Beanstalk Longshot test farm
Menu

Report rp_e1b0cef198c4c0e7 fail

#1131 perf(lift): on Windows the discovery's walks run in new processes, with the kit's result: a busy PC's lift about a quarter faster

windows on Razytos/agent-d8f902,

Machine

Commit
2a65abcb95c3efb849ec9d784f1fd74e98b60e84
Baseline
71b1edd89bfda56edb6d5337b039ecc0ee8318d1
App
nightly-126-05954779, attempt 1
Game build
e0f3ccfa
os
windows
cpu
AMD Ryzen 7 5800X3D 8-Core Processor
gpu
AMD Radeon RX 9070 XT
tags
windowsamdamd-proprietarydgpu
driver
2.0.416
ram_gb
32
vram_mb
16,304
gpu_vendor
amd
os_version
Windows 10.0.26300

Builds

main (baseline)

2,150.4 s warm

from 71b1edd89bfd

lift
115.5 s
read
552.5 s
finish
96.4 s
compile
1,379.8 s

PR

3,140.6 s warm

lift
98.9 s
read
560.4 s
finish
2,337.8 s
compile
70.8 s

Program cache: warm

Scenarios

lifted-gameplay

ok
Metric (median)mainPRΔ
exit_code00±0%
deku.hitch_count01—
field.hitch_count00±0%
deku.stutter_count02—
field.stutter_count00±0%
deku.frametime_max_ms27.02164.42+508.6%
deku.frametime_p50_ms16.6616.57−0.5%
deku.frametime_p99_ms18.9524.71+30.4%
field.frametime_max_ms24.0930.3+25.8%
field.frametime_p50_ms16.716.65−0.3%
field.frametime_p99_ms19.4820.2+3.7%
deku.frametime_1pct_low_ms20.5440.98+99.6%
field.frametime_1pct_low_ms20.5223.03+12.2%
screenshot_diff—0.15—

Pictures against main: kokiri diff 0.0018, field diff 0.1505, deku diff 0

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
deku.hitch_count01
field.hitch_count00
deku.stutter_count02
field.stutter_count00
deku.frametime_max_ms27.02164.42
deku.frametime_p50_ms16.6616.57
deku.frametime_p99_ms18.9524.71
field.frametime_max_ms24.0930.3
field.frametime_p50_ms16.716.65
field.frametime_p99_ms19.4820.2
deku.frametime_1pct_low_ms20.5440.98
field.frametime_1pct_low_ms20.5223.03
screenshot_diff—0.15

Rules

ScenarioCellRuleLimitmainPRΔ %Status
lifted-gameplaycrashFalse00—pass
lifted-gameplayscreenshot.max_diff0.1—0.15—fail
lifted-gameplayfield.frametime_p99_ms.max_delta_pct2019.4820.2+3.7%pass
lifted-gameplaydeku.frametime_p99_ms.max_delta_pct2018.9524.71+30.4%fail

Log (PR side, end)

[hos host] vk: mirror syncs (600 submits): 1278 dirty pages from their submits' captures, 102089 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 3974 runs, 2589.6 MB compared; 69 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18393 in captures, 11 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 120 live (1924 MB), 1 retired still imported (8 MB), 13 retired given back (684 MB)
[hos host] vk: 2 textures retired (700 left, 1931 MB): budget 2, drawn over 0, dead targets 0
[hos Presentation Thread] vk: capture: a submit 0.23 ms (lock 0.02, scan 0.04 over 590 ranges, 3377 MB), 614 pages written, 614 copied (2.4 MB)
[hos host] vk: a capture taken (ms): all 0.101; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.097, bytes copied 0.000; 25 kept (186 MB), 8 spare buffers, 22278 texture pages copied (87 MB)
[hos host] vk: 12 textures retired (699 left, 2012 MB): budget 12, drawn over 0, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.12, writes collected 0.04
[hos host] vk: replay reads of guest memory as submitted: 1240000 (553444 pages from the captures)
[hos host] vk: mirror of pool 0x103cc00000 (4 MB) given back: idle, 6003 submissions since its sync
[hos host] vk: 25 textures retired (683 left, 1983 MB): budget 25, drawn over 0, dead targets 0
[hos host] vk: mirror syncs (600 submits): 1198 dirty pages from their submits' captures, 103501 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4198 runs, 2677.7 MB compared; 86 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18737 in captures, 14 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vi: 4516 frames presented (58.7 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 35, 15.5-17.8 244, 17.8-22 11, 22-30 2, 30-40 0, over 40 2
[hos host] vk: 1 textures retired (682 left, 1982 MB): budget 1, drawn over 0, dead targets 0
[hos host] vk: readback pages refreshed from guest memory: 1093257 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 127542 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos sead::MainThread] nvn: GPU waits in 5 s: 6 sync waits (33 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos BaseProcExecutor2] vcores: 241121 takes, 31096 waits (mean 40 us, longest 5.0 ms), 16433 kicks, 16344 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: 9 textures retired (723 left, 2033 MB): budget 9, drawn over 0, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1260000 (562460 pages from the captures)
[hos host] vk: sparse mirrors: 53, 734 MB with memory of 1148 MB (413 MB spared); blocks given memory later 4536, given back 12285; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 1926; 354 binds (178.3 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1146 dirty pages from their submits' captures, 103107 from earlier ones', 3 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4314 runs, 2594.5 MB compared; 88 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18939 in captures, 14 only in the collection, 43 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055000000 (4 MB, captured, state 2): 3;
[hos host] vk: imports: 120 live (1924 MB), 1 retired still imported (8 MB), 13 retired given back (684 MB)
[hos host] vk: 9 textures retired (715 left, 1953 MB): budget 9, drawn over 0, dead targets 0
[hos host] vk: a frame: 9 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 76 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.77, present 0.09, draw 3.18, descriptors 1.58, texture lookups 0.40, uploads 0.35, pipelines 0.11, programs 0.38, commands 0.89, dispatch 0.56, memory sync 0.93, queue submit 0.49, slot wait 0.19
[hos host] vk: a frame: mirrors 137 (734 MB): 9.37 MB synced in 1290.0 copies, 60.66 MB written back
[hos host] vk: 4560 presents, 59.9 fps; a frame: 762 draws (0 skipped), 86 dispatches, 199 passes, 146 barriers, 6 submits, 0.9 uploads (1.4 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 42668 pipelines (24270 ms), 0 programs (10 failed), 715 textures (1953 MB)
[hos host] vk: host time in GPU resource waits: 0.51 ms a frame (fences 0.03, timelines 0.48, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 462.3 full barriers recorded, 197.7 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 3.0; 0x8: 3.0; 0x9: 1.0; 0x10: 2.0; 0x11: 43.0; 0x12: 69.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 5.4 ms a frame (the longest submission 3.4 ms); 0.9 textures made a frame
[hos host] vk: a frame: 0.85 MB unswizzled on the CPU in 0.07 ms
[hos ModuleSystemWorker5] nvn: DrawMeshTasksIndirect: 120000 so far (x1 0x117b70c000 x2 0 x3 0x117b70c000 x4 0x230040)
[hos Presentation Thread] vk: capture: a submit 0.24 ms (lock 0.01, scan 0.04 over 591 ranges, 3437 MB), 640 pages written, 640 copied (2.5 MB)
[hos host] vk: a capture taken (ms): all 0.110; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.106, bytes copied 0.000; 25 kept (282 MB), 11 spare buffers, 22867 texture pages copied (89 MB)
[hos ModuleSystemWorker3] vk: readbacks of GPU results: 1369 of 3000 waited for their write-back, 0.03 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos host] vk: 40 textures retired (677 left, 1907 MB): budget 40, drawn over 0, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.10, writes collected 0.04
[hos sead::MainThread] nvn: CommandBufferWaitEvent #75000: event 0x100c31802c mode 0 value 53118 (now 53099) (lr 0x8028c554)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 678 KB; 0 made on demand (0 MB kept now); c1 stages: at most 95 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: mirror syncs (600 submits): 1186 dirty pages from their submits' captures, 103899 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4469 runs, 2666.7 MB compared; 73 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 19555 in captures, 10 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: 1 textures retired (676 left, 1896 MB): budget 1, drawn over 0, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1280000 (571573 pages from the captures)
[hos host] vk: slow submit: 16.0 ms: draw 14.5 descriptors 13.9 texture lookups 13.6 uploads 13.6
[hos host] vk: 21 textures retired (656 left, 1856 MB): budget 21, drawn over 0, dead targets 0
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 115222 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk:   (its written part: 1 of 1 KB readable, 1386 bytes not zero; levels L0)
[hos host] vk: texture 0x10e5288a00 5072x5056x1 x1 levels 1 NVN 0x4b (25043 KB of storage) re-uploaded (its memory was written; flags 0): what the CPU wrote, 1 KB of it
[hos host] vk:   (its written part: 1 of 1 KB readable, 1386 bytes not zero; levels L0)
[hos host] vk: texture 0x10e5288a00 5072x5056x1 x1 levels 1 NVN 0x4b (25043 KB of storage) re-uploaded (its memory was written; flags 0): what the CPU wrote, 1 KB of it
[hos host] vk: 1 textures retired (656 left, 1862 MB): budget 1, drawn over 0, dead targets 0
[hos host] vk: sparse mirrors: 53, 734 MB with memory of 1148 MB (413 MB spared); blocks given memory later 4536, given back 12285; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 2187; 354 binds (178.3 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1088 dirty pages from their submits' captures, 102529 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4048 runs, 2572.9 MB compared; 76 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18644 in captures, 18 only in the collection, 43 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 120 live (1924 MB), 1 retired still imported (8 MB), 13 retired given back (684 MB)
[hos host] vk: 1 textures retired (655 left, 1844 MB): budget 1, drawn over 0, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1300000 (581129 pages from the captures)
[hos host] vk: texture #7 0x1000021c200 960x540x1 x1 NVN 0xd the GPU drew goes with no copy of its texels (live, unused for 3602 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: texture #9 0x10000348400 2560x1440x1 x1 NVN 0x38 the GPU drew goes with no copy of its texels (live, unused for 3602 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: 4 textures retired (651 left, 1777 MB): budget 4, drawn over 0, dead targets 0; GPU-drawn with no copy kept: 2 live, 0 finalized, 15 MB
[hos host] vk: a frame: 9 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 76 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.58, present 0.09, draw 3.00, descriptors 1.41, texture lookups 0.25, uploads 0.20, pipelines 0.11, programs 0.38, commands 0.89, dispatch 0.56, memory sync 0.91, queue submit 0.50, slot wait 0.20
[hos host] vk: a frame: mirrors 137 (734 MB): 9.33 MB synced in 1288.8 copies, 60.24 MB written back
[hos host] vk: 4680 presents, 60.0 fps; a frame: 761 draws (0 skipped), 85 dispatches, 199 passes, 146 barriers, 6 submits, 0.1 uploads (0.7 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 42668 pipelines (24270 ms), 0 programs (10 failed), 651 textures (1777 MB)
[hos host] vk: host time in GPU resource waits: 0.25 ms a frame (fences 0.03, timelines 0.23, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 455.4 full barriers recorded, 197.8 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 3.0; 0x8: 3.0; 0x9: 1.0; 0x10: 2.0; 0x11: 43.0; 0x12: 69.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 5.3 ms a frame (the longest submission 2.0 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.72 MB unswizzled on the CPU in 0.06 ms
[hos Presentation Thread] vk: capture: a submit 0.27 ms (lock 0.02, scan 0.04 over 591 ranges, 3481 MB), 674 pages written, 674 copied (2.6 MB)
[hos host] vk: a capture taken (ms): all 0.126; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.122, bytes copied 0.000; 25 kept (186 MB), 9 spare buffers, 26530 texture pages copied (103 MB)
[hos host] vk: mirror syncs (600 submits): 1308 dirty pages from their submits' captures, 102306 from earlier ones', 1 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4053 runs, 2644.5 MB compared; 66 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18276 in captures, 12 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055000000 (4 MB, captured, state 2): 1;
[hos ModuleSystemWorker5] nvn: CommandBufferSignalEvent #55000: event 0x100c31802c mode 0 value 54999 (lr 0x80667688)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.13, writes collected 0.04
[hos host] vk: program picks: 2551703 kept (1248362 by the last picks' textures, a uniform buffer bound since), 13938 new, 834358 with other textures (of 3400000)
[hos host] vk: replay reads of guest memory as submitted: 1320000 (590619 pages from the captures)
[hos host] vk: sparse mirrors: 53, 734 MB with memory of 1148 MB (413 MB spared); blocks given memory later 4536, given back 12285; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 2448; 354 binds (178.3 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1096 dirty pages from their submits' captures, 102905 from earlier ones', 1 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4079 runs, 2617.4 MB compared; 70 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 19010 in captures, 8 only in the collection, 43 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055000000 (4 MB, captured, state 2): 1;
[hos host] vk: imports: 120 live (1924 MB), 1 retired still imported (8 MB), 13 retired given back (684 MB)
[hos host] vk: 1 textures retired (650 left, 1777 MB): budget 1, drawn over 0, dead targets 0
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 156976 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: readback pages refreshed from guest memory: 1155086 pages so far (0 held: may be written ahead in their submission)
[hos host] vk:   (its written part: 1 of 1 KB readable, 1386 bytes not zero; levels L0)
[hos host] vk: texture 0x10e5288a00 5072x5056x1 x1 levels 1 NVN 0x4b (25043 KB of storage) re-uploaded (its memory was written; flags 0): what the CPU wrote, 1 KB of it
[hos host] vk: replay reads of guest memory as submitted: 1340000 (600155 pages from the captures)
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 1328 of 3000 waited for their write-back, 0.02 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos host] vk: mirror syncs (600 submits): 934 dirty pages from their submits' captures, 103340 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4479 runs, 2684.0 MB compared; 75 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 19240 in captures, 6 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vk: capture: a submit 0.24 ms (lock 0.01, scan 0.04 over 591 ranges, 3477 MB), 644 pages written, 644 copied (2.5 MB)
[hos host] vk: a capture taken (ms): all 0.101; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.097, bytes copied 0.000; 25 kept (186 MB), 12 spare buffers, 8107 texture pages copied (31 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.09, writes collected 0.04
[hos ModuleSystemWorker3] nvn: CommandBufferWaitEvent #80000: event 0x100c31802c mode 0 value 56918 (now 56899) (lr 0x8028c554)
[hos host] vk: memory: working set 15944 MB (peak 15973), private 24359 MB, staging 256 MB in 4 arenas
[hos host] vk: memory heaps: heap 0 (system): 12 blocks 664 MB, 40 allocations 616 MB; used 2612 of 15571 MB; heap 1 (device): 108 blocks 6033 MB, 12719 allocations 3495 MB; used 7074 of 15419 MB;
[hos host] vk: image blocks: 1 blocks 128 MB, 6 images 24 MB; 1 free ranges, the largest 103 MB
[hos host] vk: image blocks (render targets): 8 blocks 1024 MB, 162 images 354 MB; 54 free ranges, the largest 114 MB
[hos host] vk: image blocks (textures, compacted): 19 blocks 2432 MB, 600 images 959 MB; 168 free ranges, the largest 128 MB
[hos host] vk: render target images: 355 reused from the pool, 478 made new, 316 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 7074 MB used of 15419 MB allowed (2537 MB of it free in the blocks); textures 1887 MB (budget 687 MB), 764 textures
[hos host] vk: textures: decoded ASTC 945 MB, rendered 382 MB (87 drawn in one frame, kept: 38148 KB), other 225 MB; used in the last 8 presents 1379 MB
[hos host] vk:   214 MB: 0x10a4772c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a0efa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x11361ac000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x113b7bd000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   47 MB: 0x10e0ea0a00 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 1 presents ago
[hos host] vk:   24 MB: 0x10e5288a00 5072x5056x1 x1 levels 1 NVN 0x4b, last used 1 presents ago
[hos host] vk:   21 MB: 0x100056a8000 2400x2400x1 x1 levels 1 NVN 0x34 rendered, last used 870 presents ago
[hos host] vk:   18 MB: 0x10e3e53a00 2536x2528x1 x3 levels 1 NVN 0x50, last used 1 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 57 (15305 KB): 260x260x1 NVN 0xd: 53 (9325 KB); 360x202x1 NVN 0x25: 1 (284 KB); 720x405x1 NVN 0x29: 2 (4556 KB); 720x405x1 NVN 0x34: 1 (1139 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 17 waited for it, 2.88 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 164.4 ms; VRAM 6.9 of 15.1 GB
[hos host] vk: program census (play): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 27 program records loaded (39.3 ms), 2 new graphics pipelines, pipelines from the driver's cache 70, from the start's cache 0, compiled 2; 36 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (play): 5657 queued, 5657 loaded (median 0.59 ms, 99th percentile 4.47 ms, worst 26.66 ms, 5251 ms and 1199.6 MB read in all); 7651 programs (1190 of them compute variants without their code, 0 graphics programs no note names), 9323 graphics pipelines (0 not in a cache), 724 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 1372 pipelines made at create taken by draws; 2751 create calls with nothing kept for their stages, 0 before the index; units 8408 (5657 ready, 0 loading), 13910 programs alive
[hos host] vk: module identifiers (play): 794 known, 794 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: 9 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 76 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.26, present 0.09, draw 2.72, descriptors 1.21, texture lookups 0.10, uploads 0.05, pipelines 0.10, programs 0.37, commands 0.83, dispatch 0.58, memory sync 0.85, queue submit 0.48, slot wait 0.17
[hos host] vk: a frame: mirrors 137 (734 MB): 9.36 MB synced in 1289.5 copies, 60.93 MB written back
[hos host] vk: 4800 presents, 60.0 fps; a frame: 762 draws (0 skipped), 88 dispatches, 199 passes, 146 barriers, 6 submits, 1.0 uploads (0.2 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 42668 pipelines (24270 ms), 0 programs (10 failed), 764 textures (1887 MB)
[hos host] vk: host time in GPU resource waits: 0.28 ms a frame (fences 0.02, timelines 0.25, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 460.7 full barriers recorded, 197.7 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 3.0; 0x8: 3.0; 0x9: 1.0; 0x10: 2.0; 0x11: 43.0; 0x12: 69.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 5.7 ms a frame (the longest submission 3.0 ms); 0.9 textures made a frame
[hos host] vk: a frame: 0.08 MB unswizzled on the CPU in 0.01 ms
[hos host] vk: the render thread's whole frames (600 presents): 0 jobs went on before their present after 1000 ms, 0 times a game's thread waited for the work handed over, 0 times for a fence of the frame in the making
[hos host] vk: the render thread (600 presents, 16.85 ms a frame): its jobs 7.25 ms a frame (submits 6.92, presents 0.10, others 0.22), idle 9.60; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 151.50 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); 17 captures at a CPU signal; the submits' waits read in 0.024 ms a frame; readbacks: 2400 next to a texture's storage (no catch-up), 0 in one, 0 read exactly (after the work handed over: among the catch-ups); a present made 7.27 ms after its hand-over (157.34 at most)
[hos host] vk: commit limit near: 1402 MB of commit left (reserve 2560 MB); the textures' budget down to 779 MB of their 1937 MB (RX_VK_COMMIT_RESERVE_MB)
[hos host] vk: replay reads of guest memory as submitted: 1360000 (609773 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 678 KB; 0 made on demand (0 MB kept now); c1 stages: at most 95 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 53, 734 MB with memory of 1148 MB (413 MB spared); blocks given memory later 4536, given back 12285; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 2709; 354 binds (178.3 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 894 dirty pages from their submits' captures, 103865 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4301 runs, 2572.9 MB compared; 74 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 19271 in captures, 9 only in the collection, 43 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 120 live (1924 MB), 1 retired still imported (8 MB), 13 retired given back (684 MB)
[hos Presentation Thread] vi: 4817 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 25, 15.5-17.8 250, 17.8-22 24, 22-30 2, 30-40 0, over 40 0
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 72607 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos BaseProcExecutor1] vcores: 236706 takes, 30284 waits (mean 36 us, longest 4.4 ms), 16077 kicks, 15937 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: 59 textures retired (755 left, 1872 MB): budget 59, drawn over 0, dead targets 0
[hos host] vk: 54 textures retired (703 left, 1853 MB): budget 54, drawn over 0, dead targets 0
[hos host] vk: mirror syncs (600 submits): 1232 dirty pages from their submits' captures, 102197 from earlier ones', 4 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4176 runs, 2635.5 MB compared; 71 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18671 in captures, 9 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055000000 (4 MB, captured, state 2): 4;
[hos host] vk: replay reads of guest memory as submitted: 1380000 (619384 pages from the captures)
[hos host] vk: 40 textures retired (665 left, 1806 MB): budget 40, drawn over 0, dead targets 0
[hos host] vk: recorder: 3120719 packets run in 10.0 s (busy 1.00 s, spun idle 2.66 s, slept 4485 times, 1317 of them under 200 us), 0 calls made directly (no room); the replay waited for it 23662 times, 77.38 ms in all, the longest 0.15 ms; vkResetDescriptorPool 8269/30.81 ms (max 0.09), replay_mu unlock 7123/24.75 ms (max 0.15), vkEndCommandBuffer 8270/21.82 ms (max 0.08)
[hos host] vk: 2 textures retired (674 left, 1854 MB): budget 2, drawn over 0, dead targets 0
[hos Presentation Thread] vk: capture: a submit 0.22 ms (lock 0.02, scan 0.04 over 595 ranges, 3511 MB), 577 pages written, 577 copied (2.3 MB)
[hos host] vk: a capture taken (ms): all 0.074; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.070, bytes copied 0.000; 25 kept (218 MB), 11 spare buffers, 15992 texture pages copied (62 MB)
[hos host] vk: 7 textures retired (682 left, 1902 MB): budget 7, drawn over 0, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.08, writes collected 0.04
[hos host] vk: 1 textures retired (681 left, 1902 MB): budget 1, drawn over 0, dead targets 0
[hos host] vk: sparse mirrors: 53, 734 MB with memory of 1148 MB (413 MB spared); blocks given memory later 4536, given back 12285; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 2973; 354 binds (178.3 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1121 dirty pages from their submits' captures, 103601 from earlier ones', 3 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4003 runs, 2657.6 MB compared; 66 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18613 in captures, 10 only in the collection, 43 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055000000 (4 MB, captured, state 2): 3;
[hos host] vk: imports: 120 live (1924 MB), 1 retired still imported (8 MB), 13 retired given back (684 MB)
[hos host] vk: 2 textures retired (681 left, 1878 MB): budget 2, drawn over 0, dead targets 0
[hos ModuleSystemWorker1] nvn: 500000 recordings made, 31430 live copies (8.1 MB), 588 texture copies; 5 command buffers' segments (4.3 MB)
[hos host] vk: replay reads of guest memory as submitted: 1400000 (628951 pages from the captures)
[hos host] vk: a frame: 9 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 76 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.26, present 0.11, draw 2.89, descriptors 1.31, texture lookups 0.15, uploads 0.09, pipelines 0.11, programs 0.37, commands 0.88, dispatch 0.55, memory sync 0.79, queue submit 0.49, slot wait 0.18
[hos host] vk: a frame: mirrors 137 (734 MB): 9.32 MB synced in 1288.5 copies, 60.09 MB written back
[hos host] vk: 4920 presents, 59.8 fps; a frame: 761 draws (0 skipped), 83 dispatches, 199 passes, 146 barriers, 6 submits, 0.7 uploads (0.3 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 42668 pipelines (24270 ms), 0 programs (10 failed), 685 textures (1885 MB)
[hos host] vk: host time in GPU resource waits: 0.12 ms a frame (fences 0.03, timelines 0.10, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 455.6 full barriers recorded, 197.7 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 3.0; 0x8: 3.0; 0x9: 1.0; 0x10: 2.0; 0x11: 43.0; 0x12: 69.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 5.2 ms a frame (the longest submission 1.9 ms); 0.7 textures made a frame
[hos host] vk: a frame: 0.09 MB unswizzled on the CPU in 0.01 ms
[hos sead::MainThread] nvn: GPU waits in 5 s: 9 sync waits (15 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: 4 textures retired (681 left, 1879 MB): budget 4, drawn over 0, dead targets 0
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 127109 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 18 textures retired (663 left, 1789 MB): budget 18, drawn over 0, dead targets 0
[hos host] vk: mirror syncs (600 submits): 1085 dirty pages from their submits' captures, 102075 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: write-back diff (600 submits): 4147 runs, 2591.2 MB compared; 70 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18767 in captures, 11 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool: 0x1064400000 (8 MB, captured, state 2): 9;
[hos host] vk: 6 textures retired (659 left, 1785 MB): budget 6, drawn over 0, dead targets 0
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1275 of 3000 waited for their write-back, 0.02 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos host] vk: program picks: 2691560 kept (1339586 by the last picks' textures, a uniform buffer bound since), 13938 new, 894501 with other textures (of 3600000)
[hos sead::MainThread] nvn: CommandBufferSignalEvent #60000: event 0x100c318038 mode 0 value 59999 (lr 0x8066189c)
[hos host] vk: readback pages refreshed from guest memory: 1216466 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 1420000 (638318 pages from the captures)
[hos host] vk: 2 textures retired (657 left, 1784 MB): budget 2, drawn over 0, dead targets 0
[hos host] vk: 4 textures retired (653 left, 1777 MB): budget 4, drawn over 0, dead targets 0
[hos sead::MainThread] vk: capture: a submit 0.23 ms (lock 0.02, scan 0.04 over 596 ranges, 3519 MB), 603 pages written, 603 copied (2.4 MB)
[hos host] vk: a capture taken (ms): all 0.103; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.099, bytes copied 0.000; 25 kept (186 MB), 8 spare buffers, 6478 texture pages copied (25 MB)
[hos host] vk: sparse mirrors: 53, 734 MB with memory of 1148 MB (413 MB spared); blocks given memory later 4536, given back 12285; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 3234; 354 binds (178.3 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1160 dirty pages from their submits' captures, 104896 from earlier ones', 1 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4449 runs, 2633.7 MB compared; 76 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 19700 in captures, 13 only in the collection, 43 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055000000 (4 MB, captured, state 2): 1;
[hos host] vk: imports: 120 live (1924 MB), 1 retired still imported (8 MB), 13 retired given back (684 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.10, writes collected 0.04
[hos sead::MainThread] nvn: CommandBufferWaitEvent #85000: event 0x100c318024 mode 0 value 60716 (now 60697) (lr 0x8028c554)
[hos sead::MainThread] nvn: EventSignal (CPU) #10000: event 0x1005a88004 mode 0 value 1 (lr 0x8348c78c)
[hos host] vk: 2 textures retired (669 left, 1786 MB): budget 2, drawn over 0, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1440000 (647419 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 105406 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 678 KB; 0 made on demand (0 MB kept now); c1 stages: at most 95 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: mirror syncs (600 submits): 1121 dirty pages from their submits' captures, 103014 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4335 runs, 2597.1 MB compared; 74 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18915 in captures, 10 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a frame: 10 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 76 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.16, present 0.09, draw 2.74, descriptors 1.20, texture lookups 0.07, uploads 0.01, pipelines 0.10, programs 0.37, commands 0.85, dispatch 0.56, memory sync 0.83, queue submit 0.47, slot wait 0.18
[hos host] vk: a frame: mirrors 137 (734 MB): 9.39 MB synced in 1295.7 copies, 60.31 MB written back
[hos host] vk: 5040 presents, 60.2 fps; a frame: 763 draws (0 skipped), 88 dispatches, 200 passes, 146 barriers, 6 submits, 0.3 uploads (0.0 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 42668 pipelines (24270 ms), 0 programs (10 failed), 688 textures (1801 MB)
[hos host] vk: host time in GPU resource waits: 0.20 ms a frame (fences 0.03, timelines 0.17, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 460.0 full barriers recorded, 197.5 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 3.0; 0x8: 3.0; 0x9: 1.0; 0x10: 2.0; 0x11: 43.0; 0x12: 69.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 5.6 ms a frame (the longest submission 3.3 ms); 0.3 textures made a frame
[hos host] vk: a frame: 0.01 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: render targets drawn over from another start, let go: 627 so far (1063 MB)
[hos host] vk: replay reads of guest memory as submitted: 1460000 (656504 pages from the captures)
[hos host] vk: 21 textures retired (717 left, 1912 MB): budget 21, drawn over 0, dead targets 0
[hos host] vk: sparse mirrors: 53, 734 MB with memory of 1148 MB (413 MB spared); blocks given memory later 4536, given back 12285; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 3495; 354 binds (178.3 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1021 dirty pages from their submits' captures, 103208 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4095 runs, 2658.0 MB compared; 68 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18933 in captures, 10 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 120 live (1924 MB), 1 retired still imported (8 MB), 13 retired given back (684 MB)
[hos host] vk: 8 textures retired (712 left, 1910 MB): budget 8, drawn over 0, dead targets 0
[hos sead::MainThread] vk: capture: a submit 0.25 ms (lock 0.02, scan 0.04 over 601 ranges, 3543 MB), 675 pages written, 675 copied (2.6 MB)
[hos host] vk: a capture taken (ms): all 0.116; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.112, bytes copied 0.000; 25 kept (312 MB), 7 spare buffers, 20902 texture pages copied (81 MB)
[hos host] vk: 5 textures retired (709 left, 1906 MB): budget 5, drawn over 0, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.12, writes collected 0.04
[hos host] vk: 3 textures retired (710 left, 1904 MB): budget 3, drawn over 0, dead targets 0
[hos host] vk: 29 textures retired (681 left, 1869 MB): budget 29, drawn over 0, dead targets 0
[hos Presentation Thread] vi: 5117 frames presented (60.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 19, 15.5-17.8 267, 17.8-22 14, 22-30 0, 30-40 0, over 40 0
[hos host] vk: 12 textures retired (669 left, 1814 MB): budget 12, drawn over 0, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1480000 (665527 pages from the captures)
[hos host] vk: mirror syncs (600 submits): 1203 dirty pages from their submits' captures, 103094 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 3995 runs, 2602.6 MB compared; 69 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18684 in captures, 10 only in the collection, 43 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: 12 textures retired (660 left, 1785 MB): budget 12, drawn over 0, dead targets 0
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1180 of 3000 waited for their write-back, 0.02 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 114218 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: texture slots registered again at another size: 200000 (of 295644 registrations; slot 13980 360x202 -> 64x36)
[hos tera::FileResourceMgr] vcores: 232520 takes, 30350 waits (mean 38 us, longest 4.7 ms), 16184 kicks, 16101 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: 2 textures retired (661 left, 1789 MB): budget 2, drawn over 0, dead targets 0
[hos host] vk: texture storage read as submitted (2001 times; 82428 pages from the copies, 0 of them rewritten in guest memory since their submit; 12921 pages kept, 304 spans, 1445 MB)
[hos host] vk: 4 textures retired (669 left, 1795 MB): budget 4, drawn over 0, dead targets 0
[hos host] vk: a frame: 9 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 76 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.18, present 0.09, draw 2.81, descriptors 1.27, texture lookups 0.13, uploads 0.07, pipelines 0.10, programs 0.36, commands 0.87, dispatch 0.55, memory sync 0.77, queue submit 0.48, slot wait 0.18
[hos host] vk: a frame: mirrors 137 (734 MB): 9.34 MB synced in 1287.5 copies, 60.63 MB written back
[hos host] vk: 5160 presents, 59.9 fps; a frame: 761 draws (0 skipped), 83 dispatches, 199 passes, 146 barriers, 6 submits, 0.7 uploads (0.3 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 42668 pipelines (24270 ms), 0 programs (10 failed), 669 textures (1795 MB)
[hos host] vk: host time in GPU resource waits: 0.12 ms a frame (fences 0.02, timelines 0.10, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 454.9 full barriers recorded, 197.9 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 3.0; 0x8: 3.0; 0x9: 1.0; 0x10: 2.0; 0x11: 43.0; 0x12: 69.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 5.2 ms a frame (the longest submission 1.7 ms); 0.6 textures made a frame
[hos host] vk: a frame: 0.07 MB unswizzled on the CPU in 0.01 ms
[hos host] vk: sparse mirrors: 53, 734 MB with memory of 1148 MB (413 MB spared); blocks given memory later 4536, given back 12285; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 3756; 354 binds (178.3 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1082 dirty pages from their submits' captures, 102611 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4212 runs, 2668.0 MB compared; 70 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18685 in captures, 7 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 120 live (1924 MB), 1 retired still imported (8 MB), 13 retired given back (684 MB)
[hos host] vk: replay reads of guest memory as submitted: 1500000 (674535 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 1278002 pages so far (0 held: may be written ahead in their submission)
[hos Presentation Thread] vk: capture: a submit 0.22 ms (lock 0.02, scan 0.04 over 602 ranges, 3559 MB), 577 pages written, 577 copied (2.3 MB)
[hos host] vk: a capture taken (ms): all 0.079; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.076, bytes copied 0.000; 25 kept (186 MB), 12 spare buffers, 15481 texture pages copied (60 MB)
[hos host] vk: queues ordered by their events: 30000 submissions stopped at a wait, 29984 went on once it was reached, 14 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 1 (stopped 51.65 ms at most); waits met 89707, not met 0; captures at a CPU event's wait 1297, 0 of them at a wait met on arrival
[hos host] vk: 6 textures retired (679 left, 1793 MB): budget 6, drawn over 0, dead targets 0
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.08, writes collected 0.03
[hos ModuleSystemWorker4] nvn: CommandBufferWaitEvent #90000: event 0x100c318024 mode 0 value 64516 (now 64497) (lr 0x8028c554)
[hos host] vk: mirror syncs (600 submits): 1279 dirty pages from their submits' captures, 105158 from earlier ones', 2 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4417 runs, 2591.6 MB compared; 76 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 19787 in captures, 5 only in the collection, 43 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055000000 (4 MB, captured, state 2): 2;
[hos host] vk: program picks: 2831458 kept (1430714 by the last picks' textures, a uniform buffer bound since), 13938 new, 954603 with other textures (of 3800000)
[hos host] vk: 12 textures retired (672 left, 1805 MB): budget 12, drawn over 0, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1520000 (683584 pages from the captures)
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #65000: event 0x100c318044 mode 0 value 64999 (lr 0x802742f0)
[hos rmbl::VibrationThread] hid: 16384 vibration values, strongest 0.00 since the last count
[hos host] vk: 8 textures retired (672 left, 1826 MB): budget 8, drawn over 0, dead targets 0
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 102609 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: 8 textures retired (670 left, 1842 MB): budget 8, drawn over 0, dead targets 0
[hos sead::MainThread] nvn: GPU waits in 5 s: 2 sync waits (3 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 678 KB; 0 made on demand (0 MB kept now); c1 stages: at most 95 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 53, 734 MB with memory of 1148 MB (413 MB spared); blocks given memory later 4536, given back 12285; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 4017; 354 binds (178.3 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 1232 dirty pages from their submits' captures, 101969 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4387 runs, 2700.8 MB compared; 73 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18812 in captures, 15 only in the collection, 43 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 120 live (1924 MB), 1 retired still imported (8 MB), 13 retired given back (684 MB)
[hos host] vk: 2 textures retired (738 left, 2162 MB): budget 2, drawn over 0, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1540000 (693364 pages from the captures)
[hos host] vk:   (its written part: 2 of 2 KB readable, 318 bytes not zero; levels L7 L8 L9 L10 L11)
[hos host] vk: texture 0x10de823e00 3072x3072x1 x1 levels 12 NVN 0x7b (7959 KB of storage) re-uploaded (its memory was written; flags 0): what the CPU wrote, 2 KB of it
[hos host] vk:   (its written part: 2 of 2 KB readable, 318 bytes not zero; levels L7 L8 L9 L10 L11)
[hos host] vk: texture 0x10de823e00 3072x3072x1 x1 levels 12 NVN 0x7b (7959 KB of storage) re-uploaded (its memory was written; flags 0): what the CPU wrote, 2 KB of it
[hos host] vk: 3 textures retired (757 left, 2227 MB): budget 3, drawn over 0, dead targets 0
[hos sead::MainThread] vk: capture: a submit 0.26 ms (lock 0.02, scan 0.04 over 603 ranges, 3607 MB), 714 pages written, 714 copied (2.8 MB)
[hos host] vk: a capture taken (ms): all 0.136; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.132, bytes copied 0.000; 25 kept (282 MB), 7 spare buffers, 23347 texture pages copied (91 MB)
[hos host] vk: a frame: 9 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 76 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.71, present 0.09, draw 3.10, descriptors 1.54, texture lookups 0.38, uploads 0.34, pipelines 0.10, programs 0.38, commands 0.86, dispatch 0.57, memory sync 0.98, queue submit 0.48, slot wait 0.17
[hos host] vk: a frame: mirrors 138 (735 MB): 9.39 MB synced in 1296.2 copies, 60.82 MB written back
[hos host] vk: 5280 presents, 59.9 fps; a frame: 764 draws (0 skipped), 89 dispatches, 199 passes, 146 barriers, 6 submits, 1.1 uploads (1.3 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 42668 pipelines (24270 ms), 0 programs (10 failed), 757 textures (2227 MB)
[hos host] vk: host time in GPU resource waits: 0.47 ms a frame (fences 0.02, timelines 0.44, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 464.3 full barriers recorded, 197.6 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 3.0; 0x8: 3.0; 0x9: 1.0; 0x10: 2.0; 0x11: 43.0; 0x12: 69.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 5.5 ms a frame (the longest submission 3.4 ms); 1.1 textures made a frame
[hos host] vk: a frame: 0.60 MB unswizzled on the CPU in 0.05 ms
[hos host] vk: 13 textures retired (774 left, 2276 MB): budget 13, drawn over 0, dead targets 0
[hos host] vk: mirror syncs (600 submits): 1219 dirty pages from their submits' captures, 101673 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: write-back diff (600 submits): 4054 runs, 2569.2 MB compared; 85 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18142 in captures, 36 only in the collection, 42 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055000000 (4 MB, captured, state 2): 10;
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.15, writes collected 0.04
[hos host] vk: slow submit: 15.5 ms: draw 15.6 descriptors 14.5 texture lookups 13.8 uploads 13.7
[hos host] vk: 2 textures retired (775 left, 2314 MB): budget 2, drawn over 0, dead targets 0
[hos host] vk: slow upload 15.6 ms: #5030 0x10e2c0be00 5072x5056x1 x1 levels 1 NVN 0x4b, 25043 of 25043 KB
[hos host] vk: slow submit: 17.1 ms: draw 16.8 descriptors 16.2 texture lookups 15.8 uploads 15.6
[hos host] vk: 31 textures retired (745 left, 2235 MB): budget 31, drawn over 0, dead targets 0
[hos host] vk:   (its written part: 2 of 2 KB readable, 2315 bytes not zero; levels L0)
[hos host] vk: texture 0x10e2c0be00 5072x5056x1 x1 levels 1 NVN 0x4b (25043 KB of storage) re-uploaded (its memory was written; flags 0): what the CPU wrote, 2 KB of it
[hos host] vk:   (its written part: 2 of 2 KB readable, 2315 bytes not zero; levels L0)
[hos host] vk: texture 0x10e2c0be00 5072x5056x1 x1 levels 1 NVN 0x4b (25043 KB of storage) re-uploaded (its memory was written; flags 0): what the CPU wrote, 2 KB of it
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 1151 of 3000 waited for their write-back, 0.04 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 host] vk: 6 textures retired (739 left, 2208 MB): budget 6, drawn over 0, dead targets 0
[hos host] vk:   (its written part: 2 of 2 KB readable, 2315 bytes not zero; levels L0)
[hos host] vk: texture 0x10e2c0be00 5072x5056x1 x1 levels 1 NVN 0x4b (25043 KB of storage) re-uploaded (its memory was written; flags 0): what the CPU wrote, 2 KB of it
[hos host] vk: replay reads of guest memory as submitted: 1560000 (702982 pages from the captures)
[hos host] vk: 16 textures retired (723 left, 2037 MB): budget 16, drawn over 0, dead targets 0
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 195099 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: sparse mirrors: 54, 735 MB with memory of 1152 MB (416 MB spared); blocks given memory later 4536, given back 12285; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 4275; 355 binds (178.4 ms); 0 made whole for want of memory
[hos host] vk: mirror syncs (600 submits): 948 dirty pages from their submits' captures, 104136 from earlier ones', 1 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4064 runs, 2575.8 MB compared; 82 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18357 in captures, 25 only in the collection, 41 only by the replay itself
[hos host] vk:   read live, by pool: 0x1055000000 (4 MB, captured, state 2): 1;
[hos host] vk: imports: 120 live (1924 MB), 1 retired still imported (8 MB), 13 retired given back (684 MB)
[hos host] vk: 49 textures retired (674 left, 1907 MB): budget 49, drawn over 0, dead targets 0
[hos host] vk: texture #1779 0x1001ec02800 1440x810x1 x1 NVN 0x3f the GPU drew goes with no copy of its texels (live, unused for 3608 frames, over the budget): made again over this storage, it starts empty
[hos host] vk: 17 textures retired (657 left, 1843 MB): budget 17, drawn over 0, dead targets 0; GPU-drawn with no copy kept: 1 live, 0 finalized, 4 MB
[hos host] vk: 2 textures retired (655 left, 1799 MB): budget 2, drawn over 0, dead targets 0
[hos host] vk: 1 textures retired (654 left, 1774 MB): budget 1, drawn over 0, dead targets 0
[hos Presentation Thread] vk: capture: a submit 0.25 ms (lock 0.01, scan 0.04 over 603 ranges, 3607 MB), 710 pages written, 710 copied (2.8 MB)
[hos host] vk: a capture taken (ms): all 0.145; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.141, bytes copied 0.000; 25 kept (186 MB), 11 spare buffers, 8567 texture pages copied (33 MB)
[hos host] vk: replay reads of guest memory as submitted: 1580000 (712479 pages from the captures)
[hos host] vk: mirror syncs (600 submits): 953 dirty pages from their submits' captures, 102971 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: write-back diff (600 submits): 4202 runs, 2644.6 MB compared; 72 pages read into the shadows; 292 MB of shadows; pages the shadows knew that the CPU wrote: 18668 in captures, 5 only in the collection, 44 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: 2 textures retired (652 left, 1774 MB): budget 2, drawn over 0, dead targets 0
[hos host] vk: readback pages refreshed from guest memory: 1339286 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 0.13, writes collected 0.04
[hos ModuleSystemWorker5] nvn: CommandBufferWaitEvent #95000: event 0x100c318024 mode 0 value 68316 (now 68297) (lr 0x8028c554)
[hos host] vk: memory: working set 15898 MB (peak 16054), private 24314 MB, staging 256 MB in 4 arenas
[hos host] vk: memory heaps: heap 0 (system): 12 blocks 664 MB, 40 allocations 616 MB; used 2612 of 15571 MB; heap 1 (device): 108 blocks 6033 MB, 12616 allocations 3382 MB; used 7074 of 15419 MB;
[hos host] vk: image blocks: 1 blocks 128 MB, 6 images 24 MB; 1 free ranges, the largest 103 MB
[hos host] vk: image blocks (render targets): 8 blocks 1024 MB, 162 images 354 MB; 54 free ranges, the largest 114 MB
[hos host] vk: image blocks (textures, compacted): 19 blocks 2432 MB, 488 images 845 MB; 194 free ranges, the largest 128 MB
[hos host] vk: render target images: 355 reused from the pool, 478 made new, 316 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 7074 MB used of 15419 MB allowed (2651 MB of it free in the blocks); textures 1774 MB (budget 616 MB), 652 textures
[hos host] vk: textures: decoded ASTC 850 MB, rendered 378 MB (87 drawn in one frame, kept: 38148 KB), other 218 MB; used in the last 8 presents 1338 MB
[hos host] vk:   214 MB: 0x10a4772c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a0efa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x11361ac000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x113b7bd000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   47 MB: 0x10de823e00 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 1 presents ago
[hos host] vk:   24 MB: 0x10e2c0be00 5072x5056x1 x1 levels 1 NVN 0x4b, last used 1 presents ago
[hos host] vk:   21 MB: 0x100056a8000 2400x2400x1 x1 levels 1 NVN 0x34 rendered, last used 1470 presents ago
[hos host] vk:   18 MB: 0x10e17d6e00 2536x2528x1 x3 levels 1 NVN 0x50, last used 1 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 57 (15305 KB): 260x260x1 NVN 0xd: 53 (9325 KB); 360x202x1 NVN 0x25: 1 (284 KB); 720x405x1 NVN 0x29: 2 (4556 KB); 720x405x1 NVN 0x34: 1 (1139 KB);
[hos ModuleSystemWorker3] nvn: DrawMeshTasksIndirect: 140000 so far (x1 0x11775dc000 x2 0x1 x3 0x11775dc000 x4 0x4040)
[hos Presentation Thread] vk: presents after the CPU's signal (600): 18 waited for it, 3.41 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[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: 27 program records loaded (39.3 ms), 2 new graphics pipelines, pipelines from the driver's cache 70, from the start's cache 0, compiled 2; 36 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (close): 5657 queued, 5657 loaded (median 0.59 ms, 99th percentile 4.47 ms, worst 26.66 ms, 5251 ms and 1199.6 MB read in all); 7651 programs (1190 of them compute variants without their code, 0 graphics programs no note names), 9323 graphics pipelines (0 not in a cache), 724 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 1372 pipelines made at create taken by draws; 2751 create calls with nothing kept for their stages, 0 before the index; units 8408 (5657 ready, 0 loading), 13910 programs alive
[hos host] vk: module identifiers (close): 794 known, 794 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: status: 600 frames, the slowest 31.6 ms; VRAM 6.9 of 15.1 GB
[hos host] vk: program census (play): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 27 program records loaded (39.3 ms), 2 new graphics pipelines, pipelines from the driver's cache 70, from the start's cache 0, compiled 2; 36 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (play): 5657 queued, 5657 loaded (median 0.59 ms, 99th percentile 4.47 ms, worst 26.66 ms, 5251 ms and 1199.6 MB read in all); 7651 programs (1190 of them compute variants without their code, 0 graphics programs no note names), 9323 graphics pipelines (0 not in a cache), 724 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 1372 pipelines made at create taken by draws; 2751 create calls with nothing kept for their stages, 0 before the index; units 8408 (5657 ready, 0 loading), 13910 programs alive
[hos host] vk: page table census (close): no pool outside its windows, 64-72 GB and 508-512 GB (4341 pools in the table now)
[hos host] vk: module identifiers (play): 794 known, 794 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: 9 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 76 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 6.80, present 0.10, draw 3.08, descriptors 1.59, texture lookups 0.51, uploads 0.46, pipelines 0.10, programs 0.36, commands 0.83, dispatch 0.55, memory sync 1.07, queue submit 0.49, slot wait 0.16
[hos host] vk: a frame: mirrors 138 (735 MB): 9.36 MB synced in 1287.3 copies, 59.96 MB written back
[hos host] vk: 5400 presents, 60.1 fps; a frame: 766 draws (0 skipped), 84 dispatches, 199 passes, 146 barriers, 6 submits, 0.5 uploads (1.1 MB), 82 ub updates (57 overlays), GPU-bound wait 0.0 ms (recording slots only); 42668 pipelines (24270 ms), 0 programs (10 failed), 652 textures (1774 MB)
[hos host] vk: host time in GPU resource waits: 0.45 ms a frame (fences 0.03, timelines 0.43, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 456.1 full barriers recorded, 197.7 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 3.0; 0x8: 3.0; 0x9: 1.0; 0x10: 2.0; 0x11: 43.0; 0x12: 69.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 5.3 ms a frame (the longest submission 3.3 ms); 0.3 textures made a frame
[hos host] vk: a frame: 1.05 MB unswizzled on the CPU in 0.08 ms
[hos host] vk: the render thread's whole frames (600 presents): 0 jobs went on before their present after 1000 ms, 0 times a game's thread waited for the work handed over, 0 times for a fence of the frame in the making
[hos host] vk: the render thread (600 presents, 16.67 ms a frame): its jobs 6.96 ms a frame (submits 6.63, presents 0.10, others 0.24), idle 9.70; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 152.23 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); 18 captures at a CPU signal; the submits' waits read in 0.024 ms a frame; readbacks: 2400 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 6.99 ms after its hand-over (23.15 at most)
[hos host] vk: pipeline cache: play file 5 saved (150 KB): 5 pipelines, 5 from the driver's cache, 0 compiled; 685 more from the start's cache
[hos host] control: exiting
[hos host] stop: close work done (the game's saves are not waited for)