Beanstalk Longshot test farm
Menu

Report rp_969642bb2f9b90e4 fail

#2004 WIP: perf(vk): a compute program's c0 layout kept by each thread between the words it reads (RX_VK_CS_LAYOUT_FAST)

any on enum/agent-5773ed,

This report disagrees with its peers.

kokiri-walk forest.frametime_max_ms: +5.0% against peers' -11.0%
kokiri-walk run_frametime_max_ms: +26.3% against peers' +8.2%
kokiri-walk#1 run_frametime_max_ms: -12.8% against peers' +2.6%

Machine

Commit
614cc3ee060a1523a896d74b52bb5428d3cb7724
Baseline
9dd41326e71094d8b774459dac5154eb5a3306d9
App
nightly-221-a9505cc3, attempt 1
Game build
e0f3ccfa
os
windows
cpu
AMD Ryzen 9 7900X 12-Core Processor
gpu
NVIDIA GeForce RTX 4070
tags
windowsnvidianvidia-proprietarydgpu
driver
616.56.0
ram_gb
64
vram_mb
12,010
gpu_vendor
nvidia
os_version
Windows 10.0.26300

Builds

main (baseline)

0 s warm

from 9dd41326e710

a base kept from before

PR

41.5 s warm

Program cache: warm

Scenarios

kokiri-walk RX_VK_CS_LAYOUT_FAST=0 (PR only)

ok
Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count1512−20%
forest.hitch_count00±0%
log_forbidden_hits00±0%
forest.stutter_count00±0%
log_expected_missing00±0%
run_frametime_max_ms578.05729.84+26.3%
forest.frametime_max_ms19.6820.65+5%
forest.frametime_p50_ms16.616.58−0.1%
forest.frametime_p99_ms18.5718.55−0.1%
forest.frametime_1pct_low_ms18.9618.94−0.1%
screenshot_diff—0.12—

Pictures against main: forest diff 0.1162 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
run_hitch_count1512
forest.hitch_count00
log_forbidden_hits00
forest.stutter_count00
log_expected_missing00
run_frametime_max_ms578.05729.84
forest.frametime_max_ms19.6820.65
forest.frametime_p50_ms16.616.58
forest.frametime_p99_ms18.5718.55
forest.frametime_1pct_low_ms18.9618.94
screenshot_diff—0.12

kokiri-walk RX_VK_CS_LAYOUT_FAST=1 (PR only)

ok
Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count1211−8.3%
forest.hitch_count00±0%
log_forbidden_hits00±0%
forest.stutter_count00±0%
log_expected_missing00±0%
run_frametime_max_ms566.1493.84−12.8%
forest.frametime_max_ms21.1820.35−3.9%
forest.frametime_p50_ms16.6116.6−0%
forest.frametime_p99_ms18.4618.53+0.4%
forest.frametime_1pct_low_ms1919−0%
screenshot_diff—0.09—

Pictures against main: forest diff 0.0896 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
run_hitch_count1211
forest.hitch_count00
log_forbidden_hits00
forest.stutter_count00
log_expected_missing00
run_frametime_max_ms566.1493.84
forest.frametime_max_ms21.1820.35
forest.frametime_p50_ms16.6116.6
forest.frametime_p99_ms18.4618.53
forest.frametime_1pct_low_ms1919
screenshot_diff—0.09

Rules

ScenarioCellRuleLimitmainPRΔ %Status
kokiri-walkRX_VK_CS_LAYOUT_FAST=0 (PR only)crashFalse00—pass
kokiri-walkRX_VK_CS_LAYOUT_FAST=0 (PR only)screenshot.max_diff0.02—0.12—fail
kokiri-walkRX_VK_CS_LAYOUT_FAST=0 (PR only)forest.frametime_p99_ms.max_delta_pct1018.5718.55−0.1%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
kokiri-walkRX_VK_CS_LAYOUT_FAST=0 (PR only)log_forbidden_hits.max000—pass
kokiri-walkRX_VK_CS_LAYOUT_FAST=1 (PR only)crashFalse00—pass
kokiri-walkRX_VK_CS_LAYOUT_FAST=1 (PR only)screenshot.max_diff0.02—0.09—fail
kokiri-walkRX_VK_CS_LAYOUT_FAST=1 (PR only)forest.frametime_p99_ms.max_delta_pct1018.4618.53+0.4%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
kokiri-walkRX_VK_CS_LAYOUT_FAST=1 (PR only)log_forbidden_hits.max000—pass
kokiri-walkRX_VK_CS_LAYOUT_FAST=1 (PR only)forest.frametime_p99_ms: RX_VK_CS_LAYOUT_FAST=1 vs RX_VK_CS_LAYOUT_FAST=0 (PR)518.5518.53−0.1%passthe PR's own cells: RX_VK_CS_LAYOUT_FAST=1 against RX_VK_CS_LAYOUT_FAST=0 (RX_VK_CS_LAYOUT_FAST=0 (PR only))

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 .

kokiri-walk-c0-forest-main-failed.png
kokiri-walk-c0-forest-main-failed.png
kokiri-walk-c0-forest-pr-failed.png
kokiri-walk-c0-forest-pr-failed.png

Log (PR side, end)

[hos host] vk: mirror syncs (600 submits): 107 dirty pages from their submits' captures, 58822 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): 1046 dispatches, 44889 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6923 runs, 778.6 MB compared; 147 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19164 in captures, 2 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 2200000 (983258 pages from the captures)
[hos host] vk: a submission's parts ended: 19901 where the game had ended its render pass, 99 at 512 draws
[hos sead::MainThread] nvn: CommandBufferWaitEvent #95000: event 0x100d6c8048 mode 0 value 69258 (now 69239) (lr 0x806661f8)
[hos host] vk: replay reads of guest memory as submitted: 2220000 (992099 pages from the captures)
[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 3580 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.58, present 0.09, draw 2.64, descriptors 1.14, texture lookups 0.08, uploads 0.00, pipelines 0.11, programs 0.49, commands 0.75, dispatch 0.52, memory sync 0.23, queue submit 0.00, slot wait 0.04
[hos host] vk: a frame: mirrors 84 (796 MB): 4.85 MB synced in 942.7 copies, 16.56 MB written back
[hos host] vk: 5040 presents, 60.0 fps; a frame: 1074 draws (0 skipped), 120 dispatches, 120 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); 892 pipelines (110850 ms), 614 programs (0 failed), 1311 textures (3405 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.03, timelines 0.00, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 361.2 full barriers recorded, 213.4 left out (nothing recorded since the one before), 18.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 41.0 from 25.0 batches at the game's barriers (41.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 134.5, an op 848.0, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 5: 22.0; 48: 7.0;
[hos host] vk: GPU time 7.1 ms a frame (the longest submission 3.8 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.1 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: 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: 55, 796 MB with memory of 1024 MB (227 MB spared); blocks given memory later 126, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 105 binds (17.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 67 made new; 29 kept as spares, 0 destroyed; 12 spares now, 324 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1059, 4324.1 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 99 dirty pages from their submits' captures, 57486 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): 1036 dispatches, 43688 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6887 runs, 778.3 MB compared; 149 pages read into the shadows; 308 MB of shadows (172 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19040 in captures, 0 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1120 MB), 0 retired still imported (0 MB), 22 retired given back (424 MB)
[hos host] vk: render targets drawn over from another start, let go: 136 so far (737 MB), 102 of them (530 MB) between the passes
[hos sead::MainThread] nvn: CommandBufferSignalEvent #70000: event 0x100d6c8048 mode 0 value 69999 (lr 0x80666598)
[hos host] vk: program picks: 2829664 kept (1495991 by the last picks' textures, a uniform buffer bound since), 1304 new, 1369031 with other textures (of 4200000)
[hos host] vk: 1 textures retired (1311 left, 3405 MB): budget 0, drawn over 1, dead targets 0
[hos sead::MainThread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 1255 ranges, 1840 MB), 216 pages written, 216 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.004; 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 (134 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: 1 textures retired (1310 left, 3396 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 2240000 (1000922 pages from the captures)
[hos host] vk: 1 textures retired (1309 left, 3388 MB): budget 0, drawn over 1, dead targets 0
[hos sead::MainThread] vk: readbacks of GPU results: 1064 of 3000 waited for their write-back, 0.00 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: 4272 kept their virtual core (nothing there to wait for), 3152 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: write-backs left out (600 submits): 1103, 4508.6 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 136 dirty pages from their submits' captures, 58609 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1043 dispatches, 44781 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 7018 runs, 794.9 MB compared; 141 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19119 in captures, 1 only in the collection, 47 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 0.03
[hos host] vk: replay reads of guest memory as submitted: 2260000 (1009746 pages from the captures)
[hos ModuleSystemWorker1] nvn: DrawMeshTasksIndirect: 160000 so far (x1 0x117cf20000 x2 0 x3 0x117cf20000 x4 0x9b040)
[hos host] vk: replay reads of guest memory as submitted: 2280000 (1018566 pages from the captures)
[hos host] vk: sparse mirrors: 55, 796 MB with memory of 1024 MB (227 MB spared); blocks given memory later 126, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 105 binds (17.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 67 made new; 29 kept as spares, 0 destroyed; 12 spares now, 324 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1081, 4416.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 101 dirty pages from their submits' captures, 58104 from earlier ones', 0 from guest memory as it was at the replay; 24 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1040 dispatches, 44346 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6901 runs, 778.4 MB compared; 134 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19206 in captures, 0 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1120 MB), 0 retired still imported (0 MB), 22 retired given back (424 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 Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 1255 ranges, 1840 MB), 212 pages written, 212 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.004; 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 (98 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos ModuleSystemWorker4] nvn: 600000 recordings made, 48041 live copies (15.1 MB), 366 texture copies; 4 command buffers' segments (4.2 MB)
[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 3580 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.69, present 0.10, draw 2.64, descriptors 1.13, texture lookups 0.07, uploads 0.00, pipelines 0.10, programs 0.49, commands 0.75, dispatch 0.53, memory sync 0.24, queue submit 0.00, slot wait 0.05
[hos host] vk: a frame: mirrors 84 (796 MB): 4.84 MB synced in 941.9 copies, 16.56 MB written back
[hos host] vk: 5160 presents, 60.0 fps; a frame: 1072 draws (0 skipped), 120 dispatches, 120 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); 892 pipelines (110850 ms), 614 programs (0 failed), 1310 textures (3396 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.03, timelines 0.00, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 361.2 full barriers recorded, 213.4 left out (nothing recorded since the one before), 18.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 41.0 from 25.0 batches at the game's barriers (41.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 134.5, an op 848.0, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 5: 22.0; 48: 7.0;
[hos host] vk: GPU time 7.1 ms a frame (the longest submission 3.9 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.1 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: 1 textures retired (1309 left, 3388 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 2300000 (1027392 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 1059, 4324.1 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 132 dirty pages from their submits' captures, 57493 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): 1035 dispatches, 43746 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6781 runs, 762.0 MB compared; 145 pages read into the shadows; 308 MB of shadows (172 MB of pages known, in 198 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19249 in captures, 0 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: readback pages refreshed from guest memory: 1174533 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: replay reads of guest memory as submitted: 2320000 (1036244 pages from the captures)
[hos rmbl::VibrationThread] hid: 20480 vibration values, strongest 0.00 since the last count
[hos sead::MainThread] nvn: CommandBufferWaitEvent #100000: event 0x100d6c8048 mode 0 value 73058 (now 73039) (lr 0x806661f8)
[hos Presentation Thread] vi: 5226 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 301, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos BaseProcExecutor3] sync: object cache (HOS_OBJ_CACHE=16384 a thread, 34 made): 15287685 lookups, 122768 missed (0.80 %)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1054 of 3000 waited for their write-back, 0.00 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker2] vk: invalidates since the last readbacks line: 3756 kept their virtual core (nothing there to wait for), 3156 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: program picks: 2965371 kept (1569140 by the last picks' textures, a uniform buffer bound since), 1304 new, 1433324 with other textures (of 4400000)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 193 KB; 0 made on demand (0 MB kept now); c1 stages: at most 122 a recording; 0 descriptor pools made when a recording's own ran out (0 slots' pools made whole); a recording's pool took at most 89 sets (ub 1119, sb 593, cis 1543, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 55, 796 MB with memory of 1024 MB (227 MB spared); blocks given memory later 126, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 105 binds (17.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 67 made new; 29 kept as spares, 0 destroyed; 12 spares now, 324 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1081, 4416.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 122 dirty pages from their submits' captures, 58050 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): 1040 dispatches, 44272 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6897 runs, 778.4 MB compared; 132 pages read into the shadows; 308 MB of shadows (171 MB of pages known, in 197 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19237 in captures, 0 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1120 MB), 0 retired still imported (0 MB), 22 retired given back (424 MB)
[hos host] vk: replay reads of guest memory as submitted: 2340000 (1045234 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 1255 ranges, 1840 MB), 215 pages written, 215 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.004; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (122 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: 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: 2360000 (1054362 pages from the captures)
[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 3580 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.56, present 0.10, draw 2.54, descriptors 1.08, texture lookups 0.07, uploads 0.00, pipelines 0.10, programs 0.48, commands 0.73, dispatch 0.52, memory sync 0.23, queue submit 0.00, slot wait 0.04
[hos host] vk: a frame: mirrors 84 (796 MB): 4.85 MB synced in 942.8 copies, 16.56 MB written back
[hos host] vk: 5280 presents, 60.0 fps; a frame: 1074 draws (0 skipped), 120 dispatches, 120 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); 892 pipelines (110850 ms), 614 programs (0 failed), 1310 textures (3396 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.03, timelines 0.00, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 361.3 full barriers recorded, 213.5 left out (nothing recorded since the one before), 18.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 41.0 from 25.0 batches at the game's barriers (41.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 134.5, an op 848.0, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 5: 22.0; 48: 7.0;
[hos host] vk: GPU time 7.1 ms a frame (the longest submission 3.8 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.1 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos BaseProcExecutor1] vcores: 138278 takes, 17175 waits (mean 61 us, longest 5.3 ms, 0 over HOS_VCORE_LOG_MS), 7886 kicks, 7920 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: write-backs left out (600 submits): 1081, 4416.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 160 dirty pages from their submits' captures, 57815 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): 1028 dispatches, 44059 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6906 runs, 778.4 MB compared; 145 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18844 in captures, 0 only in the collection, 47 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 0.03
[hos host] vk: replay reads of guest memory as submitted: 2380000 (1063359 pages from the captures)
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #75000: event 0x100d6c800c mode 0 value 75000 (lr 0x802915c8)
[hos host] vk: 1 textures retired (1309 left, 3388 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: sparse mirrors: 55, 796 MB with memory of 1024 MB (227 MB spared); blocks given memory later 126, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 105 binds (17.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 67 made new; 29 kept as spares, 0 destroyed; 12 spares now, 324 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1081, 4416.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 226 dirty pages from their submits' captures, 58065 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): 1040 dispatches, 44283 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6901 runs, 778.4 MB compared; 131 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19286 in captures, 0 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1120 MB), 0 retired still imported (0 MB), 22 retired given back (424 MB)
[hos host] vk: replay reads of guest memory as submitted: 2400000 (1072509 pages from the captures)
[hos host] vk: 1 textures retired (1308 left, 3379 MB): budget 0, drawn over 1, dead targets 0
[hos sead::MainThread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 1255 ranges, 1840 MB), 212 pages written, 212 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.004; 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 (126 MB), 11 spare buffers, 0 texture pages copied (0 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: 2420000 (1081802 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 1081, 4416.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 150 dirty pages from their submits' captures, 58122 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): 1046 dispatches, 44324 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6900 runs, 778.4 MB compared; 148 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19227 in captures, 0 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1058 of 3000 waited for their write-back, 0.00 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker2] vk: invalidates since the last readbacks line: 3943 kept their virtual core (nothing there to wait for), 2969 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: memory: working set 9461 MB (peak 9620), private 16131 MB, staging 32 MB in 2 arenas, 19 made so far (2 ms)
[hos host] vk: memory heaps: heap 0 (device): 164 blocks 5280 MB, 14708 allocations 4733 MB; used 5459 of 11242 MB; heap 1 (system): 18 blocks 864 MB, 39 allocations 760 MB; used 2049 of 32299 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 5 images 16 MB; 1 free ranges, the largest 15 MB
[hos host] vk: image blocks (render targets): 16 blocks 512 MB, 103 images 337 MB; 36 free ranges, the largest 29 MB
[hos host] vk: image blocks (textures, compacted): 58 blocks 1856 MB, 1197 images 1855 MB; 498 free ranges, the largest 0 MB
[hos host] vk: render target images: 50 reused from the pool, 319 made new, 213 destroyed from it, 23 reused by size (on a pooled one's memory, RX_VK_TARGET_REUSE_MEMORY=1), 4 missed after their image was taken by size; it holds 0 (0 MB)
[hos host] vk: render targets not reused by size: 265 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, 7 with all smaller (the largest 72% short on average), 6 with all bigger (over 1.5 times), 21 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 213 idle
[hos host] vk: VRAM: 5459 MB used of 11242 MB allowed (546 MB of it free in the blocks); textures 3396 MB (budget 7005 MB), 1310 textures
[hos host] vk: budget: VRAM 5280 of 11242 MB (5459 MB by the driver, 178 MB unaccounted): textures 3396, mirrors 796, blocks free 546, shadows 308, other buffers 136, other images 63, staging 32; RAM 1906 MB (the process's private memory 16131 MB): imports 1120, capture spares 628, captures 114, capture table 26, other buffers 18, others 0
[hos host] vk: textures: decoded ASTC 1900 MB, rendered 927 MB (102 drawn in one frame, kept: 54738 KB), other 349 MB; used in the last 8 presents 2337 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; 385 MB of blocks hashed in 19 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 0xa, last used 3951 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: 0x113cba9000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10005192200 3072x3072x1 x3 levels 1 NVN 0x32 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: 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 0.03
[hos Presentation Thread] vk: presents after the CPU's signal (600): 85 waited for it, 0.10 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 20.1 ms; VRAM 5.3 of 11.0 GB
[hos host] vk: program census (play): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 0 program records loaded (0.0 ms), 733 new graphics pipelines, pipelines from the driver's cache 228, from the start's cache 43, compiled 571; 34960 draws and dispatches left out, 505 waits (2703.2 ms)
[hos host] vk: program loads at create (play): 0 queued, 0 loaded (median 0.00 ms, 99th percentile 0.00 ms, worst 0.00 ms, 0 ms and 0.0 MB read in all); 0 programs (0 of them compute variants without their code, 0 graphics programs no note names), 0 graphics pipelines (0 not in a cache), 0 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 0 pipelines made at create taken by draws; 7015 create calls with nothing kept for their stages, 0 before the index; units 7015 (0 ready, 0 loading), 12472 programs alive; 0 noted graphics pipeline probes deferred to first use
[hos host] vk: program loads at create (play): 0 shader modules let go once their programs' noted pipelines were built, 0 programs' modules made again for a new pipeline state (RX_VK_DROP_LOADED_MODULES=0 keeps them)
[hos host] vk: module identifiers (play): 0 known, 0 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 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 3580 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.00, present 0.11, draw 2.83, descriptors 1.22, texture lookups 0.08, uploads 0.00, pipelines 0.11, programs 0.54, commands 0.79, dispatch 0.56, memory sync 0.24, queue submit 0.00, slot wait 0.05
[hos host] vk: a frame: mirrors 84 (796 MB): 4.84 MB synced in 941.9 copies, 16.56 MB written back
[hos host] vk: 5400 presents, 60.0 fps; a frame: 1085 draws (0 skipped), 120 dispatches, 120 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); 892 pipelines (110850 ms), 614 programs (0 failed), 1310 textures (3396 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.03, timelines 0.00, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 361.4 full barriers recorded, 213.4 left out (nothing recorded since the one before), 18.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 41.0 from 25.0 batches at the game's barriers (41.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 134.5, an op 853.1, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 5: 22.0; 48: 7.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 4.0 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.1 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: the render thread (600 presents, 16.67 ms a frame): its jobs 5.02 ms a frame (submits 4.16, presents 0.11, others 0.75), idle 11.64; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 1.89 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); 93 captures at a CPU signal; the submits' waits read in 0.029 ms a frame; readbacks: 2204 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 4.42 ms after its hand-over (7.68 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 7.78 + 0.13 (the window's copy is in the idle), after it 1.25 + 0.40; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.02 ms of GPU time), after its last that reads it 1.00 (0.02 ms)
[hos host] vk: frame ahead (600 presents): 6.66 parts a frame recorded in the signal's job, 83.0% of them took nothing that came with the signal; of the others, 1.13 a frame took pages of the signal's capture into a mirror (5.9 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 52 of 93 frames, its first three in 50
[hos host] vk: frame ahead, what came with the signal: 93 signals with a capture, each of 16.6 pages in 3.1 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: 0.91+0.15, 0.05+0.07, 0.24+0.09, 0.02+0.02, 0.01+0.03, 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
[hos host] vk: replay reads of guest memory as submitted: 2440000 (1090905 pages from the captures)
[hos host] vk: program picks: 3101222 kept (1642424 by the last picks' textures, a uniform buffer bound since), 1304 new, 1497473 with other textures (of 4600000)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #105000: event 0x100d6c8048 mode 0 value 76858 (now 76839) (lr 0x806661f8)
[hos host] vk: sparse mirrors: 55, 796 MB with memory of 1024 MB (227 MB spared); blocks given memory later 126, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 105 binds (17.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 67 made new; 29 kept as spares, 0 destroyed; 12 spares now, 324 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1081, 4416.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 124 dirty pages from their submits' captures, 58049 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1043 dispatches, 44231 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6902 runs, 778.4 MB compared; 157 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19128 in captures, 0 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1120 MB), 0 retired still imported (0 MB), 22 retired given back (424 MB)
[hos host] vk: readback pages refreshed from guest memory: 1248512 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 2460000 (1100146 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.08 ms (lock 0.00, scan 0.03 over 1255 ranges, 1840 MB), 212 pages written, 212 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.004; 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 (150 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 2480000 (1109644 pages from the captures)
[hos host] vk: 1 textures retired (1309 left, 3388 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 193 KB; 0 made on demand (0 MB kept now); c1 stages: at most 122 a recording; 0 descriptor pools made when a recording's own ran out (0 slots' pools made whole); a recording's pool took at most 89 sets (ub 1119, sb 606, cis 1543, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 1080, 4414.6 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 143 dirty pages from their submits' captures, 58161 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): 1044 dispatches, 44308 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6891 runs, 778.4 MB compared; 139 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 198 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19157 in captures, 0 only in the collection, 46 only by the replay itself
[hos host] vk:   read live, by pool:
[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: reaper: 9 images (76 MB) destroyed in 10 s, 0 of them where they were handed over (its cap or VRAM pressure); the queue at most 1 (8 MB)
[hos host] vk: replay reads of guest memory as submitted: 2500000 (1119019 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: sparse mirrors: 55, 796 MB with memory of 1024 MB (227 MB spared); blocks given memory later 126, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 105 binds (17.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 67 made new; 29 kept as spares, 0 destroyed; 12 spares now, 324 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1059, 4324.1 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 138 dirty pages from their submits' captures, 57688 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1041 dispatches, 43874 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6783 runs, 762.0 MB compared; 148 pages read into the shadows; 308 MB of shadows (172 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19243 in captures, 1 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1120 MB), 0 retired still imported (0 MB), 22 retired given back (424 MB)
[hos host] vk: a frame: 54 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 3580 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.65, present 0.10, draw 2.65, descriptors 1.13, texture lookups 0.07, uploads 0.00, pipelines 0.10, programs 0.50, commands 0.75, dispatch 0.52, memory sync 0.24, queue submit 0.00, slot wait 0.05
[hos host] vk: a frame: mirrors 84 (796 MB): 4.84 MB synced in 942.4 copies, 16.56 MB written back
[hos host] vk: 5520 presents, 60.0 fps; a frame: 1082 draws (0 skipped), 120 dispatches, 120 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); 892 pipelines (110850 ms), 614 programs (0 failed), 1310 textures (3396 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.03, timelines 0.00, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 361.4 full barriers recorded, 213.4 left out (nothing recorded since the one before), 18.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 41.0 from 25.0 batches at the game's barriers (41.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 134.5, an op 850.8, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 5: 22.0; 48: 7.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 3.8 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.1 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos Presentation Thread] vi: 5526 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 300, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos host] vk: replay reads of guest memory as submitted: 2520000 (1128237 pages from the captures)
[hos sead::MainThread] vk: readbacks of GPU results: 1057 of 3000 waited for their write-back, 0.00 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: 3955 kept their virtual core (nothing there to wait for), 2957 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 1255 ranges, 1840 MB), 213 pages written, 213 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.004; 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 (116 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos sead::MainThread] nvn: 650000 recordings made, 47980 live copies (14.7 MB), 366 texture copies; 3 command buffers' segments (4.1 MB)
[hos host] vk: 1 textures retired (1309 left, 3388 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: write-backs left out (600 submits): 1081, 4416.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 105 dirty pages from their submits' captures, 58362 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): 1044 dispatches, 44347 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6911 runs, 778.4 MB compared; 135 pages read into the shadows; 308 MB of shadows (172 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19216 in captures, 1 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 2540000 (1137508 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 Presentation Thread] vcores: 138248 takes, 17639 waits (mean 62 us, longest 5.1 ms, 0 over HOS_VCORE_LOG_MS), 8202 kicks, 8146 yields, 0 starved grants, 0 escapes in 5 s
[hos sead::MainThread] nvn: CommandBufferSignalEvent #80000: event 0x100d6c8018 mode 0 value 80000 (lr 0x80660c94)
[hos host] vk: program picks: 3237069 kept (1715484 by the last picks' textures, a uniform buffer bound since), 1304 new, 1561626 with other textures (of 4800000)
[hos host] vk: texture slots registered again at another size: 400000 (of 482168 registrations; slot 16141 1280x720 -> 256x256)
[hos host] vk: recorder: 4614687 packets run in 10.0 s (busy 1.20 s, spun idle 0.11 s, slept 10168 times, 3749 of them under 200 us), 0 calls made directly (no room); 15836 batches over 600 frames (26.4 a frame, 291 packets a batch), 7679 parts ended and submitted by it; the render thread (the replay lock's holder) waited for it 1 times, 0.15 ms (0.0 times, 0.00 ms a frame); every thread 15 times, 1.57 ms in all, the longest 0.20 ms; vkDestroyImageView 15/1.57 ms (max 0.20)
[hos host] vk: recorder: of its busy 1.20 s, 0.29 s in parts' ends and submits (the longest 2.14 ms), 0.00 s of that blocked on the queue's lock (0 times, the longest 0.00 ms), so 0.29 s in the ends, submits and holds; 0.91 s running the other packets; 7691 packets a frame
[hos ModuleSystemWorker3] nvn: DrawMeshTasksIndirect: 180000 so far (x1 0x117b427040 x2 0x6001d40 x3 0x117b427000 x4 0xf33080)
[hos host] vk: replay reads of guest memory as submitted: 2560000 (1146396 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: sparse mirrors: 55, 796 MB with memory of 1024 MB (227 MB spared); blocks given memory later 126, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 105 binds (17.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 67 made new; 29 kept as spares, 0 destroyed; 12 spares now, 324 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1081, 4416.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 160 dirty pages from their submits' captures, 57505 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): 1027 dispatches, 43895 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6891 runs, 778.4 MB compared; 138 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18911 in captures, 0 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1120 MB), 0 retired still imported (0 MB), 22 retired given back (424 MB)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #110000: event 0x100d6c8048 mode 0 value 80658 (now 80639) (lr 0x806661f8)
[hos host] vk: 1 textures retired (1308 left, 3379 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 2580000 (1155292 pages from the captures)
[hos host] vk: a frame: 55 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 3580 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.69, present 0.10, draw 2.65, descriptors 1.15, texture lookups 0.07, uploads 0.00, pipelines 0.10, programs 0.50, commands 0.74, dispatch 0.53, memory sync 0.24, queue submit 0.00, slot wait 0.05
[hos host] vk: a frame: mirrors 84 (796 MB): 4.85 MB synced in 942.6 copies, 16.56 MB written back
[hos host] vk: 5640 presents, 60.0 fps; a frame: 1081 draws (0 skipped), 120 dispatches, 120 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); 892 pipelines (110850 ms), 614 programs (0 failed), 1309 textures (3388 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.03, timelines 0.00, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 361.6 full barriers recorded, 213.5 left out (nothing recorded since the one before), 18.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 41.0 from 25.0 batches at the game's barriers (41.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 134.5, an op 849.8, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 5: 22.0; 48: 7.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 3.8 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.1 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos sead::MainThread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 1255 ranges, 1840 MB), 210 pages written, 210 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.004; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (122 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: render targets drawn over from another start, let go: 136 so far (737 MB), 102 of them (530 MB) between the passes
[hos host] vk: write-backs left out (600 submits): 1080, 4414.6 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 110 dirty pages from their submits' captures, 58186 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1037 dispatches, 44319 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6772 runs, 762.0 MB compared; 140 pages read into the shadows; 308 MB of shadows (172 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19148 in captures, 1 only in the collection, 46 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: readback pages refreshed from guest memory: 1321803 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: queues ordered by their events: 36000 submissions stopped at a wait, 35986 went on once it was reached, 12 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 3 (stopped 50.00 ms at most); waits met 111155, not met 0; captures at a CPU event's wait 2148, 0 of them at a wait met on arrival
[hos host] vk: replay reads of guest memory as submitted: 2600000 (1164164 pages from the captures)
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1061 of 3000 waited for their write-back, 0.00 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: 3898 kept their virtual core (nothing there to wait for), 3014 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: replay reads of guest memory as submitted: 2620000 (1173236 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 193 KB; 0 made on demand (0 MB kept now); c1 stages: at most 122 a recording; 0 descriptor pools made when a recording's own ran out (0 slots' pools made whole); a recording's pool took at most 89 sets (ub 1119, sb 606, cis 1543, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 55, 796 MB with memory of 1024 MB (227 MB spared); blocks given memory later 126, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 105 binds (17.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 67 made new; 29 kept as spares, 0 destroyed; 12 spares now, 324 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1059, 4324.1 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 112 dirty pages from their submits' captures, 56890 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): 1024 dispatches, 43337 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6887 runs, 778.2 MB compared; 130 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18694 in captures, 0 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1120 MB), 0 retired still imported (0 MB), 22 retired given back (424 MB)
[hos host] vk: replay reads of guest memory as submitted: 2640000 (1182049 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 1255 ranges, 1840 MB), 210 pages written, 210 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.004; 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 (128 MB), 8 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: write-backs left out (600 submits): 1081, 4416.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 102 dirty pages from their submits' captures, 58251 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): 1042 dispatches, 44335 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6904 runs, 778.4 MB compared; 136 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19219 in captures, 1 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: program picks: 3372676 kept (1788379 by the last picks' textures, a uniform buffer bound since), 1304 new, 1626019 with other textures (of 5000000)
[hos host] vk: 1 textures retired (1309 left, 3388 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: a frame: 54 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 3580 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.71, present 0.10, draw 2.62, descriptors 1.12, texture lookups 0.07, uploads 0.00, pipelines 0.10, programs 0.49, commands 0.74, dispatch 0.53, memory sync 0.24, queue submit 0.00, slot wait 0.05
[hos host] vk: a frame: mirrors 84 (796 MB): 4.85 MB synced in 942.2 copies, 16.56 MB written back
[hos host] vk: 5760 presents, 60.0 fps; a frame: 1085 draws (0 skipped), 120 dispatches, 120 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); 892 pipelines (110850 ms), 614 programs (0 failed), 1309 textures (3388 MB)
[hos host] vk: host time in GPU resource waits: 0.03 ms a frame (fences 0.03, timelines 0.00, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 361.5 full barriers recorded, 213.5 left out (nothing recorded since the one before), 18.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 45.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 0.0 converted one by one, 41.0 from 25.0 batches at the game's barriers (41.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 0.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 134.5, an op 855.2, the end 7.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: 5: 22.0; 48: 7.0;
[hos host] vk: GPU time 7.3 ms a frame (the longest submission 3.9 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.1 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: replay reads of guest memory as submitted: 2660000 (1191184 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: sparse mirrors: 55, 796 MB with memory of 1024 MB (227 MB spared); blocks given memory later 126, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 105 binds (17.5 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 17 taken from the spares, 67 made new; 29 kept as spares, 0 destroyed; 12 spares now, 324 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1080, 4414.6 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 90 dirty pages from their submits' captures, 58205 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1037 dispatches, 44278 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6888 runs, 778.3 MB compared; 155 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 198 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19118 in captures, 0 only in the collection, 46 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1120 MB), 0 retired still imported (0 MB), 22 retired given back (424 MB)
[hos host] vk: replay reads of guest memory as submitted: 2680000 (1199989 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos sead::MainThread] nvn: CommandBufferWaitEvent #115000: event 0x100d6c8048 mode 0 value 84458 (now 84439) (lr 0x806661f8)
[hos Presentation Thread] vi: 5827 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 301, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos ModuleSystemWorker1] sync: object cache (HOS_OBJ_CACHE=16384 a thread, 34 made): 15512038 lookups, 126119 missed (0.81 %)
[hos host] vk: replay reads of guest memory as submitted: 2700000 (1209196 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 1255 ranges, 1840 MB), 212 pages written, 212 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.004; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (122 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos ModuleSystemWorker5] vk: readbacks of GPU results: 1060 of 3000 waited for their write-back, 0.00 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker5] vk: invalidates since the last readbacks line: 3958 kept their virtual core (nothing there to wait for), 2954 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: write-backs left out (600 submits): 1082, 4418.1 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 116 dirty pages from their submits' captures, 58592 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): 1051 dispatches, 44532 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6909 runs, 778.5 MB compared; 141 pages read into the shadows; 308 MB of shadows (173 MB of pages known, in 199 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19150 in captures, 1 only in the collection, 47 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] nvn: CommandBufferSignalEvent #85000: event 0x100d6c8024 mode 0 value 85000 (lr 0x80665068)
[hos host] control: quit asked by a client
[hos host] vk: program census (close): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 0 program records loaded (0.0 ms), 733 new graphics pipelines, pipelines from the driver's cache 228, from the start's cache 43, compiled 571; 34960 draws and dispatches left out, 505 waits (2703.2 ms)
[hos host] vk: program loads at create (close): 0 queued, 0 loaded (median 0.00 ms, 99th percentile 0.00 ms, worst 0.00 ms, 0 ms and 0.0 MB read in all); 0 programs (0 of them compute variants without their code, 0 graphics programs no note names), 0 graphics pipelines (0 not in a cache), 0 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 0 pipelines made at create taken by draws; 7015 create calls with nothing kept for their stages, 0 before the index; units 7015 (0 ready, 0 loading), 12472 programs alive; 0 noted graphics pipeline probes deferred to first use
[hos host] vk: program loads at create (close): 0 shader modules let go once their programs' noted pipelines were built, 0 programs' modules made again for a new pipeline state (RX_VK_DROP_LOADED_MODULES=0 keeps them)
[hos host] vk: module identifiers (close): 0 known, 0 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: indexed descriptor arrays (close): 20 in 20 of 1147 program stages, 0 of them near a limit
[hos host] vk: indexed descriptor arrays (close), fragment stages: 20 texture (17-64: 20); the longest 64
[hos host] vk: page table census (close): no pool outside its windows, 64-72 GB and 508-512 GB (4524 pools in the table now)
[hos host] vk: pipeline cache: play file 8 saved (3383 KB): 24 pipelines, 12 from the driver's cache, 12 compiled; 0 more from the start's cache
[hos host] control: exiting
[hos host] stop: close work done (a file the game was writing waited for, at most 3 s)