Beanstalk Longshot test farm
Menu

Report rp_405be0c8dd6ca57d error

#1574 perf(vk): ask for the window's next image on the presenter thread right after a present, so the render thread stops waiting for it

windows+nvidia on Flashman4/agent-de36c5,

This report disagrees with its peers.

hyrule-field run_frametime_max_ms: -54.7% against peers' +28.3%
hyrule-field run_hitch_count: -78.9% against peers' +0.0%
hyrule-field#1 run_hitch_count: +20.0% against peers' +100.0%
no scenario could be judged: a step ran out of time: timeout: not met in time

Machine

Commit
2f24a499ca2a8fc708008cdb350c09fea17e9ea0
Baseline
430eef4f2cc94e398b06d0daeb585a679d3ebcda
App
nightly-177-99542c21, attempt 2
Game build
e0f3ccfa
os
windows
cpu
AMD Ryzen 9 5900X 12-Core Processor
gpu
NVIDIA GeForce RTX 3090
tags
windowsnvidianvidia-proprietarydgpu
driver
621.24.0
ram_gb
128
vram_mb
24,322
gpu_vendor
nvidia
os_version
Windows 10.0.26200

Builds

main (baseline)

1,265 s warm

from 430eef4f2cc9

lift
92.7 s
read
574.7 s
finish
150.9 s
compile
398.5 s

PR

156.4 s warm

Program cache: warm

Scenarios

hyrule-field RX_VK_ACQUIRE_AHEAD=0 (PR only)

timeout

a step ran out of time: timeout: not met in time

Metric (median)mainPRΔ
run_hitch_count194−78.9%
log_forbidden_hits00±0%
log_expected_missing00±0%
run_frametime_max_ms173.6778.62−54.7%
Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code——
run_hitch_count194
log_forbidden_hits00
value_image_wait_ms——
log_expected_missing00
run_frametime_max_ms173.6778.62

hyrule-field RX_VK_ACQUIRE_AHEAD=1 (PR only)

timeout

a step ran out of time: timeout: not met in time

Metric (median)mainPRΔ
run_hitch_count56+20%
log_forbidden_hits00±0%
log_expected_missing00±0%
run_frametime_max_ms161.1179.87−50.4%
Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code——
run_hitch_count56
log_forbidden_hits00
value_image_wait_ms——
log_expected_missing00
run_frametime_max_ms161.1179.87

Rules

ScenarioCellRuleLimitmainPRΔ %Status
hyrule-fieldRX_VK_ACQUIRE_AHEAD=0 (PR only)crashFalse00—pass
hyrule-fieldRX_VK_ACQUIRE_AHEAD=0 (PR only)field.fps_mean.min_delta_pct10———skippedthe metric was not measured
hyrule-fieldRX_VK_ACQUIRE_AHEAD=0 (PR only)field.frametime_p99_ms.max_delta_pct20———skippedthe metric was not measured
hyrule-fieldRX_VK_ACQUIRE_AHEAD=0 (PR only)log_forbidden_hits.max000—pass
hyrule-fieldRX_VK_ACQUIRE_AHEAD=1 (PR only)crashFalse00—pass
hyrule-fieldRX_VK_ACQUIRE_AHEAD=1 (PR only)field.fps_mean.min_delta_pct10———skippedthe metric was not measured
hyrule-fieldRX_VK_ACQUIRE_AHEAD=1 (PR only)field.frametime_p99_ms.max_delta_pct20———skippedthe metric was not measured
hyrule-fieldRX_VK_ACQUIRE_AHEAD=1 (PR only)log_forbidden_hits.max000—pass
hyrule-fieldRX_VK_ACQUIRE_AHEAD=1 (PR only)field.fps_mean: RX_VK_ACQUIRE_AHEAD=1 vs RX_VK_ACQUIRE_AHEAD=0 (PR)10———skippedthe metric was not measured in both cells
hyrule-fieldRX_VK_ACQUIRE_AHEAD=1 (PR only)field.frametime_p99_ms: RX_VK_ACQUIRE_AHEAD=1 vs RX_VK_ACQUIRE_AHEAD=0 (PR)20———skippedthe metric was not measured in both cells

Pictures of the failure

This report had 2 pictures of its failure (the last frame of a stuck step, or main's and the PR's picture where they differed too much). The farm keeps those for a day: they were deleted.

Log (PR side, end)

[hos host] vk: host time in GPU resource waits: 0.59 ms a frame (fences 0.02, timelines 0.57, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 350.2 full barriers recorded, 182.6 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: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 43.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 3.2 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos ModuleSystemWorker3] nvn: CommandBufferWaitEvent #105000: event 0x10103c8044 mode 0 value 67563 (now 67544) (lr 0x80274390)
[hos ModuleSystemWorker5] vk: readbacks of GPU results: 1297 of 3000 waited for their write-back, 0.07 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker5] vk: invalidates since the last readbacks line: 3850 kept their virtual core (nothing there to wait for), 2550 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: program picks: 1868167 kept (780909 by the last picks' textures, a uniform buffer bound since), 776 new, 1131056 with other textures (of 3000000)
[hos host] vk: replay reads of guest memory as submitted: 1720000 (882804 pages from the captures)
[hos host] vk: sparse mirrors: 55, 801 MB with memory of 1016 MB (214 MB spared); blocks given memory later 117, given back 1231; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (31.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 62 made new; 14 kept as spares, 0 destroyed; 7 spares now, 92 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1289, 5276.2 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1079 dirty pages from their submits' captures, 103963 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): 1460 dispatches, 61482 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6733 runs, 914.2 MB compared; 106 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 31782 in captures, 11 only in the collection, 224 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1108 MB), 0 retired still imported (0 MB), 8 retired given back (112 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos host] vk: replay reads of guest memory as submitted: 1740000 (893367 pages from the captures)
[hos BaseProcExecutor2] vcores: 313829 takes, 51663 waits (mean 29 us, longest 6.3 ms, 0 over HOS_VCORE_LOG_MS), 25871 kicks, 25703 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos Presentation Thread] vk: capture: a submit 0.18 ms (lock 0.00, scan 0.05 over 1301 ranges, 1779 MB), 387 pages written, 387 copied (1.5 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (132 MB), 9 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: recorder: 4786163 packets run in 10.0 s (busy 1.35 s, spun idle 3.06 s, slept 3988 times, 869 of them under 200 us), 0 calls made directly (no room); the replay waited for it 23671 times, 84.66 ms in all, the longest 0.18 ms; vkEndCommandBuffer 7580/36.87 ms (max 0.18), replay_mu unlock 8553/26.91 ms (max 0.15), vkResetDescriptorPool 7538/20.88 ms (max 0.08)
[hos host] vk: write-backs left out (600 submits): 1288, 5274.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1048 dirty pages from their submits' captures, 105196 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): 1471 dispatches, 61964 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6612 runs, 898.2 MB compared; 105 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 31783 in captures, 10 only in the collection, 214 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1760000 (903853 pages from the captures)
[hos host] vk: a frame: 51 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.47, present 0.20, draw 3.07, descriptors 1.28, texture lookups 0.07, uploads 0.00, pipelines 0.12, programs 0.57, commands 0.88, dispatch 0.62, memory sync 0.31, queue submit 0.38, slot wait 0.04
[hos host] vk: a frame: mirrors 69 (801 MB): 7.35 MB synced in 1081.8 copies, 16.04 MB written back
[hos host] vk: 9360 presents, 73.5 fps; a frame: 823 draws (0 skipped), 103 dispatches, 142 passes, 141 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 41146 pipelines (4757475 ms), 0 programs (0 failed), 1216 textures (3478 MB)
[hos host] vk: host time in GPU resource waits: 0.72 ms a frame (fences 0.02, timelines 0.70, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 349.9 full barriers recorded, 182.5 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 42.8; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 3.3 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #70000: event 0x10103c800c mode 0 value 70000 (lr 0x802915c8)
[hos host] vk: readback pages refreshed from guest memory: 885896 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 1780000 (914331 pages from the captures)
[hos host] vk: sparse mirrors: 55, 801 MB with memory of 1016 MB (214 MB spared); blocks given memory later 117, given back 1231; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (31.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 62 made new; 14 kept as spares, 0 destroyed; 7 spares now, 92 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1311, 5368.5 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 971 dirty pages from their submits' captures, 105110 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): 1467 dispatches, 62106 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6740 runs, 914.2 MB compared; 107 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 32332 in captures, 9 only in the collection, 223 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1108 MB), 0 retired still imported (0 MB), 8 retired given back (112 MB)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 1308 of 3000 waited for their write-back, 0.08 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: 4027 kept their virtual core (nothing there to wait for), 3141 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: 1800000 (924807 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos sead::MainThread] vk: capture: a submit 0.18 ms (lock 0.00, scan 0.05 over 1301 ranges, 1779 MB), 386 pages written, 386 copied (1.5 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (136 MB), 8 spare buffers, 0 texture pages copied (0 MB)
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #110000: event 0x10103c8044 mode 0 value 71363 (now 71344) (lr 0x80274390)
[hos host] vk: write-backs left out (600 submits): 1288, 5274.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1112 dirty pages from their submits' captures, 104074 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): 1456 dispatches, 61464 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6621 runs, 898.2 MB compared; 105 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 31773 in captures, 12 only in the collection, 217 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 1820000 (935251 pages from the captures)
[hos host] vk: a frame: 51 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.59, present 0.20, draw 3.05, descriptors 1.25, texture lookups 0.07, uploads 0.00, pipelines 0.13, programs 0.56, commands 0.89, dispatch 0.65, memory sync 0.32, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 69 (801 MB): 7.33 MB synced in 1081.2 copies, 16.04 MB written back
[hos host] vk: 9480 presents, 72.0 fps; a frame: 822 draws (0 skipped), 103 dispatches, 142 passes, 142 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 41146 pipelines (4757475 ms), 0 programs (0 failed), 1217 textures (3483 MB)
[hos host] vk: host time in GPU resource waits: 0.58 ms a frame (fences 0.02, timelines 0.56, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 350.2 full barriers recorded, 182.6 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: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 43.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 3.2 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: program picks: 1993065 kept (833834 by the last picks' textures, a uniform buffer bound since), 776 new, 1206158 with other textures (of 3200000)
[hos host] vk: replay reads of guest memory as submitted: 1840000 (945657 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 344 KB; 0 made on demand (0 MB kept now); c1 stages: at most 114 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 55, 801 MB with memory of 1016 MB (214 MB spared); blocks given memory later 117, given back 1231; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (31.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 62 made new; 14 kept as spares, 0 destroyed; 7 spares now, 92 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1311, 5368.5 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1081 dirty pages from their submits' captures, 106106 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): 1482 dispatches, 62575 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6740 runs, 914.2 MB compared; 107 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 32396 in captures, 11 only in the collection, 218 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1108 MB), 0 retired still imported (0 MB), 8 retired given back (112 MB)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos sead::MainThread] vk: capture: a submit 0.18 ms (lock 0.00, scan 0.05 over 1301 ranges, 1779 MB), 386 pages written, 386 copied (1.5 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 0.003, bytes copied 0.000; 25 kept (132 MB), 8 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 1860000 (956136 pages from the captures)
[hos Presentation Thread] vi: 9550 frames presented (72.5 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 350, 15.5-17.8 13, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos sead::MainThread] vk: readbacks of GPU results: 1282 of 3000 waited for their write-back, 0.07 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos sead::MainThread] vk: invalidates since the last readbacks line: 4222 kept their virtual core (nothing there to wait for), 3202 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos ModuleSystemWorker1] nvn: 650000 recordings made, 44998 live copies (13.2 MB), 297 texture copies; 7 command buffers' segments (4.4 MB)
[hos host] vk: write-backs left out (600 submits): 1288, 5274.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1054 dirty pages from their submits' captures, 104168 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): 1456 dispatches, 61503 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6629 runs, 898.2 MB compared; 105 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 31825 in captures, 13 only in the collection, 220 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: 1 textures retired (1216 left, 3478 MB): budget 0, drawn over 1, dead targets 0
[hos host] vk: replay reads of guest memory as submitted: 1880000 (966614 pages from the captures)
[hos host] vk: memory: working set 10817 MB (peak 11090), private 17739 MB, staging 32 MB in 2 arenas, 23 made so far (5 ms)
[hos host] vk: memory heaps: heap 0 (device): 164 blocks 5198 MB, 14189 allocations 4792 MB; used 5624 of 23554 MB; heap 1 (system): 14 blocks 592 MB, 39 allocations 557 MB; used 1780 of 104029 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): 20 blocks 640 MB, 98 images 390 MB; 42 free ranges, the largest 32 MB
[hos host] vk: image blocks (textures, compacted): 56 blocks 1792 MB, 1106 images 1781 MB; 463 free ranges, the largest 9 MB
[hos host] vk: render target images: 48 reused from the pool, 262 made new, 160 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 5624 MB used of 23554 MB allowed (405 MB of it free in the blocks); textures 3478 MB (budget 11000 MB), 1216 textures
[hos host] vk: textures: decoded ASTC 1868 MB, rendered 1028 MB (101 drawn in one frame, kept: 62267 KB), other 350 MB; used in the last 8 presents 2391 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; 409 MB of blocks hashed in 52 ms
[hos host] vk:   288 MB: 0x1150c06000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10aea22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10ab1aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x114045c000 4096x4096x1 x1 levels 1 NVN 0xa, last used 3594 presents ago
[hos host] vk:   64 MB: 0x1145b99000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x114045c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x10006146200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x1000558d800 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 3581 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 2 (177 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 260x260x1 NVN 0xd: 1 (175 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 3 waited for it, 0.09 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: status: 600 frames, the slowest 17.9 ms; VRAM 5.5 of 23.0 GB
[hos host] vk: program census (play): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 91 program records loaded (9.3 ms), 0 new graphics pipelines, pipelines from the driver's cache 0, from the start's cache 0, compiled 17; 368 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (play): 4613 queued, 4613 loaded (median 4.01 ms, 99th percentile 23.16 ms, worst 43.52 ms, 25789 ms and 978.9 MB read in all); 6518 programs (1543 of them compute variants without their code, 0 graphics programs no note names), 8179 graphics pipelines (0 not in a cache), 337 compute pipelines (267 not in a cache); a first use ran 23 itself and waited for 1 (103.3 ms); 623 pipelines made at create taken by draws; 2402 create calls with nothing kept for their stages, 0 before the index; units 7015 (4613 ready, 0 loading), 12472 programs alive; 21 noted graphics pipeline probes deferred to first use
[hos host] vk: module identifiers (play): 605 known, 0 learned in this run; 337 compute pipelines made by them at a create call, 267 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 51 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.63, present 0.22, draw 3.22, descriptors 1.35, texture lookups 0.07, uploads 0.00, pipelines 0.13, programs 0.59, commands 0.92, dispatch 0.65, memory sync 0.32, queue submit 0.38, slot wait 0.04
[hos host] vk: a frame: mirrors 69 (801 MB): 7.34 MB synced in 1081.4 copies, 16.04 MB written back
[hos host] vk: 9600 presents, 71.3 fps; a frame: 823 draws (0 skipped), 103 dispatches, 142 passes, 142 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 41146 pipelines (4757475 ms), 0 programs (0 failed), 1216 textures (3478 MB)
[hos host] vk: host time in GPU resource waits: 0.54 ms a frame (fences 0.02, timelines 0.52, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 350.1 full barriers recorded, 182.7 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: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 43.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 3.5 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: the render thread's whole frames (600 presents): 0 jobs went on before their present after 1000 ms, 0 times a game's thread waited for the work handed over, 0 times for a fence of the frame in the making
[hos host] vk: the render thread (600 presents, 13.82 ms a frame): its jobs 6.22 ms a frame (submits 5.88, presents 0.21, others 0.13), idle 7.60; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 160.88 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); 3 captures at a CPU signal; the submits' waits read in 0.024 ms a frame; readbacks: 3000 next to a texture's storage (no catch-up), 0 in one, 0 read exactly (after the work handed over: among the catch-ups); a present made 6.25 ms after its hand-over (9.80 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 5.16 + 0.39 (the window's copy is in the idle), after it 0.72 + 0.36; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.03 ms of GPU time), after its last that reads it 1.00 (0.03 ms)
[hos host] vk: frame ahead (600 presents): 4.33 parts a frame recorded in the signal's job, 53.8% of them took nothing that came with the signal; of the others, 2.00 a frame took pages of the signal's capture into a mirror (5.3 pages a frame) and 0.00 pages read as guest memory had them (0.0 pages a frame); a frame's first part took nothing in 0 of 3 frames, its first three in 0
[hos host] vk: frame ahead, what came with the signal: 3 signals with a capture, each of 23.3 pages in 5.7 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: 0.33+0.17, 0.36+0.12, 0.02+0.03, 0.02+0.03, 0.01+0.01, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00
[hos host] vk: replay reads of guest memory as submitted: 1900000 (977283 pages from the captures)
[hos host] vk: sparse mirrors: 55, 801 MB with memory of 1016 MB (214 MB spared); blocks given memory later 117, given back 1231; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (31.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 62 made new; 14 kept as spares, 0 destroyed; 7 spares now, 92 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1311, 5368.5 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 997 dirty pages from their submits' captures, 106278 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): 1482 dispatches, 62641 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6742 runs, 914.2 MB compared; 106 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 32358 in captures, 8 only in the collection, 215 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1108 MB), 0 retired still imported (0 MB), 8 retired given back (112 MB)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: queues ordered by their events: 48000 submissions stopped at a wait, 47996 went on once it was reached, 0 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 0 (stopped 0.00 ms at most); waits met 114656, not met 0; captures at a CPU event's wait 4602, 0 of them at a wait met on arrival
[hos sead::MainThread] vk: capture: a submit 0.18 ms (lock 0.00, scan 0.05 over 1301 ranges, 1779 MB), 387 pages written, 387 copied (1.5 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (134 MB), 8 spare buffers, 0 texture pages copied (0 MB)
[hos sead::MainThread] nvn: CommandBufferSignalEvent #75000: event 0x10103c8018 mode 0 value 75000 (lr 0x80660c94)
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #115000: event 0x10103c8044 mode 0 value 75163 (now 75144) (lr 0x80274390)
[hos host] vk: replay reads of guest memory as submitted: 1920000 (987631 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 960061 pages so far (0 held: may be written ahead in their submission)
[hos nn.npln.Shared] vcores: 309874 takes, 48500 waits (mean 31 us, longest 7.4 ms, 0 over HOS_VCORE_LOG_MS), 24315 kicks, 24138 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: write-backs left out (600 submits): 1288, 5274.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1021 dirty pages from their submits' captures, 103923 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): 1456 dispatches, 61380 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6628 runs, 898.2 MB compared; 106 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 31800 in captures, 8 only in the collection, 220 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker2] nvn: DrawMeshTasksIndirect: 160000 so far (x1 0x1185f58000 x2 0 x3 0x1185f58000 x4 0x44b040)
[hos host] vk: replay reads of guest memory as submitted: 1940000 (997985 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos sead::MainThread] vk: readbacks of GPU results: 1291 of 3000 waited for their write-back, 0.06 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos sead::MainThread] vk: invalidates since the last readbacks line: 4009 kept their virtual core (nothing there to wait for), 2903 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: program picks: 2117726 kept (886638 by the last picks' textures, a uniform buffer bound since), 776 new, 1281497 with other textures (of 3400000)
[hos host] vk: a frame: 51 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.56, present 0.21, draw 3.14, descriptors 1.29, texture lookups 0.07, uploads 0.00, pipelines 0.13, programs 0.59, commands 0.91, dispatch 0.64, memory sync 0.31, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 69 (801 MB): 7.34 MB synced in 1080.4 copies, 16.04 MB written back
[hos host] vk: 9720 presents, 71.7 fps; a frame: 824 draws (0 skipped), 103 dispatches, 142 passes, 142 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 41146 pipelines (4757475 ms), 0 programs (0 failed), 1216 textures (3478 MB)
[hos host] vk: host time in GPU resource waits: 0.59 ms a frame (fences 0.02, timelines 0.57, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 350.3 full barriers recorded, 182.6 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: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 43.1; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 3.2 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: replay reads of guest memory as submitted: 1960000 (1008364 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos Presentation Thread] vk: capture: a submit 0.18 ms (lock 0.00, scan 0.05 over 1301 ranges, 1779 MB), 383 pages written, 383 copied (1.5 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (154 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: sparse mirrors: 55, 801 MB with memory of 1016 MB (214 MB spared); blocks given memory later 117, given back 1231; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (31.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 62 made new; 14 kept as spares, 0 destroyed; 7 spares now, 92 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1288, 5274.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1002 dirty pages from their submits' captures, 104801 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): 1474 dispatches, 61755 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6641 runs, 898.2 MB compared; 106 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 32401 in captures, 8 only in the collection, 220 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1108 MB), 0 retired still imported (0 MB), 8 retired given back (112 MB)
[hos host] vk: replay reads of guest memory as submitted: 1980000 (1018741 pages from the captures)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 344 KB; 0 made on demand (0 MB kept now); c1 stages: at most 114 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: write-backs left out (600 submits): 1311, 5368.5 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1034 dirty pages from their submits' captures, 104757 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): 1464 dispatches, 61898 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6734 runs, 914.2 MB compared; 106 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 31804 in captures, 7 only in the collection, 218 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: replay reads of guest memory as submitted: 2000000 (1029153 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos Presentation Thread] vk: capture: a submit 0.18 ms (lock 0.00, scan 0.05 over 1301 ranges, 1779 MB), 387 pages written, 387 copied (1.5 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (150 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: a frame: 51 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.45, present 0.21, draw 3.04, descriptors 1.24, texture lookups 0.07, uploads 0.00, pipelines 0.12, programs 0.58, commands 0.87, dispatch 0.63, memory sync 0.30, queue submit 0.38, slot wait 0.04
[hos host] vk: a frame: mirrors 69 (801 MB): 7.31 MB synced in 1078.0 copies, 16.04 MB written back
[hos host] vk: 9840 presents, 72.9 fps; a frame: 823 draws (0 skipped), 103 dispatches, 142 passes, 141 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 41146 pipelines (4757475 ms), 0 programs (0 failed), 1216 textures (3478 MB)
[hos host] vk: host time in GPU resource waits: 0.63 ms a frame (fences 0.02, timelines 0.61, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 350.1 full barriers recorded, 182.4 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 42.8; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 3.2 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: replay reads of guest memory as submitted: 2020000 (1039500 pages from the captures)
[hos ModuleSystemWorker5] nvn: CommandBufferWaitEvent #120000: event 0x10103c8044 mode 0 value 78963 (now 78944) (lr 0x80274390)
[hos host] vk: sparse mirrors: 55, 801 MB with memory of 1016 MB (214 MB spared); blocks given memory later 117, given back 1231; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (31.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 62 made new; 14 kept as spares, 0 destroyed; 7 spares now, 92 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1288, 5274.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1016 dirty pages from their submits' captures, 105035 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): 1474 dispatches, 61800 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6614 runs, 898.2 MB compared; 105 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 31854 in captures, 7 only in the collection, 219 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1108 MB), 0 retired still imported (0 MB), 8 retired given back (112 MB)
[hos sead::MainThread] vk: readbacks of GPU results: 1304 of 3000 waited for their write-back, 0.07 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos sead::MainThread] vk: invalidates since the last readbacks line: 3887 kept their virtual core (nothing there to wait for), 3025 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: 2040000 (1049867 pages from the captures)
[hos ModuleSystemWorker2] nvn: CommandBufferSignalEvent #80000: event 0x10103c8024 mode 0 value 80000 (lr 0x80665068)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos host] vk: write-backs left out (600 submits): 1311, 5368.5 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 975 dirty pages from their submits' captures, 105509 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: mirror syncs' pass (600 submits): 1464 dispatches, 62104 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6747 runs, 914.2 MB compared; 106 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 32328 in captures, 10 only in the collection, 218 only by the replay itself
[hos host] vk:   read live, by pool: 0x106ec00000 (8 MB, captured, state 2): 2;
[hos Presentation Thread] vi: 9911 frames presented (72.1 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 339, 15.5-17.8 22, 17.8-22 0, 22-30 0, 30-40 0, over 40 0
[hos host] vk: replay reads of guest memory as submitted: 2060000 (1060420 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.18 ms (lock 0.00, scan 0.05 over 1300 ranges, 1777 MB), 387 pages written, 387 copied (1.5 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.003, bytes copied 0.000; 25 kept (148 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: readback pages refreshed from guest memory: 1034226 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: program picks: 2242666 kept (939581 by the last picks' textures, a uniform buffer bound since), 776 new, 1356557 with other textures (of 3600000)
[hos ModuleSystemWorker1] nvn: 700000 recordings made, 45011 live copies (12.2 MB), 297 texture copies; 3 command buffers' segments (4.6 MB)
[hos host] vk: replay reads of guest memory as submitted: 2080000 (1070877 pages from the captures)
[hos host] vk: a frame: 51 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.51, present 0.21, draw 3.03, descriptors 1.26, texture lookups 0.07, uploads 0.00, pipelines 0.12, programs 0.56, commands 0.87, dispatch 0.65, memory sync 0.31, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 69 (801 MB): 7.35 MB synced in 1080.9 copies, 16.04 MB written back
[hos host] vk: 9960 presents, 72.1 fps; a frame: 823 draws (0 skipped), 103 dispatches, 142 passes, 142 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 41146 pipelines (4757475 ms), 0 programs (0 failed), 1216 textures (3478 MB)
[hos host] vk: host time in GPU resource waits: 0.57 ms a frame (fences 0.02, timelines 0.55, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 350.4 full barriers recorded, 182.5 left out (nothing recorded since the one before), 0.0 pass starts ordered in their fragment stages only (RX_VK_PASS_BARRIER=game)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 43.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 3.2 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: sparse mirrors: 55, 801 MB with memory of 1016 MB (214 MB spared); blocks given memory later 117, given back 1231; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (31.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 62 made new; 14 kept as spares, 0 destroyed; 7 spares now, 92 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1288, 5274.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1007 dirty pages from their submits' captures, 104068 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): 1456 dispatches, 61453 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6618 runs, 898.2 MB compared; 105 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 31818 in captures, 7 only in the collection, 217 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1108 MB), 0 retired still imported (0 MB), 8 retired given back (112 MB)
[hos host] vk: replay reads of guest memory as submitted: 2100000 (1081334 pages from the captures)
[hos sead::MainThread] nvn: QueueSubmitCommands #60000: queue 00001010103BF020 (its #10000), 1 handles (lr 0x80666be0)
[hos sead::MainThread] nvn: EventSignal (CPU) #20000: event 0x1009b38004 mode 0 value 1 (lr 0x8348c78c)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: write-backs left out (600 submits): 1310, 5366.5 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 981 dirty pages from their submits' captures, 105905 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): 1480 dispatches, 62490 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6719 runs, 914.1 MB compared; 106 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 32416 in captures, 9 only in the collection, 222 only by the replay itself
[hos host] vk:   read live, by pool:
[hos sead::MainThread] vk: readbacks of GPU results: 1281 of 3000 waited for their write-back, 0.06 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos sead::MainThread] vk: invalidates since the last readbacks line: 4146 kept their virtual core (nothing there to wait for), 3022 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos Presentation Thread] vk: capture: a submit 0.18 ms (lock 0.00, scan 0.05 over 1300 ranges, 1777 MB), 386 pages written, 386 copied (1.5 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (150 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 2120000 (1091743 pages from the captures)
[hos tera::FileResourceMgr] vcores: 313920 takes, 49332 waits (mean 31 us, longest 7.2 ms, 0 over HOS_VCORE_LOG_MS), 24813 kicks, 24607 yields, 0 starved grants, 0 escapes in 5 s
[hos ModuleSystemWorker3] nvn: CommandBufferWaitEvent #125000: event 0x10103c8044 mode 0 value 82763 (now 82744) (lr 0x80274390)
[hos host] vk: recorder: 4825197 packets run in 10.0 s (busy 1.33 s, spun idle 3.05 s, slept 4010 times, 862 of them under 200 us), 0 calls made directly (no room); the replay waited for it 23878 times, 83.68 ms in all, the longest 0.19 ms; vkEndCommandBuffer 7644/36.55 ms (max 0.09), replay_mu unlock 8632/26.66 ms (max 0.19), vkResetDescriptorPool 7602/20.48 ms (max 0.06)
[hos host] vk: the recording streams: 48 of 2 MB, the most one held 344 KB; 0 made on demand (0 MB kept now); c1 stages: at most 114 a recording; 0 descriptor pools made when a recording's own ran out
[hos host] vk: sparse mirrors: 55, 801 MB with memory of 1016 MB (214 MB spared); blocks given memory later 117, given back 1231; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (31.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 62 made new; 14 kept as spares, 0 destroyed; 7 spares now, 92 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1289, 5276.2 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1097 dirty pages from their submits' captures, 103968 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): 1458 dispatches, 61478 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6644 runs, 898.3 MB compared; 106 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 31886 in captures, 12 only in the collection, 214 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1108 MB), 0 retired still imported (0 MB), 8 retired given back (112 MB)
[hos host] vk: replay reads of guest memory as submitted: 2140000 (1102194 pages from the captures)
[hos host] vk: a frame: 51 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.70, present 0.21, draw 3.22, descriptors 1.35, texture lookups 0.07, uploads 0.00, pipelines 0.13, programs 0.59, commands 0.91, dispatch 0.67, memory sync 0.30, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 69 (801 MB): 7.32 MB synced in 1080.7 copies, 16.04 MB written back
[hos host] vk: 10080 presents, 70.6 fps; a frame: 823 draws (0 skipped), 103 dispatches, 142 passes, 142 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 41146 pipelines (4757475 ms), 0 programs (0 failed), 1216 textures (3478 MB)
[hos host] vk: host time in GPU resource waits: 0.49 ms a frame (fences 0.02, timelines 0.47, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 350.3 full barriers recorded, 182.8 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: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 43.3; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.2 ms a frame (the longest submission 3.4 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.04
[hos host] vk: replay reads of guest memory as submitted: 2160000 (1112734 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 1310, 5366.5 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1057 dirty pages from their submits' captures, 105768 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): 1480 dispatches, 62497 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6638 runs, 898.2 MB compared; 106 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 32361 in captures, 12 only in the collection, 219 only by the replay itself
[hos host] vk:   read live, by pool:
[hos Presentation Thread] vk: capture: a submit 0.18 ms (lock 0.00, scan 0.05 over 1300 ranges, 1777 MB), 386 pages written, 386 copied (1.5 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (152 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 2180000 (1123191 pages from the captures)
[hos ModuleSystemWorker5] nvn: DrawMeshTasksIndirect: 180000 so far (x1 0x1182d9c000 x2 0 x3 0x1182d9c000 x4 0x654040)
[hos host] vk: slow present: 18.9 ms:
[hos sead::MainThread] nvn: GPU waits in 5 s: 1 sync waits (10 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: present 10150 written to <home>\AppData\Local\<user>\data\jobs\jb_eff73394de0a68ff\runs\hyrule-field-g1-01-pr\shots/hyrule-field-pr-01-last-frame.png (the control bridge)
[hos ModuleSystemWorker1] nvn: CommandBufferSignalEvent #85000: event 0x10103c802c mode 0 value 84999 (lr 0x80667688)
[hos host] vk: program picks: 2367426 kept (992413 by the last picks' textures, a uniform buffer bound since), 776 new, 1431797 with other textures (of 3800000)
[hos sead::MainThread] vk: readbacks of GPU results: 1278 of 3000 waited for their write-back, 0.07 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos sead::MainThread] vk: invalidates since the last readbacks line: 4043 kept their virtual core (nothing there to wait for), 3125 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: sparse mirrors: 55, 801 MB with memory of 1016 MB (214 MB spared); blocks given memory later 117, given back 1231; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 89 binds (31.1 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 62 made new; 14 kept as spares, 0 destroyed; 7 spares now, 92 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 1288, 5274.3 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1055 dirty pages from their submits' captures, 104192 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): 1456 dispatches, 61367 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6732 runs, 914.1 MB compared; 126 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 31356 in captures, 23 only in the collection, 222 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 63 live (1108 MB), 0 retired still imported (0 MB), 8 retired given back (112 MB)
[hos host] vk: replay reads of guest memory as submitted: 2200000 (1133655 pages from the captures)
[hos host] vk: memory: working set 10873 MB (peak 11090), private 17695 MB, staging 32 MB in 2 arenas, 23 made so far (5 ms)
[hos host] vk: memory heaps: heap 0 (device): 164 blocks 5198 MB, 14191 allocations 4801 MB; used 5624 of 23554 MB; heap 1 (system): 15 blocks 608 MB, 40 allocations 560 MB; used 1796 of 104029 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 5 images 16 MB; 1 free ranges, the largest 15 MB
[hos host] vk: image blocks (render targets): 20 blocks 640 MB, 99 images 395 MB; 42 free ranges, the largest 32 MB
[hos host] vk: image blocks (textures, compacted): 56 blocks 1792 MB, 1106 images 1781 MB; 463 free ranges, the largest 9 MB
[hos host] vk: render target images: 48 reused from the pool, 263 made new, 160 destroyed from it; it holds 0 (0 MB)
[hos host] vk: VRAM: 5624 MB used of 23554 MB allowed (396 MB of it free in the blocks); textures 3483 MB (budget 11000 MB), 1217 textures
[hos host] vk: textures: decoded ASTC 1868 MB, rendered 1033 MB (101 drawn in one frame, kept: 62267 KB), other 350 MB; used in the last 8 presents 2385 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; 409 MB of blocks hashed in 52 ms
[hos host] vk:   288 MB: 0x1150c06000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10aea22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10ab1aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x114045c000 4096x4096x1 x1 levels 1 NVN 0xa, last used 4194 presents ago
[hos host] vk:   64 MB: 0x114045c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x1145b99000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   54 MB: 0x1000558d800 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 4181 presents ago
[hos host] vk:   54 MB: 0x10006146200 3072x3072x1 x3 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 2 (177 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 260x260x1 NVN 0xd: 1 (175 KB);
[hos Presentation Thread] vk: presents after the CPU's signal (600): 6 waited for it, 1.71 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.01, writes collected 0.05
[hos host] vk: status: 600 frames, the slowest 25.9 ms; VRAM 5.5 of 23.0 GB
[hos host] vk: program census (play): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 91 program records loaded (9.3 ms), 0 new graphics pipelines, pipelines from the driver's cache 0, from the start's cache 0, compiled 17; 368 draws and dispatches left out, 0 waits (0.0 ms)
[hos host] vk: program loads at create (play): 4613 queued, 4613 loaded (median 4.01 ms, 99th percentile 23.16 ms, worst 43.52 ms, 25789 ms and 978.9 MB read in all); 6518 programs (1543 of them compute variants without their code, 0 graphics programs no note names), 8179 graphics pipelines (0 not in a cache), 337 compute pipelines (267 not in a cache); a first use ran 23 itself and waited for 1 (103.3 ms); 623 pipelines made at create taken by draws; 2402 create calls with nothing kept for their stages, 0 before the index; units 7015 (4613 ready, 0 loading), 12472 programs alive; 21 noted graphics pipeline probes deferred to first use
[hos host] vk: module identifiers (play): 605 known, 0 learned in this run; 337 compute pipelines made by them at a create call, 267 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 51 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 0 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 5.57, present 0.37, draw 3.19, descriptors 1.33, texture lookups 0.07, uploads 0.00, pipelines 0.13, programs 0.60, commands 0.91, dispatch 0.64, memory sync 0.31, queue submit 0.37, slot wait 0.04
[hos host] vk: a frame: mirrors 69 (801 MB): 7.34 MB synced in 1079.8 copies, 16.04 MB written back
[hos host] vk: 10200 presents, 71.6 fps; a frame: 823 draws (0 skipped), 103 dispatches, 142 passes, 142 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.0 ms (recording slots only); 41146 pipelines (4757475 ms), 0 programs (0 failed), 1217 textures (3483 MB)
[hos host] vk: host time in GPU resource waits: 0.71 ms a frame (fences 0.02, timelines 0.69, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 350.4 full barriers recorded, 182.6 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: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 43.0; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: GPU time 7.4 ms a frame (the longest submission 3.2 ms); 0.0 textures made a frame
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: the render thread's whole frames (600 presents): 0 jobs went on before their present after 1000 ms, 0 times a game's thread waited for the work handed over, 0 times for a fence of the frame in the making
[hos host] vk: the render thread (600 presents, 13.93 ms a frame): its jobs 6.30 ms a frame (submits 5.88, presents 0.25, others 0.17), idle 7.63; the game waited for room at 1 presents, 11.05 ms each (0.02 a frame; 160.90 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); 8 captures at a CPU signal; the submits' waits read in 0.024 ms a frame; readbacks: 3000 next to a texture's storage (no catch-up), 0 in one, 0 read exactly (after the work handed over: among the catch-ups); a present made 6.35 ms after its hand-over (25.27 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 5.23 + 0.39 (the window's copy is in the idle), after it 0.73 + 0.36; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.03 ms of GPU time), after its last that reads it 1.00 (0.03 ms)
[hos host] vk: frame ahead (600 presents): 4.25 parts a frame recorded in the signal's job, 47.1% of them took nothing that came with the signal; of the others, 2.25 a frame took pages of the signal's capture into a mirror (16.9 pages a frame) and 0.00 pages read as guest memory had them (0.0 pages a frame); a frame's first part took nothing in 0 of 8 frames, its first three in 0
[hos host] vk: frame ahead, what came with the signal: 8 signals with a capture, each of 32.2 pages in 6.5 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: 0.33+0.18, 0.36+0.12, 0.02+0.03, 0.02+0.03, 0.01+0.01, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00, 0.00+0.00
[hos host] vk: readback pages refreshed from guest memory: 1107930 pages so far (0 held: may be written ahead in their submission)
[hos host] vk: replay reads of guest memory as submitted: 2220000 (1144112 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.18 ms (lock 0.00, scan 0.05 over 1300 ranges, 1777 MB), 383 pages written, 383 copied (1.5 MB)
[hos host] vk: a capture taken (ms): all 0.006; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (154 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: write-backs left out (600 submits): 1289, 5276.2 MB; 31 bindings followed, 46 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 1034 dirty pages from their submits' captures, 104423 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): 1458 dispatches, 61581 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6631 runs, 898.2 MB compared; 106 pages read into the shadows; 304 MB of shadows; pages the shadows knew that the CPU wrote: 31836 in captures, 9 only in the collection, 216 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker1] nvn: CommandBufferWaitEvent #130000: event 0x10103c8044 mode 0 value 86563 (now 86544) (lr 0x80274390)
[hos host] vk: replay reads of guest memory as submitted: 2240000 (1154508 pages from the captures)
[hos Presentation Thread] vi: 10267 frames presented (71.2 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 323, 15.5-17.8 31, 17.8-22 1, 22-30 1, 30-40 0, over 40 0