Beanstalk Longshot test farm
Menu

Report rp_2eacdc0f799b5a3f fail

#1702 perf(vk): the presenter thread on by default with NVIDIA on Linux and AMD on Windows, with NVIDIA Reflex; after #1121

windows+nvidia on Flashman4/agent-de36c5,

Machine

Commit
28f5fb2b1e774dbeb53968b7c7b2345695086176
Baseline
d5794c82dfa44fd35c041f2dabc919d0ce85a591
App
nightly-177-99542c21, attempt 1
Game build
e0f3ccfa
os
windows
cpu
AMD Ryzen 9 5900X 12-Core Processor
gpu
NVIDIA GeForce RTX 3090
tags
windowsnvidianvidia-proprietarydgpu
driver
621.24.0
ram_gb
128
vram_mb
24,322
gpu_vendor
nvidia
os_version
Windows 10.0.26200

Builds

main (baseline)

1,144.6 s warm

from d5794c82dfa4

lift
100.6 s
read
581.4 s
finish
155 s
compile
290.9 s

PR

57.4 s warm

Program cache: warm

Scenarios

forest-and-market

error

the game refused a step: refused: the game refused the teleport: an event is playing (a cutscene, a map change's)

Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count9.514.5+52.6%
log_forbidden_hits00±0%
log_expected_missing11±0%
run_frametime_max_ms270.8282.01+4.1%
forest-standing.fps_mean73.5479.87+8.6%
market-standing.fps_mean62.6763.1+0.7%
forest-standing.draws_mean866.5871.95+0.6%
market-standing.draws_mean2,555.32,518.5−1.4%
forest-standing.hitch_count00±0%
market-standing.hitch_count00±0%
forest-after-moving.fps_mean77.4280.53+4%
market-after-moving.fps_mean60.4460.56+0.2%
forest-standing.stutter_count24.5+125%
market-standing.stutter_count10−100%
forest-after-moving.draws_mean801.45804.55+0.4%
market-after-moving.draws_mean2,800.12,806.9+0.2%
forest-after-moving.hitch_count00±0%
market-after-moving.hitch_count00±0%
forest-standing.frametime_max_ms31.8729.57−7.2%
forest-standing.frametime_p50_ms13.0811.98−8.4%
forest-standing.frametime_p99_ms20.8619.42−6.9%
market-standing.frametime_max_ms46.4824.38−47.6%
market-standing.frametime_p50_ms15.6915.58−0.7%
market-standing.frametime_p99_ms20.4821.14+3.3%
forest-after-moving.stutter_count22.5+25%
market-after-moving.stutter_count00±0%
forest-after-moving.frametime_max_ms28.3426.72−5.7%
forest-after-moving.frametime_p50_ms12.8212.35−3.7%
forest-after-moving.frametime_p99_ms18.8217.3−8%
market-after-moving.frametime_max_ms25.4823.97−5.9%
market-after-moving.frametime_p50_ms16.0916.19+0.6%
market-after-moving.frametime_p99_ms22.7922.78−0%
forest-standing.frametime_1pct_low_ms24.3822.42−8%
market-standing.frametime_1pct_low_ms24.7122.17−10.3%
forest-after-moving.frametime_1pct_low_ms22.2921.04−5.6%
market-after-moving.frametime_1pct_low_ms23.8523.6−1.1%
screenshot_diff—0.05—

Pictures against main: forest diff 0.0637 (main against itself 0.6307), market diff 0.0392 (main against itself )

Runs (4)
Metricmain 0pr 1pr 2main 3
hangnononono
crashednononono
exit_code00——
run_hitch_count172542
log_forbidden_hits0000
log_expected_missing1111
run_frametime_max_ms471.64459.29104.7369.96
forest-standing.fps_mean63.0370.9588.7984.05
market-standing.fps_mean62.6763.1——
forest-standing.draws_mean827.5820.4923.5905.5
market-standing.draws_mean2,555.32,518.5——
forest-standing.hitch_count0000
market-standing.hitch_count00——
forest-after-moving.fps_mean72.9774.8486.2281.88
market-after-moving.fps_mean60.4460.56——
forest-standing.stutter_count4900
market-standing.stutter_count10——
forest-after-moving.draws_mean910.4927.6681.5692.5
market-after-moving.draws_mean2,800.12,806.9——
forest-after-moving.hitch_count0000
market-after-moving.hitch_count00——
forest-standing.frametime_max_ms45.3142.1217.0218.42
forest-standing.frametime_p50_ms14.5313.0310.9311.64
forest-standing.frametime_p99_ms27.0224.7714.0614.71
market-standing.frametime_max_ms46.4824.38——
market-standing.frametime_p50_ms15.6915.58——
market-standing.frametime_p99_ms20.4821.14——
forest-after-moving.stutter_count0054
market-after-moving.stutter_count00——
forest-after-moving.frametime_max_ms19.9218.6434.836.77
forest-after-moving.frametime_p50_ms13.6113.2711.4312.03
forest-after-moving.frametime_p99_ms18.3517.0417.5719.29
market-after-moving.frametime_max_ms25.4823.97——
market-after-moving.frametime_p50_ms16.0916.19——
market-after-moving.frametime_p99_ms22.7922.78——
forest-standing.frametime_1pct_low_ms33.129.7215.1215.66
market-standing.frametime_1pct_low_ms24.7122.17——
forest-after-moving.frametime_1pct_low_ms18.917.6924.3925.67
market-after-moving.frametime_1pct_low_ms23.8523.6——
screenshot_diff—0.040.06—

Rules

ScenarioCellRuleLimitmainPRΔ %Status
forest-and-marketcrashFalse00—pass
forest-and-marketforest-standing.fps_mean.min_delta_pct073.5479.87+8.6%pass
forest-and-marketforest-after-moving.fps_mean.min_delta_pct077.4280.53+4%pass
forest-and-marketmarket-standing.fps_mean.min_delta_pct062.6763.1+0.7%pass
forest-and-marketmarket-after-moving.fps_mean.min_delta_pct060.4460.56+0.2%pass
forest-and-marketscreenshot.max_diff0.20.630.06—pass
forest-and-marketlog_forbidden_hits.max000—pass
forest-and-marketlog_expected_missing.max011—fail

Log (PR side, end)

[hos host] vk: write-back diff (600 submits): 6058 runs, 1438.2 MB compared; 165 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 17303 in captures, 6 only in the collection, 416 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1300000 (572531 pages from the captures)
[hos host] vk: 1 textures retired (1639 left, 4378 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: 4 textures retired (1636 left, 4363 MB): budget 0, drawn over 4, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos host] vk: 4 textures retired (1636 left, 4362 MB): budget 0, drawn over 4, dead targets 0
[hos sead::MainThread] nvn: GPU waits in 5 s: 43 sync waits (46 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos Presentation Thread] vk: capture: a submit 0.12 ms (lock 0.00, scan 0.06 over 1060 ranges, 1971 MB), 175 pages written, 175 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.005; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (108 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: readback pages refreshed from guest memory: 729103 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: render targets drawn over from another start, let go: 739 so far (3814 MB), 718 of them (3679 MB) between the passes
[hos host] vk: 3 textures retired (1636 left, 4361 MB): budget 0, drawn over 3, dead targets 0
[hos ModuleSystemWorker3] vk: readbacks of GPU results: 1318 of 3000 waited for their write-back, 0.23 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 ModuleSystemWorker3] vk: invalidates since the last readbacks line: 2748 kept their virtual core (nothing there to wait for), 2884 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: 5 textures retired (1634 left, 4350 MB): budget 0, drawn over 5, dead targets 0
[hos host] vk: texture #1501 0x1002331fc00 3840x1080 x1 NVN 0x3f the GPU drew, finalized by the game, sampled again at frame 9533 (drawn at 9027)
[hos host] vk: NVN2 program e61be02173ea53e7 (stage 0) lowered with extended registers (1144 REGMAP words)
[hos host] vk: texture #2274 0x10004e48e00 3840x1080 x1 NVN 0x25 the GPU drew, finalized by the game, sampled again at frame 9534 (drawn at 9028)
[hos host] control: teleport to destination 3 asked for
[hos host] vk: upload over a drawn image: #288 0x10005dc0e00 3840x1080x1 x1 NVN 0x25 at present 9534 (texture frame 9535), finalized, 2028 KB dirty in 1 ranges
[hos host] vk: texture 0x10005dc0e00 3840x1080x1 x1 levels 1 NVN 0x25 (17280 KB of storage) re-uploaded (its memory was written; flags 0x8): whole, 17280 KB of it, over what the GPU wrote (finalized since)
[hos BaseProcExecutor3] cheats: teleport waits until the current event ends
[hos host] vk: NVN2 program d4d5fb09c1cd2097 (stage 0) lowered with extended registers (1132 REGMAP words)
[hos host] vk: replay reads of guest memory as submitted: 1320000 (581708 pages from the captures)
[hos host] vk: 8 textures retired (1646 left, 4515 MB): budget 0, drawn over 8, dead targets 0
[hos host] vk: sparse mirrors: 61, 814 MB with memory of 1104 MB (289 MB spared); blocks given memory later 1656, given back 3227; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 290 binds (39.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 83 made new; 39 kept as spares, 0 destroyed; 22 spares now, 208 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1471, 5604.5 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 518 dirty pages from their submits' captures, 64274 from earlier ones', 3 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): 1485 dispatches, 52196 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6183 runs, 1421.6 MB compared; 160 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 17485 in captures, 8 only in the collection, 391 only by the replay itself
[hos host] vk:   read live, by pool: 0x1060000000 (4 MB, captured, state 2): 3;
[hos host] vk: imports: 87 live (1316 MB), 0 retired still imported (0 MB), 12 retired given back (168 MB)
[hos host] vk: 8 textures retired (1638 left, 4477 MB): budget 0, drawn over 8, dead targets 0
[hos host] vk: 6 textures retired (1632 left, 4448 MB): budget 0, drawn over 6, dead targets 0
[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: 3 textures retired (1629 left, 4434 MB): budget 0, drawn over 3, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1340000 (593511 pages from the captures)
[hos host] vk: 2 textures retired (1627 left, 4425 MB): budget 0, drawn over 2, dead targets 0
[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: 1 textures retired (1626 left, 4420 MB): budget 0, drawn over 1, dead targets 0
[hos sead::MainThread] nvn: CommandBufferWaitEvent #95000: event 0x10103c800c mode 0 value 58174 (now 58155) (lr 0x80660e38)
[hos host] vk: memory: working set 21105 MB (peak 21109), private 29747 MB, staging 64 MB in 4 arenas, 25 made so far (4 ms)
[hos host] vk: memory heaps: heap 0 (device): 195 blocks 6264 MB, 14920 allocations 5846 MB; used 7346 of 23554 MB; heap 1 (system): 16 blocks 672 MB, 39 allocations 578 MB; used 2097 of 104029 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 5 images 16 MB; 1 free ranges, the largest 15 MB
[hos host] vk: image blocks (render targets): 25 blocks 800 MB, 110 images 494 MB; 59 free ranges, the largest 32 MB
[hos host] vk: image blocks (textures, compacted): 78 blocks 2496 MB, 1501 images 2469 MB; 623 free ranges, the largest 19 MB
[hos host] vk: render target images: 346 reused from the pool, 802 made new, 687 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 7346 MB used of 23554 MB allowed (417 MB of it free in the blocks); textures 4420 MB (budget 11000 MB), 1626 textures
[hos host] vk: budget: VRAM 6264 of 23554 MB (7346 MB by the driver, 1081 MB unaccounted): textures 4420, mirrors 814, blocks free 417, shadows 360, other buffers 155, staging 64, other images 31; RAM 1924 MB (the process's private memory 29747 MB): imports 1316, capture spares 470, captures 94, capture table 30, other buffers 14, others 0
[hos host] vk: textures: decoded ASTC 2428 MB, rendered 1237 MB (49 drawn in one frame, kept: 43035 KB), other 464 MB; used in the last 8 presents 2271 MB
[hos host] vk: decoded ASTC used again: 0 textures (0 MB) copied from ones the game finalized, not decoded; 0 of those retired, 9 listed; 503 MB of blocks hashed in 51 ms
[hos host] vk:   288 MB: 0x1150c06000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10aea22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10ab1aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x1145b99000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x114045c000 4096x4096x1 x1 levels 1 NVN 0xa, last used 3185 presents ago
[hos host] vk:   64 MB: 0x114045c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10004571a00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 855 presents ago
[hos host] vk:   54 MB: 0x10005b2da00 3072x3072x1 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: 39 (9570 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 50x57x1 NVN 0x25: 1 (11 KB); 62x41x1 NVN 0x25: 1 (9 KB); 74x32x1 NVN 0x25: 1 (9 KB); 90x34x1 NVN 0x25: 1 (11 KB); 101x34x1 NVN 0x25: 1 (13 KB); 101x48x1 NVN 0x25: 1 (18 KB); 114x121x1 NVN 0x1: 1 (13 KB); 126x105x1 NVN 0x1: 1 (12 KB); 138x96x1 NVN 0x1: 1 (12 KB); 154x98x1 NVN 0x1: 1 (14 KB); 165x98x1 NVN 0x1: 1 (15 KB); 165x112x1 NVN 0x1: 1 (18 KB); 198x40x1 NVN 0x25: 2 (61 KB); 242x40x1 NVN 0x25: 3 (113 KB); 244x44x1 NVN 0x25: 1 (41 KB);
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos Presentation Thread] vk: presents after the CPU's signal (600): 85 waited for it, 1.31 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 28.8 ms; VRAM 7.2 of 23.0 GB
[hos host] vk: a frame: 32 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.35, present 0.35, draw 3.03, descriptors 1.28, texture lookups 0.19, uploads 0.04, pipelines 0.11, programs 0.59, commands 0.83, dispatch 0.62, memory sync 0.37, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 100 (814 MB): 4.06 MB synced in 828.8 copies, 22.32 MB written back
[hos host] vk: 9600 presents, 83.8 fps; a frame: 638 draws (0 skipped), 86 dispatches, 138 passes, 140 barriers, 6 submits, 0.2 uploads (0.4 MB), 89 ub updates (61 overlays), GPU-bound wait 0.0 ms (recording slots only); 37481 pipelines (7184477 ms), 0 programs (0 failed), 1626 textures (4420 MB)
[hos host] vk: host time in GPU resource waits: 2.14 ms a frame (fences 0.02, timelines 2.12, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 329.4 full barriers recorded, 180.0 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 38.6; 0x12: 71.9; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.9 ms a frame (the longest submission 3.8 ms); 0.5 textures made a frame
[hos host] vk: a frame: 0.06 MB unswizzled on the CPU in 0.01 ms
[hos host] vk: the render thread (600 presents, 11.52 ms a frame): its jobs 6.13 ms a frame (submits 4.93, presents 0.37, others 0.82), idle 5.40; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 3.31 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); 88 captures at a CPU signal; the submits' waits read in 0.022 ms a frame; readbacks: 2951 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 5.31 ms after its hand-over (14.70 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 3.39 + 0.31 (the window's copy is in the idle), after it 0.79 + 0.24; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.03 ms of GPU time), after its last that reads it 1.00 (0.03 ms)
[hos host] vk: frame ahead (600 presents): 3.65 parts a frame recorded in the signal's job, 53.6% of them took nothing that came with the signal; of the others, 1.69 a frame took pages of the signal's capture into a mirror (40.1 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 13 of 88 frames, its first three in 13
[hos host] vk: frame ahead, what came with the signal: 88 signals with a capture, each of 58.3 pages in 8.7 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: 0.66+0.15, 0.08+0.04, 0.02+0.02, 0.01+0.01, 0.01+0.00, 0.00+0.00, 0.00+0.01, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00
[hos host] vk: write-backs left out (600 submits): 1380, 5256.1 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 644 dirty pages from their submits' captures, 62165 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): 1393 dispatches, 50726 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6706 runs, 1346.5 MB compared; 155 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16830 in captures, 8 only in the collection, 420 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: 1 textures retired (1625 left, 4415 MB): budget 0, drawn over 1, dead targets 0
[hos Presentation Thread] vk: capture: a submit 0.12 ms (lock 0.00, scan 0.06 over 1063 ranges, 2013 MB), 180 pages written, 180 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (114 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 1360000 (605078 pages from the captures)
[hos ModuleSystemWorker1] nvn: DrawMeshTasksIndirect: 120000 so far (x1 0x11820f0000 x2 0 x3 0x11820f0000 x4 0x654040)
[hos host] vk: 2 textures retired (1623 left, 4406 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: 4 textures retired (1619 left, 4387 MB): budget 0, drawn over 4, dead targets 0
[hos sead::MainThread] nvn: 550000 recordings made, 48717 live copies (12.1 MB), 788 texture copies; 1 command buffers' segments (4.0 MB)
[hos host] vk: 2 textures retired (1617 left, 4378 MB): budget 0, drawn over 2, dead targets 0
[hos Presentation Thread] vi: 9640 frames presented (86.1 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 425, 15.5-17.8 2, 17.8-22 4, 22-30 0, 30-40 0, over 40 0
[hos host] vk: 4 textures retired (1613 left, 4360 MB): budget 0, drawn over 4, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1380000 (615868 pages from the captures)
[hos host] vk: 3 textures retired (1610 left, 4337 MB): budget 0, drawn over 3, dead targets 0
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 301 KB; 0 made on demand (0 MB kept now); c1 stages: at most 112 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 61, 814 MB with memory of 1104 MB (289 MB spared); blocks given memory later 1656, given back 3227; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 290 binds (39.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 83 made new; 39 kept as spares, 0 destroyed; 22 spares now, 208 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1357, 5168.5 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 402 dirty pages from their submits' captures, 61727 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): 1359 dispatches, 50081 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6909 runs, 1331.8 MB compared; 133 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16644 in captures, 6 only in the collection, 411 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 87 live (1316 MB), 0 retired still imported (0 MB), 12 retired given back (168 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 ModuleSystemWorker3] vk: readbacks of GPU results: 1290 of 3000 waited for their write-back, 0.25 ms each; 0 looked at the stopped queues first, 1 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker3] vk: invalidates since the last readbacks line: 2633 kept their virtual core (nothing there to wait for), 3255 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #60000: event 0x10103c8014 mode 0 value 60000 (lr 0x802915c8)
[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: 1400000 (625614 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos sead::MainThread] vk: capture: a submit 0.13 ms (lock 0.00, scan 0.06 over 1063 ranges, 2013 MB), 184 pages written, 184 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (102 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: program picks: 1738474 kept (740006 by the last picks' textures, a uniform buffer bound since), 1133 new, 860392 with other textures (of 2600000)
[hos host] vk: a frame: 35 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.39, present 0.24, draw 2.94, descriptors 1.19, texture lookups 0.07, uploads 0.00, pipelines 0.12, programs 0.57, commands 0.84, dispatch 0.65, memory sync 0.37, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 100 (814 MB): 4.11 MB synced in 849.1 copies, 22.65 MB written back
[hos host] vk: 9720 presents, 83.8 fps; a frame: 704 draws (0 skipped), 101 dispatches, 141 passes, 142 barriers, 6 submits, 0.0 uploads (0.0 MB), 94 ub updates (64 overlays), GPU-bound wait 0.0 ms (recording slots only); 37481 pipelines (7184477 ms), 0 programs (0 failed), 1610 textures (4337 MB)
[hos host] vk: host time in GPU resource waits: 2.26 ms a frame (fences 0.02, timelines 2.24, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 350.8 full barriers recorded, 183.0 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 38.6; 0x12: 73.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.5 ms a frame (the longest submission 3.6 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: write-backs left out (600 submits): 1356, 5166.5 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 437 dirty pages from their submits' captures, 61674 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): 1356 dispatches, 50102 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 7210 runs, 1319.3 MB compared; 132 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16748 in captures, 10 only in the collection, 392 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1420000 (635404 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: sparse mirrors: 61, 814 MB with memory of 1104 MB (289 MB spared); blocks given memory later 1656, given back 3227; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 290 binds (39.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 83 made new; 39 kept as spares, 0 destroyed; 22 spares now, 208 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1312, 4995.3 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 427 dirty pages from their submits' captures, 59571 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1314 dispatches, 48298 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 7191 runs, 1315.5 MB compared; 137 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16016 in captures, 6 only in the collection, 409 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 87 live (1316 MB), 0 retired still imported (0 MB), 12 retired given back (168 MB)
[hos host] vk: replay reads of guest memory as submitted: 1440000 (645706 pages from the captures)
[hos host] control: teleport to destination 3 asked for
[hos ModuleSystemWorker3] nvn: CommandBufferWaitEvent #100000: event 0x10103c800c mode 0 value 61974 (now 61955) (lr 0x80660e38)
[hos BaseProcExecutor3] cheats: teleport waits until the current event ends
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos Presentation Thread] vk: capture: a submit 0.12 ms (lock 0.00, scan 0.06 over 1063 ranges, 2013 MB), 182 pages written, 182 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (114 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: readback pages refreshed from guest memory: 803281 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 1460000 (656184 pages from the captures)
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 1345 of 3000 waited for their write-back, 0.24 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: 2505 kept their virtual core (nothing there to wait for), 2871 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: write-backs left out (600 submits): 1311, 4993.3 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 450 dirty pages from their submits' captures, 59609 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): 1311 dispatches, 48315 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 7114 runs, 1296.5 MB compared; 136 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16136 in captures, 8 only in the collection, 389 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a frame: 40 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.47, present 0.19, draw 2.99, descriptors 1.21, texture lookups 0.06, uploads 0.00, pipelines 0.12, programs 0.57, commands 0.85, dispatch 0.65, memory sync 0.37, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 100 (814 MB): 4.11 MB synced in 847.0 copies, 22.72 MB written back
[hos host] vk: 9840 presents, 85.7 fps; a frame: 733 draws (0 skipped), 104 dispatches, 141 passes, 142 barriers, 6 submits, 0.0 uploads (0.0 MB), 94 ub updates (64 overlays), GPU-bound wait 0.0 ms (recording slots only); 37481 pipelines (7184477 ms), 0 programs (0 failed), 1610 textures (4337 MB)
[hos host] vk: host time in GPU resource waits: 2.15 ms a frame (fences 0.02, timelines 2.13, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 355.8 full barriers recorded, 183.2 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 38.4; 0x12: 73.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.6 ms a frame (the longest submission 3.6 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos ModuleSystemWorker3] vcores: 198713 takes, 24489 waits (mean 33 us, longest 6.3 ms, 0 over HOS_VCORE_LOG_MS), 12080 kicks, 11998 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: replay reads of guest memory as submitted: 1480000 (666650 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: upload over a drawn image: #288 0x10005dc0e00 3840x1080x1 x1 NVN 0x25 at present 9891 (texture frame 9892), finalized, 2028 KB dirty in 1 ranges
[hos host] vk: texture 0x10005dc0e00 3840x1080x1 x1 levels 1 NVN 0x25 (17280 KB of storage) re-uploaded (its memory was written; flags 0x8): whole, 17280 KB of it, over what the GPU wrote (finalized since)
[hos host] vk: sparse mirrors: 61, 814 MB with memory of 1104 MB (289 MB spared); blocks given memory later 1656, given back 3227; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 290 binds (39.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 83 made new; 39 kept as spares, 0 destroyed; 22 spares now, 208 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1310, 4991.3 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 391 dirty pages from their submits' captures, 59585 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1314 dispatches, 48208 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 7028 runs, 1297.1 MB compared; 140 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 15952 in captures, 9 only in the collection, 396 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 87 live (1316 MB), 0 retired still imported (0 MB), 12 retired given back (168 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos host] vk: late pages found at the replay (600 takes): 1 pages in 1 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: replay reads of guest memory as submitted: 1500000 (678513 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.13 ms (lock 0.00, scan 0.06 over 1063 ranges, 2017 MB), 181 pages written, 181 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (122 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 39 sync waits (38 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: replay reads of guest memory as submitted: 1520000 (690837 pages from the captures)
[hos host] vk: queues ordered by their events: 48000 submissions stopped at a wait, 47998 went on once it was reached, 0 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 0 (stopped 0.00 ms at most); waits met 103839, not met 0; captures at a CPU event's wait 5962, 0 of them at a wait met on arrival
[hos ModuleSystemWorker2] nvn: CommandBufferSignalEvent #65000: event 0x10103c8020 mode 0 value 65000 (lr 0x80664f14)
[hos host] vk: texture #2638 0x1000b01e000 1920x1920 x1 NVN 0x34 the GPU drew, finalized by the game, sampled again at frame 9951 (drawn at 9890)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 301 KB; 0 made on demand (0 MB kept now); c1 stages: at most 112 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: write-backs left out (600 submits): 1426, 5431.3 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 460 dirty pages from their submits' captures, 62335 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): 1368 dispatches, 51748 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6681 runs, 1385.1 MB compared; 166 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16979 in captures, 8 only in the collection, 441 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a frame: 33 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.21, present 0.23, draw 2.86, descriptors 1.18, texture lookups 0.08, uploads 0.01, pipelines 0.11, programs 0.55, commands 0.81, dispatch 0.63, memory sync 0.36, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 100 (814 MB): 4.03 MB synced in 839.7 copies, 22.53 MB written back
[hos host] vk: 9960 presents, 85.8 fps; a frame: 672 draws (0 skipped), 95 dispatches, 139 passes, 141 barriers, 6 submits, 0.1 uploads (0.2 MB), 94 ub updates (64 overlays), GPU-bound wait 0.0 ms (recording slots only); 37481 pipelines (7184477 ms), 0 programs (0 failed), 1619 textures (4380 MB)
[hos host] vk: host time in GPU resource waits: 2.02 ms a frame (fences 0.02, timelines 2.00, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 344.4 full barriers recorded, 181.6 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 38.4; 0x12: 73.0; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 4.3 ms); 0.2 textures made a frame
[hos host] vk: a frame: 0.01 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: replay reads of guest memory as submitted: 1540000 (702778 pages from the captures)
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 1299 of 3000 waited for their write-back, 0.23 ms each; 0 looked at the stopped queues first, 1 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker4] vk: invalidates since the last readbacks line: 2474 kept their virtual core (nothing there to wait for), 2902 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[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 ModuleSystemWorker1] nvn: CommandBufferWaitEvent #105000: event 0x10103c8010 mode 0 value 65775 (now 65756) (lr 0x80660fdc)
[hos host] vk: program picks: 1872975 kept (795981 by the last picks' textures, a uniform buffer bound since), 1136 new, 925888 with other textures (of 2800000)
[hos sead::MainThread] nvn: QueueSubmitCommands #60000: queue 00001010103BF020 (its #10000), 1 handles (lr 0x80666be0)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos sead::MainThread] nvn: EventSignal (CPU) #20000: event 0x1009b38004 mode 0 value 1 (lr 0x8348c78c)
[hos sead::MainThread] vk: capture: a submit 0.12 ms (lock 0.00, scan 0.05 over 1063 ranges, 2017 MB), 176 pages written, 176 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (122 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 1560000 (714637 pages from the captures)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos ModuleSystemWorker5] barrier: 120474 looks: 22170 found the masks moved already, 85548 waited until they moved, 12756 for the whole 1 ms (a wait 353 us on average); 103762 wakes
[hos host] vk: sparse mirrors: 61, 814 MB with memory of 1104 MB (289 MB spared); blocks given memory later 1656, given back 3227; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 290 binds (39.0 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 83 made new; 39 kept as spares, 0 destroyed; 22 spares now, 208 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1403, 5343.7 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 598 dirty pages from their submits' captures, 61314 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): 1346 dispatches, 50951 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6551 runs, 1361.4 MB compared; 162 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16608 in captures, 5 only in the collection, 432 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 87 live (1316 MB), 0 retired still imported (0 MB), 12 retired given back (168 MB)
[hos host] vk: 2 textures retired (1617 left, 4362 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1580000 (726500 pages from the captures)
[hos host] control: teleport to destination 3 asked for
[hos BaseProcExecutor4] cheats: teleport waits until the current event ends
[hos host] vk: texture slots registered again at another size: 200000 (of 279017 registrations; slot 27082 128x72 -> 1920x540)
[hos Presentation Thread] vi: 10074 frames presented (86.7 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 430, 15.5-17.8 3, 17.8-22 1, 22-30 0, 30-40 0, over 40 0
[hos host] vk: write-backs left out (600 submits): 1403, 5343.7 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 628 dirty pages from their submits' captures, 61248 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): 1342 dispatches, 50958 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6549 runs, 1361.4 MB compared; 168 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16554 in captures, 6 only in the collection, 426 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a frame: 29 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.79, present 0.20, draw 2.61, descriptors 1.09, texture lookups 0.06, uploads 0.00, pipelines 0.10, programs 0.51, commands 0.71, dispatch 0.59, memory sync 0.36, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 100 (814 MB): 3.96 MB synced in 835.4 copies, 22.32 MB written back
[hos host] vk: 10080 presents, 90.8 fps; a frame: 616 draws (0 skipped), 89 dispatches, 137 passes, 140 barriers, 6 submits, 0.0 uploads (0.0 MB), 94 ub updates (64 overlays), GPU-bound wait 0.0 ms (recording slots only); 37481 pipelines (7184477 ms), 0 programs (0 failed), 1617 textures (4362 MB)
[hos host] vk: host time in GPU resource waits: 1.73 ms a frame (fences 0.02, timelines 1.71, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 335.0 full barriers recorded, 180.1 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 37.9; 0x12: 72.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.9 ms a frame (the longest submission 3.7 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos ModuleSystemWorker2] nvn: DrawMeshTasksIndirect: 140000 so far (x1 0x1185e20000 x2 0x2 x3 0x1185e20000 x4 0x9b040)
[hos host] vk: replay reads of guest memory as submitted: 1600000 (738943 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos Presentation Thread] vk: capture: a submit 0.12 ms (lock 0.00, scan 0.05 over 1063 ranges, 2017 MB), 173 pages written, 173 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (106 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: readback pages refreshed from guest memory: 877543 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 1620000 (751459 pages from the captures)
[hos host] vk: render targets drawn over from another start, let go: 785 so far (3985 MB), 764 of them (3850 MB) between the passes
[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: mirror of pool 0x7f88800000 (4 MB) given back: idle, 6032 submissions since its sync
[hos host] vk: mirror of pool 0x7fd3800000 (4 MB) given back: idle, 6032 submissions since its sync
[hos sead::MainThread] vk: readbacks of GPU results: 1297 of 3000 waited for their write-back, 0.19 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: 2617 kept their virtual core (nothing there to wait for), 2759 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: sparse mirrors: 59, 812 MB with memory of 1096 MB (283 MB spared); blocks given memory later 1656, given back 3227; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 291 binds (39.2 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 83 made new; 41 kept as spares, 0 destroyed; 24 spares now, 216 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1403, 5343.7 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 572 dirty pages from their submits' captures, 61299 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): 1344 dispatches, 50960 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6560 runs, 1361.4 MB compared; 170 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16590 in captures, 10 only in the collection, 442 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 87 live (1316 MB), 0 retired still imported (0 MB), 12 retired given back (168 MB)
[hos ModuleSystemWorker2] nvn: 600000 recordings made, 48690 live copies (12.1 MB), 798 texture copies; 5 command buffers' segments (4.3 MB)
[hos host] vk: replay reads of guest memory as submitted: 1640000 (763950 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 1660000 (776810 pages from the captures)
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #110000: event 0x10103c8010 mode 0 value 69575 (now 69556) (lr 0x80660fdc)
[hos Presentation Thread] vk: capture: a submit 0.12 ms (lock 0.00, scan 0.05 over 1063 ranges, 2017 MB), 172 pages written, 172 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 24 kept (124 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: memory: working set 21103 MB (peak 21109), private 29427 MB, staging 48 MB in 3 arenas, 25 made so far (4 ms)
[hos host] vk: memory heaps: heap 0 (device): 193 blocks 6200 MB, 14875 allocations 5770 MB; used 7282 of 23554 MB; heap 1 (system): 16 blocks 672 MB, 39 allocations 590 MB; used 2097 of 104029 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 5 images 16 MB; 1 free ranges, the largest 15 MB
[hos host] vk: image blocks (render targets): 23 blocks 736 MB, 96 images 422 MB; 51 free ranges, the largest 32 MB
[hos host] vk: image blocks (textures, compacted): 78 blocks 2496 MB, 1506 images 2483 MB; 628 free ranges, the largest 7 MB
[hos host] vk: render target images: 357 reused from the pool, 813 made new, 712 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 7282 MB used of 23554 MB allowed (430 MB of it free in the blocks); textures 4362 MB (budget 11000 MB), 1617 textures
[hos host] vk: budget: VRAM 6200 of 23554 MB (7282 MB by the driver, 1081 MB unaccounted): textures 4362, mirrors 812, blocks free 430, shadows 360, other buffers 155, staging 48, other images 31; RAM 1940 MB (the process's private memory 29427 MB): imports 1316, capture spares 478, captures 116, capture table 30, others 0
[hos host] vk: textures: decoded ASTC 2443 MB, rendered 1169 MB (49 drawn in one frame, kept: 43035 KB), other 465 MB; used in the last 8 presents 2207 MB
[hos host] vk: decoded ASTC used again: 0 textures (0 MB) copied from ones the game finalized, not decoded; 0 of those retired, 9 listed; 506 MB of blocks hashed in 52 ms
[hos host] vk:   288 MB: 0x1150c06000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10aea22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10ab1aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x114045c000 4096x4096x1 x1 levels 1 NVN 0xa, last used 3785 presents ago
[hos host] vk:   64 MB: 0x114045c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x1145b99000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10004571a00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1455 presents ago
[hos host] vk:   54 MB: 0x10005b2da00 3072x3072x1 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: 39 (9570 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 50x57x1 NVN 0x25: 1 (11 KB); 62x41x1 NVN 0x25: 1 (9 KB); 74x32x1 NVN 0x25: 1 (9 KB); 90x34x1 NVN 0x25: 1 (11 KB); 101x34x1 NVN 0x25: 1 (13 KB); 101x48x1 NVN 0x25: 1 (18 KB); 114x121x1 NVN 0x1: 1 (13 KB); 126x105x1 NVN 0x1: 1 (12 KB); 138x96x1 NVN 0x1: 1 (12 KB); 154x98x1 NVN 0x1: 1 (14 KB); 165x98x1 NVN 0x1: 1 (15 KB); 165x112x1 NVN 0x1: 1 (18 KB); 198x40x1 NVN 0x25: 2 (61 KB); 242x40x1 NVN 0x25: 3 (113 KB); 244x44x1 NVN 0x25: 1 (41 KB);
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos Presentation Thread] vk: presents after the CPU's signal (600): 83 waited for it, 0.31 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: write-backs left out (600 submits): 1402, 5341.7 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 633 dirty pages from their submits' captures, 61323 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): 1344 dispatches, 50951 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6563 runs, 1361.4 MB compared; 186 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16450 in captures, 2 only in the collection, 427 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: status: 600 frames, the slowest 17.5 ms; VRAM 7.1 of 23.0 GB
[hos host] vk: a frame: 29 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.85, present 0.21, draw 2.56, descriptors 1.04, texture lookups 0.06, uploads 0.00, pipelines 0.09, programs 0.49, commands 0.72, dispatch 0.60, memory sync 0.37, queue submit 0.35, slot wait 0.04
[hos host] vk: a frame: mirrors 100 (812 MB): 3.97 MB synced in 835.5 copies, 22.32 MB written back
[hos host] vk: 10200 presents, 89.6 fps; a frame: 620 draws (0 skipped), 89 dispatches, 137 passes, 141 barriers, 6 submits, 0.0 uploads (0.0 MB), 94 ub updates (64 overlays), GPU-bound wait 0.0 ms (recording slots only); 37481 pipelines (7184477 ms), 0 programs (0 failed), 1617 textures (4362 MB)
[hos host] vk: host time in GPU resource waits: 1.63 ms a frame (fences 0.02, timelines 1.61, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 335.4 full barriers recorded, 180.2 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 38.0; 0x12: 72.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.9 ms a frame (the longest submission 3.8 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: the render thread (600 presents, 11.49 ms a frame): its jobs 5.89 ms a frame (submits 4.87, presents 0.22, others 0.80), idle 5.59; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 3.53 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); 90 captures at a CPU signal; the submits' waits read in 0.024 ms a frame; readbacks: 2882 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 5.07 ms after its hand-over (11.79 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 3.17 + 0.28 (the window's copy is in the idle), after it 0.59 + 0.25; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.03 ms of GPU time), after its last that reads it 1.00 (0.03 ms)
[hos host] vk: frame ahead (600 presents): 4.02 parts a frame recorded in the signal's job, 66.0% of them took nothing that came with the signal; of the others, 1.37 a frame took pages of the signal's capture into a mirror (24.6 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 23 of 90 frames, its first three in 22
[hos host] vk: frame ahead, what came with the signal: 90 signals with a capture, each of 39.6 pages in 6.3 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: 0.53+0.18, 0.02+0.03, 0.02+0.02, 0.01+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, 0.00+0.00, 0.00+0.00
[hos ModuleSystemWorker3] nvn: CommandBufferSignalEvent #70000: event 0x10103c8028 mode 0 value 69999 (lr 0x80667410)
[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: replay reads of guest memory as submitted: 1680000 (789324 pages from the captures)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: replay reads of guest memory as submitted: 1700000 (801890 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 301 KB; 0 made on demand (0 MB kept now); c1 stages: at most 112 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 61, 824 MB with memory of 1116 MB (291 MB spared); blocks given memory later 1656, given back 3227; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 293 binds (39.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 19 taken from the spares, 83 made new; 41 kept as spares, 0 destroyed; 22 spares now, 196 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1404, 5345.7 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 487 dirty pages from their submits' captures, 61484 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): 1363 dispatches, 51043 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6615 runs, 1364.9 MB compared; 163 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16683 in captures, 10 only in the collection, 426 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 87 live (1316 MB), 0 retired still imported (0 MB), 12 retired given back (168 MB)
[hos host] vk: replay reads of guest memory as submitted: 1720000 (813961 pages from the captures)
[hos ModuleSystemWorker3] vk: readbacks of GPU results: 1288 of 3000 waited for their write-back, 0.19 ms each; 0 looked at the stopped queues first, 1 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker3] vk: invalidates since the last readbacks line: 2683 kept their virtual core (nothing there to wait for), 3205 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos Presentation Thread] vk: capture: a submit 0.12 ms (lock 0.00, scan 0.06 over 1063 ranges, 2017 MB), 174 pages written, 174 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.008; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (110 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos tera::FileResourceMgr] vcores: 206649 takes, 26548 waits (mean 32 us, longest 5.0 ms, 0 over HOS_VCORE_LOG_MS), 13230 kicks, 13165 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos host] vk: program picks: 2011485 kept (859403 by the last picks' textures, a uniform buffer bound since), 1136 new, 987378 with other textures (of 3000000)
[hos host] vk: a frame: 29 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.85, present 0.20, draw 2.64, descriptors 1.09, texture lookups 0.06, uploads 0.00, pipelines 0.10, programs 0.51, commands 0.73, dispatch 0.60, memory sync 0.37, queue submit 0.35, slot wait 0.04
[hos host] vk: a frame: mirrors 102 (824 MB): 4.07 MB synced in 836.5 copies, 22.35 MB written back
[hos host] vk: 10320 presents, 88.8 fps; a frame: 619 draws (0 skipped), 89 dispatches, 138 passes, 141 barriers, 6 submits, 0.0 uploads (0.0 MB), 94 ub updates (64 overlays), GPU-bound wait 0.0 ms (recording slots only); 37481 pipelines (7184477 ms), 0 programs (0 failed), 1602 textures (4362 MB)
[hos host] vk: host time in GPU resource waits: 1.57 ms a frame (fences 0.02, timelines 1.55, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 335.3 full barriers recorded, 180.3 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x10: 0.0; 0x11: 38.1; 0x12: 72.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.9 ms a frame (the longest submission 3.7 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: 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: write-backs left out (600 submits): 1403, 5343.7 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 472 dirty pages from their submits' captures, 61528 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): 1403 dispatches, 51022 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6551 runs, 1361.4 MB compared; 158 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16628 in captures, 9 only in the collection, 415 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] control: teleport to destination 3 asked for
[hos BaseProcExecutor3] cheats: teleport waits until the current event ends
[hos host] vk: recorder: 4839042 packets run in 10.0 s (busy 1.50 s, spun idle 3.40 s, slept 4437 times, 1258 of them under 200 us), 0 calls made directly (no room); the replay waited for it 27811 times, 112.53 ms in all, the longest 0.53 ms; vkEndCommandBuffer 8664/48.01 ms (max 0.53), replay_mu unlock 10540/37.18 ms (max 0.26), vkResetDescriptorPool 8588/27.24 ms (max 0.05), vmaInvalidateAllocation 14/0.08 ms (max 0.01), vkQueueBindSparse 5/0.01 ms (max 0.00)
[hos host] vk: replay reads of guest memory as submitted: 1740000 (826110 pages from the captures)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: replay reads of guest memory as submitted: 1760000 (838390 pages from the captures)
[hos host] vk: 1 textures retired (1601 left, 4361 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: sparse mirrors: 61, 824 MB with memory of 1116 MB (291 MB spared); blocks given memory later 1656, given back 3227; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 293 binds (39.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 19 taken from the spares, 83 made new; 41 kept as spares, 0 destroyed; 22 spares now, 196 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1403, 5343.7 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 540 dirty pages from their submits' captures, 61469 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): 1405 dispatches, 50991 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6556 runs, 1361.4 MB compared; 179 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16544 in captures, 8 only in the collection, 407 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 87 live (1316 MB), 0 retired still imported (0 MB), 12 retired given back (168 MB)
[hos ModuleSystemWorker5] nvn: CommandBufferWaitEvent #115000: event 0x10103c8010 mode 0 value 73375 (now 73356) (lr 0x80660fdc)
[hos Presentation Thread] vk: capture: a submit 0.12 ms (lock 0.00, scan 0.05 over 1063 ranges, 2017 MB), 174 pages written, 174 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (100 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 92 sync waits (19 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: replay reads of guest memory as submitted: 1780000 (850618 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos host] vk: readback pages refreshed from guest memory: 951598 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 1800000 (863307 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: a frame: 29 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.76, present 0.19, draw 2.52, descriptors 1.05, texture lookups 0.05, uploads 0.00, pipelines 0.09, programs 0.49, commands 0.68, dispatch 0.60, memory sync 0.36, queue submit 0.34, slot wait 0.04
[hos host] vk: a frame: mirrors 102 (824 MB): 3.97 MB synced in 836.6 copies, 22.32 MB written back
[hos host] vk: 10440 presents, 91.2 fps; a frame: 621 draws (0 skipped), 89 dispatches, 138 passes, 140 barriers, 6 submits, 0.0 uploads (0.0 MB), 94 ub updates (64 overlays), GPU-bound wait 0.0 ms (recording slots only); 37481 pipelines (7184477 ms), 0 programs (0 failed), 1601 textures (4361 MB)
[hos host] vk: host time in GPU resource waits: 1.66 ms a frame (fences 0.02, timelines 1.64, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 336.1 full barriers recorded, 180.0 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 37.9; 0x12: 72.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.9 ms a frame (the longest submission 3.7 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: write-backs left out (600 submits): 1403, 5343.7 MB; 38 bindings followed, 51 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 1686 so far
[hos host] vk: mirror syncs (600 submits): 528 dirty pages from their submits' captures, 61561 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): 1407 dispatches, 51095 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6566 runs, 1361.4 MB compared; 183 pages read into the shadows; 360 MB of shadows; pages the shadows knew that the CPU wrote: 16537 in captures, 7 only in the collection, 408 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1299 of 3000 waited for their write-back, 0.19 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: 2609 kept their virtual core (nothing there to wait for), 3023 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[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: 1820000 (875899 pages from the captures)
[hos ModuleSystemWorker3] nvn: CommandBufferSignalEvent #75000: event 0x10103c8034 mode 0 value 74999 (lr 0x806616b0)