Beanstalk Longshot test farm
Menu

Report rp_c020f2e072f697bb fail

#1996 WIP: perf(vk): the render target pool's bound spares the images that came in the last presents (RX_VK_TARGET_POOL_GRACE=0 to disable)

linux+radv on Simon-Olivier/agent-1b2b4f,

This report disagrees with its peers.

kokiri-camera run_frametime_max_ms: -18.0% against peers' +44.9%

Machine

Commit
e88ea1b701b3aeedf155a86e6567e75a2ee4ae6d
Baseline
f1f2dc5bd2821cef061798ba40a043d281ac1707
App
nightly-221-a9505cc3, attempt 2
Game build
e0f3ccfa
os
linux
cpu
AMD Ryzen 7 5700X3D 8-Core Processor
gpu
AMD Radeon RX 9070 XT (RADV GFX1201)
tags
linuxamdradvdgpu
driver
26.1.99
ram_gb
32
vram_mb
16,304
gpu_vendor
amd
os_version
SteamOS (kernel 7.2.7)

Builds

main (baseline)

189.3 s warm

from f1f2dc5bd282

lift
109.8 s
read
9.1 s
finish
15.4 s
compile
46.3 s
shaders
7.4 s

PR

41.2 s warm

Program cache: warm

Scenarios

kokiri-camera RX_VK_IMAGE_REUSE=0 (PR only)

ok
Metric (median)mainPRΔ
exit_code00±0%
value_made8,4958,588+1.1%
value_reused6,7296,980+3.7%
run_hitch_count2927−6.9%
log_forbidden_hits00±0%
log_expected_missing00±0%
run_frametime_max_ms344.28282.42−18%
value_reused_by_size1,8181,758−3.3%
after-turn.hitch_count00±0%
after-turn.vram_mean_mb5,792.65,687.1−1.8%
after-turn.vram_peak_mb6,9336,742−2.8%
after-turn.stutter_count00±0%
after-turn.frametime_max_ms19.0118.39−3.2%
after-turn.frametime_p50_ms16.4716.58+0.7%
after-turn.frametime_p99_ms18.8218.26−3%
after-turn.frametime_1pct_low_ms18.9318.33−3.2%
screenshot_diff—0.8—
value_over_bound_mb—0—

Pictures against main: forest diff 0.8031 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
value_made8,4958,588
value_reused6,7296,980
run_hitch_count2927
log_forbidden_hits00
value_over_bound_mb—0
log_expected_missing00
run_frametime_max_ms344.28282.42
value_reused_by_size1,8181,758
after-turn.hitch_count00
after-turn.vram_mean_mb5,792.65,687.1
after-turn.vram_peak_mb6,9336,742
after-turn.stutter_count00
after-turn.frametime_max_ms19.0118.39
after-turn.frametime_p50_ms16.4716.58
after-turn.frametime_p99_ms18.8218.26
after-turn.frametime_1pct_low_ms18.9318.33
screenshot_diff—0.8

kokiri-camera RX_VK_IMAGE_REUSE=1 (PR only)

ok
Metric (median)mainPRΔ
exit_code00±0%
value_made8,5798,651+0.8%
value_reused7,0447,066+0.3%
run_hitch_count2828±0%
log_forbidden_hits00±0%
log_expected_missing00±0%
run_frametime_max_ms358.64389.1+8.5%
value_reused_by_size1,7541,782+1.6%
after-turn.hitch_count00±0%
after-turn.vram_mean_mb5,733.35,749.1+0.3%
after-turn.vram_peak_mb7,2216,965−3.5%
after-turn.stutter_count00±0%
after-turn.frametime_max_ms18.3518.59+1.3%
after-turn.frametime_p50_ms16.616.51−0.5%
after-turn.frametime_p99_ms18.1618.22+0.4%
after-turn.frametime_1pct_low_ms18.2718.35+0.4%
screenshot_diff—0.87—
value_over_bound_mb—0—

Pictures against main: forest diff 0.8745 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
value_made8,5798,651
value_reused7,0447,066
run_hitch_count2828
log_forbidden_hits00
value_over_bound_mb—0
log_expected_missing00
run_frametime_max_ms358.64389.1
value_reused_by_size1,7541,782
after-turn.hitch_count00
after-turn.vram_mean_mb5,733.35,749.1
after-turn.vram_peak_mb7,2216,965
after-turn.stutter_count00
after-turn.frametime_max_ms18.3518.59
after-turn.frametime_p50_ms16.616.51
after-turn.frametime_p99_ms18.1618.22
after-turn.frametime_1pct_low_ms18.2718.35
screenshot_diff—0.87

Rules

ScenarioCellRuleLimitmainPRΔ %Status
kokiri-cameraRX_VK_IMAGE_REUSE=0 (PR only)crashFalse00—pass
kokiri-cameraRX_VK_IMAGE_REUSE=0 (PR only)screenshot.max_diff0—0.8—fail
kokiri-cameraRX_VK_IMAGE_REUSE=0 (PR only)after-turn.vram_peak_mb.max_delta_pct106,9336,742−2.8%pass
kokiri-cameraRX_VK_IMAGE_REUSE=0 (PR only)after-turn.frametime_p99_ms.max_delta_pct018.8218.26−3%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
kokiri-cameraRX_VK_IMAGE_REUSE=0 (PR only)log_forbidden_hits.max000—pass
kokiri-cameraRX_VK_IMAGE_REUSE=1 (PR only)crashFalse00—pass
kokiri-cameraRX_VK_IMAGE_REUSE=1 (PR only)screenshot.max_diff0—0.87—fail
kokiri-cameraRX_VK_IMAGE_REUSE=1 (PR only)after-turn.vram_peak_mb.max_delta_pct107,2216,965−3.5%pass
kokiri-cameraRX_VK_IMAGE_REUSE=1 (PR only)after-turn.frametime_p99_ms.max_delta_pct018.1618.22+0.4%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
kokiri-cameraRX_VK_IMAGE_REUSE=1 (PR only)log_forbidden_hits.max000—pass
kokiri-cameraRX_VK_IMAGE_REUSE=1 (PR only)value_reused: RX_VK_IMAGE_REUSE=1 vs RX_VK_IMAGE_REUSE=0 (PR)16,9807,066+1.2%passthe PR's own cells: RX_VK_IMAGE_REUSE=1 against RX_VK_IMAGE_REUSE=0 (RX_VK_IMAGE_REUSE=0 (PR only))
kokiri-cameraRX_VK_IMAGE_REUSE=1 (PR only)value_made: RX_VK_IMAGE_REUSE=1 vs RX_VK_IMAGE_REUSE=0 (PR)-18,5888,651+0.7%failthe PR's own cells: RX_VK_IMAGE_REUSE=1 against RX_VK_IMAGE_REUSE=0 (RX_VK_IMAGE_REUSE=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-camera-c0-forest-main-failed.png
kokiri-camera-c0-forest-main-failed.png
kokiri-camera-c0-forest-pr-failed.png
kokiri-camera-c0-forest-pr-failed.png

Log (PR side, end)

[hos host] vk: replay reads of guest memory as submitted: 11400000 (6421940 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 1035, 4228.4 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 2869 dirty pages from their submits' captures, 55250 from earlier ones', 9 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): 1003 dispatches, 42903 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6457 runs, 738.5 MB compared; 137 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 21032 in captures, 0 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055800000 (4 MB, captured, state 2): 9;
[hos host] vk: 4 textures retired (1756 left, 4508 MB): budget 0, drawn over 4, dead targets 0
[hos ModuleSystemWorker1] nvn: DrawMeshTasksIndirect: 600000 so far (x1 0x117a4f2040 x2 0x6001d40 x3 0x117a4f2000 x4 0xf33080)
[hos host] vk: 1 textures retired (1755 left, 4507 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 11420000 (6433248 pages from the captures)
[hos ModuleSystemWorker1] vcores: 98965 takes, 13823 waits (mean 72 us, longest 5.4 ms, 0 over HOS_VCORE_LOG_MS), 5934 kicks, 6161 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: program picks: 11808271 kept (6014587 by the last picks' textures, a uniform buffer bound since), 22389 new, 5769339 with other textures (of 17600000)
[hos host] vk: replay reads of guest memory as submitted: 11440000 (6444703 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: 63, 763 MB with memory of 992 MB (228 MB spared); blocks given memory later 1368, given back 19677; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 110 binds (163.4 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 19 taken from the spares, 67 made new; 23 kept as spares, 0 destroyed; 4 spares now, 264 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 874, 3570.7 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5349 dirty pages from their submits' captures, 50823 from earlier ones', 10 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): 867 dispatches, 40435 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6479 runs, 641.3 MB compared; 81 pages read into the shadows; 260 MB of shadows (125 MB of pages known, in 164 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18625 in captures, 0 only in the collection, 39 only by the replay itself
[hos host] vk:   read live, by pool: 0x1053c00000 (4 MB, captured, state 2): 10;
[hos host] vk: imports: 68 live (1148 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: a frame: 62 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 3674 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.58, present 0.13, draw 4.08, descriptors 1.89, texture lookups 0.15, uploads 0.00, pipelines 0.17, programs 0.58, commands 1.18, dispatch 0.61, memory sync 0.24, queue submit 0.57, slot wait 0.16
[hos host] vk: a frame: mirrors 86 (763 MB): 5.18 MB synced in 976.4 copies, 16.47 MB written back
[hos host] vk: 15960 presents, 59.1 fps; a frame: 1337 draws (0 skipped), 118 dispatches, 157 passes, 144 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.2 ms (recording slots only); 1111 pipelines (118 ms), 651 programs (10 failed), 1771 textures (4692 MB)
[hos host] vk: host time in GPU resource waits: 3.56 ms a frame (fences 0.00, timelines 3.56, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 559.6 full barriers recorded, 289.6 left out (nothing recorded since the one before), 17.1 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 2.9; 0x9: 1.0; 0x11: 45.0; 0x12: 69.1; 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): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 12.1 ms a frame (the longest submission 5.5 ms); 1.9 textures made a frame: colour targets 0.8 (4.6 MB), depth 1.2 (9.5 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.01 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: replay reads of guest memory as submitted: 11460000 (6455765 pages from the captures)
[hos host] vk: 3 textures retired (1768 left, 4655 MB): budget 0, drawn over 3, dead targets 0
[hos host] vk: 9 textures retired (1759 left, 4538 MB): budget 0, drawn over 9, dead targets 0
[hos Presentation Thread] vi: 15982 frames presented (59.7 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 27, 15.5-17.8 237, 17.8-22 35, 22-30 0, 30-40 0, over 40 0
[hos Presentation Thread] vk: capture: a submit 0.08 ms (lock 0.00, scan 0.02 over 1315 ranges, 1858 MB), 221 pages written, 221 copied (0.9 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 0.001, bytes copied 0.000; 24 kept (166 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 11480000 (6467806 pages from the captures)
[hos host] vk: 9 textures retired (1751 left, 4455 MB): budget 0, drawn over 9, dead targets 0
[hos host] vk: texture #18783 0x10009692600 1475x1475 x1 NVN 0x34 the GPU drew, finalized by the game, sampled again at frame 15990 (drawn at 15929)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 184 KB; 0 made on demand (0 MB kept now); c1 stages: at most 134 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 181 sets (ub 2209, sb 1296, cis 6706, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 897, 3664.6 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 4244 dirty pages from their submits' captures, 48912 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): 870 dispatches, 38841 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5813 runs, 642.0 MB compared; 76 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18667 in captures, 0 only in the collection, 39 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: 10 textures retired (1751 left, 4482 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: 7 textures retired (1754 left, 4558 MB): budget 0, drawn over 7, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 11500000 (6478856 pages from the captures)
[hos host] vk: 13 textures retired (1751 left, 4559 MB): budget 0, drawn over 13, dead targets 0
[hos host] vk: readback pages refreshed from guest memory: 4662192 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: 10 textures retired (1760 left, 4656 MB): budget 0, drawn over 0, dead targets 10
[hos host] vk: 22 textures retired (1749 left, 4463 MB): budget 0, drawn over 0, dead targets 22
[hos host] vk: sparse mirrors: 63, 763 MB with memory of 992 MB (228 MB spared); blocks given memory later 1368, given back 19677; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 110 binds (163.4 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 19 taken from the spares, 67 made new; 23 kept as spares, 0 destroyed; 4 spares now, 264 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1058, 4322.4 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 11 dirty pages from their submits' captures, 55244 from earlier ones', 20 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): 1025 dispatches, 41973 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5775 runs, 738.3 MB compared; 242 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 20354 in captures, 7 only in the collection, 46 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055800000 (4 MB, captured, state 2): 20;
[hos sead::MainThread] nvn: GPU waits in 5 s: 197 sync waits (611 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: imports: 68 live (1148 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: replay reads of guest memory as submitted: 11520000 (6490254 pages from the captures)
[hos host] vk: 2 textures retired (1756 left, 4535 MB): budget 0, drawn over 2, dead targets 0
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 972 of 3000 waited for their write-back, 0.27 ms each; 0 looked at the stopped queues first, 2 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: 3617 kept their virtual core (nothing there to wait for), 3295 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: 3 textures retired (1758 left, 4554 MB): budget 0, drawn over 3, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 11540000 (6502158 pages from the captures)
[hos host] vk: 6 textures retired (1756 left, 4525 MB): budget 0, drawn over 6, dead targets 0
[hos Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.02 over 1315 ranges, 1858 MB), 189 pages written, 189 copied (0.7 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 0.001, bytes copied 0.000; 24 kept (188 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: 2 textures retired (1755 left, 4512 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: a frame: 60 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 3674 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.25, present 0.12, draw 3.78, descriptors 1.75, texture lookups 0.14, uploads 0.01, pipelines 0.16, programs 0.55, commands 1.09, dispatch 0.60, memory sync 0.24, queue submit 0.54, slot wait 0.16
[hos host] vk: a frame: mirrors 86 (763 MB): 5.01 MB synced in 950.8 copies, 16.32 MB written back
[hos host] vk: 16080 presents, 60.8 fps; a frame: 1206 draws (0 skipped), 110 dispatches, 155 passes, 144 barriers, 6 submits, 0.0 uploads (0.1 MB), 80 ub updates (56 overlays), GPU-bound wait 0.2 ms (recording slots only); 1111 pipelines (118 ms), 651 programs (10 failed), 1755 textures (4512 MB)
[hos host] vk: host time in GPU resource waits: 3.29 ms a frame (fences 0.00, timelines 3.29, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 544.5 full barriers recorded, 287.8 left out (nothing recorded since the one before), 16.8 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: 2.8; 0x9: 1.0; 0x11: 45.0; 0x12: 69.0; 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): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 11.5 ms a frame (the longest submission 5.6 ms); 1.8 textures made a frame: colour targets 0.7 (4.3 MB), depth 1.1 (8.6 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.12 MB unswizzled on the CPU in 0.01 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: replay reads of guest memory as submitted: 11560000 (6513704 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 1012, 4134.5 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 3325 dirty pages from their submits' captures, 55025 from earlier ones', 6 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): 980 dispatches, 42888 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6694 runs, 729.2 MB compared; 108 pages read into the shadows; 260 MB of shadows (125 MB of pages known, in 164 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 20788 in captures, 4 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool: 0x1053c00000 (4 MB, captured, state 2): 6;
[hos ModuleSystemWorker3] nvn: CommandBufferWaitEvent #360000: event 0x100d6c8030 mode 0 value 269506 (now 269487) (lr 0x80662a68)
[hos host] vk: program picks: 11944261 kept (6085930 by the last picks' textures, a uniform buffer bound since), 22389 new, 5833349 with other textures (of 17800000)
[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: 11580000 (6524996 pages from the captures)
[hos ModuleSystemWorker4] nvn: CommandBufferSignalEvent #270000: event 0x100d6c802c mode 0 value 269999 (lr 0x80667688)
[hos host] vk: replay reads of guest memory as submitted: 11600000 (6536352 pages from the captures)
[hos host] vk: 13 textures retired (1766 left, 4646 MB): budget 0, drawn over 3, dead targets 10
[hos host] vk: sparse mirrors: 63, 763 MB with memory of 992 MB (228 MB spared); blocks given memory later 1368, given back 19677; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 110 binds (163.4 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 19 taken from the spares, 67 made new; 23 kept as spares, 0 destroyed; 4 spares now, 264 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 828, 3382.7 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5943 dirty pages from their submits' captures, 49188 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): 841 dispatches, 39313 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6281 runs, 609.4 MB compared; 71 pages read into the shadows; 260 MB of shadows (123 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18366 in captures, 0 only in the collection, 36 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 68 live (1148 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: 4 textures retired (1762 left, 4596 MB): budget 0, drawn over 4, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 11620000 (6548434 pages from the captures)
[hos host] vk: 10 textures retired (1752 left, 4479 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: texture #19052 0x10009692600 1473x1473 x1 NVN 0x34 the GPU drew, finalized by the game, sampled again at frame 16145 (drawn at 16084)
[hos host] vk: 10 textures retired (1749 left, 4472 MB): budget 0, drawn over 10, dead targets 0
[hos Presentation Thread] vk: capture: a submit 0.08 ms (lock 0.00, scan 0.02 over 1315 ranges, 1858 MB), 216 pages written, 216 copied (0.8 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 0.001, bytes copied 0.000; 25 kept (182 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: 9 textures retired (1750 left, 4512 MB): budget 0, drawn over 9, dead targets 0
[hos host] vk: texture #18984 0x10009692600 1515x1515 x1 NVN 0x34 the GPU drew, finalized by the game, sampled again at frame 16161 (drawn at 16099)
[hos host] vk: write-backs left out (600 submits): 966, 3946.5 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 2620 dirty pages from their submits' captures, 50397 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): 917 dispatches, 39591 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5740 runs, 681.8 MB compared; 131 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18777 in captures, 0 only in the collection, 42 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 11640000 (6559685 pages from the captures)
[hos host] vk: 6 textures retired (1752 left, 4574 MB): budget 0, drawn over 6, dead targets 0
[hos host] vk: 8 textures retired (1752 left, 4590 MB): budget 0, drawn over 8, dead targets 0
[hos host] vk: 10 textures retired (1762 left, 4702 MB): budget 0, drawn over 0, dead targets 10
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 968 of 3000 waited for their write-back, 0.27 ms each; 0 looked at the stopped queues first, 2 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: 3777 kept their virtual core (nothing there to wait for), 2879 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: replay reads of guest memory as submitted: 11660000 (6570477 pages from the captures)
[hos host] vk: memory: working set 6352 MB (peak 6496), private 1286 MB, staging 48 MB in 3 arenas, 26 made so far (0 ms)
[hos host] vk: memory heaps: heap 0 (system): 22 blocks 744 MB, 38 allocations 639 MB; used 1908 of 15793 MB; heap 1 (device): 210 blocks 6589 MB, 198353 allocations 6211 MB; used 6773 of 15264 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 4 images 12 MB; 1 free ranges, the largest 19 MB
[hos host] vk: image blocks (render targets): 47 blocks 1504 MB, 185 images 1273 MB; 122 free ranges, the largest 10 MB
[hos host] vk: image blocks (textures, compacted): 73 blocks 2336 MB, 1582 images 2270 MB; 307 free ranges, the largest 28 MB
[hos host] vk: render target images: 7066 reused from the pool, 8651 made new, 8463 destroyed from it, 1782 reused by size (on a pooled one's memory, RX_VK_TARGET_REUSE_MEMORY=1), 219 missed after their image was taken by size; it holds 31 (366 MB); at most 0 MB over its bound after a trim since the last line (RX_VK_IMAGE_REUSE's grace)
[hos host] vk: render targets not reused by size: 2572 with the pool empty (0 of them with targets still in the dead queue), 1008 with none in it a present yet, 0 with none of a kind they can use, 3019 with all smaller (the largest 24% short on average), 1150 with all bigger (over 1.5 times), 902 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 8463 idle
[hos host] vk: VRAM: 6773 MB used of 15264 MB allowed (378 MB of it free in the blocks); textures 4479 MB (budget 11000 MB), 1750 textures
[hos host] vk: budget: VRAM 6589 of 15264 MB (6773 MB by the driver, 183 MB unaccounted): textures 4479, mirrors 763, blocks free 378, target pool 366, shadows 260, other buffers 233, other images 60, staging 48; RAM 1813 MB (the process's private memory 1286 MB): imports 1148, capture spares 450, captures 182, capture table 26, other buffers 7, others 0
[hos host] vk: textures: decoded ASTC 2472 MB, rendered 1424 MB (73 drawn in one frame, kept: 45510 KB), other 401 MB; used in the last 8 presents 2488 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; 511 MB of blocks hashed in 52 ms
[hos host] vk:   288 MB: 0x1147c16000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a5b22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a22aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x113cba9000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0xa, last used 14187 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0x34 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: 0x10b4ad6c00 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 60 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 31 (9405 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 62x41x1 NVN 0x25: 1 (9 KB); 74x32x1 NVN 0x25: 1 (9 KB); 90x34x1 NVN 0x25: 1 (11 KB); 126x105x1 NVN 0x1: 1 (12 KB); 138x96x1 NVN 0x1: 1 (12 KB); 154x98x1 NVN 0x1: 1 (14 KB); 198x40x1 NVN 0x25: 2 (61 KB); 242x40x1 NVN 0x25: 3 (113 KB); 260x260x1 NVN 0xd: 2 (351 KB); 262x104x1 NVN 0x1: 3 (79 KB); 306x104x1 NVN 0x1: 3 (93 KB); 592x49x1 NVN 0x25: 3 (339 KB); 656x113x1 NVN 0x1: 3 (217 KB); 1466x282x1 NVN 0x25: 5 (8074 KB);
[hos host] vk: 2 textures retired (1748 left, 4456 MB): budget 0, drawn over 1, dead targets 1
[hos host] vk: malloc: 1457 MB in arenas (1107 MB in use, 349 MB free, 0 MB of it at the tops), 26 MB mapped
[hos host] vk: malloc_trim: 349 MB free in the arenas now, the process's private memory 1286 -> 1281 MB (4 ms)
[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): 503 waited for it, 2.54 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 20.8 ms; VRAM 6.6 of 14.9 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), 806 new graphics pipelines, pipelines from the driver's cache 987, from the start's cache 0, compiled 0; 2899 draws and dispatches left out, 552 waits (2880.8 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: 63 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 3674 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.72, present 0.13, draw 4.18, descriptors 1.96, texture lookups 0.15, uploads 0.01, pipelines 0.18, programs 0.58, commands 1.21, dispatch 0.61, memory sync 0.25, queue submit 0.57, slot wait 0.16
[hos host] vk: a frame: mirrors 86 (763 MB): 5.20 MB synced in 980.3 copies, 16.41 MB written back
[hos host] vk: 16200 presents, 59.9 fps; a frame: 1386 draws (0 skipped), 116 dispatches, 159 passes, 144 barriers, 6 submits, 0.0 uploads (0.1 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 1111 pipelines (118 ms), 651 programs (10 failed), 1750 textures (4479 MB)
[hos host] vk: host time in GPU resource waits: 3.76 ms a frame (fences 0.00, timelines 3.76, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 565.9 full barriers recorded, 292.0 left out (nothing recorded since the one before), 17.2 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: 2.8; 0x9: 1.0; 0x11: 45.0; 0x12: 69.1; 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): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 11.8 ms a frame (the longest submission 5.4 ms); 1.8 textures made a frame: colour targets 0.8 (4.6 MB), depth 1.1 (9.0 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.12 MB unswizzled on the CPU in 0.01 ms
[hos host] vk: the render thread (600 presents, 16.68 ms a frame): its jobs 7.01 ms a frame (submits 0.70, presents 0.13, others 6.17), idle 9.67; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 1.94 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); 592 captures at a CPU signal; the submits' waits read in 0.040 ms a frame; readbacks: 2994 next to a texture's storage (no catch-up), 0 in one, 0 read exactly (after the work handed over: among the catch-ups); a present made 6.38 ms after its hand-over (9.86 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 1.91 + 0.47 (the window's copy is in the idle), after it 1.62 + 0.71; after a frame's last recording that reads or writes the game's memory come 1.01 recordings (0.02 ms of GPU time), after its last that reads it 1.01 (0.02 ms)
[hos host] vk: frame ahead (600 presents): 7.42 parts a frame recorded in the signal's job, 75.6% of them took nothing that came with the signal; of the others, 1.80 a frame took pages of the signal's capture into a mirror (80.0 pages a frame) and 0.01 pages read as guest memory had them (0.1 pages a frame); a frame's first part took nothing in 194 of 592 frames, its first three in 194
[hos host] vk: frame ahead, what came with the signal: 592 signals with a capture, each of 101.2 pages in 6.6 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: 0.84+0.28, 0.33+0.16, 0.18+0.10, 0.20+0.07, 0.04+0.04, 0.02+0.03, 0.00+0.01, 0.00+0.01, 0.00+0.01, 0.00+0.00, 0.00+0.00, 0.00+0.00
[hos host] vk: 4 textures retired (1754 left, 4514 MB): budget 0, drawn over 4, dead targets 0
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 184 KB; 0 made on demand (0 MB kept now); c1 stages: at most 134 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 181 sets (ub 2209, sb 1296, cis 6706, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 63, 763 MB with memory of 992 MB (228 MB spared); blocks given memory later 1368, given back 19677; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 110 binds (163.4 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 19 taken from the spares, 67 made new; 23 kept as spares, 0 destroyed; 4 spares now, 264 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1080, 4414.6 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1266 dirty pages from their submits' captures, 56209 from earlier ones', 9 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, 43301 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6038 runs, 743.3 MB compared; 206 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 21273 in captures, 0 only in the collection, 46 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055800000 (4 MB, captured, state 2): 9;
[hos host] vk: imports: 68 live (1148 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: replay reads of guest memory as submitted: 11680000 (6582324 pages from the captures)
[hos host] vk: 2 textures retired (1757 left, 4542 MB): budget 0, drawn over 2, dead targets 0
[hos ModuleSystemWorker1] nvn: 1950000 recordings made, 49488 live copies (15.5 MB), 848 texture copies; 7 command buffers' segments (4.4 MB)
[hos host] vk: 3 textures retired (1757 left, 4537 MB): budget 0, drawn over 3, dead targets 0
[hos host] vk: readback pages refreshed from guest memory: 4735791 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 11700000 (6593579 pages from the captures)
[hos BaseProcExecutor0] vcores: 98106 takes, 13684 waits (mean 72 us, longest 4.3 ms, 0 over HOS_VCORE_LOG_MS), 5807 kicks, 6092 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: program picks: 12080882 kept (6158026 by the last picks' textures, a uniform buffer bound since), 22389 new, 5896728 with other textures (of 18000000)
[hos host] vk: 12 textures retired (1755 left, 4511 MB): budget 0, drawn over 2, dead targets 10
[hos Presentation Thread] vk: capture: a submit 0.08 ms (lock 0.00, scan 0.02 over 1315 ranges, 1858 MB), 193 pages written, 193 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.005; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 24 kept (186 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: write-backs left out (600 submits): 944, 3854.3 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 3487 dirty pages from their submits' captures, 53334 from earlier ones', 6 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): 923 dispatches, 41528 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6687 runs, 701.6 MB compared; 98 pages read into the shadows; 260 MB of shadows (125 MB of pages known, in 164 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 19590 in captures, 0 only in the collection, 42 only by the replay itself
[hos host] vk:   read live, by pool: 0x1053c00000 (4 MB, captured, state 2): 6;
[hos host] vk: replay reads of guest memory as submitted: 11720000 (6604756 pages from the captures)
[hos host] vk: queues ordered by their events: 120000 submissions stopped at a wait, 119760 went on once it was reached, 236 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 17 (stopped 50.06 ms at most); waits met 363900, not met 0; captures at a CPU event's wait 14155, 0 of them at a wait met on arrival
[hos host] vk: 10 textures retired (1761 left, 4557 MB): budget 0, drawn over 0, dead targets 10
[hos host] vk: texture slots registered again at another size: 1400000 (of 1617916 registrations; slot 28798 512x512 -> 1280x720)
[hos host] vk: replay reads of guest memory as submitted: 11740000 (6615971 pages from the captures)
[hos host] vk: 12 textures retired (1765 left, 4643 MB): budget 0, drawn over 2, dead targets 10
[hos Presentation Thread] vi: 16283 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 30, 15.5-17.8 241, 17.8-22 30, 22-30 0, 30-40 0, over 40 0
[hos ModuleSystemWorker1] sync: object cache (HOS_OBJ_CACHE=16384 a thread, 35 made): 15931071 lookups, 131570 missed (0.83 %)
[hos host] vk: sparse mirrors: 63, 763 MB with memory of 992 MB (228 MB spared); blocks given memory later 1368, given back 19677; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 110 binds (163.4 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 19 taken from the spares, 67 made new; 23 kept as spares, 0 destroyed; 4 spares now, 264 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 828, 3382.7 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5444 dirty pages from their submits' captures, 48126 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): 820 dispatches, 38411 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6146 runs, 606.5 MB compared; 66 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 162 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17768 in captures, 0 only in the collection, 36 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 68 live (1148 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: 8 textures retired (1758 left, 4539 MB): budget 0, drawn over 8, dead targets 0
[hos ModuleSystemWorker3] nvn: CommandBufferWaitEvent #365000: event 0x100d6c8038 mode 0 value 273308 (now 273289) (lr 0x80661fac)
[hos host] vk: replay reads of guest memory as submitted: 11760000 (6627916 pages from the captures)
[hos host] vk: 10 textures retired (1750 left, 4459 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: 8 textures retired (1752 left, 4510 MB): budget 0, drawn over 8, dead targets 0
[hos host] vk: texture #18283 0x10009692600 1477x1477 x1 NVN 0x34 the GPU drew, finalized by the game, sampled again at frame 16310 (drawn at 16248)
[hos host] vk: 8 textures retired (1752 left, 4568 MB): budget 0, drawn over 8, dead targets 0
[hos host] vk: a frame: 64 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 3674 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 7.05, present 0.13, draw 4.44, descriptors 2.07, texture lookups 0.15, uploads 0.03, pipelines 0.19, programs 0.62, commands 1.28, dispatch 0.64, memory sync 0.25, queue submit 0.59, slot wait 0.14
[hos host] vk: a frame: mirrors 86 (763 MB): 5.33 MB synced in 995.7 copies, 16.57 MB written back
[hos host] vk: 16320 presents, 60.2 fps; a frame: 1493 draws (0 skipped), 123 dispatches, 160 passes, 144 barriers, 6 submits, 0.0 uploads (0.1 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 1111 pipelines (118 ms), 651 programs (10 failed), 1753 textures (4581 MB)
[hos host] vk: host time in GPU resource waits: 4.44 ms a frame (fences 0.00, timelines 4.44, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 579.3 full barriers recorded, 293.6 left out (nothing recorded since the one before), 17.3 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: 2.8; 0x9: 1.0; 0x11: 45.0; 0x12: 68.9; 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): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 12.3 ms a frame (the longest submission 5.5 ms); 1.2 textures made a frame: colour targets 0.5 (2.8 MB), depth 0.7 (6.3 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.06 MB unswizzled on the CPU in 0.03 ms
[hos host] vk: replay reads of guest memory as submitted: 11780000 (6639297 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.08 ms (lock 0.00, scan 0.02 over 1315 ranges, 1858 MB), 211 pages written, 211 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.007; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (216 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: 10 textures retired (1751 left, 4587 MB): budget 0, drawn over 10, dead targets 0
[hos sead::MainThread] vk: readbacks of GPU results: 974 of 3000 waited for their write-back, 0.26 ms each; 0 looked at the stopped queues first, 1 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos sead::MainThread] vk: invalidates since the last readbacks line: 3663 kept their virtual core (nothing there to wait for), 2993 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: write-backs left out (600 submits): 1035, 4228.4 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1660 dirty pages from their submits' captures, 53154 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): 994 dispatches, 41402 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5742 runs, 723.5 MB compared; 176 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 20198 in captures, 0 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: 2 textures retired (1758 left, 4666 MB): budget 0, drawn over 2, dead targets 0
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: present 16342 written to /<user>/<user>/Beanstalk/jobs/jb_88e0af6e692dea69/runs/kokiri-camera-g1-01-pr/shots/kokiri-camera-c1-pr-01-forest.png (the control bridge)
[hos host] vk: recorder: 5127369 packets run in 10.0 s (busy 1.37 s, spun idle 0.13 s, slept 12286 times, 4650 of them under 200 us), 0 calls made directly (no room); 17693 batches over 601 frames (29.4 a frame, 290 packets a batch), 8646 parts ended and submitted by it; the render thread (the replay lock's holder) waited for it 112 times, 15.07 ms (0.2 times, 0.03 ms a frame); every thread 362 times, 37.56 ms in all, the longest 0.72 ms; vkDestroyImageView 215/24.99 ms (max 0.72), vkQueueSubmit2 101/12.21 ms (max 0.48), the queue's lock 45/0.34 ms (max 0.02), vmaDestroyImage 1/0.01 ms (max 0.01)
[hos host] vk: recorder: 72 waits given up under the queue's lock, the recorder waiting for it
[hos host] vk: recorder: of its busy 1.37 s, 0.25 s in parts' ends and submits (the longest 0.16 ms), 0.01 s of that blocked on the queue's lock (84 times, the longest 0.15 ms), so 0.24 s in the ends, submits and holds; 1.11 s running the other packets; 8531 packets a frame
[hos host] vk: 1 textures retired (1761 left, 4683 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 11800000 (6650001 pages from the captures)
[hos host] vk: texture #19272 0x10009692600 1920x1920 x1 NVN 0x34 the GPU drew, finalized by the game, sampled again at frame 16358 (drawn at 16297)
[hos host] vk: 1 textures retired (1760 left, 4674 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: 5 textures retired (1755 left, 4628 MB): budget 0, drawn over 5, dead targets 0
[hos host] vk: 8 textures retired (1747 left, 4546 MB): budget 0, drawn over 8, dead targets 0
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #275000: event 0x100d6c8038 mode 0 value 274999 (lr 0x8066189c)
[hos host] vk: sparse mirrors: 63, 763 MB with memory of 992 MB (228 MB spared); blocks given memory later 1368, given back 19677; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 110 binds (163.4 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 19 taken from the spares, 67 made new; 23 kept as spares, 0 destroyed; 4 spares now, 264 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1127, 4604.3 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 43 dirty pages from their submits' captures, 57141 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): 1049 dispatches, 44106 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6239 runs, 787.7 MB compared; 249 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 21815 in captures, 2 only in the collection, 50 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 68 live (1148 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: 6 textures retired (1741 left, 4459 MB): budget 0, drawn over 6, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 11820000 (6660021 pages from the captures)
[hos sead::MainThread] nvn: DrawMeshTasksIndirect: 620000 so far (x1 0x117da08010 x2 0 x3 0x117da08000 x4 0x9b040)
[hos host] vk: program picks: 12215423 kept (6226556 by the last picks' textures, a uniform buffer bound since), 22389 new, 5962187 with other textures (of 18200000)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: texture #19293 0x1000a4a2800 384x384 x1 NVN 0x34 the GPU drew, finalized by the game, sampled again at frame 16404 (drawn at 16343)
[hos Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.03 over 1315 ranges, 1858 MB), 181 pages written, 181 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.002; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (198 MB), 9 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: 3 textures retired (1738 left, 4415 MB): budget 0, drawn over 3, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 11840000 (6670003 pages from the captures)
[hos host] vk: reaper: 457 images (4175 MB) destroyed in 10 s, 0 of them where they were handed over (its cap or VRAM pressure); the queue at most 16 (246 MB)
[hos host] vk: render targets drawn over from another start, let go: 13093 so far (87826 MB), 11901 of them (79720 MB) between the passes
[hos host] vk: 9 textures retired (1730 left, 4305 MB): budget 0, drawn over 9, dead targets 0
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 184 KB; 0 made on demand (0 MB kept now); c1 stages: at most 134 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 181 sets (ub 2209, sb 1296, cis 6706, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 1173, 4792.2 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 56 dirty pages from their submits' captures, 59506 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): 1077 dispatches, 45919 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6517 runs, 819.8 MB compared; 258 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 22964 in captures, 1 only in the collection, 50 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: 10 textures retired (1720 left, 4165 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: a frame: 59 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 3674 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.62, present 0.21, draw 2.62, descriptors 1.21, texture lookups 0.08, uploads 0.01, pipelines 0.11, programs 0.39, commands 0.75, dispatch 0.57, memory sync 0.23, queue submit 0.53, slot wait 0.23
[hos host] vk: a frame: mirrors 86 (763 MB): 4.57 MB synced in 901.6 copies, 16.07 MB written back
[hos host] vk: 16440 presents, 60.0 fps; a frame: 754 draws (0 skipped), 98 dispatches, 145 passes, 144 barriers, 6 submits, 0.0 uploads (0.1 MB), 80 ub updates (56 overlays), GPU-bound wait 0.2 ms (recording slots only); 1111 pipelines (118 ms), 651 programs (10 failed), 1720 textures (4165 MB)
[hos host] vk: host time in GPU resource waits: 2.70 ms a frame (fences 0.00, timelines 2.70, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 487.7 full barriers recorded, 276.4 left out (nothing recorded since the one before), 17.2 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: 68.7; 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): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 10.3 ms a frame (the longest submission 3.8 ms); 0.4 textures made a frame: colour targets 0.1 (0.9 MB), depth 0.3 (2.4 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.06 MB unswizzled on the CPU in 0.01 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: 10 textures retired (1710 left, 4041 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: 10 textures retired (1700 left, 3934 MB): budget 0, drawn over 10, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 11860000 (6680045 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 4809317 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: 1 textures retired (1699 left, 3923 MB): budget 0, drawn over 1, dead targets 0
[hos ModuleSystemWorker3] vk: readbacks of GPU results: 980 of 3000 waited for their write-back, 0.40 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker3] vk: invalidates since the last readbacks line: 3435 kept their virtual core (nothing there to wait for), 2965 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: sparse mirrors: 63, 763 MB with memory of 992 MB (228 MB spared); blocks given memory later 1368, given back 19677; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 110 binds (163.4 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 19 taken from the spares, 67 made new; 23 kept as spares, 0 destroyed; 4 spares now, 264 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1150, 4698.2 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 69 dirty pages from their submits' captures, 58293 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1056 dispatches, 45047 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6399 runs, 803.8 MB compared; 254 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 22471 in captures, 0 only in the collection, 51 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 68 live (1148 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos host] vk: mirror of pool 0x7f78800000 (4 MB) given back: idle, 6027 submissions since its sync
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #370000: event 0x100d6c8030 mode 0 value 277106 (now 277087) (lr 0x80662a68)
[hos host] vk: replay reads of guest memory as submitted: 11880000 (6690089 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] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.02 over 1315 ranges, 1858 MB), 180 pages written, 180 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.002; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (210 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: 1 textures retired (1698 left, 3912 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: write-backs left out (600 submits): 1150, 4698.2 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 25 dirty pages from their submits' captures, 57968 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1056 dispatches, 44886 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6398 runs, 803.8 MB compared; 256 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 22185 in captures, 0 only in the collection, 50 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 11900000 (6700121 pages from the captures)
[hos tera::FileResourceMgr] vcores: 97695 takes, 12984 waits (mean 64 us, longest 5.4 ms, 0 over HOS_VCORE_LOG_MS), 5621 kicks, 5804 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: a frame: 60 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 3674 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 4.42, present 0.11, draw 2.50, descriptors 1.13, texture lookups 0.05, uploads 0.00, pipelines 0.11, programs 0.38, commands 0.73, dispatch 0.57, memory sync 0.22, queue submit 0.52, slot wait 0.22
[hos host] vk: a frame: mirrors 86 (763 MB): 4.54 MB synced in 899.6 copies, 16.08 MB written back
[hos host] vk: 16560 presents, 60.0 fps; a frame: 721 draws (0 skipped), 98 dispatches, 145 passes, 144 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.2 ms (recording slots only); 1111 pipelines (118 ms), 651 programs (10 failed), 1698 textures (3912 MB)
[hos host] vk: host time in GPU resource waits: 2.93 ms a frame (fences 0.00, timelines 2.93, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 485.6 full barriers recorded, 275.6 left out (nothing recorded since the one before), 17.2 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): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 10.3 ms a frame (the longest submission 3.7 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 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: 11920000 (6710174 pages from the captures)
[hos host] vk: sparse mirrors: 62, 763 MB with memory of 988 MB (224 MB spared); blocks given memory later 1368, given back 19677; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 111 binds (163.4 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 19 taken from the spares, 67 made new; 24 kept as spares, 0 destroyed; 5 spares now, 268 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1172, 4790.5 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 50 dirty pages from their submits' captures, 59248 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 1076 dispatches, 45885 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6505 runs, 819.8 MB compared; 251 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 22706 in captures, 1 only in the collection, 50 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 68 live (1148 MB), 0 retired still imported (0 MB), 10 retired given back (260 MB)
[hos Presentation Thread] vi: 16585 frames presented (60.4 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 12, 15.5-17.8 287, 17.8-22 3, 22-30 0, 30-40 0, over 40 0
[hos Presentation Thread] vk: capture: a submit 0.07 ms (lock 0.00, scan 0.02 over 1315 ranges, 1858 MB), 178 pages written, 178 copied (0.7 MB)
[hos host] vk: a capture taken (ms): all 0.002; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.001, bytes copied 0.000; 25 kept (202 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.03
[hos host] vk: replay reads of guest memory as submitted: 11940000 (6720193 pages from the captures)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 980 of 3000 waited for their write-back, 0.44 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker2] vk: invalidates since the last readbacks line: 3792 kept their virtual core (nothing there to wait for), 3120 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos sead::MainThread] nvn: 2000000 recordings made, 49508 live copies (14.4 MB), 848 texture copies; 4 command buffers' segments (4.2 MB)
[hos host] vk: write-backs left out (600 submits): 1151, 4700.0 MB; 26 bindings followed, 40 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 9 dirty pages from their submits' captures, 58513 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): 1061 dispatches, 45140 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6390 runs, 803.3 MB compared; 268 pages read into the shadows; 260 MB of shadows (124 MB of pages known, in 161 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 22174 in captures, 0 only in the collection, 51 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker5] nvn: CommandBufferSignalEvent #280000: event 0x100d6c8044 mode 0 value 279999 (lr 0x802742f0)
[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: 11960000 (6730553 pages from the captures)
[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), 806 new graphics pipelines, pipelines from the driver's cache 987, from the start's cache 0, compiled 0; 2899 draws and dispatches left out, 552 waits (2880.8 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): 29 in 29 of 1285 program stages, 0 of them near a limit
[hos host] vk: indexed descriptor arrays (close), fragment stages: 29 texture (17-64: 29); 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 (2 KB): 1 pipelines, 1 from the driver's cache, 0 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)