Beanstalk Longshot test farm
Menu

Report rp_d887d308d2c8d328 fail

#2043 feat(vk): enable Switch 2 AI upscaler on AMD RDNA 3/3.5/4 via fused 16x16x16 WMMA (+9-12% native FPS, sharper in-game upscaling)

linux+amd on jabbaxipil/RX7900XTX-Linux,

This report disagrees with its peers.

forest-driver-cold forest.frametime_max_ms: +18.3% against peers' +74.8%
forest-driver-cold run_frametime_max_ms: +44.7% against peers' +110.3%
forest-driver-cold run_hitch_count: +120.0% against peers' -33.3%
forest-driver-warm forest.frametime_max_ms: -25.0% against peers' +32.1%
forest-driver-warm run_frametime_max_ms: -29.3% against peers' +319.0%
forest-driver-warm run_hitch_count: -55.6% against peers' -25.0%
forest-driver-warm shader_compile_s: -86.3% against peers' +0.0%

Machine

Commit
0f38b3c0c2836a814b62a5c02999f96a4dee76d6
Baseline
89c3d21ea169ed46124e6c97acd90587f660b75a
App
nightly-217-14cc1aa3, attempt 1
Game build
e0f3ccfa
os
linux
cpu
AMD Ryzen 7 5800X3D 8-Core Processor
gpu
AMD Radeon RX 7900 XTX (RADV NAVI31)
tags
linuxamdradvdgpu
driver
26.2.7
ram_gb
32
vram_mb
24,576
gpu_vendor
amd
os_version
Ubuntu 24.04.5 LTS (kernel 7.0.0)

Builds

main (baseline)

134.7 s warm

from 89c3d21ea169

lift
0.6 s
read
3.7 s
finish
16.9 s
compile
47.8 s
shaders
25 s

PR

90.3 s warm

Program cache: warm

Scenarios

forest-driver-cold resolution=0

error

the driver's shader cache was not empty at the start (2 MB)

Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count511+120%
shader_compile_s57.7458.65+1.6%
shader_pipelines38,14037,754−1%
forest.hitch_count00±0%
log_forbidden_hits00±0%
forest.stutter_count00±0%
log_expected_missing10−100%
run_frametime_max_ms84.27121.95+44.7%
driver_cache_start_mb22±0%
forest.frametime_max_ms18.5221.91+18.3%
forest.frametime_p50_ms16.5516.55+0%
forest.frametime_p99_ms17.8217.87+0.3%
shader_compute_pipelines4,9064,947+0.8%
forest.frametime_1pct_low_ms17.9518.38+2.4%
screenshot_diff—0.71—

Pictures against main: forest-cold diff 0.7076 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
run_hitch_count511
shader_compile_s57.7458.65
shader_pipelines38,14037,754
forest.hitch_count00
log_forbidden_hits00
forest.stutter_count00
log_expected_missing10
run_frametime_max_ms84.27121.95
driver_cache_start_mb22
forest.frametime_max_ms18.5221.91
forest.frametime_p50_ms16.5516.55
forest.frametime_p99_ms17.8217.87
shader_compute_pipelines4,9064,947
forest.frametime_1pct_low_ms17.9518.38
screenshot_diff—0.71

forest-driver-cold resolution=9

error

the driver's shader cache was not empty at the start (2 MB)

Not compared: what ran is not what the plan asked. resolution: asked 9, main's game had 0; resolution: asked 9, the PR's game had 0

Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count118−27.3%
shader_compile_s57.7458.65+1.6%
shader_pipelines38,14037,754−1%
forest.hitch_count00±0%
log_forbidden_hits00±0%
forest.stutter_count00±0%
log_expected_missing10−100%
run_frametime_max_ms109.79148.73+35.5%
driver_cache_start_mb22±0%
forest.frametime_max_ms18.0320.29+12.5%
forest.frametime_p50_ms16.6116.54−0.5%
forest.frametime_p99_ms17.6217.83+1.2%
shader_compute_pipelines4,9064,947+0.8%
forest.frametime_1pct_low_ms17.7418.17+2.4%
screenshot_diff—0.07—

Pictures against main: forest-cold diff 0.0646 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
run_hitch_count118
shader_compile_s57.7458.65
shader_pipelines38,14037,754
forest.hitch_count00
log_forbidden_hits00
forest.stutter_count00
log_expected_missing10
run_frametime_max_ms109.79148.73
driver_cache_start_mb22
forest.frametime_max_ms18.0320.29
forest.frametime_p50_ms16.6116.54
forest.frametime_p99_ms17.6217.83
shader_compute_pipelines4,9064,947
forest.frametime_1pct_low_ms17.7418.17
screenshot_diff—0.06

forest-driver-warm resolution=0

ok
Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count94−55.6%
shader_compile_s2.710.37−86.3%
shader_pipelines00±0%
forest.hitch_count00±0%
log_forbidden_hits00±0%
forest.stutter_count00±0%
log_expected_missing10−100%
run_frametime_max_ms123.1887.09−29.3%
forest.frametime_max_ms24.8418.63−25%
forest.frametime_p50_ms16.6116.49−0.7%
forest.frametime_p99_ms17.7218.05+1.9%
shader_compute_pipelines00±0%
forest.frametime_1pct_low_ms18.4818.18−1.6%
screenshot_diff—0.7—

Pictures against main: forest-warm diff 0.7044 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
run_hitch_count94
shader_compile_s2.710.37
shader_pipelines00
forest.hitch_count00
log_forbidden_hits00
forest.stutter_count00
log_expected_missing10
run_frametime_max_ms123.1887.09
driver_cache_start_mb——
forest.frametime_max_ms24.8418.63
forest.frametime_p50_ms16.6116.49
forest.frametime_p99_ms17.7218.05
shader_compute_pipelines00
forest.frametime_1pct_low_ms18.4818.18
screenshot_diff—0.7

forest-driver-warm resolution=9

mismatch

what ran is not what was asked: resolution: asked 9, main's game had 0; resolution: asked 9, the PR's game had 0

Not compared: what ran is not what the plan asked. resolution: asked 9, main's game had 0; resolution: asked 9, the PR's game had 0

Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count76−14.3%
shader_compile_s2.710.37−86.3%
shader_pipelines00±0%
forest.hitch_count00±0%
log_forbidden_hits00±0%
forest.stutter_count00±0%
log_expected_missing10−100%
run_frametime_max_ms178.38142.32−20.2%
forest.frametime_max_ms21.1417.98−15%
forest.frametime_p50_ms16.6116.55−0.3%
forest.frametime_p99_ms17.717.75+0.2%
shader_compute_pipelines00±0%
forest.frametime_1pct_low_ms18.1517.85−1.7%
screenshot_diff—0.08—

Pictures against main: forest-warm diff 0.0776 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
run_hitch_count76
shader_compile_s2.710.37
shader_pipelines00
forest.hitch_count00
log_forbidden_hits00
forest.stutter_count00
log_expected_missing10
run_frametime_max_ms178.38142.32
driver_cache_start_mb——
forest.frametime_max_ms21.1417.98
forest.frametime_p50_ms16.6116.55
forest.frametime_p99_ms17.717.75
shader_compute_pipelines00
forest.frametime_1pct_low_ms18.1517.85
screenshot_diff—0.08

Rules

ScenarioCellRuleLimitmainPRΔ %Status
forest-driver-coldresolution=0crashFalse00—pass
forest-driver-coldresolution=0screenshot.max_diff0.15—0.71—fail
forest-driver-coldresolution=0forest.frametime_p99_ms.max_delta_pct40———inconclusiveneed two usable starts per side (baseline: 0, PR: 0)
forest-driver-coldresolution=0log_forbidden_hits.max000—pass
forest-driver-coldresolution=0log_expected_missing.max010—pass
forest-driver-coldresolution=9crashFalse00—pass
forest-driver-coldresolution=9screenshot.max_diff0.15———mismatchmismatch: resolution: asked 9, main's game had 0; resolution: asked 9, the PR's game had 0
forest-driver-coldresolution=9forest.frametime_p99_ms.max_delta_pct40———mismatchmismatch: resolution: asked 9, main's game had 0; resolution: asked 9, the PR's game had 0
forest-driver-coldresolution=9log_forbidden_hits.max0———mismatchmismatch: resolution: asked 9, main's game had 0; resolution: asked 9, the PR's game had 0
forest-driver-coldresolution=9log_expected_missing.max0———mismatchmismatch: resolution: asked 9, main's game had 0; resolution: asked 9, the PR's game had 0
forest-driver-warmresolution=0crashFalse00—pass
forest-driver-warmresolution=0screenshot.max_diff0.15—0.7—fail
forest-driver-warmresolution=0forest.frametime_p99_ms.max_delta_pct4017.7218.05+1.9%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
forest-driver-warmresolution=0log_forbidden_hits.max000—pass
forest-driver-warmresolution=0log_expected_missing.max010—pass
forest-driver-warmresolution=9crashFalse00—pass
forest-driver-warmresolution=9screenshot.max_diff0.15———mismatchmismatch: resolution: asked 9, main's game had 0; resolution: asked 9, the PR's game had 0
forest-driver-warmresolution=9forest.frametime_p99_ms.max_delta_pct40———mismatchmismatch: resolution: asked 9, main's game had 0; resolution: asked 9, the PR's game had 0
forest-driver-warmresolution=9log_forbidden_hits.max0———mismatchmismatch: resolution: asked 9, main's game had 0; resolution: asked 9, the PR's game had 0
forest-driver-warmresolution=9log_expected_missing.max0———mismatchmismatch: resolution: asked 9, main's game had 0; resolution: asked 9, the PR's game had 0

Pictures of the failure

This report had 2 pictures of its failure (the last frame of a stuck step, or main's and the PR's picture where they differed too much). The farm keeps those for a day: they were deleted.

Log (PR side, end)

[hos host] vk:   read live, by pool:
[hos host] vk: imports: 80 live (1248 MB), 0 retired still imported (0 MB), 19 retired given back (372 MB)
[hos Presentation Thread] vi: 3994 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 1, 15.5-17.8 299, 17.8-22 1, 22-30 0, 30-40 0, over 40 0
[hos host] vk: replay reads of guest memory as submitted: 1420000 (698384 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 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 sead::MainThread] vk: capture: a submit 0.09 ms (lock 0.00, scan 0.05 over 1100 ranges, 2651 MB), 158 pages written, 158 copied (0.6 MB)
[hos host] vk: a capture taken (ms): all 0.003; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (152 MB), 9 spare buffers, 848 texture pages copied (3 MB)
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 181 dirty pages from their submits' captures, 48574 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): 1088 dispatches, 40234 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6416 runs, 1178.8 MB compared; 99 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15778 in captures, 5 only in the collection, 50 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1069 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 ModuleSystemWorker1] vk: invalidates since the last readbacks line: 2994 kept their virtual core (nothing there to wait for), 3150 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: replay reads of guest memory as submitted: 1440000 (706390 pages from the captures)
[hos host] vk: program picks: 1397227 kept (513117 by the last picks' textures, a uniform buffer bound since), 1002 new, 801770 with other textures (of 2200000)
[hos ModuleSystemWorker4] nvn: 500000 recordings made, 39924 live copies (10.4 MB), 427 texture copies; 7 command buffers' segments (4.4 MB)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 199 KB; 0 made on demand (0 MB kept now); c1 stages: at most 97 a recording; 8 descriptor pools made when a recording's own ran out (8 slots' pools made whole); a recording's pool took at most 400 sets (ub 4476, sb 2017, cis 33443, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 53, 756 MB with memory of 1036 MB (279 MB spared); blocks given memory later 749, given back 2179; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 182 binds (54.9 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 14 taken from the spares, 85 made new; 46 kept as spares, 0 destroyed; 32 spares now, 448 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 143 dirty pages from their submits' captures, 48651 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): 1085 dispatches, 40219 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6317 runs, 1155.5 MB compared; 98 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15702 in captures, 3 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 80 live (1248 MB), 0 retired still imported (0 MB), 19 retired given back (372 MB)
[hos rmbl::VibrationThread] hid: 12288 vibration values, strongest 0.00 since the last count
[hos host] vk: replay reads of guest memory as submitted: 1460000 (714752 pages from the captures)
[hos host] vk: a frame: 67 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 3.94, present 0.08, draw 1.93, descriptors 0.84, texture lookups 0.03, uploads 0.00, pipelines 0.07, programs 0.29, commands 0.59, dispatch 0.38, memory sync 0.27, queue submit 1.23, slot wait 0.04
[hos host] vk: a frame: mirrors 99 (756 MB): 3.89 MB synced in 820.6 copies, 23.58 MB written back
[hos host] vk: 4080 presents, 60.0 fps; a frame: 782 draws (0 skipped), 90 dispatches, 148 passes, 126 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 10861 pipelines (256 ms), 0 programs (0 failed), 1503 textures (4008 MB)
[hos host] vk: host time in GPU resource waits: 3.51 ms a frame (fences 0.00, timelines 3.51, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 483.9 full barriers recorded, 227.2 left out (nothing recorded since the one before), 14.2 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: 29.0; 0x12: 67.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.1 ms a frame (the longest submission 2.8 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 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 ModuleSystemWorker5] nvn: DrawMeshTasksIndirect: 120000 so far (x1 0x1178aa8000 x2 0x2 x3 0x1178aa8000 x4 0xf0040)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos sead::MainThread] vk: capture: a submit 0.09 ms (lock 0.00, scan 0.05 over 1100 ranges, 2651 MB), 159 pages written, 159 copied (0.6 MB)
[hos host] vk: a capture taken (ms): all 0.003; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (134 MB), 9 spare buffers, 848 texture pages copied (3 MB)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 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): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 102 dirty pages from their submits' captures, 48674 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): 1084 dispatches, 40225 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6318 runs, 1155.5 MB compared; 98 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15855 in captures, 2 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1480000 (722535 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 760666 pages so far (0 held: may be written ahead in their submission)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #75000: event 0x100d6c8044 mode 0 value 54249 (now 54211) (lr 0x80274390)
[hos rmbl::VibrationThread] vcores: 80504 takes, 10791 waits (mean 40 us, longest 1.8 ms, 0 over HOS_VCORE_LOG_MS), 4311 kicks, 4169 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: replay reads of guest memory as submitted: 1500000 (730264 pages from the captures)
[hos host] vk: sparse mirrors: 53, 756 MB with memory of 1036 MB (279 MB spared); blocks given memory later 749, given back 2179; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 182 binds (54.9 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 14 taken from the spares, 85 made new; 46 kept as spares, 0 destroyed; 32 spares now, 448 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 110 dirty pages from their submits' captures, 48616 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): 1086 dispatches, 40184 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6320 runs, 1155.4 MB compared; 98 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15795 in captures, 4 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 80 live (1248 MB), 0 retired still imported (0 MB), 19 retired given back (372 MB)
[hos ModuleSystemWorker5] nvn: CommandBufferSignalEvent #55000: event 0x100d6c8020 mode 0 value 55000 (lr 0x80667410)
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 1079 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 ModuleSystemWorker4] vk: invalidates since the last readbacks line: 2993 kept their virtual core (nothing there to wait for), 2895 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos sead::MainThread] vk: capture: a submit 0.09 ms (lock 0.00, scan 0.05 over 1100 ranges, 2651 MB), 157 pages written, 157 copied (0.6 MB)
[hos host] vk: a capture taken (ms): all 0.003; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (152 MB), 11 spare buffers, 848 texture pages copied (3 MB)
[hos host] vk: replay reads of guest memory as submitted: 1520000 (738014 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: memory: working set 7266 MB (peak 7343), private 2119 MB, staging 80 MB in 5 arenas, 25 made so far (26 ms)
[hos host] vk: memory heaps: heap 0 (device): 130 blocks 5816 MB, 14217 allocations 5301 MB; used 5868 of 23541 MB; heap 1 (system): 18 blocks 824 MB, 39 allocations 730 MB; used 2133 of 15167 MB; heap 2 (device): 60 blocks 173 MB, 152 allocations 173 MB; used 386 of 256 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 5 images 16 MB; 2 free ranges, the largest 15 MB
[hos host] vk: image blocks (render targets): 14 blocks 448 MB, 91 images 204 MB; 53 free ranges, the largest 25 MB
[hos host] vk: image blocks (textures, compacted): 73 blocks 2336 MB, 1397 images 2302 MB; 246 free ranges, the largest 20 MB
[hos host] vk: render target images: 227 reused from the pool, 346 made new, 252 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 6255 MB used of 23797 MB allowed (514 MB of it free in the blocks); textures 4008 MB (budget 11000 MB), 1503 textures
[hos host] vk: budget: VRAM 5989 of 23797 MB (6255 MB by the driver, 265 MB unaccounted): textures 4008, mirrors 756, blocks free 514, shadows 352, other buffers 213, staging 80, other images 63; RAM 2004 MB (the process's private memory 2119 MB): imports 1248, capture spares 586, captures 134, capture table 26, other buffers 10, others 0
[hos host] vk: textures: decoded ASTC 2360 MB, rendered 844 MB (88 drawn in one frame, kept: 38405 KB), other 431 MB; used in the last 8 presents 1804 MB
[hos host] vk: decoded ASTC used again: 0 textures (0 MB) copied from ones the game finalized, not decoded; 0 of those retired, 5 listed; 474 MB of blocks hashed in 43 ms
[hos host] vk:   288 MB: 0x1147c16000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a5b22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a22aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0xa, last used 2956 presents ago
[hos host] vk:   64 MB: 0x113cba9000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10002d97a00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x100030a4200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 964 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 Presentation Thread] vk: presents after the CPU's signal (600): 26 waited for it, 0.17 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 18.3 ms; VRAM 6.1 of 23.2 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: 5 program records loaded (1.5 ms), 0 new graphics pipelines, pipelines from the driver's cache 0, from the start's cache 0, compiled 0; 0 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (play): 4846 queued, 4846 loaded (median 0.15 ms, 99th percentile 1.41 ms, worst 8.55 ms, 1190 ms and 961.1 MB read in all); 6857 programs (1794 of them compute variants without their code, 0 graphics programs no note names), 8908 graphics pipelines (0 not in a cache), 582 compute pipelines (25 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 788 pipelines made at create taken by draws; 2191 create calls with nothing kept for their stages, 0 before the index; units 7037 (4846 ready, 0 loading), 12254 programs alive; 0 noted graphics pipeline probes deferred to first use
[hos host] vk: program loads at create (play): 9898 shader modules let go once their programs' noted pipelines were built, 0 programs' modules made again for a new pipeline state (RX_VK_DROP_LOADED_MODULES=0 keeps them)
[hos host] vk: module identifiers (play): 607 known, 0 learned in this run; 582 compute pipelines made by them at a create call, 25 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 67 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 3.96, present 0.07, draw 1.99, descriptors 0.87, texture lookups 0.03, uploads 0.00, pipelines 0.07, programs 0.30, commands 0.61, dispatch 0.38, memory sync 0.27, queue submit 1.23, slot wait 0.04
[hos host] vk: a frame: mirrors 99 (756 MB): 3.89 MB synced in 820.5 copies, 23.58 MB written back
[hos host] vk: 4200 presents, 60.0 fps; a frame: 782 draws (0 skipped), 90 dispatches, 148 passes, 126 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 10861 pipelines (256 ms), 0 programs (0 failed), 1503 textures (4008 MB)
[hos host] vk: host time in GPU resource waits: 3.45 ms a frame (fences 0.00, timelines 3.45, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 484.1 full barriers recorded, 227.3 left out (nothing recorded since the one before), 14.2 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: 29.0; 0x12: 67.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.1 ms a frame (the longest submission 2.8 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 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, 16.67 ms a frame): its jobs 4.49 ms a frame (submits 2.12, presents 0.08, others 2.28), idle 12.18; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 4.41 jobs queued at a present); 0 catch-ups (a finish, a texture read back, a write-back not recorded yet: 0), 0.00 ms each; 0 presents held for the CPU's signal, 0.00 ms each (0 let go by RX_VK_HOLD_MS); 377 captures at a CPU signal; the submits' waits read in 0.022 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 3.98 ms after its hand-over (4.73 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 6.76 + 0.14 (the window's copy is in the idle), after it 2.32 + 0.37; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.01 ms of GPU time), after its last that reads it 1.00 (0.01 ms)
[hos host] vk: frame ahead (600 presents): 5.64 parts a frame recorded in the signal's job, 93.0% of them took nothing that came with the signal; of the others, 0.39 a frame took pages of the signal's capture into a mirror (0.6 pages a frame) and 0.00 pages read as guest memory had them (0.0 pages a frame); a frame's first part took nothing in 241 of 377 frames, its first three in 241
[hos host] vk: frame ahead, what came with the signal: 377 signals with a capture, each of 1.8 pages in 1.4 pools; between a frame's last submit and that signal 0.00 texture and sampler registrations, 0.00 releases and 0.00 other CPU signals a signal (0 of them with any)
[hos host] vk: frame overlap, by place: idle + uploads before a frame's 2nd to 12th recording with a draw, then all later ones: 0.80+0.10, 0.84+0.11, 0.02+0.13, 0.65+0.02, 0.00+0.02, 0.00+0.01, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 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): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 145 dirty pages from their submits' captures, 48528 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): 1084 dispatches, 40140 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6319 runs, 1155.4 MB compared; 98 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15657 in captures, 5 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker5] barrier: 53760 looks: 20992 found the masks moved already, 32324 waited until they moved, 444 for the whole 1 ms (a wait 324 us on average); 31453 wakes
[hos host] vk: replay reads of guest memory as submitted: 1540000 (745830 pages from the captures)
[hos host] vk: render targets drawn over from another start, let go: 408 so far (1105 MB), 405 of them (1090 MB) between the passes
[hos host] vk: sparse mirrors: 53, 756 MB with memory of 1036 MB (279 MB spared); blocks given memory later 749, given back 2179; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 182 binds (54.9 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 14 taken from the spares, 85 made new; 46 kept as spares, 0 destroyed; 32 spares now, 448 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1100, 4281.4 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 271 dirty pages from their submits' captures, 49576 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): 1114 dispatches, 41103 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6440 runs, 1179.1 MB compared; 100 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 16119 in captures, 9 only in the collection, 50 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 80 live (1248 MB), 0 retired still imported (0 MB), 19 retired given back (372 MB)
[hos host] vk: replay reads of guest memory as submitted: 1560000 (753558 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.09 ms (lock 0.00, scan 0.05 over 1100 ranges, 2651 MB), 160 pages written, 160 copied (0.6 MB)
[hos host] vk: a capture taken (ms): all 0.003; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (112 MB), 12 spare buffers, 848 texture pages copied (3 MB)
[hos host] vk: program picks: 1540445 kept (571615 by the last picks' textures, a uniform buffer bound since), 1002 new, 858552 with other textures (of 2400000)
[hos Presentation Thread] vi: 4295 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 301, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 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: the recording streams: 48 of 1 MB, the most one held 199 KB; 0 made on demand (0 MB kept now); c1 stages: at most 97 a recording; 8 descriptor pools made when a recording's own ran out (8 slots' pools made whole); a recording's pool took at most 400 sets (ub 4476, sb 2017, cis 33443, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 99 dirty pages from their submits' captures, 48504 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): 1079 dispatches, 40099 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6320 runs, 1155.5 MB compared; 98 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15743 in captures, 2 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1580000 (761296 pages from the captures)
[hos host] vk: a frame: 67 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 3.95, present 0.08, draw 1.97, descriptors 0.86, texture lookups 0.03, uploads 0.00, pipelines 0.07, programs 0.29, commands 0.60, dispatch 0.38, memory sync 0.27, queue submit 1.22, slot wait 0.04
[hos host] vk: a frame: mirrors 99 (756 MB): 3.88 MB synced in 819.8 copies, 23.58 MB written back
[hos host] vk: 4320 presents, 60.0 fps; a frame: 783 draws (0 skipped), 90 dispatches, 148 passes, 126 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 10861 pipelines (256 ms), 0 programs (0 failed), 1503 textures (4008 MB)
[hos host] vk: host time in GPU resource waits: 3.46 ms a frame (fences 0.00, timelines 3.46, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 483.7 full barriers recorded, 227.2 left out (nothing recorded since the one before), 14.2 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: 29.0; 0x12: 67.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.1 ms a frame (the longest submission 2.8 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 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 ModuleSystemWorker2] vk: readbacks of GPU results: 1063 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 ModuleSystemWorker2] vk: invalidates since the last readbacks line: 2793 kept their virtual core (nothing there to wait for), 3095 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos ModuleSystemWorker2] nvn: CommandBufferWaitEvent #80000: event 0x100d6c8044 mode 0 value 58049 (now 58011) (lr 0x80664f90)
[hos host] vk: replay reads of guest memory as submitted: 1600000 (769904 pages from the captures)
[hos host] vk: sparse mirrors: 53, 756 MB with memory of 1036 MB (279 MB spared); blocks given memory later 749, given back 2179; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 182 binds (54.9 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 14 taken from the spares, 85 made new; 46 kept as spares, 0 destroyed; 32 spares now, 448 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 105 dirty pages from their submits' captures, 48711 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): 1086 dispatches, 40161 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6321 runs, 1155.4 MB compared; 98 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15680 in captures, 1 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 80 live (1248 MB), 0 retired still imported (0 MB), 19 retired given back (372 MB)
[hos host] vk: readback pages refreshed from guest memory: 827973 pages so far (0 held: may be written ahead in their submission)
[hos sead::MainThread] vk: capture: a submit 0.09 ms (lock 0.00, scan 0.05 over 1100 ranges, 2651 MB), 157 pages written, 157 copied (0.6 MB)
[hos host] vk: a capture taken (ms): all 0.003; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (154 MB), 11 spare buffers, 848 texture pages copied (3 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: replay reads of guest memory as submitted: 1620000 (777879 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 145 dirty pages from their submits' captures, 48528 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): 1080 dispatches, 40127 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6207 runs, 1132.1 MB compared; 97 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 232 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15629 in captures, 2 only in the collection, 48 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: a frame: 67 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 3.96, present 0.07, draw 1.98, descriptors 0.86, texture lookups 0.03, uploads 0.00, pipelines 0.07, programs 0.29, commands 0.61, dispatch 0.38, memory sync 0.27, queue submit 1.21, slot wait 0.04
[hos host] vk: a frame: mirrors 99 (756 MB): 3.89 MB synced in 819.2 copies, 23.58 MB written back
[hos host] vk: 4440 presents, 60.0 fps; a frame: 787 draws (0 skipped), 90 dispatches, 148 passes, 126 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 10861 pipelines (256 ms), 0 programs (0 failed), 1503 textures (4008 MB)
[hos host] vk: host time in GPU resource waits: 3.59 ms a frame (fences 0.00, timelines 3.59, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 484.3 full barriers recorded, 227.4 left out (nothing recorded since the one before), 14.2 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: 29.0; 0x12: 67.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.1 ms a frame (the longest submission 2.8 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 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 ModuleSystemWorker5] nvn: CommandBufferSignalEvent #60000: event 0x100d6c8028 mode 0 value 59999 (lr 0x80667410)
[hos host] vk: replay reads of guest memory as submitted: 1640000 (785890 pages from the captures)
[hos BaseProcExecutor1] vcores: 80017 takes, 10574 waits (mean 42 us, longest 1.8 ms, 0 over HOS_VCORE_LOG_MS), 4125 kicks, 4036 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: sparse mirrors: 53, 756 MB with memory of 1036 MB (279 MB spared); blocks given memory later 749, given back 2179; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 182 binds (54.9 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 14 taken from the spares, 85 made new; 46 kept as spares, 0 destroyed; 32 spares now, 448 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1056, 4110.2 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 108 dirty pages from their submits' captures, 48022 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): 1082 dispatches, 39629 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6300 runs, 1155.3 MB compared; 98 pages read into the shadows; 352 MB of shadows (196 MB of pages known, in 232 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15749 in captures, 3 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 80 live (1248 MB), 0 retired still imported (0 MB), 19 retired given back (372 MB)
[hos sead::MainThread] vk: capture: a submit 0.09 ms (lock 0.00, scan 0.05 over 1100 ranges, 2651 MB), 158 pages written, 158 copied (0.6 MB)
[hos host] vk: a capture taken (ms): all 0.003; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (148 MB), 11 spare buffers, 848 texture pages copied (3 MB)
[hos host] vk: recorder: 3391378 packets run in 10.0 s (busy 0.60 s, spun idle 2.05 s, slept 5104 times, 1488 of them under 200 us), 0 calls made directly (no room); the replay waited for it 22188 times, 48.47 ms in all, the longest 0.18 ms; replay_mu unlock 7566/19.89 ms (max 0.18), vkResetDescriptorPool 7281/15.97 ms (max 0.02), vkEndCommandBuffer 7341/12.61 ms (max 0.02)
[hos host] vk: replay reads of guest memory as submitted: 1660000 (793906 pages from the captures)
[hos sead::MainThread] vk: readbacks of GPU results: 1076 of 3000 waited for their write-back, 0.53 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: 3072 kept their virtual core (nothing there to wait for), 2816 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 40 dirty pages from their submits' captures, 48240 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): 1066 dispatches, 39858 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6215 runs, 1132.0 MB compared; 98 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 232 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15374 in captures, 2 only in the collection, 48 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: replay reads of guest memory as submitted: 1680000 (801901 pages from the captures)
[hos sead::MainThread] nvn: 550000 recordings made, 39941 live copies (10.6 MB), 427 texture copies; 4 command buffers' segments (4.2 MB)
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #85000: event 0x100d6c8044 mode 0 value 61849 (now 61811) (lr 0x80274390)
[hos host] vk: program picks: 1683445 kept (629984 by the last picks' textures, a uniform buffer bound since), 1002 new, 915552 with other textures (of 2600000)
[hos sead::MainThread] nvn: DrawMeshTasksIndirect: 140000 so far (x1 0x1178b9a000 x2 0x2 x3 0x1178b9a000 x4 0x654040)
[hos sead::MainThread] vk: capture: a submit 0.09 ms (lock 0.00, scan 0.05 over 1100 ranges, 2651 MB), 156 pages written, 156 copied (0.6 MB)
[hos host] vk: a capture taken (ms): all 0.003; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (144 MB), 9 spare buffers, 848 texture pages copied (3 MB)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 199 KB; 0 made on demand (0 MB kept now); c1 stages: at most 97 a recording; 8 descriptor pools made when a recording's own ran out (8 slots' pools made whole); a recording's pool took at most 400 sets (ub 4476, sb 2017, cis 33443, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 53, 756 MB with memory of 1036 MB (279 MB spared); blocks given memory later 749, given back 2179; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 182 binds (54.9 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 14 taken from the spares, 85 made new; 46 kept as spares, 0 destroyed; 32 spares now, 448 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 172 dirty pages from their submits' captures, 48435 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): 1088 dispatches, 40108 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6422 runs, 1178.8 MB compared; 99 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15783 in captures, 6 only in the collection, 50 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 80 live (1248 MB), 0 retired still imported (0 MB), 19 retired given back (372 MB)
[hos host] vk: a frame: 67 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 3.96, present 0.08, draw 1.95, descriptors 0.85, texture lookups 0.03, uploads 0.00, pipelines 0.07, programs 0.29, commands 0.60, dispatch 0.39, memory sync 0.27, queue submit 1.23, slot wait 0.04
[hos host] vk: a frame: mirrors 99 (756 MB): 3.88 MB synced in 819.1 copies, 23.58 MB written back
[hos host] vk: 4560 presents, 60.0 fps; a frame: 781 draws (0 skipped), 90 dispatches, 148 passes, 126 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 10861 pipelines (256 ms), 0 programs (0 failed), 1503 textures (4008 MB)
[hos host] vk: host time in GPU resource waits: 3.62 ms a frame (fences 0.00, timelines 3.62, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 484.2 full barriers recorded, 227.4 left out (nothing recorded since the one before), 14.2 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: 29.0; 0x12: 67.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.1 ms a frame (the longest submission 2.8 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 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: replay reads of guest memory as submitted: 1700000 (809930 pages from the captures)
[hos Presentation Thread] vi: 4596 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 301, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 111 dirty pages from their submits' captures, 48490 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): 1084 dispatches, 40053 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6320 runs, 1155.4 MB compared; 98 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15767 in captures, 5 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: readback pages refreshed from guest memory: 895927 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: replay reads of guest memory as submitted: 1720000 (817880 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.09 ms (lock 0.00, scan 0.05 over 1100 ranges, 2651 MB), 158 pages written, 158 copied (0.6 MB)
[hos host] vk: a capture taken (ms): all 0.003; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (140 MB), 9 spare buffers, 848 texture pages copied (3 MB)
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1072 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 ModuleSystemWorker1] vk: invalidates since the last readbacks line: 2917 kept their virtual core (nothing there to wait for), 3227 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: sparse mirrors: 53, 756 MB with memory of 1036 MB (279 MB spared); blocks given memory later 749, given back 2179; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 182 binds (54.9 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 14 taken from the spares, 85 made new; 46 kept as spares, 0 destroyed; 32 spares now, 448 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 121 dirty pages from their submits' captures, 48613 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): 1086 dispatches, 40180 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6317 runs, 1155.5 MB compared; 98 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15828 in captures, 5 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 80 live (1248 MB), 0 retired still imported (0 MB), 19 retired given back (372 MB)
[hos host] vk: replay reads of guest memory as submitted: 1740000 (826270 pages from the captures)
[hos host] vk: a frame: 67 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 3.94, present 0.07, draw 1.97, descriptors 0.86, texture lookups 0.03, uploads 0.00, pipelines 0.07, programs 0.29, commands 0.61, dispatch 0.38, memory sync 0.27, queue submit 1.23, slot wait 0.04
[hos host] vk: a frame: mirrors 99 (756 MB): 3.88 MB synced in 818.7 copies, 23.58 MB written back
[hos host] vk: 4680 presents, 60.0 fps; a frame: 781 draws (0 skipped), 90 dispatches, 148 passes, 126 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 10861 pipelines (256 ms), 0 programs (0 failed), 1503 textures (4008 MB)
[hos host] vk: host time in GPU resource waits: 3.48 ms a frame (fences 0.00, timelines 3.48, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 483.9 full barriers recorded, 227.3 left out (nothing recorded since the one before), 14.2 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: 29.0; 0x12: 67.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.1 ms a frame (the longest submission 2.8 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 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: replay reads of guest memory as submitted: 1760000 (834053 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 104 dirty pages from their submits' captures, 48667 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): 1089 dispatches, 40157 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6317 runs, 1155.5 MB compared; 98 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 235 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15866 in captures, 4 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker4] nvn: CommandBufferSignalEvent #65000: event 0x100d6c8034 mode 0 value 64999 (lr 0x806616b0)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 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 0.09 ms (lock 0.00, scan 0.05 over 1100 ranges, 2651 MB), 158 pages written, 158 copied (0.6 MB)
[hos host] vk: a capture taken (ms): all 0.003; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (156 MB), 10 spare buffers, 848 texture pages copied (3 MB)
[hos host] vk: replay reads of guest memory as submitted: 1780000 (841796 pages from the captures)
[hos ModuleSystemWorker2] nvn: CommandBufferWaitEvent #90000: event 0x100d6c8044 mode 0 value 65649 (now 65611) (lr 0x80664f90)
[hos host] vk: queues ordered by their events: 30000 submissions stopped at a wait, 29957 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 50.06 ms at most); waits met 90163, not met 0; captures at a CPU event's wait 2175, 0 of them at a wait met on arrival
[hos BaseProcExecutor2] vcores: 79605 takes, 10707 waits (mean 41 us, longest 1.4 ms, 0 over HOS_VCORE_LOG_MS), 4178 kicks, 4105 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: sparse mirrors: 53, 756 MB with memory of 1036 MB (279 MB spared); blocks given memory later 749, given back 2179; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 182 binds (54.9 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 14 taken from the spares, 85 made new; 46 kept as spares, 0 destroyed; 32 spares now, 448 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1100, 4281.4 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 172 dirty pages from their submits' captures, 49356 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): 1101 dispatches, 40835 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6447 runs, 1179.0 MB compared; 100 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 16086 in captures, 7 only in the collection, 50 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 80 live (1248 MB), 0 retired still imported (0 MB), 19 retired given back (372 MB)
[hos host] vk: replay reads of guest memory as submitted: 1800000 (849516 pages from the captures)
[hos host] vk: program picks: 1826943 kept (688555 by the last picks' textures, a uniform buffer bound since), 1002 new, 972054 with other textures (of 2800000)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: memory: working set 7266 MB (peak 7343), private 2119 MB, staging 80 MB in 5 arenas, 25 made so far (26 ms)
[hos host] vk: memory heaps: heap 0 (device): 130 blocks 5816 MB, 14217 allocations 5301 MB; used 5868 of 23541 MB; heap 1 (system): 18 blocks 824 MB, 39 allocations 734 MB; used 2133 of 15167 MB; heap 2 (device): 60 blocks 173 MB, 152 allocations 173 MB; used 386 of 256 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 5 images 16 MB; 2 free ranges, the largest 15 MB
[hos host] vk: image blocks (render targets): 14 blocks 448 MB, 91 images 204 MB; 53 free ranges, the largest 25 MB
[hos host] vk: image blocks (textures, compacted): 73 blocks 2336 MB, 1397 images 2302 MB; 246 free ranges, the largest 20 MB
[hos host] vk: render target images: 227 reused from the pool, 346 made new, 252 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 6255 MB used of 23797 MB allowed (514 MB of it free in the blocks); textures 4008 MB (budget 11000 MB), 1503 textures
[hos host] vk: budget: VRAM 5989 of 23797 MB (6255 MB by the driver, 265 MB unaccounted): textures 4008, mirrors 756, blocks free 514, shadows 352, other buffers 213, staging 80, other images 63; RAM 2020 MB (the process's private memory 2119 MB): imports 1248, capture spares 592, captures 154, capture table 26, others 0
[hos host] vk: textures: decoded ASTC 2360 MB, rendered 844 MB (88 drawn in one frame, kept: 38405 KB), other 431 MB; used in the last 8 presents 1804 MB
[hos host] vk: decoded ASTC used again: 0 textures (0 MB) copied from ones the game finalized, not decoded; 0 of those retired, 5 listed; 474 MB of blocks hashed in 43 ms
[hos host] vk:   288 MB: 0x1147c16000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a5b22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a22aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0xa, last used 3556 presents ago
[hos host] vk:   64 MB: 0x113cba9000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10002d97a00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x100030a4200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1564 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 Presentation Thread] vk: presents after the CPU's signal (600): 25 waited for it, 0.07 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 17.9 ms; VRAM 6.1 of 23.2 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: 5 program records loaded (1.5 ms), 0 new graphics pipelines, pipelines from the driver's cache 0, from the start's cache 0, compiled 0; 0 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (play): 4846 queued, 4846 loaded (median 0.15 ms, 99th percentile 1.41 ms, worst 8.55 ms, 1190 ms and 961.1 MB read in all); 6857 programs (1794 of them compute variants without their code, 0 graphics programs no note names), 8908 graphics pipelines (0 not in a cache), 582 compute pipelines (25 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 788 pipelines made at create taken by draws; 2191 create calls with nothing kept for their stages, 0 before the index; units 7037 (4846 ready, 0 loading), 12254 programs alive; 0 noted graphics pipeline probes deferred to first use
[hos host] vk: program loads at create (play): 9898 shader modules let go once their programs' noted pipelines were built, 0 programs' modules made again for a new pipeline state (RX_VK_DROP_LOADED_MODULES=0 keeps them)
[hos host] vk: module identifiers (play): 607 known, 0 learned in this run; 582 compute pipelines made by them at a create call, 25 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 67 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 3.91, present 0.08, draw 1.92, descriptors 0.83, texture lookups 0.03, uploads 0.00, pipelines 0.07, programs 0.29, commands 0.59, dispatch 0.38, memory sync 0.27, queue submit 1.21, slot wait 0.04
[hos host] vk: a frame: mirrors 99 (756 MB): 3.88 MB synced in 818.1 copies, 23.58 MB written back
[hos host] vk: 4800 presents, 60.0 fps; a frame: 780 draws (0 skipped), 90 dispatches, 148 passes, 126 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 10861 pipelines (256 ms), 0 programs (0 failed), 1503 textures (4008 MB)
[hos host] vk: host time in GPU resource waits: 3.52 ms a frame (fences 0.00, timelines 3.52, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 483.9 full barriers recorded, 227.3 left out (nothing recorded since the one before), 14.2 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: 29.0; 0x12: 67.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.1 ms a frame (the longest submission 2.8 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 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, 16.67 ms a frame): its jobs 4.47 ms a frame (submits 2.20, presents 0.08, others 2.18), idle 12.20; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 4.37 jobs queued at a present); 0 catch-ups (a finish, a texture read back, a write-back not recorded yet: 0), 0.00 ms each; 0 presents held for the CPU's signal, 0.00 ms each (0 let go by RX_VK_HOLD_MS); 362 captures at a CPU signal; the submits' waits read in 0.021 ms a frame; readbacks: 2997 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 3.97 ms after its hand-over (4.78 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 6.80 + 0.14 (the window's copy is in the idle), after it 2.32 + 0.37; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.01 ms of GPU time), after its last that reads it 1.00 (0.01 ms)
[hos host] vk: frame ahead (600 presents): 5.63 parts a frame recorded in the signal's job, 93.7% of them took nothing that came with the signal; of the others, 0.35 a frame took pages of the signal's capture into a mirror (0.4 pages a frame) and 0.00 pages read as guest memory had them (0.0 pages a frame); a frame's first part took nothing in 236 of 362 frames, its first three in 236
[hos host] vk: frame ahead, what came with the signal: 362 signals with a capture, each of 1.4 pages in 1.4 pools; between a frame's last submit and that signal 0.00 texture and sampler registrations, 0.00 releases and 0.00 other CPU signals a signal (0 of them with any)
[hos host] vk: frame overlap, by place: idle + uploads before a frame's 2nd to 12th recording with a draw, then all later ones: 0.81+0.10, 0.84+0.11, 0.01+0.13, 0.65+0.02, 0.00+0.02, 0.00+0.01, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 1070 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 ModuleSystemWorker4] vk: invalidates since the last readbacks line: 2841 kept their virtual core (nothing there to wait for), 2791 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 199 KB; 0 made on demand (0 MB kept now); c1 stages: at most 97 a recording; 8 descriptor pools made when a recording's own ran out (8 slots' pools made whole); a recording's pool took at most 400 sets (ub 4476, sb 2017, cis 33443, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 1078, 4195.8 MB; 31 bindings followed, 39 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 105 dirty pages from their submits' captures, 48557 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): 1084 dispatches, 40124 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6319 runs, 1155.5 MB compared; 98 pages read into the shadows; 352 MB of shadows (197 MB of pages known, in 234 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 15778 in captures, 4 only in the collection, 49 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1820000 (857291 pages from the captures)
[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: 5 program records loaded (1.5 ms), 0 new graphics pipelines, pipelines from the driver's cache 0, from the start's cache 0, compiled 0; 0 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (close): 4846 queued, 4846 loaded (median 0.15 ms, 99th percentile 1.41 ms, worst 8.55 ms, 1190 ms and 961.1 MB read in all); 6857 programs (1794 of them compute variants without their code, 0 graphics programs no note names), 8908 graphics pipelines (0 not in a cache), 582 compute pipelines (25 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 788 pipelines made at create taken by draws; 2191 create calls with nothing kept for their stages, 0 before the index; units 7037 (4846 ready, 0 loading), 12254 programs alive; 0 noted graphics pipeline probes deferred to first use
[hos host] vk: program loads at create (close): 9898 shader modules let go once their programs' noted pipelines were built, 0 programs' modules made again for a new pipeline state (RX_VK_DROP_LOADED_MODULES=0 keeps them)
[hos host] vk: module identifiers (close): 607 known, 0 learned in this run; 582 compute pipelines made by them at a create call, 25 asked that way were in no cache (their code read, as before)
[hos host] vk: indexed descriptor arrays (close): 945 in 945 of 16805 program stages, 0 of them near a limit
[hos host] vk: indexed descriptor arrays (close), fragment stages: 945 texture (17-64: 945); the longest 64
[hos host] vk: page table census (close): no pool outside its windows, 64-72 GB and 508-512 GB (4267 pools in the table now)
[hos host] control: exiting
[hos host] stop: close work done (a file the game was writing waited for, at most 3 s)