Beanstalk Longshot test farm
Menu

Report rp_b501478316215d49 error

#1681 perf(vk): render targets take a pooled image's memory by size, and the pool holds longer on integrated GPUs

linux+nvidia on Nerdtality/agent-0452e3,

This report disagrees with its peers.

field-turn-and-stand run_hitch_count: -40.0% against peers' +100.0%
field-turn-and-stand#1 run_hitch_count: +0.0% against peers' -100.0%
no scenario could be judged: a step ran out of time at step 2 (wait map HyruleField_Child): timeout: not met in time (then the game was in play on KokiriForest_Child)

Machine

Commit
742b053ae6b0a34dca75a574db424c559f22850f
Baseline
068fb8ebe232d455ef48e817702b03ccc62d8520
App
nightly-211-845707b2, attempt 1
Game build
e0f3ccfa
os
linux
cpu
Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz
gpu
NVIDIA GeForce GTX 1070
tags
linuxnvidianvidia-proprietarydgpu
driver
580.196.7
ram_gb
32
vram_mb
8,438
gpu_vendor
nvidia
os_version
CachyOS (kernel 7.3.4)

Builds

main (baseline)

1,778.1 s warm

from 068fb8ebe232

lift
1.4 s
read
4 s
finish
24.1 s
compile
18.8 s
shaders
1,712.2 s

PR

55.5 s warm

Program cache: warm

Scenarios

field-turn-and-stand RX_VK_TARGET_REUSE_MEMORY=0 (PR only)

timeout

a step ran out of time at step 2 (wait map HyruleField_Child): timeout: not met in time (then the game was in play on KokiriForest_Child)

Metric (median)mainPRΔ
value_pool_mb00±0%
run_hitch_count53−40%
log_forbidden_hits00±0%
log_expected_missing00±0%
run_frametime_max_ms119.77138.34+15.5%
value_reused_by_size—0—
Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code——
value_pool_mb00
run_hitch_count53
log_forbidden_hits00
log_expected_missing00
run_frametime_max_ms119.77138.34
value_reused_by_size—0

field-turn-and-stand RX_VK_TARGET_REUSE_MEMORY=1 (PR only)

timeout

a step ran out of time at step 2 (wait map HyruleField_Child): timeout: not met in time (then the game was in play on KokiriForest_Child)

Metric (median)mainPRΔ
value_pool_mb560−100%
run_hitch_count55±0%
log_forbidden_hits00±0%
log_expected_missing00±0%
run_frametime_max_ms125.13122.97−1.7%
value_reused_by_size—13—
Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code——
value_pool_mb560
run_hitch_count55
log_forbidden_hits00
log_expected_missing00
run_frametime_max_ms125.13122.97
value_reused_by_size—13

field-turn-and-stand RX_VK_TARGET_REUSE_MEMORY=2 (PR only)

timeout

a step ran out of time at step 2 (wait map HyruleField_Child): timeout: not met in time (then the game was in play on KokiriForest_Child)

Metric (median)mainPRΔ
value_pool_mb040—
run_hitch_count37+133.3%
log_forbidden_hits00±0%
log_expected_missing00±0%
run_frametime_max_ms100.59135.51+34.7%
value_reused_by_size—12—
Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code——
value_pool_mb040
run_hitch_count37
log_forbidden_hits00
log_expected_missing00
run_frametime_max_ms100.59135.51
value_reused_by_size—12

Rules

ScenarioCellRuleLimitmainPRΔ %Status
field-turn-and-standRX_VK_TARGET_REUSE_MEMORY=0 (PR only)crashFalse00—pass
field-turn-and-standRX_VK_TARGET_REUSE_MEMORY=0 (PR only)screenshot.max_diff0.15———skippedthe plan has no screenshot probe
field-turn-and-standRX_VK_TARGET_REUSE_MEMORY=0 (PR only)log_forbidden_hits.max000—pass
field-turn-and-standRX_VK_TARGET_REUSE_MEMORY=1 (PR only)crashFalse00—pass
field-turn-and-standRX_VK_TARGET_REUSE_MEMORY=1 (PR only)screenshot.max_diff0.15———skippedthe plan has no screenshot probe
field-turn-and-standRX_VK_TARGET_REUSE_MEMORY=1 (PR only)log_forbidden_hits.max000—pass
field-turn-and-standRX_VK_TARGET_REUSE_MEMORY=2 (PR only)crashFalse00—pass
field-turn-and-standRX_VK_TARGET_REUSE_MEMORY=2 (PR only)screenshot.max_diff0.15———skippedthe plan has no screenshot probe
field-turn-and-standRX_VK_TARGET_REUSE_MEMORY=2 (PR only)log_forbidden_hits.max000—pass
field-turn-and-standRX_VK_TARGET_REUSE_MEMORY=1 (PR only)standing.frametime_p99_ms: RX_VK_TARGET_REUSE_MEMORY=1 vs RX_VK_TARGET_REUSE_MEMORY=0 (PR)20———skippedthe metric was not measured in both cells
field-turn-and-standRX_VK_TARGET_REUSE_MEMORY=2 (PR only)standing.frametime_p99_ms: RX_VK_TARGET_REUSE_MEMORY=2 vs RX_VK_TARGET_REUSE_MEMORY=0 (PR)20———skippedthe metric was not measured in both cells

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 .

field-turn-and-stand-c0-baseline0-last-frame.png
field-turn-and-stand-c0-baseline0-last-frame.png
field-turn-and-stand-c0-pr1-last-frame.png
field-turn-and-stand-c0-pr1-last-frame.png
field-turn-and-stand-c2-baseline0-last-frame.png
field-turn-and-stand-c2-baseline0-last-frame.png

Log (PR side, end)

[hos host] vk: mirror syncs' pass (600 submits): 1371 dispatches, 56289 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6688 runs, 1032.3 MB compared; 154 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 24157 in captures, 0 only in the collection, 61 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 60 live (1100 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: replay reads of guest memory as submitted: 1160000 (613029 pages from the captures)
[hos host] vk: texture slots registered again at another size: 200000 (of 275953 registrations; slot 26166 480x270 -> 960x540)
[hos Presentation Thread] vi: 4148 frames presented (54.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 65, 17.8-22 206, 22-30 0, 30-40 0, over 40 0
[hos ModuleSystemWorker4] nvn: CommandBufferSignalEvent #45000: event 0x100d6c8014 mode 0 value 45000 (lr 0x80660bcc)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos sead::MainThread] vk: readbacks of GPU results: 1466 of 3000 waited for their write-back, 0.45 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: 4684 kept their virtual core (nothing there to wait for), 3252 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos sead::MainThread] nvn: GPU waits in 5 s: 35 sync waits (93 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: replay reads of guest memory as submitted: 1180000 (623711 pages from the captures)
[hos sead::MainThread] nvn: DrawMeshTasksIndirect: 100000 so far (x1 0x117b427010 x2 0 x3 0x117b427000 x4 0xf33080)
[hos host] vk: write-backs left out (600 submits): 1342, 5625.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 779 dirty pages from their submits' captures, 69924 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): 1352 dispatches, 55179 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6784 runs, 1049.2 MB compared; 150 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 23761 in captures, 1 only in the collection, 62 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker2] nvn: CommandBufferWaitEvent #65000: event 0x100d6c8004 mode 0 value 45566 (now 45547) (lr 0x82c0326c)
[hos host] vk: memory: working set 11549 MB (peak 11768), private 5129 MB, staging 32 MB in 2 arenas, 24 made so far (105 ms)
[hos host] vk: memory heaps: heap 0 (device): 105 blocks 3826 MB, 14370 allocations 3609 MB; used 4265 of 7233 MB; heap 1 (system): 27 blocks 820 MB, 38 allocations 703 MB; used 1942 of 24004 MB; heap 2 (device): 119 blocks 173 MB, 149 allocations 173 MB; used 190 of 55 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 5 images 27 MB; 1 free ranges, the largest 4 MB
[hos host] vk: image blocks (render targets): 25 blocks 400 MB, 95 images 254 MB; 32 free ranges, the largest 16 MB
[hos host] vk: image blocks (textures, compacted): 32 blocks 1024 MB, 1210 images 1007 MB; 451 free ranges, the largest 9 MB
[hos host] vk: render target images: 64 reused from the pool, 258 made new, 161 destroyed from it, 12 reused by size (on a pooled one's memory, RX_VK_TARGET_REUSE_MEMORY=2), 1 missed after their image was taken by size; it holds 0 (0 MB)
[hos host] vk: render targets not reused by size: 193 with the pool empty (0 of them with targets still in the dead queue), 20 with none in it a present yet, 0 with none of a kind they can use, 5 with all smaller (the largest 28% short on average), 18 with all bigger (over 1.5 times), 22 with some of each, 0 over a block, 0 refused by the driver, 0 not bound; the pool let go of 0 over its bound and 161 idle
[hos host] vk: VRAM: 4456 MB used of 7288 MB allowed (216 MB of it free in the blocks); textures 2411 MB (budget 4778 MB), 1314 textures
[hos host] vk: budget: VRAM 3999 of 7288 MB (4456 MB by the driver, 456 MB unaccounted): textures 2411, mirrors 810, shadows 260, other buffers 225, blocks free 216, other images 43, staging 32; RAM 1830 MB (the process's private memory 5129 MB): imports 1100, capture spares 486, captures 182, other buffers 35, capture table 27, others 0
[hos host] vk: textures: decoded ASTC 1055 MB, rendered 768 MB (103 drawn in one frame, kept: 47857 KB), other 331 MB; used in the last 8 presents 1726 MB
[hos host] vk: decoded ASTC used again: 0 textures (0 MB) copied from ones the game finalized, not decoded; 0 of those retired, 3 listed; 372 MB of blocks hashed in 44 ms
[hos host] vk:   288 MB: 0x1147c16000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a22aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a5b22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, 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:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0xa, last used 2315 presents ago
[hos host] vk:   64 MB: 0x113cba9000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   47 MB: 0x10afd32400 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 1 presents ago
[hos host] vk:   47 MB: 0x10c2c99000 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 3 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 2 (177 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 260x260x1 NVN 0xd: 1 (175 KB);
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.10
[hos Presentation Thread] vk: presents after the CPU's signal (600): 82 waited for it, 2.42 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 21.4 ms; VRAM 4.4 of 7.1 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: 72 program records loaded (3.5 ms), 0 new graphics pipelines, pipelines from the driver's cache 0, from the start's cache 0, compiled 0; 0 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (play): 4770 queued, 4770 loaded (median 4.05 ms, 99th percentile 28.44 ms, worst 74.03 ms, 27170 ms and 1130.5 MB read in all); 7228 programs (1556 of them compute variants without their code, 0 graphics programs no note names), 8838 graphics pipelines (0 not in a cache), 569 compute pipelines (25 not in a cache); a first use ran 21 itself and waited for 3 (18.8 ms); 618 pipelines made at create taken by draws; 2245 create calls with nothing kept for their stages, 0 before the index; units 7015 (4770 ready, 0 loading), 12472 programs alive; 41 noted graphics pipeline probes deferred to first use
[hos host] vk: module identifiers (play): 594 known, 0 learned in this run; 569 compute pipelines made by them at a create call, 25 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 52 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 228 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 7.41, present 0.14, draw 4.25, descriptors 1.41, texture lookups 0.08, uploads 0.00, pipelines 0.15, programs 0.59, commands 1.84, dispatch 0.92, memory sync 0.64, queue submit 0.21, slot wait 0.02
[hos host] vk: a frame: mirrors 68 (810 MB): 4.53 MB synced in 905.5 copies, 16.92 MB written back
[hos host] vk: 4200 presents, 54.0 fps; a frame: 666 draws (0 skipped), 98 dispatches, 147 passes, 144 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 52365 pipelines (643401 ms), 0 programs (11 failed), 1314 textures (2411 MB)
[hos host] vk: host time in GPU resource waits: 4.23 ms a frame (fences 0.01, timelines 4.22, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 457.2 full barriers recorded, 230.1 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: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 13.4 ms a frame (the longest submission 7.6 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: the render thread (600 presents, 18.50 ms a frame): its jobs 8.08 ms a frame (submits 6.77, presents 0.17, others 1.14), idle 10.42; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 1.66 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 0.066 ms a frame; readbacks: 3000 next to a texture's storage (no catch-up), 0 in one, 0 read exactly (after the work handed over: among the catch-ups); a present made 7.01 ms after its hand-over (9.76 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 3.55 + 0.63 (the window's copy is in the idle), after it 0.49 + 0.45; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.11 ms of GPU time), after its last that reads it 1.00 (0.11 ms)
[hos host] vk: frame ahead (600 presents): 3.14 parts a frame recorded in the signal's job, 41.3% of them took nothing that came with the signal; of the others, 1.84 a frame took pages of the signal's capture into a mirror (61.9 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 6 of 88 frames, its first three in 6
[hos host] vk: frame ahead, what came with the signal: 88 signals with a capture, each of 88.9 pages in 8.2 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: 0.46+0.36, 0.01+0.04, 0.01+0.03, 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 sead::MainThread] vk: capture: a submit 0.17 ms (lock 0.00, scan 0.09 over 1258 ranges, 1855 MB), 201 pages written, 201 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.004, bytes copied 0.000; 25 kept (212 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 1200000 (634362 pages from the captures)
[hos rmbl::VibrationThread] hid: 12288 vibration values, strongest 0.00 since the last count
[hos host] vk: program picks: 1020690 kept (331941 by the last picks' textures, a uniform buffer bound since), 21047 new, 758262 with other textures (of 1800000)
[hos host] vk: sparse mirrors: 51, 810 MB with memory of 964 MB (153 MB spared); blocks given memory later 86, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 88 binds (16.2 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 6 taken from the spares, 62 made new; 17 kept as spares, 0 destroyed; 11 spares now, 296 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1342, 5625.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 792 dirty pages from their submits' captures, 70054 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): 1350 dispatches, 55282 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6677 runs, 1032.3 MB compared; 149 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 23777 in captures, 6 only in the collection, 61 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 60 live (1100 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos aal::LowPrioWorkerThread] vcores: 101702 takes, 15036 waits (mean 64 us, longest 7.5 ms, 0 over HOS_VCORE_LOG_MS), 6845 kicks, 7002 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: replay reads of guest memory as submitted: 1220000 (645048 pages from the captures)
[hos host] vk: 1 textures retired (1314 left, 2411 MB): budget 0, drawn over 1, 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: mirror of pool 0x1042c00000 (8 MB) given back: idle, 6012 submissions since its sync
[hos ModuleSystemWorker2] barrier: 106674 looks: 24754 found the masks moved already, 32724 waited until they moved, 49196 for the whole 1 ms (a wait 772 us on average); 34336 wakes
[hos host] vk: replay reads of guest memory as submitted: 1240000 (655708 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 1364, 5717.3 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 682 dirty pages from their submits' captures, 71219 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): 1372 dispatches, 56116 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6787 runs, 1049.2 MB compared; 150 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 24291 in captures, 2 only in the collection, 62 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: render targets drawn over from another start, let go: 161 so far (533 MB), 159 of them (525 MB) between the passes
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.10
[hos Presentation Thread] vk: capture: a submit 0.17 ms (lock 0.00, scan 0.09 over 1258 ranges, 1855 MB), 199 pages written, 199 copied (0.8 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; 24 kept (196 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: readback pages refreshed from guest memory: 567881 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: a frame: 52 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 228 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 7.46, present 0.16, draw 4.26, descriptors 1.41, texture lookups 0.08, uploads 0.00, pipelines 0.15, programs 0.59, commands 1.84, dispatch 0.92, memory sync 0.65, queue submit 0.21, slot wait 0.02
[hos host] vk: a frame: mirrors 68 (807 MB): 4.53 MB synced in 905.4 copies, 16.92 MB written back
[hos host] vk: 4320 presents, 53.9 fps; a frame: 666 draws (0 skipped), 98 dispatches, 147 passes, 144 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 52365 pipelines (643401 ms), 0 programs (11 failed), 1315 textures (2417 MB)
[hos host] vk: host time in GPU resource waits: 4.28 ms a frame (fences 0.01, timelines 4.26, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 457.3 full barriers recorded, 230.1 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: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 13.4 ms a frame (the longest submission 8.0 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 1463 of 3000 waited for their write-back, 0.44 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: 4339 kept their virtual core (nothing there to wait for), 2829 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: 1260000 (666377 pages from the captures)
[hos host] vk: queues ordered by their events: 24000 submissions stopped at a wait, 23981 went on once it was reached, 18 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 2 (stopped 50.06 ms at most); waits met 68805, not met 0; captures at a CPU event's wait 1180, 0 of them at a wait met on arrival
[hos host] vk: sparse mirrors: 50, 807 MB with memory of 956 MB (148 MB spared); blocks given memory later 86, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (16.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 6 taken from the spares, 62 made new; 18 kept as spares, 0 destroyed; 12 spares now, 304 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1342, 5625.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 891 dirty pages from their submits' captures, 69838 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): 1349 dispatches, 55222 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6696 runs, 1032.2 MB compared; 148 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 23812 in captures, 4 only in the collection, 61 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 60 live (1100 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: replay reads of guest memory as submitted: 1280000 (677080 pages from the captures)
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #70000: event 0x100d6c8004 mode 0 value 49366 (now 49347) (lr 0x82c0326c)
[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 ModuleSystemWorker1] nvn: 450000 recordings made, 43958 live copies (12.4 MB), 332 texture copies; 4 command buffers' segments (4.2 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.10
[hos Presentation Thread] vk: capture: a submit 0.16 ms (lock 0.00, scan 0.09 over 1258 ranges, 1855 MB), 200 pages written, 200 copied (0.8 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; 24 kept (168 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 1300000 (687810 pages from the captures)
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #50000: event 0x100d6c8020 mode 0 value 50000 (lr 0x80664f14)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 355 KB; 0 made on demand (0 MB kept now); c1 stages: at most 105 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: write-backs left out (600 submits): 1342, 5625.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 777 dirty pages from their submits' captures, 70369 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): 1351 dispatches, 55496 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6680 runs, 1032.2 MB compared; 165 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 23676 in captures, 1 only in the collection, 61 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vi: 4419 frames presented (54.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 64, 17.8-22 206, 22-30 1, 30-40 0, over 40 0
[hos sead::MainThread] nvn: GPU waits in 5 s: 39 sync waits (101 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: a frame: 53 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 228 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 7.44, present 0.13, draw 4.30, descriptors 1.42, texture lookups 0.08, uploads 0.00, pipelines 0.15, programs 0.59, commands 1.86, dispatch 0.93, memory sync 0.63, queue submit 0.21, slot wait 0.02
[hos host] vk: a frame: mirrors 68 (807 MB): 4.54 MB synced in 907.1 copies, 16.92 MB written back
[hos host] vk: 4440 presents, 54.1 fps; a frame: 665 draws (0 skipped), 98 dispatches, 147 passes, 144 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 52365 pipelines (643401 ms), 0 programs (11 failed), 1316 textures (2422 MB)
[hos host] vk: host time in GPU resource waits: 4.44 ms a frame (fences 0.01, timelines 4.43, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 457.5 full barriers recorded, 230.1 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: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 13.4 ms a frame (the longest submission 8.1 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: replay reads of guest memory as submitted: 1320000 (698348 pages from the captures)
[hos host] vk: sparse mirrors: 50, 807 MB with memory of 956 MB (148 MB spared); blocks given memory later 86, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (16.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 6 taken from the spares, 62 made new; 18 kept as spares, 0 destroyed; 12 spares now, 304 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1364, 5717.3 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 695 dirty pages from their submits' captures, 71142 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): 1372 dispatches, 56092 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6792 runs, 1049.2 MB compared; 156 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 24161 in captures, 2 only in the collection, 62 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 60 live (1100 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1463 of 3000 waited for their write-back, 0.45 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker1] vk: invalidates since the last readbacks line: 4345 kept their virtual core (nothing there to wait for), 3079 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: 1340000 (708879 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.17 ms (lock 0.00, scan 0.10 over 1258 ranges, 1855 MB), 201 pages written, 201 copied (0.8 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 (212 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: 1 textures retired (1315 left, 2417 MB): budget 0, drawn over 1, 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: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.10
[hos host] vk: program picks: 1135489 kept (368670 by the last picks' textures, a uniform buffer bound since), 21326 new, 843184 with other textures (of 2000000)
[hos aal::LowPrioWorkerThread] vcores: 103138 takes, 15346 waits (mean 68 us, longest 4.6 ms, 0 over HOS_VCORE_LOG_MS), 7010 kicks, 7125 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: replay reads of guest memory as submitted: 1360000 (719419 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 1342, 5625.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 805 dirty pages from their submits' captures, 70305 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, 55398 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6701 runs, 1032.3 MB compared; 138 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 23818 in captures, 2 only in the collection, 61 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a frame: 53 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 228 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 7.39, present 0.15, draw 4.25, descriptors 1.40, texture lookups 0.08, uploads 0.00, pipelines 0.15, programs 0.59, commands 1.84, dispatch 0.92, memory sync 0.64, queue submit 0.23, slot wait 0.02
[hos host] vk: a frame: mirrors 68 (807 MB): 4.53 MB synced in 904.3 copies, 16.92 MB written back
[hos host] vk: 4560 presents, 53.9 fps; a frame: 665 draws (0 skipped), 98 dispatches, 147 passes, 144 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 52365 pipelines (643401 ms), 0 programs (11 failed), 1315 textures (2417 MB)
[hos host] vk: host time in GPU resource waits: 4.26 ms a frame (fences 0.01, timelines 4.24, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 457.2 full barriers recorded, 230.0 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: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 13.4 ms a frame (the longest submission 8.2 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: 1 textures retired (1314 left, 2411 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1380000 (729970 pages from the captures)
[hos ModuleSystemWorker5] nvn: CommandBufferWaitEvent #75000: event 0x100d6c8004 mode 0 value 53166 (now 53147) (lr 0x82c0326c)
[hos Presentation Thread] vk: capture: a submit 0.17 ms (lock 0.00, scan 0.09 over 1258 ranges, 1855 MB), 201 pages written, 201 copied (0.8 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 (198 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.10
[hos host] vk: sparse mirrors: 50, 807 MB with memory of 956 MB (148 MB spared); blocks given memory later 86, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (16.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 6 taken from the spares, 62 made new; 18 kept as spares, 0 destroyed; 12 spares now, 304 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1364, 5717.3 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 828 dirty pages from their submits' captures, 70627 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): 1358 dispatches, 55857 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6797 runs, 1049.2 MB compared; 161 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 24122 in captures, 4 only in the collection, 62 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 60 live (1100 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: replay reads of guest memory as submitted: 1400000 (740530 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: readback pages refreshed from guest memory: 639259 pages so far (0 held: may be written ahead in their submission)
[hos ModuleSystemWorker1] nvn: DrawMeshTasksIndirect: 120000 so far (x1 0x117e3d8010 x2 0 x3 0x117e3d8000 x4 0x8b040)
[hos host] vk: 1 textures retired (1314 left, 2411 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1420000 (751057 pages from the captures)
[hos sead::MainThread] vk: readbacks of GPU results: 1466 of 3000 waited for their write-back, 0.43 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: 4530 kept their virtual core (nothing there to wait for), 3150 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: write-backs left out (600 submits): 1342, 5625.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 737 dirty pages from their submits' captures, 69958 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): 1350 dispatches, 55206 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6681 runs, 1032.3 MB compared; 150 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 23677 in captures, 2 only in the collection, 61 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] nvn: CommandBufferSignalEvent #55000: event 0x100d6c8028 mode 0 value 54999 (lr 0x80667410)
[hos host] vk: a frame: 53 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 228 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 7.48, present 0.15, draw 4.37, descriptors 1.44, texture lookups 0.09, uploads 0.00, pipelines 0.15, programs 0.60, commands 1.91, dispatch 0.93, memory sync 0.64, queue submit 0.21, slot wait 0.02
[hos host] vk: a frame: mirrors 68 (807 MB): 4.54 MB synced in 905.9 copies, 16.92 MB written back
[hos host] vk: 4680 presents, 53.8 fps; a frame: 665 draws (0 skipped), 98 dispatches, 147 passes, 144 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 52365 pipelines (643401 ms), 0 programs (11 failed), 1314 textures (2411 MB)
[hos host] vk: host time in GPU resource waits: 4.32 ms a frame (fences 0.01, timelines 4.31, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 457.2 full barriers recorded, 230.1 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: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 13.5 ms a frame (the longest submission 8.1 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos Presentation Thread] vi: 4689 frames presented (53.9 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 61, 17.8-22 209, 22-30 0, 30-40 0, over 40 0
[hos host] vk: replay reads of guest memory as submitted: 1440000 (761854 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.17 ms (lock 0.00, scan 0.09 over 1258 ranges, 1855 MB), 202 pages written, 202 copied (0.8 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 (222 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.10
[hos sead::MainThread] nvn: GPU waits in 5 s: 35 sync waits (100 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 355 KB; 0 made on demand (0 MB kept now); c1 stages: at most 105 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 50, 807 MB with memory of 956 MB (148 MB spared); blocks given memory later 86, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (16.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 6 taken from the spares, 62 made new; 18 kept as spares, 0 destroyed; 12 spares now, 304 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1364, 5717.3 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 780 dirty pages from their submits' captures, 71296 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): 1371 dispatches, 56195 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6776 runs, 1047.6 MB compared; 157 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 24131 in captures, 4 only in the collection, 61 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 60 live (1100 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[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: 1460000 (772503 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 1480000 (783187 pages from the captures)
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #80000: event 0x100d6c8004 mode 0 value 56966 (now 56947) (lr 0x82c0326c)
[hos host] vk: program picks: 1250134 kept (405380 by the last picks' textures, a uniform buffer bound since), 21605 new, 928260 with other textures (of 2200000)
[hos host] vk: write-backs left out (600 submits): 1342, 5625.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 835 dirty pages from their submits' captures, 69984 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): 1351 dispatches, 55195 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6710 runs, 1033.8 MB compared; 134 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 23917 in captures, 4 only in the collection, 62 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] nvn: 500000 recordings made, 43966 live copies (11.9 MB), 332 texture copies; 1 command buffers' segments (4.0 MB)
[hos SceneLowPriority] vcores: 101665 takes, 15629 waits (mean 68 us, longest 7.6 ms, 0 over HOS_VCORE_LOG_MS), 7104 kicks, 7284 yields, 0 starved grants, 0 escapes in 5 s
[hos Presentation Thread] vk: capture: a submit 0.16 ms (lock 0.00, scan 0.09 over 1258 ranges, 1855 MB), 199 pages written, 199 copied (0.8 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; 24 kept (196 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: memory: working set 11550 MB (peak 11768), private 5129 MB, staging 32 MB in 2 arenas, 24 made so far (105 ms)
[hos host] vk: memory heaps: heap 0 (device): 105 blocks 3826 MB, 14326 allocations 3612 MB; used 4265 of 7233 MB; heap 1 (system): 27 blocks 820 MB, 38 allocations 703 MB; used 1942 of 24004 MB; heap 2 (device): 119 blocks 173 MB, 149 allocations 173 MB; used 190 of 55 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 5 images 27 MB; 1 free ranges, the largest 4 MB
[hos host] vk: image blocks (render targets): 25 blocks 400 MB, 96 images 259 MB; 32 free ranges, the largest 16 MB
[hos host] vk: image blocks (textures, compacted): 32 blocks 1024 MB, 1210 images 1007 MB; 451 free ranges, the largest 9 MB
[hos host] vk: render target images: 64 reused from the pool, 263 made new, 165 destroyed from it, 12 reused by size (on a pooled one's memory, RX_VK_TARGET_REUSE_MEMORY=2), 1 missed after their image was taken by size; it holds 0 (0 MB)
[hos host] vk: render targets not reused by size: 198 with the pool empty (0 of them with targets still in the dead queue), 20 with none in it a present yet, 0 with none of a kind they can use, 5 with all smaller (the largest 28% short on average), 18 with all bigger (over 1.5 times), 22 with some of each, 0 over a block, 0 refused by the driver, 0 not bound; the pool let go of 0 over its bound and 165 idle
[hos host] vk: VRAM: 4456 MB used of 7288 MB allowed (213 MB of it free in the blocks); textures 2417 MB (budget 4778 MB), 1315 textures
[hos host] vk: budget: VRAM 3999 of 7288 MB (4456 MB by the driver, 456 MB unaccounted): textures 2417, mirrors 807, shadows 260, other buffers 225, blocks free 213, other images 43, staging 32; RAM 1830 MB (the process's private memory 5129 MB): imports 1100, capture spares 500, captures 200, capture table 27, other buffers 3, others 0
[hos host] vk: textures: decoded ASTC 1055 MB, rendered 773 MB (103 drawn in one frame, kept: 47857 KB), other 331 MB; used in the last 8 presents 1731 MB
[hos host] vk: decoded ASTC used again: 0 textures (0 MB) copied from ones the game finalized, not decoded; 0 of those retired, 3 listed; 372 MB of blocks hashed in 44 ms
[hos host] vk:   288 MB: 0x1147c16000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a5b22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a22aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0xa, last used 2915 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:   47 MB: 0x10c2c99000 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 3 presents ago
[hos host] vk:   47 MB: 0x10bba9a200 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 2 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 2 (177 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 260x260x1 NVN 0xd: 1 (175 KB);
[hos host] vk: 1 textures retired (1314 left, 2411 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.10
[hos Presentation Thread] vk: presents after the CPU's signal (600): 83 waited for it, 2.62 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 22.2 ms; VRAM 4.4 of 7.1 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: 72 program records loaded (3.5 ms), 0 new graphics pipelines, pipelines from the driver's cache 0, from the start's cache 0, compiled 0; 0 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (play): 4770 queued, 4770 loaded (median 4.05 ms, 99th percentile 28.44 ms, worst 74.03 ms, 27170 ms and 1130.5 MB read in all); 7228 programs (1556 of them compute variants without their code, 0 graphics programs no note names), 8838 graphics pipelines (0 not in a cache), 569 compute pipelines (25 not in a cache); a first use ran 21 itself and waited for 3 (18.8 ms); 618 pipelines made at create taken by draws; 2245 create calls with nothing kept for their stages, 0 before the index; units 7015 (4770 ready, 0 loading), 12472 programs alive; 41 noted graphics pipeline probes deferred to first use
[hos host] vk: module identifiers (play): 594 known, 0 learned in this run; 569 compute pipelines made by them at a create call, 25 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 53 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 228 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 7.52, present 0.13, draw 4.35, descriptors 1.44, texture lookups 0.08, uploads 0.00, pipelines 0.15, programs 0.59, commands 1.89, dispatch 0.93, memory sync 0.65, queue submit 0.21, slot wait 0.02
[hos host] vk: a frame: mirrors 68 (807 MB): 4.53 MB synced in 904.8 copies, 16.92 MB written back
[hos host] vk: 4800 presents, 53.9 fps; a frame: 665 draws (0 skipped), 98 dispatches, 147 passes, 144 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 52365 pipelines (643401 ms), 0 programs (11 failed), 1314 textures (2411 MB)
[hos host] vk: host time in GPU resource waits: 4.49 ms a frame (fences 0.01, timelines 4.48, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 457.1 full barriers recorded, 230.1 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: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 13.4 ms a frame (the longest submission 8.1 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, 18.55 ms a frame): its jobs 8.07 ms a frame (submits 6.77, presents 0.15, others 1.14), idle 10.48; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 1.68 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); 89 captures at a CPU signal; the submits' waits read in 0.067 ms a frame; readbacks: 2995 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 6.98 ms after its hand-over (10.62 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 3.55 + 0.63 (the window's copy is in the idle), after it 0.49 + 0.44; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.11 ms of GPU time), after its last that reads it 1.00 (0.11 ms)
[hos host] vk: frame ahead (600 presents): 3.08 parts a frame recorded in the signal's job, 40.9% of them took nothing that came with the signal; of the others, 1.82 a frame took pages of the signal's capture into a mirror (62.5 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 7 of 89 frames, its first three in 7
[hos host] vk: frame ahead, what came with the signal: 89 signals with a capture, each of 90.7 pages in 8.2 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: 0.46+0.36, 0.01+0.04, 0.01+0.03, 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 host] vk: replay reads of guest memory as submitted: 1500000 (793864 pages from the captures)
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 1466 of 3000 waited for their write-back, 0.46 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: 4333 kept their virtual core (nothing there to wait for), 2835 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: sparse mirrors: 50, 807 MB with memory of 956 MB (148 MB spared); blocks given memory later 86, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (16.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 6 taken from the spares, 62 made new; 18 kept as spares, 0 destroyed; 12 spares now, 304 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1342, 5625.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 785 dirty pages from their submits' captures, 70302 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): 1350 dispatches, 55239 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6682 runs, 1032.2 MB compared; 153 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 24170 in captures, 4 only in the collection, 61 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 60 live (1100 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: replay reads of guest memory as submitted: 1520000 (804517 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.17 ms (lock 0.00, scan 0.09 over 1258 ranges, 1855 MB), 201 pages written, 201 copied (0.8 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 (192 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 1540000 (815223 pages from the captures)
[hos host] vk: render targets drawn over from another start, let go: 161 so far (533 MB), 159 of them (525 MB) between the passes
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.10
[hos host] vk: write-backs left out (600 submits): 1342, 5625.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 904 dirty pages from their submits' captures, 70155 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): 1349 dispatches, 55211 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6671 runs, 1032.2 MB compared; 155 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 24159 in captures, 2 only in the collection, 61 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker3] barrier: 128368 looks: 30064 found the masks moved already, 36150 waited until they moved, 62154 for the whole 1 ms (a wait 805 us on average); 39059 wakes
[hos host] vk: a frame: 52 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 228 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 7.48, present 0.14, draw 4.39, descriptors 1.50, texture lookups 0.08, uploads 0.00, pipelines 0.15, programs 0.60, commands 1.88, dispatch 0.93, memory sync 0.65, queue submit 0.22, slot wait 0.02
[hos host] vk: a frame: mirrors 68 (807 MB): 4.55 MB synced in 906.0 copies, 16.92 MB written back
[hos host] vk: 4920 presents, 53.9 fps; a frame: 666 draws (0 skipped), 98 dispatches, 147 passes, 144 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 52365 pipelines (643401 ms), 0 programs (11 failed), 1316 textures (2422 MB)
[hos host] vk: host time in GPU resource waits: 4.21 ms a frame (fences 0.01, timelines 4.20, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 457.3 full barriers recorded, 230.2 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: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 13.4 ms a frame (the longest submission 8.0 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: replay reads of guest memory as submitted: 1560000 (825867 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 710129 pages so far (0 held: may be written ahead in their submission)
[hos ModuleSystemWorker5] nvn: CommandBufferSignalEvent #60000: event 0x100d6c8034 mode 0 value 59999 (lr 0x806616b0)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos Presentation Thread] vi: 4959 frames presented (53.9 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 59, 17.8-22 211, 22-30 0, 30-40 0, over 40 0
[hos host] vk: sparse mirrors: 50, 807 MB with memory of 956 MB (148 MB spared); blocks given memory later 86, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (16.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 6 taken from the spares, 62 made new; 18 kept as spares, 0 destroyed; 12 spares now, 304 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1364, 5717.3 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 801 dirty pages from their submits' captures, 71546 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): 1372 dispatches, 56261 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6788 runs, 1048.1 MB compared; 143 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 24665 in captures, 1 only in the collection, 61 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 60 live (1100 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: replay reads of guest memory as submitted: 1580000 (836550 pages from the captures)
[hos ModuleSystemWorker3] vk: readbacks of GPU results: 1461 of 3000 waited for their write-back, 0.43 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: 4367 kept their virtual core (nothing there to wait for), 3057 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #85000: event 0x100d6c8004 mode 0 value 60766 (now 60747) (lr 0x82c0326c)
[hos sead::MainThread] nvn: GPU waits in 5 s: 39 sync waits (109 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos Presentation Thread] vk: capture: a submit 0.17 ms (lock 0.00, scan 0.09 over 1258 ranges, 1855 MB), 203 pages written, 203 copied (0.8 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 (198 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.10
[hos sead::MainThread] nvn: EventSignal (CPU) #10000: event 0x1006e38004 mode 0 value 1 (lr 0x8348c78c)
[hos host] vk: replay reads of guest memory as submitted: 1600000 (847350 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 355 KB; 0 made on demand (0 MB kept now); c1 stages: at most 105 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: write-backs left out (600 submits): 1342, 5625.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 720 dirty pages from their submits' captures, 70590 from earlier ones', 71 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): 1351 dispatches, 55323 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6688 runs, 1033.3 MB compared; 164 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 24093 in captures, 5 only in the collection, 62 only by the replay itself
[hos host] vk:   read live, by pool: 0x1065800000 (8 MB, captured, state 2): 71;
[hos host] vk: present 5029 written to /<user>/<user>/Beanstalk/jobs/jb_ff7b9571e212161e/runs/field-turn-and-stand-g2-01-pr/shots/field-turn-and-stand-pr-01-last-frame.png (the control bridge)
[hos host] vk: 1 textures retired (1315 left, 2417 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a frame: 52 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 228 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 7.46, present 0.22, draw 4.34, descriptors 1.43, texture lookups 0.08, uploads 0.00, pipelines 0.15, programs 0.59, commands 1.90, dispatch 0.93, memory sync 0.63, queue submit 0.21, slot wait 0.02
[hos host] vk: a frame: mirrors 68 (807 MB): 4.56 MB synced in 906.5 copies, 16.92 MB written back
[hos host] vk: 5040 presents, 53.9 fps; a frame: 666 draws (0 skipped), 98 dispatches, 147 passes, 144 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 52365 pipelines (643401 ms), 0 programs (11 failed), 1315 textures (2417 MB)
[hos host] vk: host time in GPU resource waits: 4.29 ms a frame (fences 0.01, timelines 4.28, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 457.3 full barriers recorded, 230.0 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: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 13.4 ms a frame (the longest submission 8.1 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: replay reads of guest memory as submitted: 1620000 (857890 pages from the captures)
[hos sead::MainThread] vcores: 102084 takes, 15732 waits (mean 65 us, longest 3.0 ms, 0 over HOS_VCORE_LOG_MS), 7241 kicks, 7445 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: program picks: 1364656 kept (442032 by the last picks' textures, a uniform buffer bound since), 21885 new, 1013458 with other textures (of 2400000)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos Presentation Thread] vk: capture: a submit 0.16 ms (lock 0.00, scan 0.09 over 1258 ranges, 1855 MB), 200 pages written, 200 copied (0.8 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; 24 kept (188 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 1640000 (868435 pages from the captures)
[hos host] vk: sparse mirrors: 50, 807 MB with memory of 956 MB (148 MB spared); blocks given memory later 86, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (16.3 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 6 taken from the spares, 62 made new; 18 kept as spares, 0 destroyed; 12 spares now, 304 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1342, 5625.1 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 827 dirty pages from their submits' captures, 70540 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): 1363 dispatches, 55403 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6694 runs, 1032.3 MB compared; 154 pages read into the shadows; 260 MB of shadows; pages the shadows knew that the CPU wrote: 23933 in captures, 1 only in the collection, 61 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 60 live (1100 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.10
[hos ModuleSystemWorker4] nvn: DrawMeshTasksIndirect: 140000 so far (x1 0x11789a4000 x2 0x1 x3 0x11789a4000 x4 0x4040)
[hos host] vk: 1 textures retired (1314 left, 2411 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1660000 (879003 pages from the captures)
[hos ModuleSystemWorker3] vk: readbacks of GPU results: 1465 of 3000 waited for their write-back, 0.45 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: 4393 kept their virtual core (nothing there to wait for), 3031 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)