Beanstalk Longshot test farm
Menu

Report rp_e7e2c1e04417c5d0 fail

#1987 WIP: perf(vk): a BC3-kept ASTC texture's levels re-encode in one run, and a GPU-drawn target is not uploaded again for a buffer write (test switches)

vram8gb on manasuazo/agent-03253d,

Machine

Commit
623507146b63cdc503ddcb382bbad7ab6cd59c26
Baseline
ed602b9468143506917b35488cd61d7c523156f1
App
nightly-217-14cc1aa3, attempt 1
Game build
e0f3ccfa
os
linux
cpu
AMD Ryzen 5 5500F 6-Core Processor
gpu
AMD Radeon RX 580 Series (RADV POLARIS10)
tags
linuxamdradvdgpu
driver
26.2.7
ram_gb
32
vram_mb
8,192
gpu_vendor
amd
os_version
<host> (kernel 7.3.4)

Builds

main (baseline)

47.1 s warm

from ed602b946814

lift
0.7 s
read
4.7 s
finish
16.2 s
compile
17.2 s
shaders
7.2 s

PR

42.4 s warm

Program cache: warm

Scenarios

texture-turns-and-entries RX_VK_BC_LEVELS=1 (PR only)

ok
Metric (median)mainPRΔ
exit_code00±0%
load_time_s4.873.82−21.6%
load_time_max_s5.686.09+7.2%
run_hitch_count139113−18.7%
market.hitch_count59+80%
market.stutter_count110−100%
run_frametime_max_ms1,873.591,746.15−6.8%
market.frametime_max_ms63.9456.86−11.1%
market.frametime_p50_ms20.1730.92+53.3%
market.frametime_p99_ms42.3151.39+21.5%
market.frametime_1pct_low_ms50.3152.67+4.7%
screenshot_diff—0.88—

Pictures against main: forest diff 0.0002 (main against itself ), field diff 0.3174 (main against itself ), market diff 0.8804 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
load_time_s4.873.82
load_time_max_s5.686.09
run_hitch_count139113
market.hitch_count59
market.stutter_count110
run_frametime_max_ms1,873.591,746.15
market.frametime_max_ms63.9456.86
market.frametime_p50_ms20.1730.92
market.frametime_p99_ms42.3151.39
market.frametime_1pct_low_ms50.3152.67
screenshot_diff—0.88
loadsKokiriForest_Child 3.29 sHyruleField_Child 5.65 sMarket_Child 5.68 sKokiriForest_Child 0.06 sHyruleField_Child 5.32 sMarket_Child 6.09 s

Rules

ScenarioCellRuleLimitmainPRΔ %Status
texture-turns-and-entriesRX_VK_BC_LEVELS=1 (PR only)crashFalse00—pass
texture-turns-and-entriesRX_VK_BC_LEVELS=1 (PR only)screenshot.max_diff0.05—0.88—fail
texture-turns-and-entriesRX_VK_BC_LEVELS=1 (PR only)load_time_max_s.max_delta_pct255.686.09+7.2%pass
texture-turns-and-entriesRX_VK_BC_LEVELS=1 (PR only)market.frametime_p99_ms.max_delta_pct1542.3151.39+21.5%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
texture-turns-and-entriesRX_VK_BC_LEVELS=1 (PR only)run_frametime_max_ms.max_delta_pct301,873.591,746.15−6.8%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)

Pictures of the failure

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

texture-turns-and-entries-c0-field-main-failed.png
texture-turns-and-entries-c0-field-main-failed.png
texture-turns-and-entries-c0-field-pr-failed.png
texture-turns-and-entries-c0-field-pr-failed.png
texture-turns-and-entries-c0-market-main-failed.png
texture-turns-and-entries-c0-market-main-failed.png

Log (PR side, end)

[hos host] vk: sparse mirrors: 129, 1193 MB with memory of 1616 MB (422 MB spared); blocks given memory later 8507, given back 11606; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 192; 574 binds (332.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 108 taken from the spares, 201 made new; 180 kept as spares, 0 destroyed; 72 spares now, 1476 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 756, 3708.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5416 dirty pages from their submits' captures, 36292 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): 1041 dispatches, 30650 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5243 runs, 636.3 MB compared; 143 pages read into the shadows; 264 MB of shadows (35 MB of pages known, in 75 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13452 in captures, 0 only in the collection, 15 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 167 live (1964 MB), 0 retired still imported (0 MB), 132 retired given back (2136 MB)
[hos host] vk: replay reads of guest memory as submitted: 820000 (495532 pages from the captures)
[hos host] vk: mirror of pool 0x7f70c00000 (4 MB) given back: idle, 6023 submissions since its sync
[hos host] vk: mirror of pool 0x7f71000000 (4 MB) given back: idle, 6023 submissions since its sync
[hos host] vk: mirror of pool 0x7f36800000 (4 MB) given back: idle, 6022 submissions since its sync
[hos host] vk: mirror of pool 0x7fa5400000 (4 MB) given back: idle, 6021 submissions since its sync
[hos host] vk: mirror of pool 0x7fa8000000 (8 MB) given back: idle, 6021 submissions since its sync
[hos host] vk: mirror of pool 0x7fb5800000 (12 MB) given back: idle, 6021 submissions since its sync
[hos host] vk: mirror of pool 0x1165c00000 (8 MB) given back: idle, 6021 submissions since its sync
[hos host] vk: mirror of pool 0x1167000000 (8 MB) given back: idle, 6021 submissions since its sync
[hos host] vk: mirror of pool 0x7f57800000 (4 MB) given back: idle, 6020 submissions since its sync
[hos host] vk: mirror of pool 0x7fae400000 (28 MB) given back: idle, 6020 submissions since its sync
[hos sead::MainThread] vk: capture: a submit 0.20 ms (lock 0.00, scan 0.04 over 944 ranges, 4888 MB), 266 pages written, 266 copied (1.0 MB)
[hos host] vk: a capture taken (ms): all 0.043; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.029, bytes copied 0.000; 25 kept (138 MB), 12 spare buffers, 10930 texture pages copied (42 MB)
[hos host] vk: program picks: 1020082 kept (414307 by the last picks' textures, a uniform buffer bound since), 17123 new, 362794 with other textures (of 1400000)
[hos host] vk: reaper: 1254 images (3663 MB) destroyed in 10 s, 0 of them where they were handed over (its cap or VRAM pressure); the queue at most 460 (796 MB)
[hos host] vk: 2 textures retired (2359 left, 2967 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 840000 (507945 pages from the captures)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 283 KB; 0 made on demand (0 MB kept now); c1 stages: at most 128 a recording; 26 descriptor pools made when a recording's own ran out (26 slots' pools made whole); a recording's pool took at most 505 sets (ub 5550, sb 2563, cis 42181, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 735, 3605.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5594 dirty pages from their submits' captures, 35848 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): 1022 dispatches, 30599 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5319 runs, 652.6 MB compared; 60 pages read into the shadows; 264 MB of shadows (34 MB of pages known, in 73 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13535 in captures, 0 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: 860000 (520435 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 46799 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): 185 pages in 19 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #30000: event 0x100d6c8038 mode 0 value 19177 (now 19139) (lr 0x80661fac)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.05, writes collected 0.03
[hos BaseProcExecutor1] vcores: 58777 takes, 9244 waits (mean 90 us, longest 14.3 ms, 0 over HOS_VCORE_LOG_MS), 4115 kicks, 4394 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: pipeline cache: play file 5 saved (4766 KB): 122 pipelines, 122 from the driver's cache, 0 compiled; 0 more from the start's cache
[hos host] vk: replay reads of guest memory as submitted: 880000 (532985 pages from the captures)
[hos host] vk: 1 textures retired (2364 left, 2964 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 900000 (545447 pages from the captures)
[hos host] vk: sparse mirrors: 119, 1146 MB with memory of 1532 MB (385 MB spared); blocks given memory later 8507, given back 11606; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 192; 575 binds (333.4 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 108 taken from the spares, 201 made new; 190 kept as spares, 0 destroyed; 82 spares now, 1560 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 756, 3708.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5359 dirty pages from their submits' captures, 36618 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): 1020 dispatches, 30922 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5491 runs, 655.4 MB compared; 72 pages read into the shadows; 264 MB of shadows (35 MB of pages known, in 73 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13448 in captures, 0 only in the collection, 21 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 167 live (1964 MB), 0 retired still imported (0 MB), 132 retired given back (2136 MB)
[hos host] vk: readback pages refreshed from guest memory: 386115 pages so far (0 held: may be written ahead in their submission)
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #20000: event 0x100d6c8000 mode 0 value 19999 (lr 0x80291440)
[hos ModuleSystemWorker3] vk: readbacks of GPU results: 704 of 3000 waited for their write-back, 0.99 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 ModuleSystemWorker3] vk: invalidates since the last readbacks line: 8217 kept their virtual core (nothing there to wait for), 3047 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: 920000 (557997 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.19 ms (lock 0.00, scan 0.04 over 944 ranges, 4888 MB), 274 pages written, 274 copied (1.1 MB)
[hos host] vk: a capture taken (ms): all 0.045; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.031, bytes copied 0.000; 24 kept (172 MB), 12 spare buffers, 15766 texture pages copied (61 MB)
[hos host] vk: program picks: 1171911 kept (466003 by the last picks' textures, a uniform buffer bound since), 17203 new, 410885 with other textures (of 1600000)
[hos host] vk: replay reads of guest memory as submitted: 940000 (570505 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 756, 3708.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5522 dirty pages from their submits' captures, 36549 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): 1018 dispatches, 30968 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5509 runs, 655.6 MB compared; 73 pages read into the shadows; 264 MB of shadows (34 MB of pages known, in 73 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13551 in captures, 0 only in the collection, 20 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a frame: 409 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 10986 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 18.50, present 0.33, draw 7.80, descriptors 2.61, texture lookups 0.38, uploads 0.17, pipelines 0.29, programs 1.04, commands 3.34, dispatch 0.83, memory sync 0.47, queue submit 7.29, slot wait 0.08
[hos host] vk: a frame: mirrors 309 (1146 MB): 4.57 MB synced in 860.5 copies, 18.18 MB written back
[hos host] vk: 2760 presents, 30.9 fps; a frame: 2418 draws (0 skipped), 127 dispatches, 149 passes, 158 barriers, 6 submits, 1.0 uploads (0.1 MB), 82 ub updates (57 overlays), GPU-bound wait 0.1 ms (recording slots only); 2739 pipelines (479 ms), 1593 programs (12 failed), 2378 textures (2978 MB)
[hos host] vk: host time in GPU resource waits: 19.16 ms a frame (fences 0.04, timelines 19.12, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 525.7 full barriers recorded, 284.2 left out (nothing recorded since the one before), 15.2 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: 49.3; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 34.4 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 34.4 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 23.6 ms a frame (the longest submission 9.6 ms); 1.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.1 MB), ASTC or BC 0.9 (0.6 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.07 MB unswizzled on the CPU in 0.05 ms
[hos host] vk: mirror of pool 0x7fd6800000 (36 MB) given back: idle, 6017 submissions since its sync
[hos host] vk: mirror of pool 0x7fe2000000 (4 MB) given back: idle, 6015 submissions since its sync
[hos host] vk: 192 mirrors given back (1808 MB in all); 1114 MB of mirrors now, 44 MB of them idle and kept for what the GPU wrote there alone
[hos host] vk: texture #6874 0x10001518e00 1920x1080 NVN 0x34: the GPU wrote its storage (submission 19376): uploaded again
[hos host] vk: upload over a drawn image: #6874 0x10001518e00 1920x1080x1 x1 NVN 0x34 at present 2764 (texture frame 2765), finalized, 0 KB dirty in 0 ranges, a GPU buffer wrote it
[hos host] vk: texture 0x10001518e00 1920x1080x1 x1 levels 1 NVN 0x34 (8640 KB of storage) re-uploaded (its memory was written; flags 0x8): a GPU write into its storage, 8640 KB of it, over what the GPU wrote (finalized since)
[hos host] vk: replay reads of guest memory as submitted: 960000 (582958 pages from the captures)
[hos host] vk: mirror of pool 0x7fb7000000 (4 MB) given back: idle, 6012 submissions since its sync
[hos host] vk: mirror of pool 0x7f7f800000 (4 MB) given back: idle, 6012 submissions since its sync
[hos host] vk: mirror of pool 0x7fe2c00000 (4 MB) given back: idle, 6008 submissions since its sync
[hos host] vk: mirror of pool 0x105b800000 (4 MB) given back: idle, 6012 submissions since its sync
[hos Presentation Thread] vi: 2785 frames presented (30.9 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 0, 22-30 54, 30-40 93, over 40 8
[hos sead::MainThread] nvn: GPU waits in 5 s: 153 sync waits (2243 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: mirror of pool 0x7f92400000 (4 MB) given back: idle, 6018 submissions since its sync
[hos host] vk: mirror of pool 0x105cc00000 (4 MB) given back: idle, 6015 submissions since its sync
[hos host] vk: replay reads of guest memory as submitted: 980000 (595467 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.05, writes collected 0.03
[hos host] vk: mirror of pool 0x105b000000 (4 MB) given back: idle, 6051 submissions since its sync
[hos host] vk: mirror of pool 0x105c000000 (4 MB) given back: idle, 6051 submissions since its sync
[hos host] vk: sparse mirrors: 108, 1107 MB with memory of 1456 MB (348 MB spared); blocks given memory later 8507, given back 11606; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 192; 579 binds (334.7 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 108 taken from the spares, 201 made new; 198 kept as spares, 0 destroyed; 90 spares now, 1624 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 756, 3708.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5626 dirty pages from their submits' captures, 36438 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): 1040 dispatches, 31055 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5480 runs, 655.2 MB compared; 157 pages read into the shadows; 264 MB of shadows (35 MB of pages known, in 73 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13456 in captures, 0 only in the collection, 19 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 167 live (1964 MB), 0 retired still imported (0 MB), 132 retired given back (2136 MB)
[hos host] vk: replay reads of guest memory as submitted: 1000000 (607916 pages from the captures)
[hos host] vk: VRAM compaction 2 begins: the textures' blocks hold 2624 MB, 605 MB of it free space
[hos host] vk: program picks: 1323651 kept (517447 by the last picks' textures, a uniform buffer bound since), 17284 new, 459064 with other textures (of 1800000)
[hos host] vk: replay reads of guest memory as submitted: 1020000 (620438 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.13 ms (lock 0.00, scan 0.04 over 944 ranges, 4888 MB), 243 pages written, 243 copied (1.0 MB)
[hos host] vk: a capture taken (ms): all 0.030; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.017, bytes copied 0.000; 24 kept (98 MB), 10 spare buffers, 13615 texture pages copied (53 MB)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5750 dirty pages from their submits' captures, 37508 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): 1062 dispatches, 31740 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5596 runs, 673.0 MB compared; 267 pages read into the shadows; 264 MB of shadows (34 MB of pages known, in 73 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13700 in captures, 0 only in the collection, 28 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1040000 (632992 pages from the captures)
[hos AudioFrameworkThread] vcores: 61467 takes, 8991 waits (mean 83 us, longest 9.0 ms, 0 over HOS_VCORE_LOG_MS), 3958 kicks, 4268 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 49653 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 2 textures retired (2418 left, 2997 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: late pages found at the replay (600 takes): 376 pages in 20 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: 209 mirrors given back (1852 MB in all); 1071 MB of mirrors now, 44 MB of them idle and kept for what the GPU wrote there alone
[hos host] vk: replay reads of guest memory as submitted: 1060000 (645505 pages from the captures)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 283 KB; 0 made on demand (0 MB kept now); c1 stages: at most 128 a recording; 26 descriptor pools made when a recording's own ran out (26 slots' pools made whole); a recording's pool took at most 505 sets (ub 5550, sb 2563, cis 42181, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 93, 1039 MB with memory of 1336 MB (296 MB spared); blocks given memory later 8507, given back 11606; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 192; 586 binds (338.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 108 taken from the spares, 201 made new; 216 kept as spares, 0 destroyed; 108 spares now, 1756 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5794 dirty pages from their submits' captures, 37780 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): 1076 dispatches, 32049 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5560 runs, 672.4 MB compared; 71 pages read into the shadows; 264 MB of shadows (35 MB of pages known, in 75 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13860 in captures, 0 only in the collection, 31 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 167 live (1964 MB), 0 retired still imported (0 MB), 132 retired given back (2136 MB)
[hos ModuleSystemWorker5] vk: readbacks of GPU results: 625 of 3000 waited for their write-back, 1.18 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 ModuleSystemWorker5] vk: invalidates since the last readbacks line: 8211 kept their virtual core (nothing there to wait for), 3053 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: VRAM compaction 2 over: 350 textures moved (152 MB) in 31 passes, 0 left where they were; 13 blocks (416 MB) given back to the driver, 1088 MB in all so far
[hos host] vk: 2 textures retired (2439 left, 2995 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1080000 (657953 pages from the captures)
[hos host] vk: a frame: 406 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 10986 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 18.55, present 0.80, draw 7.87, descriptors 2.78, texture lookups 0.58, uploads 0.24, pipelines 0.29, programs 1.03, commands 3.26, dispatch 0.93, memory sync 0.44, queue submit 6.90, slot wait 0.08
[hos host] vk: a frame: mirrors 309 (1039 MB): 4.55 MB synced in 858.5 copies, 18.18 MB written back
[hos host] vk: 2880 presents, 30.9 fps; a frame: 2395 draws (0 skipped), 127 dispatches, 148 passes, 158 barriers, 6 submits, 1.3 uploads (0.2 MB), 82 ub updates (57 overlays), GPU-bound wait 0.1 ms (recording slots only); 2739 pipelines (479 ms), 1593 programs (12 failed), 2439 textures (2995 MB)
[hos host] vk: host time in GPU resource waits: 19.87 ms a frame (fences 0.04, timelines 19.83, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 524.1 full barriers recorded, 283.4 left out (nothing recorded since the one before), 15.2 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: 49.2; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 34.3 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 34.3 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 23.5 ms a frame (the longest submission 10.4 ms); 1.2 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.1 MB), ASTC or BC 1.2 (0.6 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.16 MB unswizzled on the CPU in 0.12 ms
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #35000: event 0x100d6c8038 mode 0 value 22977 (now 22939) (lr 0x80661fac)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.05, writes collected 0.02
[hos host] vk: replay reads of guest memory as submitted: 1100000 (670368 pages from the captures)
[hos host] vk: 5 textures retired (2447 left, 2986 MB): budget 0, drawn over 5, dead targets 0
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5684 dirty pages from their submits' captures, 36825 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): 1056 dispatches, 31431 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5462 runs, 669.6 MB compared; 140 pages read into the shadows; 264 MB of shadows (36 MB of pages known, in 76 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13546 in captures, 0 only in the collection, 29 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: program picks: 1475731 kept (569427 by the last picks' textures, a uniform buffer bound since), 17365 new, 506903 with other textures (of 2000000)
[hos host] vk: 2 textures retired (2456 left, 2982 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1120000 (682900 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.14 ms (lock 0.00, scan 0.04 over 944 ranges, 4888 MB), 265 pages written, 265 copied (1.0 MB)
[hos host] vk: a capture taken (ms): all 0.044; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.029, bytes copied 0.000; 25 kept (100 MB), 10 spare buffers, 6990 texture pages copied (27 MB)
[hos host] vk: readback pages refreshed from guest memory: 459213 pages so far (0 held: may be written ahead in their submission)
[hos BaseProcExecutor2] sync: object cache (HOS_OBJ_CACHE=16384 a thread, 44 made): 8990315 lookups, 79488 missed (0.88 %)
[hos host] vk: replay reads of guest memory as submitted: 1140000 (695447 pages from the captures)
[hos Presentation Thread] vi: 2941 frames presented (31.1 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 0, 22-30 42, 30-40 108, over 40 6
[hos host] vk: sparse mirrors: 93, 1039 MB with memory of 1336 MB (296 MB spared); blocks given memory later 8507, given back 11606; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 192; 586 binds (338.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 108 taken from the spares, 201 made new; 216 kept as spares, 0 destroyed; 108 spares now, 1756 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 756, 3708.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5767 dirty pages from their submits' captures, 36787 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): 1037 dispatches, 31244 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5458 runs, 655.8 MB compared; 70 pages read into the shadows; 264 MB of shadows (35 MB of pages known, in 75 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13872 in captures, 0 only in the collection, 27 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 167 live (1964 MB), 0 retired still imported (0 MB), 132 retired given back (2136 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 156 sync waits (2193 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: replay reads of guest memory as submitted: 1160000 (707948 pages from the captures)
[hos ModuleSystemWorker1] nvn: 550000 recordings made, 67120 live copies (22.6 MB), 1453 texture copies; 3 command buffers' segments (4.1 MB)
[hos host] vk: replay reads of guest memory as submitted: 1180000 (720496 pages from the captures)
[hos host] vk: 1 textures retired (2476 left, 2995 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5239 dirty pages from their submits' captures, 37518 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): 1058 dispatches, 31381 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5663 runs, 674.2 MB compared; 73 pages read into the shadows; 264 MB of shadows (36 MB of pages known, in 76 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13509 in captures, 0 only in the collection, 28 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1200000 (733082 pages from the captures)
[hos host] vk: program picks: 1627765 kept (621160 by the last picks' textures, a uniform buffer bound since), 17446 new, 554788 with other textures (of 2200000)
[hos host] vk: reaper: 206 images (237 MB) destroyed in 10 s, 0 of them where they were handed over (its cap or VRAM pressure); the queue at most 21 (24 MB)
[hos host] vk: 1 textures retired (2475 left, 2990 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.05, writes collected 0.02
[hos host] vk: memory: working set 8613 MB (peak 8625), private 1991 MB, staging 144 MB in 9 arenas, 30 made so far (25 ms)
[hos host] vk: memory heaps: heap 0 (device): 173 blocks 5266 MB, 19703 allocations 4493 MB; used 5303 of 7294 MB; heap 1 (system): 16 blocks 920 MB, 39 allocations 866 MB; used 2966 of 15633 MB; heap 2 (device): 64 blocks 237 MB, 156 allocations 237 MB; used 436 of 256 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 6 images 31 MB; 3 free ranges, the largest 0 MB
[hos host] vk: image blocks (render targets): 35 blocks 560 MB, 162 images 346 MB; 151 free ranges, the largest 16 MB
[hos host] vk: image blocks (textures, compacted): 70 blocks 2240 MB, 2306 images 2045 MB; 1523 free ranges, the largest 10 MB
[hos host] vk: render target images: 1081 reused from the pool, 1297 made new, 1130 destroyed from it, 127 reused by size (on a pooled one's memory, RX_VK_TARGET_REUSE_MEMORY=1), 32 missed after their image was taken by size; it holds 0 (0 MB)
[hos host] vk: render targets not reused by size: 749 with the pool empty (0 of them with targets still in the dead queue), 180 with none in it a present yet, 31 with none of a kind they can use, 99 with all smaller (the largest 46% short on average), 161 with all bigger (over 1.5 times), 68 with some of each, 9 over a block, 0 refused by the driver, 0 not bound; the pool let go of 236 over its bound and 894 idle
[hos host] vk: VRAM: 5739 MB used of 7550 MB allowed (772 MB of it free in the blocks); textures 2990 MB (budget 4629 MB), 2477 textures
[hos host] vk: budget: VRAM 5503 of 7550 MB (5739 MB by the driver, 236 MB unaccounted): textures 2990, mirrors 1039, blocks free 772, shadows 264, other buffers 213, staging 144, other images 79; RAM 2866 MB (the process's private memory 1991 MB): imports 1964, capture spares 710, captures 128, capture table 36, other buffers 28, others 0
[hos host] vk: textures: decoded ASTC 753 MB, rendered 857 MB (64 drawn in one frame, kept: 51560 KB), other 536 MB; used in the last 8 presents 1403 MB
[hos host] vk: decoded ASTC used again: 354 textures (326 MB) copied from ones the game finalized, not decoded; 354 of those retired, 70 listed; 1192 MB of blocks hashed in 131 ms
[hos host] vk:   288 MB: 0x1147c16000 12288x12288x1 x1 levels 1 NVN 0x32 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:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   37 MB: 0x110e181600 1800x1380x1 x8 levels 1 NVN 0x6, last used 830 presents ago
[hos host] vk:   32 MB: 0x1159c17000 4096x4096x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   31 MB: 0x111ae71c00 1775x1175x1 x8 levels 1 NVN 0x6, last used 1 presents ago
[hos host] vk:   21 MB: 0x10004850600 2400x2400x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   14 MB: 0x100030a4200 1600x1600x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 48 (6815 KB): 242x40x1 NVN 0x25: 4 (151 KB); 260x260x1 NVN 0xd: 34 (5982 KB); 306x104x1 NVN 0x1: 4 (124 KB); 592x49x1 NVN 0x25: 3 (339 KB); 656x113x1 NVN 0x1: 3 (217 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 593 waited for it, 15.15 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #25000: event 0x100d6c8010 mode 0 value 25000 (lr 0x80660adc)
[hos host] vk: status: 600 frames, the slowest 213.5 ms; VRAM 5.6 of 7.4 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: 0 program records loaded (0.0 ms), 2247 new graphics pipelines, pipelines from the driver's cache 2405, from the start's cache 0, compiled 136; 4827 draws and dispatches left out, 1456 waits (10000.1 ms)
[hos host] vk: program loads at create (play): 0 queued, 0 loaded (median 0.00 ms, 99th percentile 0.00 ms, worst 0.00 ms, 0 ms and 0.0 MB read in all); 0 programs (0 of them compute variants without their code, 0 graphics programs no note names), 0 graphics pipelines (0 not in a cache), 0 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 0 pipelines made at create taken by draws; 8919 create calls with nothing kept for their stages, 0 before the index; units 8919 (0 ready, 0 loading), 14616 programs alive; 0 noted graphics pipeline probes deferred to first use
[hos host] vk: program loads at create (play): 0 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): 0 known, 0 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: 407 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 10986 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 17.86, present 0.31, draw 7.75, descriptors 2.45, texture lookups 0.26, uploads 0.11, pipelines 0.29, programs 1.03, commands 3.47, dispatch 0.89, memory sync 0.42, queue submit 6.87, slot wait 0.08
[hos host] vk: a frame: mirrors 309 (1039 MB): 4.56 MB synced in 857.4 copies, 18.19 MB written back
[hos host] vk: 3000 presents, 31.2 fps; a frame: 2391 draws (0 skipped), 127 dispatches, 149 passes, 158 barriers, 6 submits, 0.9 uploads (0.1 MB), 82 ub updates (57 overlays), GPU-bound wait 0.1 ms (recording slots only); 2739 pipelines (479 ms), 1593 programs (12 failed), 2477 textures (2990 MB)
[hos host] vk: host time in GPU resource waits: 19.43 ms a frame (fences 0.04, timelines 19.39, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 524.4 full barriers recorded, 283.2 left out (nothing recorded since the one before), 15.1 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: 49.0; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 34.4 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 34.4 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 23.4 ms a frame (the longest submission 9.4 ms); 0.9 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.2 MB), ASTC or BC 0.8 (0.3 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.06 MB unswizzled on the CPU in 0.05 ms
[hos host] vk: the render thread (600 presents, 33.12 ms a frame): its jobs 21.19 ms a frame (submits 2.25, presents 0.47, others 18.46), idle 11.94; the game waited for room at 1 presents, 1.28 ms each (0.00 a frame; 1.51 jobs queued at a present); 8 catch-ups (a finish, a texture read back, a write-back not recorded yet: 0), 21.19 ms each; 0 presents held for the CPU's signal, 0.00 ms each (0 let go by RX_VK_HOLD_MS); 593 captures at a CPU signal; the submits' waits read in 0.048 ms a frame; readbacks: 2970 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 19.26 ms after its hand-over (189.70 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 0.77 + 1.06 (the window's copy is in the idle), after it 3.36 + 4.96; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.14 ms of GPU time), after its last that reads it 1.00 (0.14 ms)
[hos host] vk: frame ahead (600 presents): 9.09 parts a frame recorded in the signal's job, 63.0% of them took nothing that came with the signal; of the others, 3.36 a frame took pages of the signal's capture into a mirror (143.1 pages a frame) and 0.02 pages read as guest memory had them (1.4 pages a frame); a frame's first part took nothing in 0 of 593 frames, its first three in 0
[hos host] vk: frame ahead, what came with the signal: 593 signals with a capture, each of 232.8 pages in 14.7 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: 1.31+1.85, 0.14+0.31, 0.71+0.40, 0.60+0.64, 0.23+0.71, 0.09+0.34, 0.15+0.37, 0.05+0.23, 0.01+0.07, 0.06+0.05, 0.00+0.00, 0.00+0.00
[hos sead::MainThread] vk: capture: a submit 0.16 ms (lock 0.00, scan 0.04 over 944 ranges, 4888 MB), 246 pages written, 246 copied (1.0 MB)
[hos host] vk: a capture taken (ms): all 0.035; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.020, bytes copied 0.000; 25 kept (120 MB), 11 spare buffers, 6344 texture pages copied (24 MB)
[hos host] vk: replay reads of guest memory as submitted: 1220000 (745615 pages from the captures)
[hos host] vk: 1 textures retired (2478 left, 2994 MB): budget 0, drawn over 1, dead targets 0
[hos BaseProcExecutor0] vcores: 58814 takes, 9172 waits (mean 92 us, longest 22.1 ms, 0 over HOS_VCORE_LOG_MS), 4075 kicks, 4397 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: sparse mirrors: 93, 1039 MB with memory of 1336 MB (296 MB spared); blocks given memory later 8507, given back 11606; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 192; 586 binds (338.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 108 taken from the spares, 201 made new; 216 kept as spares, 0 destroyed; 108 spares now, 1756 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 756, 3708.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5722 dirty pages from their submits' captures, 36835 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): 1039 dispatches, 31279 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5515 runs, 656.4 MB compared; 159 pages read into the shadows; 264 MB of shadows (36 MB of pages known, in 76 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13892 in captures, 0 only in the collection, 22 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 167 live (1964 MB), 0 retired still imported (0 MB), 132 retired given back (2136 MB)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 653 of 3000 waited for their write-back, 1.23 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: 8307 kept their virtual core (nothing there to wait for), 2957 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: 1240000 (758113 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 35864 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): 176 pages in 10 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos Presentation Thread] vi: real-time step on: ratio 1.836, game's frame rate 32
[hos host] vk: 1 textures retired (2477 left, 2990 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1260000 (770639 pages from the captures)
[hos host] vk: 1 textures retired (2476 left, 2986 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 283 KB; 0 made on demand (0 MB kept now); c1 stages: at most 128 a recording; 26 descriptor pools made when a recording's own ran out (26 slots' pools made whole); a recording's pool took at most 505 sets (ub 5550, sb 2563, cis 42181, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5469 dirty pages from their submits' captures, 36805 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): 1017 dispatches, 31160 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5702 runs, 675.1 MB compared; 73 pages read into the shadows; 264 MB of shadows (36 MB of pages known, in 76 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13545 in captures, 0 only in the collection, 28 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1280000 (783154 pages from the captures)
[hos host] vk: program picks: 1779725 kept (672828 by the last picks' textures, a uniform buffer bound since), 17527 new, 602747 with other textures (of 2400000)
[hos host] vk: replay reads of guest memory as submitted: 1300000 (795668 pages from the captures)
[hos rmbl::VibrationThread] hid: 16384 vibration values, strongest 0.00 since the last count
[hos Presentation Thread] vk: capture: a submit 0.13 ms (lock 0.00, scan 0.04 over 950 ranges, 4702 MB), 238 pages written, 238 copied (0.9 MB)
[hos host] vk: a capture taken (ms): all 0.028; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.014, bytes copied 0.000; 24 kept (116 MB), 12 spare buffers, 8584 texture pages copied (33 MB)
[hos host] vk: sparse mirrors: 93, 1039 MB with memory of 1336 MB (296 MB spared); blocks given memory later 8507, given back 11606; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 192; 586 binds (338.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 108 taken from the spares, 201 made new; 216 kept as spares, 0 destroyed; 108 spares now, 1756 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 756, 3708.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5640 dirty pages from their submits' captures, 36398 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): 1040 dispatches, 30952 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5397 runs, 655.5 MB compared; 64 pages read into the shadows; 264 MB of shadows (34 MB of pages known, in 73 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 14005 in captures, 0 only in the collection, 28 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 167 live (1964 MB), 0 retired still imported (0 MB), 132 retired given back (2136 MB)
[hos host] vk: render targets drawn over from another start, let go: 2234 so far (6409 MB), 1942 of them (5720 MB) between the passes
[hos host] vk: replay reads of guest memory as submitted: 1320000 (808112 pages from the captures)
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #40000: event 0x100d6c8038 mode 0 value 26777 (now 26739) (lr 0x80661fac)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.04, writes collected 0.02
[hos host] vk: late pages' fix (600 recordings with one): of 604 captures with late pages 604 fixed (78.5 pages, 78.6 fixes a capture), waited as before: 0 no pool, 0 no stamps or shadow yet, 0 stamps too new, 0 whole copies, 0 in place or unknown, 0 a page's earlier fix not recorded yet, 0 other; 74 read on the CPU first, 0 with no table, 0 pages whose shadow changed first; stamps 576 blocks (36864 KB); of the reads on the CPU, 5 waited for an earlier part's fix, 0 copied the page again instead (that part not waited for); the paths: reads on the CPU with the fix done 48, of an earlier recording 21, not recorded yet 0, of this part 0; pages fixed from stamps 46599, from snapshots 820; the shadows: 191120 pages synced from captures, 190569 made known by it, 551 left unknown because the replay or the collection wrote them since
[hos Presentation Thread] vi: 3098 frames presented (31.3 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 0, 22-30 61, 30-40 90, over 40 6
[hos host] vk: readback pages refreshed from guest memory: 532705 pages so far (0 held: may be written ahead in their submission)
[hos sead::MainThread] nvn: GPU waits in 5 s: 156 sync waits (2186 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: replay reads of guest memory as submitted: 1340000 (820543 pages from the captures)
[hos ModuleSystemWorker4] nvn: DrawMeshTasksIndirect: 60000 so far (x1 0x117b427030 x2 0 x3 0x117b427000 x4 0xf33080)
[hos host] vk: a frame: 410 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 10986 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 17.93, present 0.32, draw 7.55, descriptors 2.36, texture lookups 0.17, uploads 0.02, pipelines 0.29, programs 1.05, commands 3.32, dispatch 0.84, memory sync 0.37, queue submit 6.96, slot wait 0.08
[hos host] vk: a frame: mirrors 309 (1039 MB): 4.55 MB synced in 856.3 copies, 18.22 MB written back
[hos host] vk: 3120 presents, 31.3 fps; a frame: 2398 draws (0 skipped), 128 dispatches, 149 passes, 157 barriers, 6 submits, 0.1 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.1 ms (recording slots only); 2739 pipelines (479 ms), 1593 programs (12 failed), 2482 textures (3000 MB)
[hos host] vk: host time in GPU resource waits: 19.78 ms a frame (fences 0.04, timelines 19.73, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 524.3 full barriers recorded, 283.3 left out (nothing recorded since the one before), 15.2 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: 49.0; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 34.4 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 34.4 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 23.6 ms a frame (the longest submission 9.5 ms); 0.2 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.2 MB), ASTC or BC 0.1 (0.1 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: write-backs left out (600 submits): 777, 3811.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5744 dirty pages from their submits' captures, 37401 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): 1071 dispatches, 31686 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5392 runs, 654.1 MB compared; 56 pages read into the shadows; 264 MB of shadows (36 MB of pages known, in 76 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13544 in captures, 0 only in the collection, 26 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1360000 (833050 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 1380000 (845535 pages from the captures)
[hos host] vk: program picks: 1932423 kept (724916 by the last picks' textures, a uniform buffer bound since), 17607 new, 649969 with other textures (of 2600000)
[hos ModuleSystemWorker5] vk: readbacks of GPU results: 681 of 3000 waited for their write-back, 1.16 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker5] vk: invalidates since the last readbacks line: 8240 kept their virtual core (nothing there to wait for), 3024 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: 1400000 (858053 pages from the captures)
[hos host] vk: sparse mirrors: 93, 1039 MB with memory of 1336 MB (296 MB spared); blocks given memory later 8507, given back 11606; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 192; 586 binds (338.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 108 taken from the spares, 201 made new; 216 kept as spares, 0 destroyed; 108 spares now, 1756 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5588 dirty pages from their submits' captures, 37407 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): 1043 dispatches, 31662 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5785 runs, 693.1 MB compared; 75 pages read into the shadows; 264 MB of shadows (36 MB of pages known, in 76 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13979 in captures, 0 only in the collection, 34 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 167 live (1964 MB), 0 retired still imported (0 MB), 132 retired given back (2136 MB)
[hos BaseProcExecutor1] vcores: 59592 takes, 8979 waits (mean 75 us, longest 6.6 ms, 0 over HOS_VCORE_LOG_MS), 3992 kicks, 4320 yields, 0 starved grants, 0 escapes in 5 s
[hos Presentation Thread] vk: capture: a submit 0.15 ms (lock 0.00, scan 0.04 over 950 ranges, 4688 MB), 260 pages written, 260 copied (1.0 MB)
[hos host] vk: a capture taken (ms): all 0.042; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.027, bytes copied 0.000; 25 kept (118 MB), 11 spare buffers, 10032 texture pages copied (39 MB)
[hos host] vk: replay reads of guest memory as submitted: 1420000 (870496 pages from the captures)
[hos host] vk: 1 textures retired (2486 left, 2997 MB): budget 0, drawn over 1, dead targets 0
[hos ModuleSystemWorker3] barrier: 47343 looks: 14575 found the masks moved already, 28471 waited until they moved, 4297 for the whole 1 ms (a wait 346 us on average); 21734 wakes
[hos host] vk: replay reads of guest memory as submitted: 1440000 (883008 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.06, writes collected 0.03
[hos host] vk: 1 textures retired (2485 left, 2993 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 51191 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5704 dirty pages from their submits' captures, 37943 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): 1056 dispatches, 32033 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5686 runs, 675.4 MB compared; 75 pages read into the shadows; 264 MB of shadows (36 MB of pages known, in 76 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 14317 in captures, 0 only in the collection, 35 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: late pages found at the replay (600 takes): 5 pages in 5 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: 1 textures retired (2485 left, 2993 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1460000 (895460 pages from the captures)
[hos host] vk: queues ordered by their events: 18000 submissions stopped at a wait, 17966 went on once it was reached, 32 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 3 (stopped 50.57 ms at most); waits met 43078, not met 0; captures at a CPU event's wait 2575, 0 of them at a wait met on arrival
[hos host] vk: replay reads of guest memory as submitted: 1480000 (907964 pages from the captures)
[hos host] vk: program picks: 2084030 kept (776469 by the last picks' textures, a uniform buffer bound since), 17689 new, 698280 with other textures (of 2800000)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 283 KB; 0 made on demand (0 MB kept now); c1 stages: at most 128 a recording; 26 descriptor pools made when a recording's own ran out (26 slots' pools made whole); a recording's pool took at most 505 sets (ub 5550, sb 2563, cis 42181, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 93, 1039 MB with memory of 1336 MB (296 MB spared); blocks given memory later 8507, given back 11606; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 192; 586 binds (338.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 108 taken from the spares, 201 made new; 216 kept as spares, 0 destroyed; 108 spares now, 1756 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5506 dirty pages from their submits' captures, 36785 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): 1028 dispatches, 31199 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5647 runs, 674.3 MB compared; 68 pages read into the shadows; 264 MB of shadows (36 MB of pages known, in 76 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 13548 in captures, 0 only in the collection, 32 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 167 live (1964 MB), 0 retired still imported (0 MB), 132 retired given back (2136 MB)
[hos SceneLowPriority] sync: object cache (HOS_OBJ_CACHE=16384 a thread, 44 made): 9448493 lookups, 76074 missed (0.81 %)
[hos host] vk: a frame: 403 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 10986 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 16.50, present 0.34, draw 7.47, descriptors 2.35, texture lookups 0.17, uploads 0.04, pipelines 0.29, programs 1.05, commands 3.27, dispatch 0.83, memory sync 0.50, queue submit 5.68, slot wait 0.08
[hos host] vk: a frame: mirrors 309 (1039 MB): 4.54 MB synced in 854.7 copies, 18.24 MB written back
[hos host] vk: 3240 presents, 31.7 fps; a frame: 2374 draws (0 skipped), 127 dispatches, 148 passes, 157 barriers, 6 submits, 0.2 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.1 ms (recording slots only); 2739 pipelines (479 ms), 1593 programs (12 failed), 2493 textures (2994 MB)
[hos host] vk: host time in GPU resource waits: 19.02 ms a frame (fences 0.04, timelines 18.98, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 522.0 full barriers recorded, 282.9 left out (nothing recorded since the one before), 15.2 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: 48.7; 0x12: 76.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 34.7 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 34.7 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 23.6 ms a frame (the longest submission 9.6 ms); 0.2 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.2 (0.1 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.02 MB unswizzled on the CPU in 0.02 ms
[hos host] vk: replay reads of guest memory as submitted: 1500000 (920404 pages from the captures)
[hos host] vk: 1 textures retired (2493 left, 2994 MB): budget 0, drawn over 1, dead targets 0
[hos Presentation Thread] vi: 3257 frames presented (31.7 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 0, 22-30 62, 30-40 93, over 40 4
[hos sead::MainThread] vk: capture: a submit 0.15 ms (lock 0.01, scan 0.04 over 968 ranges, 4572 MB), 260 pages written, 260 copied (1.0 MB)
[hos host] vk: a capture taken (ms): all 0.042; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.026, bytes copied 0.000; 25 kept (140 MB), 10 spare buffers, 8145 texture pages copied (31 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 159 sync waits (2027 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos ModuleSystemWorker4] nvn: CommandBufferSignalEvent #30000: event 0x100d6c8008 mode 0 value 29995 (lr 0x80660f34)
[hos host] vk: replay reads of guest memory as submitted: 1520000 (932871 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 777, 3811.3 MB; 29 bindings followed, 36 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 5540 dirty pages from their submits' captures, 37497 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): 1041 dispatches, 31660 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5703 runs, 674.9 MB compared; 70 pages read into the shadows; 264 MB of shadows (36 MB of pages known, in 76 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 14006 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1540000 (945312 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 605710 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: 1 textures retired (2493 left, 2994 MB): budget 0, drawn over 1, dead targets 0
[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: 0 program records loaded (0.0 ms), 2247 new graphics pipelines, pipelines from the driver's cache 2405, from the start's cache 0, compiled 136; 4827 draws and dispatches left out, 1456 waits (10000.1 ms)
[hos host] vk: program loads at create (close): 0 queued, 0 loaded (median 0.00 ms, 99th percentile 0.00 ms, worst 0.00 ms, 0 ms and 0.0 MB read in all); 0 programs (0 of them compute variants without their code, 0 graphics programs no note names), 0 graphics pipelines (0 not in a cache), 0 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 0 pipelines made at create taken by draws; 8919 create calls with nothing kept for their stages, 0 before the index; units 8919 (0 ready, 0 loading), 14616 programs alive; 0 noted graphics pipeline probes deferred to first use
[hos host] vk: program loads at create (close): 0 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): 0 known, 0 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: indexed descriptor arrays (close): 233 in 233 of 3216 program stages, 0 of them near a limit
[hos host] vk: indexed descriptor arrays (close), fragment stages: 233 texture (17-64: 233); the longest 64
[hos host] vk: page table census (close): no pool outside its windows, 64-72 GB and 508-512 GB (4262 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)