Beanstalk Longshot test farm
Menu

Report rp_b1d123012df361b1 fail

#1073 perf(vk): use smaller VMA block sizes on < 7 GB GPUs

vram-ge8gb on Jiro113/agent-fd7c7b,

Machine

Commit
9e62352fb8101597b52d28f4337e0082fa3d2f89
Baseline
02903b04da567c7967c79b1fdd1b772b47151d91
App
nightly-126-05954779, attempt 1
Game build
e0f3ccfa
os
windows
cpu
13th Gen Intel(R) Core(TM) i9-13900K
gpu
NVIDIA GeForce RTX 4090
tags
windowsnvidianvidia-proprietarydgpu
driver
621.24.0
ram_gb
64
vram_mb
24,138
gpu_vendor
nvidia
os_version
Windows 10.0.26300

Builds

main (baseline)

1,474.1 s cold

from 02903b04da56

lift
87.7 s
read
352.4 s
finish
465.7 s
compile
566.9 s

PR

26.8 s warm

Program cache: warm

Scenarios

market-tour

ok
Metric (median)mainPRΔ
exit_code00±0%
load_time_s11.726.27−46.5%
load_time_max_s19.298.61−55.4%
log_forbidden_hits00±0%
market.hitch_count00±0%
market.vram_mean_mb12,232.213,161.6+7.6%
market.vram_peak_mb12,40013,332+7.5%
log_expected_missing00±0%
market.stutter_count10−100%
market.frametime_max_ms31.5622.6−28.4%
market.frametime_p50_ms10.4111.88+14.2%
market.frametime_p99_ms14.2517.82+25%
field-after-tour.hitch_count00±0%
market.frametime_1pct_low_ms15.4219.07+23.6%
field-after-tour.vram_mean_mb15,163.615,197.6+0.2%
field-after-tour.vram_peak_mb15,52215,428−0.6%
field-after-tour.stutter_count11±0%
field-after-tour.frametime_max_ms22.3623.56+5.4%
field-after-tour.frametime_p50_ms10.3710.04−3.2%
field-after-tour.frametime_p99_ms15.0814.85−1.5%
field-after-tour.frametime_1pct_low_ms16.4316.71+1.7%
screenshot_diff—0.72—

Pictures against main: market diff 0.7211, field diff 0.0654

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
load_time_s11.726.27
load_time_max_s19.298.61
log_forbidden_hits00
market.hitch_count00
market.vram_mean_mb12,232.213,161.6
market.vram_peak_mb12,40013,332
log_expected_missing00
market.stutter_count10
market.frametime_max_ms31.5622.6
market.frametime_p50_ms10.4111.88
market.frametime_p99_ms14.2517.82
field-after-tour.hitch_count00
market.frametime_1pct_low_ms15.4219.07
field-after-tour.vram_mean_mb15,163.615,197.6
field-after-tour.vram_peak_mb15,52215,428
field-after-tour.stutter_count11
field-after-tour.frametime_max_ms22.3623.56
field-after-tour.frametime_p50_ms10.3710.04
field-after-tour.frametime_p99_ms15.0814.85
field-after-tour.frametime_1pct_low_ms16.4316.71
screenshot_diff—0.72
loadsMarket_Child 19.29 sKakarikoVillage_Child 10.01 sHyruleField_Child 5.87 sMarket_Child 8.61 sKakarikoVillage_Child 5.64 sHyruleField_Child 4.56 s

Rules

ScenarioCellRuleLimitmainPRΔ %Status
market-tourcrashFalse00—pass
market-tourscreenshot.max_diff0.05—0.72—fail
market-tourmarket.vram_peak_mb.max_delta_pct1512,40013,332+7.5%pass
market-tourfield-after-tour.vram_peak_mb.max_delta_pct1515,52215,428−0.6%pass
market-tourload_time_max_s.max_delta_pct3019.298.61−55.4%pass
market-tourmarket.frametime_p99_ms.max_delta_pct2014.2517.82+25%fail
market-tourfield-after-tour.frametime_p99_ms.max_delta_pct2015.0814.85−1.5%pass
market-tourlog_forbidden_hits.max000—pass

Log (PR side, end)

[hos host] vk:   64 MB: 0x114045c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10005f65c00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10005fde200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 2332 presents ago
[hos host] vk:   54 MB: 0x1000657e400 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 2328 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 53 (7694 KB): 242x40x1 NVN 0x25: 4 (151 KB); 260x260x1 NVN 0xd: 39 (6862 KB); 306x104x1 NVN 0x1: 4 (124 KB); 592x49x1 NVN 0x25: 3 (339 KB); 656x113x1 NVN 0x1: 3 (217 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 127 waited for it, 0.19 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 12.3 ms; VRAM 15.1 of 22.8 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: 121 program records loaded (33.2 ms), 12 new graphics pipelines, pipelines from the driver's cache 27, from the start's cache 22, compiled 85; 1957 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (play): 6165 queued, 6165 loaded (median 3.40 ms, 99th percentile 16.85 ms, worst 63.46 ms, 26432 ms and 1349.3 MB read in all); 8420 programs (1390 of them compute variants without their code, 0 graphics programs no note names), 10740 graphics pipelines (0 not in a cache), 178 compute pipelines (526 not in a cache); a first use ran 24 itself and waited for 0 (51.7 ms); 2530 pipelines made at create taken by draws; 3288 create calls with nothing kept for their stages, 0 before the index; units 9453 (6165 ready, 0 loading), 16483 programs alive
[hos host] vk: module identifiers (play): 777 known, 777 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 130 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 5.15, present 0.17, draw 2.61, descriptors 1.00, texture lookups 0.05, uploads 0.00, pipelines 0.14, programs 0.48, commands 0.75, dispatch 0.52, memory sync 0.33, queue submit 0.37, slot wait 0.03
[hos host] vk: a frame: mirrors 377 (1089 MB): 7.38 MB synced in 1129.5 copies, 71.42 MB written back
[hos host] vk: 19800 presents, 101.9 fps; a frame: 1510 draws (0 skipped), 159 dispatches, 152 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 43831 pipelines (3922986 ms), 0 programs (0 failed), 5018 textures (10647 MB)
[hos host] vk: host time in GPU resource waits: 0.43 ms a frame (fences 0.02, timelines 0.41, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 508.0 full barriers recorded, 189.5 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.8 ms a frame (the longest submission 2.1 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's whole frames (600 presents): 0 jobs went on before their present after 1000 ms, 0 times a game's thread waited for the work handed over, 0 times for a fence of the frame in the making
[hos host] vk: the render thread (600 presents, 9.68 ms a frame): its jobs 5.61 ms a frame (submits 4.11, presents 0.17, others 1.33), idle 4.07; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 111.00 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); 167 captures at a CPU signal; the submits' waits read in 0.037 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 5.63 ms after its hand-over (8.31 at most)
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 932 binds (127.0 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 537 dirty pages from their submits' captures, 52706 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5153 runs, 1991.9 MB compared; 41 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16423 in captures, 5 only in the collection, 200 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos host] vk: program picks: 6233081 kept (2371191 by the last picks' textures, a uniform buffer bound since), 3058 new, 2963860 with other textures (of 9200000)
[hos ModuleSystemWorker5] nvn: CommandBufferWaitEvent #275000: event 0x10103c8038 mode 0 value 187566 (now 187547) (lr 0x80661a48)
[hos host] vk: mirror syncs (600 submits): 502 dirty pages from their submits' captures, 50413 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5050 runs, 1905.5 MB compared; 42 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15852 in captures, 8 only in the collection, 202 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 5020000 (2881016 pages from the captures)
[hos BaseProcExecutor1] vcores: 394688 takes, 42301 waits (mean 26 us, longest 5.1 ms), 21739 kicks, 21379 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 684 KB; 0 made on demand (0 MB kept now); c1 stages: at most 245 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 932 binds (127.0 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 543 dirty pages from their submits' captures, 52336 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5185 runs, 2008.7 MB compared; 41 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16488 in captures, 5 only in the collection, 200 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos host] vk: recorder: 10004853 packets run in 10.0 s (busy 1.65 s, spun idle 3.96 s, slept 3855 times, 1103 of them under 200 us), 0 calls made directly (no room); the replay waited for it 56251 times, 136.54 ms in all, the longest 0.27 ms; vkEndCommandBuffer 22021/82.32 ms (max 0.27), replay_mu unlock 12206/27.91 ms (max 0.15), vkResetDescriptorPool 22021/26.30 ms (max 0.13), vkQueueBindSparse 3/0.01 ms (max 0.01)
[hos host] vk: readback pages refreshed from guest memory: 3707427 pages so far (0 held: may be written ahead in their submission)
[hos sead::MainThread] vk: capture: a submit 0.11 ms (lock 0.01, scan 0.04 over 613 ranges, 2448 MB), 377 pages written, 377 copied (1.5 MB)
[hos host] vk: a capture taken (ms): all 0.009; 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 (136 MB), 7 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 5040000 (2888993 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: mirror syncs (600 submits): 618 dirty pages from their submits' captures, 50656 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5010 runs, 1921.3 MB compared; 41 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15923 in captures, 7 only in the collection, 184 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos host] vk: sparse mirrors: 70, 1086 MB with memory of 1500 MB (413 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 932 binds (127.0 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 755 dirty pages from their submits' captures, 50073 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5019 runs, 1943.3 MB compared; 41 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15860 in captures, 10 only in the collection, 197 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos host] vk: a frame: 130 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 5.45, present 0.19, draw 2.78, descriptors 1.08, texture lookups 0.05, uploads 0.00, pipelines 0.15, programs 0.49, commands 0.82, dispatch 0.54, memory sync 0.34, queue submit 0.39, slot wait 0.04
[hos host] vk: a frame: mirrors 377 (1086 MB): 7.38 MB synced in 1131.2 copies, 71.16 MB written back
[hos host] vk: 19920 presents, 98.1 fps; a frame: 1503 draws (0 skipped), 159 dispatches, 151 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 43831 pipelines (3922986 ms), 0 programs (0 failed), 5018 textures (10647 MB)
[hos host] vk: host time in GPU resource waits: 0.30 ms a frame (fences 0.03, timelines 0.27, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 506.7 full barriers recorded, 189.5 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.7 ms a frame (the longest submission 2.1 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1549 of 3000 waited for their write-back, 0.03 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos host] vk: replay reads of guest memory as submitted: 5060000 (2896855 pages from the captures)
[hos ModuleSystemWorker1] nvn: DrawMeshTasksIndirect: 400000 so far (x1 0x1186e8b060 x2 0 x3 0x1186e8b040 x4 0x44b040)
[hos host] vk: mirror syncs (600 submits): 638 dirty pages from their submits' captures, 50613 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5004 runs, 1921.2 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15837 in captures, 9 only in the collection, 194 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: program picks: 6384377 kept (2446395 by the last picks' textures, a uniform buffer bound since), 3058 new, 3012564 with other textures (of 9400000)
[hos ModuleSystemWorker3] nvn: CommandBufferSignalEvent #190000: event 0x10103c803c mode 0 value 189999 (lr 0x80661bfc)
[hos host] vk: replay reads of guest memory as submitted: 5080000 (2904878 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.10 ms (lock 0.01, scan 0.04 over 607 ranges, 2416 MB), 360 pages written, 360 copied (1.4 MB)
[hos host] vk: a capture taken (ms): all 0.012; 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 (168 MB), 6 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 933 binds (127.0 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 811 dirty pages from their submits' captures, 51721 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5096 runs, 1986.3 MB compared; 42 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16489 in captures, 12 only in the collection, 195 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 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 sead::MainThread] nvn: QueueSubmitCommands #120000: queue 0000101010395020 (its #19996), 1 handles (lr 0x80666be0)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos sead::MainThread] nvn: EventSignal (CPU) #40000: event 0x1009b38004 mode 0 value 1 (lr 0x8348c78c)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 684 KB; 0 made on demand (0 MB kept now); c1 stages: at most 245 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: mirror syncs (600 submits): 710 dirty pages from their submits' captures, 50131 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5088 runs, 1943.0 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15848 in captures, 9 only in the collection, 194 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] nvn: GPU waits in 5 s: 144 sync waits (22 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: readback pages refreshed from guest memory: 3774261 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 5100000 (2912401 pages from the captures)
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 933 binds (127.0 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 602 dirty pages from their submits' captures, 50864 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5013 runs, 1923.9 MB compared; 42 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15912 in captures, 11 only in the collection, 201 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #280000: event 0x10103c8038 mode 0 value 191366 (now 191347) (lr 0x80661a48)
[hos host] vk: a frame: 130 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 5.17, present 0.18, draw 2.63, descriptors 1.03, texture lookups 0.05, uploads 0.00, pipelines 0.14, programs 0.48, commands 0.76, dispatch 0.52, memory sync 0.33, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 378 (1089 MB): 7.41 MB synced in 1128.9 copies, 71.45 MB written back
[hos host] vk: 20040 presents, 103.0 fps; a frame: 1509 draws (0 skipped), 159 dispatches, 152 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 43831 pipelines (3922986 ms), 0 programs (0 failed), 5018 textures (10647 MB)
[hos host] vk: host time in GPU resource waits: 0.38 ms a frame (fences 0.02, timelines 0.36, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 507.8 full barriers recorded, 189.5 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.8 ms a frame (the longest submission 2.1 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: mirror syncs (600 submits): 833 dirty pages from their submits' captures, 49865 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5016 runs, 1941.3 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15851 in captures, 13 only in the collection, 187 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 5120000 (2920018 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.01, scan 0.04 over 604 ranges, 2404 MB), 365 pages written, 365 copied (1.4 MB)
[hos host] vk: a capture taken (ms): all 0.011; 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 (146 MB), 8 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: program picks: 6535016 kept (2521556 by the last picks' textures, a uniform buffer bound since), 3058 new, 3061925 with other textures (of 9600000)
[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: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 933 binds (127.0 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 693 dirty pages from their submits' captures, 52032 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5178 runs, 1988.7 MB compared; 41 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16492 in captures, 9 only in the collection, 191 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: replay reads of guest memory as submitted: 5140000 (2927540 pages from the captures)
[hos sead::MainThread] vk: readbacks of GPU results: 1548 of 3000 waited for their write-back, 0.04 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos host] vk: mirror syncs (600 submits): 638 dirty pages from their submits' captures, 50434 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5009 runs, 1923.2 MB compared; 41 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15881 in captures, 11 only in the collection, 195 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 684 KB; 0 made on demand (0 MB kept now); c1 stages: at most 245 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 933 binds (127.0 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 763 dirty pages from their submits' captures, 50121 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5016 runs, 1926.4 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15842 in captures, 12 only in the collection, 189 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos host] vk: replay reads of guest memory as submitted: 5160000 (2935200 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 3840794 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: a frame: 130 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 5.20, present 0.16, draw 2.64, descriptors 1.02, texture lookups 0.05, uploads 0.00, pipelines 0.14, programs 0.48, commands 0.77, dispatch 0.52, memory sync 0.34, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 378 (1089 MB): 7.37 MB synced in 1128.9 copies, 71.20 MB written back
[hos host] vk: 20160 presents, 103.0 fps; a frame: 1511 draws (0 skipped), 159 dispatches, 152 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 43831 pipelines (3922986 ms), 0 programs (0 failed), 5018 textures (10647 MB)
[hos host] vk: host time in GPU resource waits: 0.44 ms a frame (fences 0.02, timelines 0.42, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 507.9 full barriers recorded, 189.5 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.7 ms a frame (the longest submission 2.1 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: mirror syncs (600 submits): 747 dirty pages from their submits' captures, 52125 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5006 runs, 1907.4 MB compared; 43 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16451 in captures, 15 only in the collection, 198 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.01, scan 0.04 over 603 ranges, 2380 MB), 366 pages written, 366 copied (1.4 MB)
[hos host] vk: a capture taken (ms): all 0.012; 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 (142 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: replay reads of guest memory as submitted: 5180000 (2942790 pages from the captures)
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 933 binds (127.0 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 707 dirty pages from their submits' captures, 50308 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5103 runs, 2007.0 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15870 in captures, 14 only in the collection, 201 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: program picks: 6685971 kept (2596707 by the last picks' textures, a uniform buffer bound since), 3058 new, 3110970 with other textures (of 9800000)
[hos ModuleSystemWorker2] nvn: 2050000 recordings made, 74757 live copies (21.1 MB), 5026 texture copies; 5 command buffers' segments (4.3 MB)
[hos host] vk: mirror syncs (600 submits): 467 dirty pages from their submits' captures, 50294 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5088 runs, 1909.2 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15810 in captures, 14 only in the collection, 203 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] nvn: CommandBufferSignalEvent #195000: event 0x10103c8048 mode 0 value 194999 (lr 0x80666598)
[hos host] vk: replay reads of guest memory as submitted: 5200000 (2950322 pages from the captures)
[hos ModuleSystemWorker5] nvn: CommandBufferWaitEvent #285000: event 0x10103c8038 mode 0 value 195166 (now 195147) (lr 0x80661a48)
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 934 binds (127.1 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 720 dirty pages from their submits' captures, 50607 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5011 runs, 1923.2 MB compared; 41 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15876 in captures, 10 only in the collection, 185 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos Presentation Thread] vi: 20255 frames presented (101.7 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 509, 15.5-17.8 0, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos sead::MainThread] vk: capture: a submit 0.10 ms (lock 0.01, scan 0.04 over 595 ranges, 2354 MB), 364 pages written, 364 copied (1.4 MB)
[hos host] vk: a capture taken (ms): all 0.012; 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 (156 MB), 7 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 5220000 (2958368 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 ModuleSystemWorker1] vk: readbacks of GPU results: 1547 of 3000 waited for their write-back, 0.04 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 684 KB; 0 made on demand (0 MB kept now); c1 stages: at most 245 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: mirror syncs (600 submits): 716 dirty pages from their submits' captures, 50222 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5020 runs, 1939.9 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15875 in captures, 11 only in the collection, 187 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a frame: 130 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 5.18, present 0.19, draw 2.64, descriptors 1.04, texture lookups 0.05, uploads 0.00, pipelines 0.14, programs 0.48, commands 0.76, dispatch 0.53, memory sync 0.34, queue submit 0.36, slot wait 0.04
[hos host] vk: a frame: mirrors 379 (1089 MB): 7.40 MB synced in 1128.4 copies, 71.32 MB written back
[hos host] vk: 20280 presents, 103.4 fps; a frame: 1510 draws (0 skipped), 159 dispatches, 152 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 43831 pipelines (3922986 ms), 0 programs (0 failed), 5018 textures (10647 MB)
[hos host] vk: host time in GPU resource waits: 0.42 ms a frame (fences 0.02, timelines 0.40, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 508.1 full barriers recorded, 189.5 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: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.8 ms a frame (the longest submission 2.1 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: readback pages refreshed from guest memory: 3908171 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: queues ordered by their events: 108000 submissions stopped at a wait, 107983 went on once it was reached, 14 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 1 (stopped 51.87 ms at most); waits met 286473, not met 0; captures at a CPU event's wait 8018, 0 of them at a wait met on arrival
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 934 binds (127.1 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 697 dirty pages from their submits' captures, 52046 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5007 runs, 1921.2 MB compared; 43 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16489 in captures, 12 only in the collection, 198 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos host] vk: render targets drawn over from another start, let go: 9783 so far (53633 MB)
[hos host] vk: replay reads of guest memory as submitted: 5240000 (2966355 pages from the captures)
[hos ModuleSystemWorker4] nvn: DrawMeshTasksIndirect: 420000 so far (x1 0x118687b040 x2 0 x3 0x118687b040 x4 0x8b040)
[hos host] vk: program picks: 6836677 kept (2671892 by the last picks' textures, a uniform buffer bound since), 3058 new, 3160264 with other textures (of 10000000)
[hos host] vk: mirror syncs (600 submits): 706 dirty pages from their submits' captures, 50710 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5183 runs, 2010.1 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15891 in captures, 11 only in the collection, 193 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 934 binds (127.1 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 660 dirty pages from their submits' captures, 50169 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5012 runs, 1923.2 MB compared; 41 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15900 in captures, 8 only in the collection, 201 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.01, scan 0.04 over 594 ranges, 2352 MB), 365 pages written, 365 copied (1.4 MB)
[hos host] vk: a capture taken (ms): all 0.011; 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 (220 MB), 8 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 5260000 (2974216 pages from the captures)
[hos ModuleSystemWorker3] vcores: 398939 takes, 42537 waits (mean 27 us, longest 4.7 ms), 21496 kicks, 21158 yields, 0 starved grants, 0 escapes in 5 s
[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: mirror syncs (600 submits): 757 dirty pages from their submits' captures, 52171 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5185 runs, 2006.7 MB compared; 41 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16474 in captures, 14 only in the collection, 190 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: memory: working set 13872 MB (peak 14362), private 30771 MB, staging 128 MB in 2 arenas
[hos host] vk: memory heaps: heap 0 (device): 175 blocks 14037 MB, 22696 allocations 12592 MB; used 15300 of 23370 MB; heap 1 (system): 17 blocks 960 MB, 39 allocations 840 MB; used 3023 of 48065 MB;
[hos host] vk: image blocks: 1 blocks 128 MB, 5 images 16 MB; 1 free ranges, the largest 111 MB
[hos host] vk: image blocks (render targets): 13 blocks 1664 MB, 155 images 862 MB; 56 free ranges, the largest 80 MB
[hos host] vk: image blocks (textures, compacted): 71 blocks 9088 MB, 4860 images 8970 MB; 2053 free ranges, the largest 112 MB
[hos host] vk: render target images: 7423 reused from the pool, 2823 made new, 2667 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 15300 MB used of 23370 MB allowed (1445 MB of it free in the blocks); textures 10647 MB (budget 11000 MB), 5018 textures
[hos host] vk: textures: decoded ASTC 7743 MB, rendered 1244 MB (73 drawn in one frame, kept: 54858 KB), other 1163 MB; used in the last 8 presents 1627 MB
[hos host] vk:   288 MB: 0x1150c06000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 3698 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 0x34 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10005f65c00 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10005fde200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 2932 presents ago
[hos host] vk:   54 MB: 0x1000657e400 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 2928 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 53 (7694 KB): 242x40x1 NVN 0x25: 4 (151 KB); 260x260x1 NVN 0xd: 39 (6862 KB); 306x104x1 NVN 0x1: 4 (124 KB); 592x49x1 NVN 0x25: 3 (339 KB); 656x113x1 NVN 0x1: 3 (217 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 111 waited for it, 0.18 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 13.0 ms; VRAM 14.9 of 22.8 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: 121 program records loaded (33.2 ms), 12 new graphics pipelines, pipelines from the driver's cache 27, from the start's cache 22, compiled 85; 1957 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (play): 6165 queued, 6165 loaded (median 3.40 ms, 99th percentile 16.85 ms, worst 63.46 ms, 26432 ms and 1349.3 MB read in all); 8420 programs (1390 of them compute variants without their code, 0 graphics programs no note names), 10740 graphics pipelines (0 not in a cache), 178 compute pipelines (526 not in a cache); a first use ran 24 itself and waited for 0 (51.7 ms); 2530 pipelines made at create taken by draws; 3288 create calls with nothing kept for their stages, 0 before the index; units 9453 (6165 ready, 0 loading), 16483 programs alive
[hos host] vk: module identifiers (play): 777 known, 777 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 130 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 5.20, present 0.16, draw 2.63, descriptors 1.02, texture lookups 0.06, uploads 0.00, pipelines 0.14, programs 0.48, commands 0.76, dispatch 0.52, memory sync 0.33, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 379 (1089 MB): 7.39 MB synced in 1130.5 copies, 71.42 MB written back
[hos host] vk: 20400 presents, 102.8 fps; a frame: 1508 draws (0 skipped), 159 dispatches, 152 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 43831 pipelines (3922986 ms), 0 programs (0 failed), 5018 textures (10647 MB)
[hos host] vk: host time in GPU resource waits: 0.37 ms a frame (fences 0.02, timelines 0.34, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 507.7 full barriers recorded, 189.4 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.8 ms a frame (the longest submission 2.2 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: the render thread's whole frames (600 presents): 0 jobs went on before their present after 1000 ms, 0 times a game's thread waited for the work handed over, 0 times for a fence of the frame in the making
[hos host] vk: the render thread (600 presents, 9.80 ms a frame): its jobs 5.72 ms a frame (submits 4.42, presents 0.18, others 1.11), idle 4.08; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 111.00 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); 138 captures at a CPU signal; the submits' waits read in 0.037 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 5.73 ms after its hand-over (8.39 at most)
[hos host] vk: replay reads of guest memory as submitted: 5280000 (2982191 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 684 KB; 0 made on demand (0 MB kept now); c1 stages: at most 245 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 934 binds (127.1 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 626 dirty pages from their submits' captures, 50473 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4999 runs, 1919.8 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15882 in captures, 15 only in the collection, 192 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos host] vk: readback pages refreshed from guest memory: 3974704 pages so far (0 held: may be written ahead in their submission)
[hos ModuleSystemWorker5] nvn: CommandBufferWaitEvent #290000: event 0x10103c8038 mode 0 value 198966 (now 198947) (lr 0x80661a48)
[hos host] vk: mirror syncs (600 submits): 577 dirty pages from their submits' captures, 50193 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5001 runs, 1921.2 MB compared; 42 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15874 in captures, 11 only in the collection, 186 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] vk: readbacks of GPU results: 1548 of 3000 waited for their write-back, 0.04 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos host] vk: replay reads of guest memory as submitted: 5300000 (2990063 pages from the captures)
[hos host] vk: program picks: 6987411 kept (2746939 by the last picks' textures, a uniform buffer bound since), 3058 new, 3209530 with other textures (of 10200000)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.01, scan 0.04 over 593 ranges, 2346 MB), 365 pages written, 365 copied (1.4 MB)
[hos host] vk: a capture taken (ms): all 0.012; 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 (146 MB), 8 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 934 binds (127.1 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 686 dirty pages from their submits' captures, 51645 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5019 runs, 1943.3 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15934 in captures, 8 only in the collection, 196 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #200000: event 0x10103c800c mode 0 value 200000 (lr 0x80660adc)
[hos host] vk: mirror syncs (600 submits): 529 dirty pages from their submits' captures, 51278 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5177 runs, 1986.6 MB compared; 42 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16482 in captures, 8 only in the collection, 195 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 5320000 (2998044 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 934 binds (127.1 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 682 dirty pages from their submits' captures, 51341 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5028 runs, 1943.3 MB compared; 42 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16529 in captures, 7 only in the collection, 192 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos host] vk: a frame: 130 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 0.19, draw 2.75, descriptors 1.07, texture lookups 0.06, uploads 0.00, pipelines 0.14, programs 0.51, commands 0.79, dispatch 0.54, memory sync 0.33, queue submit 0.38, slot wait 0.04
[hos host] vk: a frame: mirrors 379 (1086 MB): 7.38 MB synced in 1130.4 copies, 71.42 MB written back
[hos host] vk: 20520 presents, 100.9 fps; a frame: 1505 draws (0 skipped), 159 dispatches, 151 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 43831 pipelines (3922986 ms), 0 programs (0 failed), 5018 textures (10647 MB)
[hos host] vk: host time in GPU resource waits: 0.26 ms a frame (fences 0.02, timelines 0.23, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 506.7 full barriers recorded, 189.4 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.7 ms a frame (the longest submission 2.2 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: replay reads of guest memory as submitted: 5340000 (3005977 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 684 KB; 0 made on demand (0 MB kept now); c1 stages: at most 245 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: mirror syncs (600 submits): 518 dirty pages from their submits' captures, 51643 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5178 runs, 1988.0 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15903 in captures, 8 only in the collection, 202 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vk: capture: a submit 0.11 ms (lock 0.01, scan 0.04 over 584 ranges, 2266 MB), 378 pages written, 378 copied (1.5 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 (204 MB), 8 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: readback pages refreshed from guest memory: 4041698 pages so far (0 held: may be written ahead in their submission)
[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: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 935 binds (127.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 460 dirty pages from their submits' captures, 52050 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5055 runs, 1966.0 MB compared; 42 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16484 in captures, 5 only in the collection, 198 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos host] vk: program picks: 7139483 kept (2822306 by the last picks' textures, a uniform buffer bound since), 3058 new, 3257458 with other textures (of 10400000)
[hos host] vk: replay reads of guest memory as submitted: 5360000 (3013920 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] nvn: GPU waits in 5 s: 127 sync waits (19 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: mirror syncs (600 submits): 483 dirty pages from their submits' captures, 50769 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5148 runs, 1963.9 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15924 in captures, 6 only in the collection, 201 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] vk: readbacks of GPU results: 1521 of 3000 waited for their write-back, 0.02 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos host] vk: replay reads of guest memory as submitted: 5380000 (3021536 pages from the captures)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #295000: event 0x10103c8038 mode 0 value 202766 (now 202747) (lr 0x80661a48)
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 935 binds (127.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 683 dirty pages from their submits' captures, 51877 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5103 runs, 1986.9 MB compared; 43 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16483 in captures, 9 only in the collection, 192 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos host] vk: a frame: 130 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 5.70, present 0.18, draw 2.82, descriptors 1.09, texture lookups 0.06, uploads 0.00, pipelines 0.15, programs 0.51, commands 0.82, dispatch 0.57, memory sync 0.33, queue submit 0.44, slot wait 0.04
[hos host] vk: a frame: mirrors 380 (1089 MB): 7.40 MB synced in 1131.5 copies, 71.42 MB written back
[hos host] vk: 20640 presents, 90.9 fps; a frame: 1499 draws (0 skipped), 159 dispatches, 150 passes, 130 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 43831 pipelines (3922986 ms), 0 programs (0 failed), 5018 textures (10647 MB)
[hos host] vk: host time in GPU resource waits: 0.16 ms a frame (fences 0.03, timelines 0.13, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 505.7 full barriers recorded, 189.3 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 29.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 16.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 6.7 ms a frame (the longest submission 2.4 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] vk: capture: a submit 0.11 ms (lock 0.01, scan 0.04 over 583 ranges, 2264 MB), 377 pages written, 377 copied (1.5 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 (162 MB), 9 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: mirror syncs (600 submits): 585 dirty pages from their submits' captures, 50384 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5099 runs, 1943.0 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15950 in captures, 6 only in the collection, 191 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: replay reads of guest memory as submitted: 5400000 (3029060 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 684 KB; 0 made on demand (0 MB kept now); c1 stages: at most 245 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 71, 1089 MB with memory of 1504 MB (414 MB spared); blocks given memory later 11675, given back 15041; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 935 binds (127.2 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 635 dirty pages from their submits' captures, 52398 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5179 runs, 1990.0 MB compared; 43 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 16548 in captures, 6 only in the collection, 200 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 112 live (1948 MB), 0 retired still imported (0 MB), 71 retired given back (1536 MB)
[hos ModuleSystemWorker5] nvn: 2100000 recordings made, 74774 live copies (20.9 MB), 5026 texture copies; 5 command buffers' segments (4.3 MB)
[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: 4108623 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: program picks: 7290577 kept (2897518 by the last picks' textures, a uniform buffer bound since), 3058 new, 3306364 with other textures (of 10600000)
[hos host] vk: replay reads of guest memory as submitted: 5420000 (3036704 pages from the captures)
[hos host] vk: mirror syncs (600 submits): 556 dirty pages from their submits' captures, 50425 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 5030 runs, 1945.4 MB compared; 40 pages read into the shadows; 408 MB of shadows; pages the shadows knew that the CPU wrote: 15951 in captures, 9 only in the collection, 195 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] nvn: DrawMeshTasksIndirect: 440000 so far (x1 0x11819a8000 x2 0 x3 0x11819a8000 x4 0xf0040)
[hos host] vk: present 20722 written to <home>\data\jobs\jb_5217997691de66da\runs\market-tour-g0-01-pr\shots/market-tour-c0-pr-01-field.png (the control bridge)
[hos host] control: quit asked by a client
[hos host] vk: program census (close): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 121 program records loaded (33.2 ms), 12 new graphics pipelines, pipelines from the driver's cache 27, from the start's cache 22, compiled 85; 1957 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (close): 6165 queued, 6165 loaded (median 3.40 ms, 99th percentile 16.85 ms, worst 63.46 ms, 26432 ms and 1349.3 MB read in all); 8420 programs (1390 of them compute variants without their code, 0 graphics programs no note names), 10740 graphics pipelines (0 not in a cache), 178 compute pipelines (526 not in a cache); a first use ran 24 itself and waited for 0 (51.7 ms); 2530 pipelines made at create taken by draws; 3288 create calls with nothing kept for their stages, 0 before the index; units 9453 (6165 ready, 0 loading), 16483 programs alive
[hos host] vk: module identifiers (close): 777 known, 777 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: page table census (close): no pool outside its windows, 64-72 GB and 508-512 GB (4303 pools in the table now)
[hos host] control: exiting
[hos host] stop: close work done (the game's saves are not waited for)