: -0.4% against peers' -97.7%
market-framegen market.hitch_count: +13.3% against peers' -100.0%
market-framegen market.stutter_count: +16.7% against peers' -100.0%
market-framegen#1 market.fps_mean: +32.6% against peers' +0.0%
market-framegen#2 market.frametime_1pct_low_ms: -0.0% against peers' +25.7%
market-framegen#2 market.frametime_max_ms: +0.5% against peers' +92.0%
market-framegen#3 market.draws_mean: -7.7% against peers' +58.4%
market-framegen#3 market.fps_mean: -44.1% against peers' -12.7%
market-framegen#3 market.frametime_1pct_low_ms: -0.2% against peers' +1244.9%
market-framegen#3 market.frametime_max_ms: -0.4% against peers' +4612.6%

Machine

Commit
0c552c97714ad41922bb616e5328cf968fc73898
Baseline
67efc4d3e305eccf11052b456783a1898d47c23d
App
nightly-156-4ec688b0, attempt 2
Game build
e0f3ccfa
os
linux
cpu
AMD Ryzen 7 5700X3D 8-Core Processor
gpu
AMD Radeon RX 7900 XT (RADV NAVI31)
tags
linuxamdradvdgpu
driver
26.2.6
ram_gb
32
vram_mb
20,464
gpu_vendor
amd
os_version
<host> (kernel 7.3.3)

Builds

main (baseline)

0 s warm

from 67efc4d3e305

the base was built already

PR

82.6 s warm

Program cache: warm

Scenarios

market-framegen vsync=1, frame_generation=2

ok
Metric (median)mainPRΔ
exit_code00±0%
market.fps_mean31.9830.2−5.6%
market.draws_mean797.4652.7−18.1%
log_forbidden_hits00±0%
market.hitch_count1517+13.3%
log_expected_missing00±0%
market.stutter_count1821+16.7%
market.frametime_max_ms1,013.281,007.4−0.6%
market.frametime_p50_ms16.5216.77+1.5%
market.frametime_p99_ms1,005.51,001.4−0.4%
market.frametime_1pct_low_ms1,007.491,004.39−0.3%
screenshot_diff—0.7—

Pictures against main: market diff 0.7004 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
market.fps_mean31.9830.2
market.draws_mean797.4652.7
log_forbidden_hits00
market.hitch_count1517
log_expected_missing00
market.stutter_count1821
market.frametime_max_ms1,013.281,007.4
market.frametime_p50_ms16.5216.77
market.frametime_p99_ms1,005.51,001.4
market.frametime_1pct_low_ms1,007.491,004.39
screenshot_diff—0.7

market-framegen vsync=1, frame_generation=3

ok
Metric (median)mainPRΔ
exit_code00±0%
market.fps_mean8.5811.38+32.6%
market.draws_mean2,154.41,864.8−13.4%
log_forbidden_hits00±0%
market.hitch_count2423−4.2%
log_expected_missing00±0%
market.stutter_count2423−4.2%
market.frametime_max_ms1,009.641,012.3+0.3%
market.frametime_p50_ms22.321.78−2.3%
market.frametime_p99_ms1,008.841,009.77+0.1%
market.frametime_1pct_low_ms1,009.171,010.96+0.2%
screenshot_diff—0.47—

Pictures against main: market diff 0.4673 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
market.fps_mean8.5811.38
market.draws_mean2,154.41,864.8
log_forbidden_hits00
market.hitch_count2423
log_expected_missing00
market.stutter_count2423
market.frametime_max_ms1,009.641,012.3
market.frametime_p50_ms22.321.78
market.frametime_p99_ms1,008.841,009.77
market.frametime_1pct_low_ms1,009.171,010.96
screenshot_diff—0.47

market-framegen vsync=0, frame_generation=2

ok
Metric (median)mainPRΔ
exit_code00±0%
market.fps_mean10.789.31−13.6%
market.draws_mean2,141.61,720.7−19.7%
log_forbidden_hits00±0%
market.hitch_count2222±0%
log_expected_missing00±0%
market.stutter_count2222±0%
market.frametime_max_ms1,0111,016.09+0.5%
market.frametime_p50_ms22.6920.12−11.3%
market.frametime_p99_ms1,009.631,007.47−0.2%
market.frametime_1pct_low_ms1,010.591,010.38−0%
screenshot_diff—0.7—

Pictures against main: market diff 0.6991 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
market.fps_mean10.789.31
market.draws_mean2,141.61,720.7
log_forbidden_hits00
market.hitch_count2222
log_expected_missing00
market.stutter_count2222
market.frametime_max_ms1,0111,016.09
market.frametime_p50_ms22.6920.12
market.frametime_p99_ms1,009.631,007.47
market.frametime_1pct_low_ms1,010.591,010.38
screenshot_diff—0.7

market-framegen vsync=0, frame_generation=3

ok
Metric (median)mainPRΔ
exit_code00±0%
market.fps_mean11.286.31−44.1%
market.draws_mean2,110.81,949.2−7.7%
log_forbidden_hits00±0%
market.hitch_count2325+8.7%
log_expected_missing00±0%
market.stutter_count2325+8.7%
market.frametime_max_ms1,019.291,014.82−0.4%
market.frametime_p50_ms21.8622.25+1.8%
market.frametime_p99_ms1,009.951,009.73−0%
market.frametime_1pct_low_ms1,013.631,011.68−0.2%
screenshot_diff—0.42—

Pictures against main: market diff 0.4221 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
market.fps_mean11.286.31
market.draws_mean2,110.81,949.2
log_forbidden_hits00
market.hitch_count2325
log_expected_missing00
market.stutter_count2325
market.frametime_max_ms1,019.291,014.82
market.frametime_p50_ms21.8622.25
market.frametime_p99_ms1,009.951,009.73
market.frametime_1pct_low_ms1,013.631,011.68
screenshot_diff—0.42

Rules

ScenarioCellRuleLimitmainPRΔ %Status
market-framegenvsync=1, frame_generation=2crashFalse00—pass
market-framegenvsync=1, frame_generation=2market.fps_mean.min_delta_pct1031.9830.2−5.6%pass
market-framegenvsync=1, frame_generation=2screenshot.max_diff0.1—0.7—fail
market-framegenvsync=1, frame_generation=2market.frametime_p99_ms.max_delta_pct151,005.51,001.4−0.4%pass
market-framegenvsync=1, frame_generation=2log_forbidden_hits.max000—pass
market-framegenvsync=1, frame_generation=3crashFalse00—pass
market-framegenvsync=1, frame_generation=3market.fps_mean.min_delta_pct108.5811.38+32.6%pass
market-framegenvsync=1, frame_generation=3screenshot.max_diff0.1—0.47—fail
market-framegenvsync=1, frame_generation=3market.frametime_p99_ms.max_delta_pct151,008.841,009.77+0.1%pass
market-framegenvsync=1, frame_generation=3log_forbidden_hits.max000—pass
market-framegenvsync=0, frame_generation=2crashFalse00—pass
market-framegenvsync=0, frame_generation=2market.fps_mean.min_delta_pct1010.789.31−13.6%fail
market-framegenvsync=0, frame_generation=2screenshot.max_diff0.1—0.7—fail
market-framegenvsync=0, frame_generation=2market.frametime_p99_ms.max_delta_pct151,009.631,007.47−0.2%pass
market-framegenvsync=0, frame_generation=2log_forbidden_hits.max000—pass
market-framegenvsync=0, frame_generation=3crashFalse00—pass
market-framegenvsync=0, frame_generation=3market.fps_mean.min_delta_pct1011.286.31−44.1%fail
market-framegenvsync=0, frame_generation=3screenshot.max_diff0.1—0.42—fail
market-framegenvsync=0, frame_generation=3market.frametime_p99_ms.max_delta_pct151,009.951,009.73±0%pass
market-framegenvsync=0, frame_generation=3log_forbidden_hits.max000—pass

Log (PR side, end)

[hos ModuleSystemWorker5] vk: readbacks of GPU results: 1062 of 3000 waited for their write-back, 0.50 ms each; 0 looked at the stopped queues first, 4 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 0.13, writes collected 0.06
[hos host] vk: 10 textures retired (4616 left, 8380 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: sparse mirrors: 94, 1076 MB with memory of 1596 MB (519 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 861, 4223.3 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1670 dirty pages from their submits' captures, 81793 from earlier ones', 9 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): 4827 runs, 728.8 MB compared; 132 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 19836 in captures, 18 only in the collection, 40 only by the replay itself
[hos host] vk:   read live, by pool: 0x1065800000 (8 MB, captured, state 2): 9;
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos host] vk: 23 textures retired (4614 left, 8418 MB): budget 0, drawn over 11, dead targets 12
[hos ModuleSystemWorker4] nvn: CommandBufferSignalEvent #160000: event 0x100d6c8000 mode 0 value 159999 (lr 0x80291440)
[hos host] vk: 10 textures retired (4623 left, 8473 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: recorder: 1518738 packets run in 10.0 s (busy 0.35 s, spun idle 0.94 s, slept 5161 times, 599 of them under 200 us), 0 calls made directly (no room); the replay waited for it 8279 times, 27.35 ms in all, the longest 0.34 ms; vkEndCommandBuffer 3162/13.70 ms (max 0.34), replay_mu unlock 1757/7.96 ms (max 0.28), vkResetDescriptorPool 3161/4.72 ms (max 0.20), vkDestroyImageView 199/0.98 ms (max 0.09)
[hos Presentation Thread] vi: 11024 frames presented (17.4 fps over the last 5.8 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 22, 15.5-17.8 30, 17.8-22 33, 22-30 11, 30-40 1, over 40 4
[hos host] vk: replay reads of guest memory as submitted: 8180000 (6884661 pages from the captures)
[hos sead::MainThread] nvn: GPU waits in 5 s: 19 sync waits (4034 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: 13 textures retired (4576 left, 8111 MB): budget 0, drawn over 10, dead targets 3
[hos host] vk: texture #40781 0x10001950600 1920x1080 NVN 0xa: the GPU wrote its storage (submission 77286): uploaded again
[hos host] vk: upload over a drawn image: #40781 0x10001950600 1920x1080x1 x1 NVN 0xa at present 11038 (texture frame 11039), finalized, 0 KB dirty in 0 ranges, a GPU buffer wrote it
[hos host] vk: a frame: 125 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 57 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 7.85, present 0.53, draw 3.73, descriptors 1.92, texture lookups 0.87, uploads 0.62, pipelines 0.14, programs 0.47, commands 0.87, dispatch 0.60, memory sync 1.13, queue submit 1.19, slot wait 0.04
[hos host] vk: a frame: mirrors 292 (1076 MB): 8.08 MB synced in 1211.4 copies, 17.81 MB written back
[hos host] vk: 11040 presents, 14.7 fps; a frame: 947 draws (0 skipped), 98 dispatches, 175 passes, 161 barriers, 6 submits, 1.9 uploads (2.7 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 43807 pipelines (3609 ms), 0 programs (10 failed), 4543 textures (7972 MB)
[hos host] vk: host time in GPU resource waits: 53.09 ms a frame (fences 0.00, timelines 53.09, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 561.8 full barriers recorded, 255.9 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: 41.0; 0x12: 88.2; 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 2.7 ms); 8.8 textures made a frame
[hos host] vk: a frame: 2.60 MB unswizzled on the CPU in 0.17 ms
[hos PreActorModelCompose] vcores: 47260 takes, 9493 waits (mean 85 us, longest 20.7 ms, 0 over HOS_VCORE_LOG_MS), 4663 kicks, 4699 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: write-backs left out (600 submits): 1071, 5253.4 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1750 dirty pages from their submits' captures, 95551 from earlier ones', 12 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): 5335 runs, 880.6 MB compared; 88 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 24411 in captures, 13 only in the collection, 51 only by the replay itself
[hos host] vk:   read live, by pool: 0x1065800000 (8 MB, captured, state 2): 12;
[hos Presentation Thread] vk: capture: a submit 0.34 ms (lock 0.01, scan 0.05 over 835 ranges, 2615 MB), 894 pages written, 894 copied (3.5 MB)
[hos host] vk: a capture taken (ms): all 0.131; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.125, bytes copied 0.000; 25 kept (266 MB), 10 spare buffers, 15151 texture pages copied (59 MB)
[hos host] vk: replay reads of guest memory as submitted: 8200000 (6901872 pages from the captures)
[hos host] control: state play -> loading at frame 11069 (0.0 draws a frame)
[hos host] control: state loading -> play at frame 11084 (485.4 draws a frame)
[hos Presentation Thread] vi: 11085 frames presented (12.2 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 22, 15.5-17.8 20, 17.8-22 10, 22-30 5, 30-40 0, over 40 4
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 168121 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.13, writes collected 0.06
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #220000: event 0x100d6c8014 mode 0 value 161619 (now 161600) (lr 0x80661124)
[hos host] vk: replay reads of guest memory as submitted: 8220000 (6919967 pages from the captures)
[hos host] vk: sparse mirrors: 94, 1076 MB with memory of 1596 MB (519 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 1071, 5253.4 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1850 dirty pages from their submits' captures, 94762 from earlier ones', 1 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): 5440 runs, 910.5 MB compared; 92 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 24190 in captures, 19 only in the collection, 52 only by the replay itself
[hos host] vk:   read live, by pool: 0x1043000000 (4 MB, captured, state 2): 1;
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos host] vk: ASTC texture #40961 0x11321db800 768x768 x1 levels 10: the image of #40952 0x110e00f600 (the same blocks, finalized by the game) copied, not decoded again (5000 so far, 8004 MB)
[hos host] vk: readback pages refreshed from guest memory: 4104147 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 8240000 (6937851 pages from the captures)
[hos host] vk: texture #41068 0x10001950600 1920x1080 NVN 0xa: the GPU wrote its storage (submission 77993): uploaded again
[hos host] vk: upload over a drawn image: #41068 0x10001950600 1920x1080x1 x1 NVN 0xa at present 11138 (texture frame 11139), finalized, 0 KB dirty in 0 ranges, a GPU buffer wrote it
[hos host] vk: 11 textures retired (4504 left, 7672 MB): budget 0, drawn over 0, dead targets 11
[hos sead::MainThread] nvn: GPU waits in 5 s: 14 sync waits (4051 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos ModuleSystemWorker4] vcores: 73526 takes, 14646 waits (mean 83 us, longest 18.9 ms, 0 over HOS_VCORE_LOG_MS), 7231 kicks, 7223 yields, 0 starved grants, 0 escapes in 5 s
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 867 of 3000 waited for their write-back, 0.48 ms each; 0 looked at the stopped queues first, 5 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: texture storage read as submitted (124001 times; 4110758 pages from the copies, 0 of them rewritten in guest memory since their submit; 15151 pages kept, 37 spans, 110 MB)
[hos host] vk: 1 textures retired (4521 left, 7800 MB): budget 0, drawn over 0, dead targets 1
[hos sead::MainThread] vk: capture: a submit 0.33 ms (lock 0.01, scan 0.05 over 835 ranges, 2615 MB), 881 pages written, 881 copied (3.4 MB)
[hos host] vk: a capture taken (ms): all 0.128; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.122, bytes copied 0.000; 25 kept (436 MB), 5 spare buffers, 15151 texture pages copied (59 MB)
[hos host] vk: write-backs left out (600 submits): 1029, 5047.4 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1713 dirty pages from their submits' captures, 92204 from earlier ones', 25 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): 5269 runs, 864.5 MB compared; 171 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 23236 in captures, 26 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool: 0x105dc00000 (4 MB, captured, state 2): 15; 0x1065800000 (8 MB, captured, state 2): 9; 0x1043000000 (4 MB, captured, state 2): 1;
[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 57 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.26, present 0.44, draw 2.74, descriptors 1.49, texture lookups 0.70, uploads 0.56, pipelines 0.10, programs 0.34, commands 0.57, dispatch 0.55, memory sync 1.14, queue submit 1.03, slot wait 0.04
[hos host] vk: a frame: mirrors 292 (1076 MB): 7.43 MB synced in 1142.0 copies, 17.58 MB written back
[hos host] vk: 11160 presents, 20.0 fps; a frame: 537 draws (0 skipped), 86 dispatches, 139 passes, 149 barriers, 6 submits, 1.3 uploads (2.6 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 43807 pipelines (3609 ms), 0 programs (10 failed), 4517 textures (7825 MB)
[hos host] vk: host time in GPU resource waits: 36.48 ms a frame (fences 0.00, timelines 36.48, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 472.9 full barriers recorded, 235.6 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: 41.5; 0x12: 75.7; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.6 ms a frame (the longest submission 2.8 ms); 3.4 textures made a frame
[hos host] vk: a frame: 2.48 MB unswizzled on the CPU in 0.15 ms
[hos host] vk: replay reads of guest memory as submitted: 8260000 (6954986 pages from the captures)
[hos host] vk: 1 textures retired (4543 left, 8041 MB): budget 0, drawn over 0, dead targets 1
[hos host] vk: replay reads of guest memory as submitted: 8280000 (6972491 pages from the captures)
[hos host] vk: recorder: 891239 packets run in 10.0 s (busy 0.26 s, spun idle 0.70 s, slept 5335 times, 601 of them under 200 us), 0 calls made directly (no room); the replay waited for it 6241 times, 21.80 ms in all, the longest 0.38 ms; vkEndCommandBuffer 2080/9.16 ms (max 0.33), replay_mu unlock 1981/8.08 ms (max 0.38), vkResetDescriptorPool 2075/3.90 ms (max 0.02), vkDestroyImageView 104/0.66 ms (max 0.33), vmaInvalidateAllocation 1/0.00 ms (max 0.00)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 154458 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 632 KB; 0 made on demand (0 MB kept now); c1 stages: at most 257 a recording; 19 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 94, 1076 MB with memory of 1596 MB (519 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 756, 3708.3 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1462 dirty pages from their submits' captures, 69490 from earlier ones', 2 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4487 runs, 640.1 MB compared; 74 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 17408 in captures, 14 only in the collection, 36 only by the replay itself
[hos host] vk:   read live, by pool: 0x7fa6800000 (20 MB, captured, state 2): 2;
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos Presentation Thread] vi: 11192 frames presented (18.4 fps over the last 5.8 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 34, 15.5-17.8 35, 17.8-22 28, 22-30 6, 30-40 0, over 40 4
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.13, writes collected 0.06
[hos host] vk: 1 textures retired (4545 left, 8162 MB): budget 0, drawn over 0, dead targets 1
[hos host] vk: replay reads of guest memory as submitted: 8300000 (6989722 pages from the captures)
[hos sead::MainThread] nvn: GPU waits in 5 s: 21 sync waits (4027 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos OptimizeChangeBanc] vcores: 47470 takes, 9769 waits (mean 94 us, longest 22.0 ms, 0 over HOS_VCORE_LOG_MS), 4840 kicks, 4866 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: texture storage read as submitted (126001 times; 4156830 pages from the copies, 0 of them rewritten in guest memory since their submit; 15151 pages kept, 37 spans, 110 MB)
[hos host] vk: write-backs left out (600 submits): 546, 2678.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1337 dirty pages from their submits' captures, 53721 from earlier ones', 4 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): 3278 runs, 464.8 MB compared; 63 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 12864 in captures, 10 only in the collection, 26 only by the replay itself
[hos host] vk:   read live, by pool: 0x7fd3800000 (8 MB, captured, state 2): 2; 0x105dc00000 (4 MB, captured, state 2): 2;
[hos host] vk: replay reads of guest memory as submitted: 8320000 (7006919 pages from the captures)
[hos host] vk: program picks: 7837167 kept (2442800 by the last picks' textures, a uniform buffer bound since), 14608 new, 3348224 with other textures (of 11200000)
[hos host] vk: replay reads of guest memory as submitted: 8340000 (7024872 pages from the captures)
[hos host] vk: sparse mirrors: 94, 1076 MB with memory of 1596 MB (519 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 420, 2060.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1094 dirty pages from their submits' captures, 46695 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): 2656 runs, 360.8 MB compared; 182 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9706 in captures, 8 only in the collection, 19 only by the replay itself
[hos host] vk:   read live, by pool: 0x105d800000 (4 MB, captured, state 2): 22; 0x7fda000000 (8 MB, captured, state 2): 1; 0x7fa0c00000 (4 MB, captured, state 2): 1;
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos host] vk: texture 0x10008558800 1056x1056x1 x1 levels 1 NVN 0x34 -> vk 126 (4500 textures, 7732 MB)
[hos host] vk: replay reads of guest memory as submitted: 8360000 (7042632 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.31 ms (lock 0.02, scan 0.04 over 835 ranges, 2615 MB), 813 pages written, 813 copied (3.2 MB)
[hos host] vk: a capture taken (ms): all 0.111; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.103, bytes copied 0.000; 25 kept (260 MB), 8 spare buffers, 15151 texture pages copied (59 MB)
[hos rmbl::VibrationThread] hid: 106496 vibration values, strongest 0.00 since the last count
[hos Presentation Thread] vi: 11246 frames presented (10.5 fps over the last 5.2 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 3, 17.8-22 27, 22-30 19, 30-40 1, over 40 4
[hos host] vk: replay reads of guest memory as submitted: 8380000 (7060371 pages from the captures)
[hos ModuleSystemWorker4] nvn: 2050000 recordings made, 71507 live copies (25.8 MB), 4567 texture copies; 5 command buffers' segments (4.2 MB)
[hos host] vk: render targets drawn over from another start, let go: 30668 so far (107863 MB), 27184 of them (94316 MB) between the passes
[hos host] vk: 11 textures retired (4518 left, 7794 MB): budget 0, drawn over 0, dead targets 11
[hos host] vk: write-backs left out (600 submits): 336, 1648.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1226 dirty pages from their submits' captures, 40200 from earlier ones', 10 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): 2231 runs, 292.0 MB compared; 57 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 8444 in captures, 5 only in the collection, 15 only by the replay itself
[hos host] vk:   read live, by pool: 0x1065800000 (8 MB, captured, state 2): 7; 0x7fe2400000 (4 MB, captured, state 2): 2; 0x7fe1000000 (4 MB, captured, state 2): 1;
[hos host] vk: readback pages refreshed from guest memory: 4177541 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 8400000 (7077988 pages from the captures)
[hos host] vk: 12 textures retired (4530 left, 7817 MB): budget 0, drawn over 1, dead targets 11
[hos host] control: state play -> menu at frame 11261 (31.0 draws a frame)
[hos sead::MainThread] nvn: GPU waits in 5 s: 36 sync waits (4086 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos BaseProcExecutor1] vcores: 45571 takes, 8941 waits (mean 110 us, longest 26.7 ms, 0 over HOS_VCORE_LOG_MS), 4366 kicks, 4443 yields, 0 starved grants, 0 escapes in 5 s
[hos host] control: state menu -> play at frame 11263 (1748.0 draws a frame)
[hos host] vk: sparse mirrors: 94, 1076 MB with memory of 1596 MB (519 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 357, 1751.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1265 dirty pages from their submits' captures, 42443 from earlier ones', 115 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): 2444 runs, 310.6 MB compared; 28 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 7970 in captures, 5 only in the collection, 15 only by the replay itself
[hos host] vk:   read live, by pool: 0x1058400000 (4 MB, captured, state 2): 115;
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos host] vk: recorder: 1184837 packets run in 10.0 s (busy 0.25 s, spun idle 0.76 s, slept 5019 times, 460 of them under 200 us), 0 calls made directly (no room); the replay waited for it 6043 times, 16.82 ms in all, the longest 0.39 ms; vkEndCommandBuffer 2476/8.90 ms (max 0.24), replay_mu unlock 1027/4.32 ms (max 0.21), vkResetDescriptorPool 2476/2.97 ms (max 0.02), vkDestroyImageView 63/0.63 ms (max 0.39), vmaInvalidateAllocation 1/0.00 ms (max 0.00)
[hos host] vk: 5 textures retired (4539 left, 7813 MB): budget 0, drawn over 5, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 8420000 (7094802 pages from the captures)
[hos Presentation Thread] vi: 11273 frames presented (4.8 fps over the last 5.6 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 0, 17.8-22 4, 22-30 17, 30-40 1, over 40 5
[hos ModuleSystemWorker2] nvn: CommandBufferSignalEvent #165000: event 0x100d6c8000 mode 0 value 164996 (lr 0x80291440)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 145266 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: present 11274 written to <data>/jobs/jb_31b2bb2f0e470894/runs/market-framegen-g0-01-pr/shots/market-framegen-c3-pr-01-market.png (the control bridge)
[hos host] vk: a frame: 351 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 57 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 11.84, present 0.64, draw 5.95, descriptors 2.67, texture lookups 0.91, uploads 0.66, pipelines 0.25, programs 0.76, commands 1.60, dispatch 0.79, memory sync 1.10, queue submit 1.73, slot wait 0.08
[hos host] vk: a frame: mirrors 292 (1076 MB): 9.04 MB synced in 1334.1 copies, 18.04 MB written back
[hos host] vk: 11280 presents, 8.3 fps; a frame: 1985 draws (0 skipped), 113 dispatches, 170 passes, 154 barriers, 6 submits, 2.3 uploads (2.9 MB), 82 ub updates (57 overlays), GPU-bound wait 0.1 ms (recording slots only); 43807 pipelines (3609 ms), 0 programs (10 failed), 4546 textures (7838 MB)
[hos host] vk: host time in GPU resource waits: 104.10 ms a frame (fences 0.00, timelines 104.10, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 597.4 full barriers recorded, 260.7 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: 41.0; 0x12: 81.1; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 11.5 ms a frame (the longest submission 3.4 ms); 4.4 textures made a frame
[hos host] vk: a frame: 2.88 MB unswizzled on the CPU in 0.20 ms
[hos host] vk: texture storage read as submitted (128001 times; 4209800 pages from the copies, 0 of them rewritten in guest memory since their submit; 15151 pages kept, 37 spans, 110 MB)
[hos host] vk: queues ordered by their events: 72000 submissions stopped at a wait, 71956 went on once it was reached, 42 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 3 (stopped 51.13 ms at most); waits met 224619, not met 0; captures at a CPU event's wait 3934, 0 of them at a wait met on arrival
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 632 KB; 0 made on demand (0 MB kept now); c1 stages: at most 257 a recording; 19 descriptor pools made when a recording's own ran out
[hos host] vk: write-backs left out (600 submits): 378, 1854.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1778 dirty pages from their submits' captures, 44494 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): 2645 runs, 327.5 MB compared; 130 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 8855 in captures, 7 only in the collection, 15 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 8440000 (7110895 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.12, writes collected 0.06
[hos host] vk: program picks: 7988554 kept (2496148 by the last picks' textures, a uniform buffer bound since), 14608 new, 3396837 with other textures (of 11400000)
[hos ModuleSystemWorker3] vk: readbacks of GPU results: 1034 of 3000 waited for their write-back, 0.46 ms each; 0 looked at the stopped queues first, 4 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #225000: event 0x100d6c8014 mode 0 value 165419 (now 165400) (lr 0x80661124)
[hos host] vk: replay reads of guest memory as submitted: 8460000 (7126993 pages from the captures)
[hos AudioUserTaskThread] vcores: 37350 takes, 7648 waits (mean 88 us, longest 15.5 ms, 0 over HOS_VCORE_LOG_MS), 3824 kicks, 3861 yields, 0 starved grants, 0 escapes in 6 s
[hos sead::MainThread] nvn: GPU waits in 5 s: 23 sync waits (5043 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: sparse mirrors: 94, 1076 MB with memory of 1596 MB (519 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1058 dirty pages from their submits' captures, 45732 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): 2769 runs, 345.5 MB compared; 134 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9327 in captures, 7 only in the collection, 18 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos host] vk: mirror of pool 0x7f71c00000 (4 MB) given back: idle, 6058 submissions since its sync
[hos host] vk: replay reads of guest memory as submitted: 8480000 (7142938 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1406 dirty pages from their submits' captures, 45324 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): 2772 runs, 346.6 MB compared; 133 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9340 in captures, 8 only in the collection, 20 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vi: 11325 frames presented (10.1 fps over the last 5.2 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 0, 17.8-22 22, 22-30 25, 30-40 1, over 40 4
[hos host] vk: replay reads of guest memory as submitted: 8500000 (7158903 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.34 ms (lock 0.01, scan 0.04 over 835 ranges, 2615 MB), 876 pages written, 876 copied (3.4 MB)
[hos host] vk: a capture taken (ms): all 0.127; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.116, bytes copied 0.000; 25 kept (248 MB), 4 spare buffers, 15151 texture pages copied (59 MB)
[hos host] vk: texture storage read as submitted (130001 times; 4255535 pages from the copies, 0 of them rewritten in guest memory since their submit; 15151 pages kept, 37 spans, 110 MB)
[hos sead::MainThread] vcores: 35551 takes, 6296 waits (mean 91 us, longest 17.9 ms, 0 over HOS_VCORE_LOG_MS), 3127 kicks, 3172 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: replay reads of guest memory as submitted: 8520000 (7174890 pages from the captures)
[hos host] vk: sparse mirrors: 93, 1073 MB with memory of 1592 MB (518 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1513 dirty pages from their submits' captures, 46535 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): 2789 runs, 347.5 MB compared; 146 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9588 in captures, 9 only in the collection, 18 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 25 sync waits (5047 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: recorder: 1212902 packets run in 10.0 s (busy 0.23 s, spun idle 0.76 s, slept 4960 times, 402 of them under 200 us), 0 calls made directly (no room); the replay waited for it 5952 times, 11.74 ms in all, the longest 0.26 ms; vkEndCommandBuffer 2465/6.62 ms (max 0.26), vkResetDescriptorPool 2464/2.90 ms (max 0.01), replay_mu unlock 983/2.17 ms (max 0.16), vkDestroyImageView 40/0.05 ms (max 0.01)
[hos host] vk: 9 textures retired (4545 left, 7801 MB): budget 0, drawn over 9, dead targets 0
[hos host] vk: readback pages refreshed from guest memory: 4250045 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 8540000 (7190919 pages from the captures)
[hos host] vk: 8 textures retired (4537 left, 7762 MB): budget 0, drawn over 8, dead targets 0
[hos host] vk: write-backs left out (600 submits): 420, 2060.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1436 dirty pages from their submits' captures, 46453 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): 2920 runs, 363.4 MB compared; 221 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9285 in captures, 7 only in the collection, 18 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 162130 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 12 textures retired (4525 left, 7713 MB): budget 0, drawn over 12, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 8560000 (7206989 pages from the captures)
[hos host] vk: program picks: 8138418 kept (2549017 by the last picks' textures, a uniform buffer bound since), 14608 new, 3446973 with other textures (of 11600000)
[hos Presentation Thread] vi: 11379 frames presented (10.5 fps over the last 5.2 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 0, 17.8-22 29, 22-30 21, 30-40 0, over 40 4
[hos host] vk: 10 textures retired (4515 left, 7680 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 632 KB; 0 made on demand (0 MB kept now); c1 stages: at most 257 a recording; 19 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 93, 1073 MB with memory of 1592 MB (518 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1137 dirty pages from their submits' captures, 45286 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): 2772 runs, 345.6 MB compared; 35 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9420 in captures, 6 only in the collection, 15 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos host] vk: replay reads of guest memory as submitted: 8580000 (7222932 pages from the captures)
[hos aal::LowPrioWorkerThread] vcores: 40951 takes, 8427 waits (mean 75 us, longest 18.3 ms, 0 over HOS_VCORE_LOG_MS), 4213 kicks, 4246 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: 11 textures retired (4504 left, 7647 MB): budget 0, drawn over 11, dead targets 0
[hos host] vk: mirror of pool 0x105dc00000 (4 MB) given back: idle, 6051 submissions since its sync
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.13, writes collected 0.06
[hos sead::MainThread] nvn: GPU waits in 5 s: 36 sync waits (4070 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: memory: working set 9769 MB (peak 15609), private 3346 MB, staging 48 MB in 3 arenas, 34 made so far (7 ms)
[hos host] vk: memory heaps: heap 0 (system): 16 blocks 944 MB, 50 allocations 908 MB; used 3040 of 15769 MB; heap 1 (device): 315 blocks 10395 MB, 21938 allocations 9381 MB; used 11110 of 18608 MB;
[hos host] vk: image blocks: 4 blocks 128 MB, 12 images 97 MB; 5 free ranges, the largest 23 MB
[hos host] vk: image blocks (render targets): 23 blocks 736 MB, 138 images 302 MB; 109 free ranges, the largest 32 MB
[hos host] vk: image blocks (textures, compacted): 191 blocks 6112 MB, 4350 images 5808 MB; 1173 free ranges, the largest 15 MB
[hos host] vk: render target images: 18783 reused from the pool, 12929 made new, 12785 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 11110 MB used of 18608 MB allowed (1014 MB of it free in the blocks); textures 7647 MB (budget 11000 MB), 4504 textures
[hos host] vk: textures: decoded ASTC 4099 MB, rendered 906 MB (50 drawn in one frame, kept: 38059 KB), other 1827 MB; used in the last 8 presents 2164 MB
[hos host] vk: decoded ASTC used again: 5204 textures (8318 MB) copied from ones the game finalized, not decoded; 5204 of those retired, 489 listed; 2528 MB of blocks hashed in 482 ms
[hos host] vk:   288 MB: 0x1147c16000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a5b22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 9625 presents ago
[hos host] vk:   214 MB: 0x10a22aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 9625 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0xa, last used 9868 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x113cba9000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x100030a4200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10002d97a00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 261 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 38 (8621 KB): 140x42x1 NVN 0x25: 1 (22 KB); 204x106x1 NVN 0x1: 1 (21 KB); 242x40x1 NVN 0x25: 4 (151 KB); 260x260x1 NVN 0xd: 21 (3695 KB); 306x104x1 NVN 0x1: 4 (124 KB); 592x49x1 NVN 0x25: 3 (339 KB); 656x113x1 NVN 0x1: 3 (217 KB); 960x540x1 NVN 0x29: 1 (4050 KB);
[hos host] vk: 3 textures retired (4501 left, 7637 MB): budget 0, drawn over 3, dead targets 0
[hos Presentation Thread] vk: presents after the CPU's signal (600): 272 waited for it, 145.24 ms each, 23 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: replay reads of guest memory as submitted: 8600000 (7238803 pages from the captures)
[hos host] vk: status: 600 frames, the slowest 1021.0 ms; VRAM 10.8 of 18.2 GB
[hos host] vk: program census (play): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 108 program records loaded (119.6 ms), 64 new graphics pipelines, pipelines from the driver's cache 100, from the start's cache 0, compiled 30; 72 draws and dispatches left out, 45 waits (1824.7 ms)
[hos host] vk: program loads at create (play): 5088 queued, 5088 loaded (median 0.51 ms, 99th percentile 6.96 ms, worst 99.86 ms, 5028 ms and 1071.3 MB read in all); 7005 programs (1124 of them compute variants without their code, 0 graphics programs no note names), 9195 graphics pipelines (0 not in a cache), 592 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 1932 pipelines made at create taken by draws; 3271 create calls with nothing kept for their stages, 0 before the index; units 8359 (5088 ready, 0 loading), 14504 programs alive
[hos host] vk: module identifiers (play): 658 known, 658 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 425 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 57 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 12.97, present 0.50, draw 6.87, descriptors 2.98, texture lookups 0.84, uploads 0.68, pipelines 0.30, programs 0.93, commands 1.90, dispatch 0.81, memory sync 1.20, queue submit 1.85, slot wait 0.08
[hos host] vk: a frame: mirrors 292 (1072 MB): 9.59 MB synced in 1416.4 copies, 18.24 MB written back
[hos host] vk: 11400 presents, 9.5 fps; a frame: 2402 draws (0 skipped), 127 dispatches, 169 passes, 150 barriers, 6 submits, 2.2 uploads (2.8 MB), 82 ub updates (57 overlays), GPU-bound wait 0.1 ms (recording slots only); 43807 pipelines (3609 ms), 0 programs (10 failed), 4501 textures (7637 MB)
[hos host] vk: host time in GPU resource waits: 88.16 ms a frame (fences 0.00, timelines 88.16, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 635.6 full barriers recorded, 268.3 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: 41.0; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 11.7 ms a frame (the longest submission 2.6 ms); 0.8 textures made a frame
[hos host] vk: a frame: 2.78 MB unswizzled on the CPU in 0.18 ms
[hos host] vk: the render thread's whole frames (600 presents): 39 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, 84.78 ms a frame): its jobs 11.43 ms a frame (submits 6.21, presents 0.54, others 4.68), idle 73.35; the game waited for room at 1 presents, 4.59 ms each (0.01 a frame; 162.61 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; 23 presents held for the CPU's signal, 12.24 ms each (0 let go by RX_VK_HOLD_MS); 278 captures at a CPU signal; the submits' waits read in 0.044 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 11.48 ms after its hand-over (27.72 at most)
[hos host] vk: write-backs left out (600 submits): 420, 2060.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1409 dirty pages from their submits' captures, 47176 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): 2807 runs, 350.6 MB compared; 124 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9850 in captures, 6 only in the collection, 19 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: texture storage read as submitted (132001 times; 4307923 pages from the copies, 0 of them rewritten in guest memory since their submit; 15151 pages kept, 37 spans, 110 MB)
[hos host] vk: replay reads of guest memory as submitted: 8620000 (7254946 pages from the captures)
[hos host] vk: sparse mirrors: 92, 1072 MB with memory of 1588 MB (515 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1357 dirty pages from their submits' captures, 45415 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): 2940 runs, 364.6 MB compared; 124 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9370 in captures, 5 only in the collection, 19 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos rmbl::VibrationThread] hid: 110592 vibration values, strongest 0.00 since the last count
[hos sead::MainThread] vk: capture: a submit 0.33 ms (lock 0.01, scan 0.04 over 835 ranges, 2615 MB), 864 pages written, 864 copied (3.4 MB)
[hos host] vk: a capture taken (ms): all 0.130; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.119, bytes copied 0.000; 25 kept (410 MB), 6 spare buffers, 15151 texture pages copied (59 MB)
[hos Presentation Thread] vi: 11427 frames presented (9.6 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 3, 17.8-22 22, 22-30 19, 30-40 0, over 40 4
[hos host] vk: replay reads of guest memory as submitted: 8640000 (7270990 pages from the captures)
[hos ModuleSystemWorker3] vcores: 40876 takes, 7811 waits (mean 94 us, longest 21.6 ms, 0 over HOS_VCORE_LOG_MS), 3927 kicks, 3965 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1378 dirty pages from their submits' captures, 45615 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): 2794 runs, 346.8 MB compared; 39 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9515 in captures, 11 only in the collection, 18 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: recorder: 1458630 packets run in 10.0 s (busy 0.28 s, spun idle 0.89 s, slept 4971 times, 478 of them under 200 us), 0 calls made directly (no room); the replay waited for it 7220 times, 16.85 ms in all, the longest 0.49 ms; vkEndCommandBuffer 2982/9.62 ms (max 0.23), vkResetDescriptorPool 2980/3.55 ms (max 0.11), replay_mu unlock 1212/3.54 ms (max 0.49), vkDestroyImageView 46/0.15 ms (max 0.05)
[hos host] vk: replay reads of guest memory as submitted: 8660000 (7286959 pages from the captures)
[hos sead::MainThread] vk: readbacks of GPU results: 1018 of 3000 waited for their write-back, 0.52 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: readback pages refreshed from guest memory: 4322556 pages so far (0 held: may be written ahead in their submission)
[hos sead::MainThread] nvn: GPU waits in 5 s: 31 sync waits (4058 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: program picks: 8288415 kept (2601841 by the last picks' textures, a uniform buffer bound since), 14608 new, 3496976 with other textures (of 11800000)
[hos host] vk: replay reads of guest memory as submitted: 8680000 (7302876 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 148990 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: sparse mirrors: 90, 1069 MB with memory of 1580 MB (510 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1331 dirty pages from their submits' captures, 45746 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): 2804 runs, 346.0 MB compared; 141 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9361 in captures, 5 only in the collection, 17 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos host] vk: texture storage read as submitted (134001 times; 4353762 pages from the copies, 0 of them rewritten in guest memory since their submit; 15151 pages kept, 37 spans, 110 MB)
[hos host] vk: replay reads of guest memory as submitted: 8700000 (7319122 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 632 KB; 0 made on demand (0 MB kept now); c1 stages: at most 257 a recording; 19 descriptor pools made when a recording's own ran out
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1336 dirty pages from their submits' captures, 45468 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): 2812 runs, 345.1 MB compared; 132 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9399 in captures, 8 only in the collection, 16 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vi: 11479 frames presented (10.2 fps over the last 5.1 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 1, 17.8-22 30, 22-30 16, 30-40 1, over 40 4
[hos host] vk: replay reads of guest memory as submitted: 8720000 (7335084 pages from the captures)
[hos host] vk: upload over a drawn image: #41161 0x10001248800 480x270x1 x1 NVN 0x3f at present 11485 (texture frame 11486), finalized, 0 KB dirty in 0 ranges, a GPU buffer wrote it
[hos BaseProcExecutor0] vcores: 44612 takes, 8351 waits (mean 102 us, longest 17.6 ms, 0 over HOS_VCORE_LOG_MS), 4105 kicks, 4181 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.14, writes collected 0.06
[hos host] vk: sparse mirrors: 85, 1025 MB with memory of 1504 MB (478 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1378 dirty pages from their submits' captures, 45436 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): 2796 runs, 346.0 MB compared; 156 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9364 in captures, 6 only in the collection, 18 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #230000: event 0x100d6c8018 mode 0 value 169220 (now 169201) (lr 0x80661124)
[hos sead::MainThread] nvn: GPU waits in 5 s: 41 sync waits (4080 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: replay reads of guest memory as submitted: 8740000 (7351053 pages from the captures)
[hos sead::MainThread] nvn: DrawMeshTasksIndirect: 340000 so far (x1 0x117d97b040 x2 0 x3 0x117d97b040 x4 0x8b040)
[hos ModuleSystemWorker5] nvn: 2100000 recordings made, 71487 live copies (24.9 MB), 4567 texture copies; 3 command buffers' segments (4.1 MB)
[hos Presentation Thread] vk: capture: a submit 0.32 ms (lock 0.01, scan 0.04 over 835 ranges, 2615 MB), 840 pages written, 840 copied (3.3 MB)
[hos host] vk: a capture taken (ms): all 0.125; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.114, bytes copied 0.000; 25 kept (438 MB), 8 spare buffers, 15151 texture pages copied (59 MB)
[hos host] vk: replay reads of guest memory as submitted: 8760000 (7367009 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1351 dirty pages from their submits' captures, 45498 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): 2799 runs, 346.9 MB compared; 34 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9509 in captures, 5 only in the collection, 17 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] control: state play -> menu at frame 11517 (81.0 draws a frame)
[hos host] vk: a frame: 432 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 57 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 12.99, present 0.50, draw 6.73, descriptors 2.91, texture lookups 0.82, uploads 0.67, pipelines 0.29, programs 0.91, commands 1.86, dispatch 0.80, memory sync 1.19, queue submit 1.88, slot wait 0.08
[hos host] vk: a frame: mirrors 292 (1025 MB): 9.63 MB synced in 1420.6 copies, 18.22 MB written back
[hos host] vk: 11520 presents, 9.6 fps; a frame: 2424 draws (0 skipped), 128 dispatches, 169 passes, 150 barriers, 6 submits, 2.3 uploads (2.8 MB), 82 ub updates (57 overlays), GPU-bound wait 0.1 ms (recording slots only); 43807 pipelines (3609 ms), 0 programs (10 failed), 4501 textures (7637 MB)
[hos host] vk: host time in GPU resource waits: 87.94 ms a frame (fences 0.00, timelines 87.94, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 639.1 full barriers recorded, 268.4 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: 41.0; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 11.9 ms a frame (the longest submission 2.6 ms); 0.7 textures made a frame
[hos host] vk: a frame: 2.79 MB unswizzled on the CPU in 0.17 ms
[hos Presentation Thread] vi: 11527 frames presented (9.6 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 0, 17.8-22 28, 22-30 15, 30-40 1, over 40 4
[hos host] vk: replay reads of guest memory as submitted: 8780000 (7383173 pages from the captures)
[hos host] control: state menu -> play at frame 11528 (2609.0 draws a frame)
[hos host] vk: texture storage read as submitted (136001 times; 4406745 pages from the copies, 0 of them rewritten in guest memory since their submit; 15151 pages kept, 37 spans, 110 MB)
[hos BaseProcExecutor1] vcores: 36841 takes, 6787 waits (mean 90 us, longest 24.7 ms, 0 over HOS_VCORE_LOG_MS), 3415 kicks, 3438 yields, 0 starved grants, 0 escapes in 6 s
[hos host] vk: sparse mirrors: 85, 1025 MB with memory of 1504 MB (478 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1601 dirty pages from their submits' captures, 45309 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): 2789 runs, 345.7 MB compared; 229 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9316 in captures, 9 only in the collection, 19 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos host] vk: recorder: 1448286 packets run in 10.0 s (busy 0.30 s, spun idle 0.87 s, slept 4965 times, 459 of them under 200 us), 0 calls made directly (no room); the replay waited for it 7147 times, 16.58 ms in all, the longest 0.27 ms; vkEndCommandBuffer 2962/10.13 ms (max 0.27), vkResetDescriptorPool 2960/3.59 ms (max 0.02), replay_mu unlock 1193/2.66 ms (max 0.08), vkDestroyImageView 32/0.20 ms (max 0.09)
[hos host] vk: program picks: 8438363 kept (2654669 by the last picks' textures, a uniform buffer bound since), 14609 new, 3547027 with other textures (of 12000000)
[hos sead::MainThread] nvn: GPU waits in 5 s: 24 sync waits (5039 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: 1 textures retired (4505 left, 7650 MB): budget 0, drawn over 1, dead targets 0
[hos ModuleSystemWorker5] nvn: CommandBufferSignalEvent #170000: event 0x100d6c8008 mode 0 value 169995 (lr 0x80660f34)
[hos host] vk: readback pages refreshed from guest memory: 4395060 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 8800000 (7399284 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 146597 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 1 textures retired (4504 left, 7647 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1173 dirty pages from their submits' captures, 45980 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): 2784 runs, 344.5 MB compared; 140 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9386 in captures, 3 only in the collection, 16 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 8820000 (7415411 pages from the captures)
[hos Presentation Thread] vi: 11557 frames presented (5.3 fps over the last 5.6 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 0, 17.8-22 16, 22-30 9, 30-40 0, over 40 5
[hos host] vk: replay reads of guest memory as submitted: 8840000 (7431489 pages from the captures)
[hos PreActorModelCompose] vcores: 34565 takes, 6598 waits (mean 89 us, longest 14.7 ms, 0 over HOS_VCORE_LOG_MS), 3288 kicks, 3325 yields, 0 starved grants, 0 escapes in 6 s
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 632 KB; 0 made on demand (0 MB kept now); c1 stages: at most 257 a recording; 19 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 85, 1025 MB with memory of 1504 MB (478 MB spared); blocks given memory later 10540, given back 12852; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1; 888 binds (283.2 ms); 0 made whole for want of memory
[hos host] vk: write-backs left out (600 submits): 399, 1957.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1342 dirty pages from their submits' captures, 45721 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): 2783 runs, 346.6 MB compared; 59 pages read into the shadows; 272 MB of shadows; pages the shadows knew that the CPU wrote: 9993 in captures, 9 only in the collection, 19 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 143 live (2072 MB), 0 retired still imported (0 MB), 51 retired given back (796 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 29 sync waits (5049 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos rmbl::VibrationThread] hid: 114688 vibration values, strongest 0.00 since the last count
[hos host] control: quit asked by a client
[hos host] vk: program census (close): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 108 program records loaded (119.6 ms), 64 new graphics pipelines, pipelines from the driver's cache 100, from the start's cache 0, compiled 30; 72 draws and dispatches left out, 45 waits (1824.7 ms)
[hos host] vk: program loads at create (close): 5088 queued, 5088 loaded (median 0.51 ms, 99th percentile 6.96 ms, worst 99.86 ms, 5028 ms and 1071.3 MB read in all); 7005 programs (1124 of them compute variants without their code, 0 graphics programs no note names), 9195 graphics pipelines (0 not in a cache), 592 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 1933 pipelines made at create taken by draws; 3271 create calls with nothing kept for their stages, 0 before the index; units 8359 (5088 ready, 0 loading), 14504 programs alive
[hos host] vk: module identifiers (close): 658 known, 658 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: page table census (close): no pool outside its windows, 64-72 GB and 508-512 GB (4256 pools in the table now)
[hos host] vk: pipeline cache: play file 4 saved (1421 KB): 29 pipelines, 23 from the driver's cache, 6 compiled; 2 more from the start's cache
[hos host] control: exiting
[hos host] stop: close work done (the game's saves are not waited for)