Commit Graph
172 Commits
Author SHA1 Message Date
Valiantsin DzerakhandClaude Opus 5.5 3b5469a83f Sample compare-sampled R32F and R16 textures of compute passes through a depth twin
TotK's compute passes sample a 4x4 R32_SFLOAT texture through a shadow sampler. Vulkan allows depth comparison only
on formats with SAMPLED_IMAGE_DEPTH_COMPARISON, which AMD's drivers do not give R32_SFLOAT, so the result there was
undefined (20 validation errors in 40 s of field play). Such an image now gets a D32_SFLOAT (or D16_UNORM) twin
copied through a buffer right before each dispatch that samples it, and the dispatch samples the twin. Formats the
driver compares as they are keep the colour image. TOTK_DEPTH_TWIN=0 turns it off. Design after Savage Wind Recomp's
depth twins.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 20:09:36 +03:00
Valiantsin DzerakhandClaude Opus 5.5 fb13c05870 Merge PR #86: drop released textures that reload intact when video memory runs low
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 19:56:08 +03:00
Valiantsin DzerakhandClaude Opus 5.5 43704c0d9f Turn off the recorder thread on AMD GPUs and GPL on AMD's RDNA 1/2 driver, and log GPU breadcrumbs on a lost device
AMD drivers lost the device at boot or at menu opens with the recorder thread; vk_recorder = 0 ended it for three
RX 9070 XT users (issue #5), and RADV saw an illegal opcode at the end of a command buffer (issue #40). It now
defaults to off on vendor 0x1002; TOTK_VK_RECORDER=1 turns it back on.

On AMD's separate driver branch for RDNA 1/2 ("AMD proprietary shader compiler") every device loss came with a cold
pipeline cache and draws waiting on fast-linked pipelines, and the recorder setting did not help there. TOTK_GPL now
has no fixed default: off on that branch, on elsewhere.

VK_AMD_buffer_marker breadcrumbs (vk_crumbs.c) mark the start and end of each submission part, upload batch, ASTC
decode chunk and present; on device lost the log names the last point the GPU reached and the last one it finished.

vk_fsr.c ensure_images held no queue lock around vkDeviceWaitIdle, which needs every queue externally synchronized.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 19:48:38 +03:00
Valiantsin DzerakhandClaude Opus 5.5 473acfda2e Merge task/nosleep: keep the display awake while the game window is up
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 15:37:26 +03:00
Valiantsin DzerakhandClaude Opus 5.5 3332250df4 Merge task/dpiaware: per-monitor DPI awareness for the window, fullscreen and the overlay
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 15:35:21 +03:00
Valiantsin Dzerakh f930e56825 Merge task/fade: floor the 1.4.3 LOD and pre-actor fade increments at 1 2026-10-30 15:30:46 +03:00
Valiantsin DzerakhandClaude Opus 5.5 afa3607259 Keep the display and the system awake while the game window is up and not minimized
Windows turned the monitor off after its idle timeout during play, since gamepad input does not count as activity (#7).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 15:30:16 +03:00
Valiantsin DzerakhandClaude Opus 5.5 39ab925588 Merge task/wbcap: diagnostic fence write-back limit and its trace log
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 15:28:18 +03:00
Valiantsin DzerakhandClaude Opus 5.5 6dbecab25d Merge task/amd-devlost: log the VK_EXT_device_fault report on device lost
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 15:27:58 +03:00
Valiantsin DzerakhandClaude Opus 5.5 bb1b675381 Floor the 1.4.3 LOD and pre-actor fade increments at 1
Above about 63.5 fps the fade step trunc(2.1 * ratio) is 0, so objects that start a fade stay translucent or
invisible until a slower frame moves them on (issues #10, #37). Switching to 60 fps and back cleared it because the
60 fps step is 1. Port of MacNativeGames' high-FPS fade fix as a fade set under TOTK_HIGHFPS_FIX.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 15:27:44 +03:00
Valiantsin DzerakhandClaude Opus 5.5 abb00c500c Add a diagnostics key for the fence write-back size limit and trace the targets it skips
wb_fence_limit (TOTK_WB_FENCE_LIMIT, default 65536) replaces the fixed 64 KiB limit of the write-back done when the CPU waits for a fence. With trace = gfx the log names each dirty target left on the GPU because of its size and each target over 64 KiB that a raised limit wrote back. Meant to test whether the fuse weapon switch bug (#18) and the Purah Pad camera bug (#13, #63) come from a target the game reads on the CPU without an invalidate.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 15:27:42 +03:00
arthurhfcardosoandClaude Sonnet 5.5 444abcbebb Drop released textures that reload intact when video memory runs low
On a 4 GiB GPU the released-texture list (up to 2048 entries, each ASTC texture decoded to RGBA8) filled video memory and the heap spilled into system memory: a Depths lava scene went from 12 ms to 60 ms of GPU time per frame and about 16 fps. Above tex_vram_pressure percent of the device-local budget (default 90, 0 = off), each submit now destroys up to 32 of the oldest released entries.

Only entries that come back intact are dropped: mirrored textures the GPU never wrote, which are uploaded again from their storage. A render target exists only in its image; dropping those made textures come back empty (black flashes in the Depths boss fight, about 170 "left uninitialised" aliases a minute).

Adds the setting to totk.ini and the Advanced tab, and a perf log line with the released count and drops.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-10-30 09:26:36 -03:00
Valiantsin DzerakhandClaude Opus 5.5 fcaecabad4 Make totk.exe per-monitor DPI aware so fullscreen and the window use physical pixels
The manifest declared no DPI awareness, so Windows ran the process DPI-unaware: on a 4K monitor at 150% scale the
fullscreen window and its swapchain were 2560x1440, the 4K game image was shrunk to it and Windows stretched the
result back up, blurred (issues #25, #56, #31).

The manifest now declares PerMonitorV2 (true/pm for older Windows 10). The window frame is measured at the window's
DPI, a framed window keeps its client size when it moves to a monitor at another scale, WM_DPICHANGED takes the
suggested rect or keeps fullscreen over the monitor, and window_size means screen pixels. The menu's minimum text
size follows the window's DPI, and the text entry dialog is measured at the game window's DPI. --probe-dpi prints
the process awareness and fails unless it is per-monitor v2.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 13:40:04 +03:00
Valiantsin DzerakhandClaude Opus 5.5 d8547a84ac Offer 90 FPS in the frame rate limit menu
GitHub issue #21 asks for a 90 fps limit. fps_limit = 90 in totk.ini already worked, because frame_rate.c accepts any whole
rate from 30 to 360 and cfg.c does not validate the values column. The menu row and the documented value lists did not
name it. Added 90 to the menu between 60 and 120 (new catalog id UI_MENU_FPS_90 in both languages), to the FPS_LIMIT
row in cfg.c and to totk.ini.example.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 13:30:09 +03:00
Valiantsin DzerakhandClaude Opus 5.5 67116a387f Map the zh-CN and zh-TW game languages to zh-Hans and zh-Hant
Choosing Simplified Chinese in the installer wrote lang = zh-CN to totk.ini and the game showed Japanese: ov_language() handed the code to nn::oe::GetDesiredLanguage unchanged, and TotK only accepts zh-Hans and zh-Hant for Chinese. Reported in #72 and #68.

ov_language() now maps both old spellings, so existing totk.ini files keep working and cfg.c still accepts them. The F1 menu offers one row per language under the canonical codes. The installer lists only zh-Hans and zh-Hant, writes them, and reads zh-CN or zh-TW from the CLI or from a recorded totk-install.json as the canonical code, so a repair or update of an old install does not fail with check.lang_unknown and rewrites the ini. --probe-services prints the configured language and checks the four Chinese codes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 13:27:19 +03:00
Valiantsin DzerakhandClaude Opus 5.5 112fb2dd74 Log the VK_EXT_device_fault report when the device is lost
vk_device_fault_report queries vkGetDeviceFaultInfoEXT once per process after VK_ERROR_DEVICE_LOST and logs the
description, the faulting addresses with type and precision, and the vendor codes. The submit, upload, occlusion
query and wait fatals call it before rt_fatal, and the present path calls it on a failed acquire, blit submit or
present. Threads that lose the device together wait until the report is written. The extension counts as usable only
when the driver reports the deviceFault feature. No cost unless the device is lost.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 13:21:14 +03:00
Valiantsin DzerakhandClaude Opus 5.5 8691f6d724 Remap keys, mouse buttons and controller buttons from the Controls tab
Every Switch button and the eight stick directions the keys drive get two key cells and a controller cell under the
button layout. A cell takes the next key, mouse button or controller button pressed; a key in use moves over, a
controller button swaps. The tab writes bind_ lines (new: bind_LSTICK_* / bind_RSTICK_*, key:default, pad:default) and
input_win.c builds the tables again when one changes, so remapping applies while the game runs. More keys can be
bound: Enter, Backspace, right Shift and Ctrl, Alt, the numpad, punctuation, F2-F7, F9, F12, mouse buttons 4 and 5.
Alt alone no longer enters the window menu loop.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-29 17:53:57 +03:00
Valiantsin DzerakhandClaude Opus 5.5 ba0f33e044 Pick the gamepad button layout by the pad's labels, with a Button layout row on the Controls tab
The gamepad map was by position only, so on an Xbox pad the button labelled A pressed Switch B. pad_layout = auto
(default) keeps the Nintendo layout for a pad whose east button SDL reports labelled A and uses the Xbox layout (south
A, east B, west X, north Y; cross A on a PlayStation pad) for the others; nintendo and xbox force one. It applies at
once; bind_ lines name physical buttons and apply on top of either layout.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-29 17:20:03 +03:00
Valiantsin DzerakhandClaude Opus 5.5 ce150130df Make code comments say what they mean instead of pointing at things a reader cannot follow
Comments cited codes from local notes (port-lessons P1, backlog), task labels (T6c1-T6c3, M4), bare upstream commit
hashes and issue numbers, a file that no longer exists (nvn_api.c, now nvn_api_*.c), a generated file (main_0103.c)
and line numbers into upstream files that refs/ has moved past, so they pointed at unrelated code. Those now name the
file, function or project and say what the cited code does. Attribution of copied code keeps its pinned commits.

Corrected facts: TOTK_REPLAY_ASYNC defaults to 0, and the shader cache, pipeline cache and pipeline keys live next to
totk.exe, not in build/v121.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-29 12:58:01 +03:00
Valiantsin DzerakhandClaude Opus 5.5 d3fd4c5944 Call the replay "GPU frame prep" in the menu and the overlay, say it adds up to one frame of latency, and name GPU time "GPU busy per frame"; the run_game.py example runs build/v121
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-29 12:49:34 +03:00
Valiantsin DzerakhandClaude Opus 5.5 c7b290f2b0 Write the log of a game run to logs\totk.log next to totk.exe when stderr is not a file
A player who starts totk.exe from Explorer or a shortcut had no log to attach to a report: the log went only to stderr.
A game run whose stderr is not a file (a console or a pipe) now also writes logs\totk.log, keeps the four earlier
runs as totk.1.log to totk.4.log, and stops the file at 64 MiB. Runs whose stderr is a file (run_game.py, 2> file) and
the tool modes write no extra file. Lines from before the file opens (mem_init comes first so the CRT heap stays out of
the guest address space) wait in a static buffer.

Crash reports, fatal errors, IPC and SVC diagnostics, guest debug output and the boot: lines now go through
rt_logf / rt_log_raw, so they reach the file too. Open log folder in the menu opens the folder of the file.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-29 12:41:12 +03:00
Valiantsin DzerakhandClaude Opus 5.5 f4619f7b45 Add a VERSION file with the port's semver, shown beside the installer title and on the game menu's title line with the game update
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-29 11:40:01 +03:00
Valiantsin DzerakhandClaude Opus 5.5 4e764cd39a Take world-clock windows in TOTK_PROFILE_FRAMES and TOTK_CRIT: w<lo>-<hi> is seconds of game time since the save loaded, so a profile or critical-path session covers the same play at any frame rate instead of a present range that lands elsewhere on a slow machine
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-29 08:56:50 +03:00
Valiantsin DzerakhandClaude Opus 5.5 b35f66207f Merge branch 'main' into task/wwatch
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-29 08:35:10 +03:00
Valiantsin DzerakhandClaude Opus 5.5 1274fb65d8 Read totk.ini in mem_init before the pool size: the size asks resolution.c, which decides output_res and render_scale once on that first call, and main read the file only after mem_init, so both settings from totk.ini and the menu were ignored and only TOTK_* environment variables reached them
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:45:48 +03:00
Valiantsin DzerakhandClaude Opus 5.5 1f7d80306e Freeze idle regions that also hold memory outside NVN pools, and freeze after 60 idle submissions instead of 300
Write-watch queries skip frozen 2 MiB regions, but a region was frozen only when it lay wholly in imported pools and
wholly committed. NVN pools are small and not 2 MiB aligned, so most regions also hold other guest memory or reserved
holes: in town 340 of the 436 MiB queried every frame lay in regions nothing had written for 300 submissions. On the
GTX 1050 Ti laptop the queries took up to 2.2 ms a frame, 1.4 ms of it in queries that found no write.

mem_freeze now makes the committed pages of a region read-only when all of them are read-write and leaves reserved
holes alone; thaw_locked already restored such regions run by run. The pool rule is gone: the only kernel writes into
guest memory are file reads, which thaw their buffer first, and every user-mode write faults and thaws as before.
TOTK_FREEZE_IDLE defaults to 60 (in a fight 100 of 257 queried MiB lay in regions idle for 30-299 submissions; a thaw
by a write costs about 60 us on the RX 9070 machine), and import mode freezes too, for the texture checks of APUs.

RX 9070, per frame after load: town 436 -> 96 MiB queried, 0.68 -> 0.38 ms; fight 168 MiB, 0.45 ms (257 MiB, 0.57 ms
with the new rule at 300 submissions), about 20 thaws a second. TOTK_MEMORY=verify in the fight: 3.65 million pages compared, 0 mismatches. The gfx-abi probe
covers a region with reserved holes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:22:30 +03:00
Valiantsin DzerakhandClaude Opus 5.5 3af201b43b Bind uniform blocks from a copy taken when the submission starts recording
TotK rewrites per-object uniform blocks on other threads soon after it submits the frame that reads them, and the
replay records for milliseconds after the submit call, so late draws could read another object's matrix (phantom
rocks, a flat far field). The guard in nvn_api_core.c obs_release_read only knows 1.2.1's producer. runtime/vk_usnap.c
copies every uniform range of imported pools that the submission binds into an arena, after MacNativeGames'
TOTK_UBO_SNAPSHOT, and ubo_info binds the copy; ranges a submission still in flight writes are left out, and GPU writes
within the submission drop the copies they touch. TOTK_UBO_SNAPSHOT=0 binds live memory; =ab alternates.

Town on an RX 9070 (in-launch A/B, 70 s): 2235 ranges, 1.9 MiB per submission in 0.49 ms; interval +0.02 ms, replay
+0.19 ms. The sort of the ranges was half of the cost with qsort (0.53 ms for about 10,000 bindings) and is a radix
sort now. fight2 interval within noise. The four probes passed on 1.4.3.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 20:25:40 +03:00
Valiantsin DzerakhandClaude Opus 5.5 c2943970cc Merge the spin wait gates: they also end the fight hangs
The generator emits a guest poll such as `ldr w9, [x19]; cmp w9, #1; b.eq` as a plain C load, and clang hoists it out
of the loop: the lock wait in f_7001fa0e2c compiled to `cmp; jne; jmp .`, so a thread that found the lock held spun
forever after its holder released it. A fight run hung this way in 3 of 13 launches on an RX 9070, with the UBO
snapshot on and off; gdb showed the thread on that `jmp` in every sample. The gate is an opaque call on every
back-edge of the 868 loops tools/spin_waits.py lists, so each pass reloads memory: the exe goes from 182 self-jumps to
148, and 144 of those are the game's own `b .` instructions. 8 of 8 fight runs then passed; tests/diff/func.py with
--patches passed 5470 cases on 878 functions, and the four probes passed on 1.4.3.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 20:05:44 +03:00
Valiantsin DzerakhandClaude Opus 5.5 1f4c6fcdb5 Decode the game's zstd frames with the host's libzstd: the patch set zstd (TOTK_ZSTD_NATIVE, on by default) sends the calls of ZSTD_decompressMultiFrame and ZSTD_decompress_usingDDict in main to runtime/ov_zstd.c, which makes host dictionaries from the guest DDicts, gives the virtual core up for inputs of 64 KiB and more, and leaves in-place decodes and every error to the guest decoder
The lifted decoder ran its copy loops as chains of small functions. On the island and fight routes TOTK_ZSTD_CHECK=1 decoded every call both ways with no difference; the host took 0.72 s where the guest decoder took 1.99 s. Load time and fps on an RX 9070 and a 7900X did not change: the gain is CPU time on the loading threads, for machines with fewer cores. The caller at main+0x547a04 stays on the guest decoder: it resets a static DCtx with which the game's zstd rejects some frames libzstd decodes, and the game relies on that result.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 18:27:12 +03:00
Valiantsin DzerakhandClaude Opus 5.5 4e9322730d Log the load time with the world clock start: the line now says how many seconds after the guest started the world clock began, so load times compare across runs without the presents count
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 18:27:11 +03:00
Valiantsin DzerakhandClaude Opus 5.5 72253abcc0 Run and measure by game time instead of presents: totk.exe --seconds S ends a run once the world clock passed S seconds (run_game.py --seconds, regress.py routes in world seconds), TOTK_AB blocks are seconds of world clock, the perf line carries the world clock, and TOTK_SHOTS=w<lo>-<hi>:<step> takes shots by it
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 16:51:36 +03:00
Valiantsin DzerakhandClaude Opus 5.5 3fb4b8d4b9 Make the replay's Vulkan calls on a recorder thread: each vkCmd* into the replay command buffer becomes a packet in a ring that the recorder makes in order, every other command buffer is called at once, and the submit waits for the ring to empty. Town on an RX 9070: replay 9.98 -> 8.29 ms a frame (in-launch A/B), no new validation messages, same picture
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 16:41:46 +03:00
Valiantsin DzerakhandClaude Opus 5.5 48dfe0cfdf Measure a change within one launch: TOTK_AB=<presents> alternates A and B blocks (A B B A), a switch set to ab is on in the B blocks, and tools/ab_summary.py compares the perf windows of each side; TOTK_ONE_PER_CORE=1 keeps one logical processor per physical core (off: no gain on a 7900X)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 16:41:46 +03:00
Valiantsin DzerakhandClaude Opus 5.5 166e5a7b7d Drop released textures oldest first, one per release past 2048, instead of all of them once the count passed 4096: that sweep destroyed thousands of images in one frame, 333-422 ms frames every 40-50 s on a GTX 1050 Ti, and the textures still in use came back as new images and uploads
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 15:14:11 +03:00
Valiantsin DzerakhandClaude Opus 5.5 01dfe5d5a6 Take staging buffers from a VMA pool of their own that keeps two 64 MiB blocks from boot: in the default pool a block made by the headroom thread or by a burst of uploads held the render thread's staging allocations under the memory type's lock, 47-50 ms for one in the first frame after a save load (that frame 138 -> 70 ms)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 15:06:40 +03:00
Valiantsin DzerakhandClaude Opus 5.5 41f201ad39 Skip full-screen draws while their pipeline builds by default (TOTK_PIPE_SKIP=fullscreen): in a fight a full-screen pass whose stage libraries were cold in the driver cache held three frames in a row at 64-72 ms, each spending the 50 ms wait budget, and some draws were skipped after all
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 15:02:18 +03:00
Valiantsin DzerakhandClaude Opus 5.5 9dc738c045 Count untimed script lines in 60 Hz ticks of game time instead of presents, so the title menus get scripted presses at the same game time at any frame limit; in a recording the untimed lines stop when the world clock starts, and the recorder cuts a button or stick held at that moment and continues it as a timed event
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 14:49:41 +03:00
Valiantsin DzerakhandClaude Opus 5.5 5da6c5bb89 Merge the frame critical path recorder: TOTK_CRIT=<first>-<last frame> with TOTK_TRACE=sched writes the guest threads' state changes and wakes, read by tools/crit_path.py
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 14:13:40 +03:00
Valiantsin DzerakhandClaude Opus 5.5 2fe6a6d4c1 Keep MXCSR in line with FPCR where FPCR changes instead of checking it in every FP op
The FP helpers compared c->fpcr with a cached copy before every operation to resync MXCSR lazily. MXCSR now
follows FPCR.FZ at once: MSR FPCR calls rt_fpcr_sync, and so do rt_call (thread, fiber and host entry into guest
code), fiber resume (FIBER_FLAG_FLOAT_SWITCH fibers keep their own MXCSR) and the differential test harness per
case. op and vop only test FPCR.RMode for the slow path, and their redo check is one ordered-not-equal compare of
|r| with the smallest normal, the same NaN-or-smallest-normal set as before.

Hot guest code shrinks by about a quarter (132 functions holding half the guest cycles, 1.4.3); fight frame time
moves 1-3%, inside the run-to-run spread. lift, simd, func, atomic and flags_check pass on 1.4.3; simd.py fails
without the harness sync.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 12:25:25 +03:00
Valiantsin DzerakhandClaude Opus 5.5 ba52fe6e89 Record the critical path of the frame with TOTK_CRIT
With TOTK_TRACE=sched, TOTK_CRIT=<lo>-<hi> records every state change and wake of the guest threads and every
present over those frames (runtime/kernel.c crit_add from acct() and wake(), crit_frame at the present) and writes
them to TOTK_CRIT_OUT at the last frame. tools/crit_path.py walks back from each present through the waits to the
thread whose signal ended them and splits the frame interval by thread and state; the sums equal the interval.

A 1.4.3 fight: with the replay on the submitting thread the path is 7.3 ms of replay on the render thread and 4.3 ms
of guest code; with TOTK_REPLAY_ASYNC=1 it is 10.1 ms of guest code through the fork-join of the main thread and five
workers, at the same frame interval.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 09:26:07 +03:00
Valiantsin DzerakhandClaude Opus 5.5 27056b212a Keep NZCV in a local of each generated function instead of c->nzcv
Every flag-setting instruction packed N, Z, C and V into c->nzcv and every reader unpacked them. recomp/flags.py
now runs a reaching-definitions pass per function: a compare stores its operands in a local Flags F, and a reader
whose definitions are all of one kind computes its condition from them, so cmp and b.cond become one host compare.
CCMP, FCMP, MSR NZCV and ADCS write the packed form, which readers reached by several kinds also use. Calls and
returns publish nothing (AAPCS64 neither passes nor preserves NZCV); tail branches, falls into the next function and
br through rt_dispatch publish c->nzcv, since discovery can start a function inside another one. Functions with an
unresolved computed br keep flags in c->nzcv.

1.4.3 main: 492k direct and 89k packed flag reads, .text 342 -> 299 MiB. Fight bench (bench-fight2-2026-10-27b,
frames 5400-7500, uncapped) 13.15 and 13.63 ms vs 13.9 before. tests/flags_check.py checks the new helpers against
the packed ones; tests/diff/func.py no longer compares NZCV, which generated code leaves untouched.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 18:07:58 +03:00
Valiantsin DzerakhandClaude Opus 5.5 101f2c8b5e Give the virtual core up in nn::fs host functions and in NVN memory pool setup, as the Switch blocks those threads in IPC: the game's loader threads held a core through CloseFile, CreateCacheStorage and nvnMemoryPoolInitialize (up to 20 ms) while job workers waited. Core diagnostics now also report waits on an owner in a host call, with its module, offset, latest SVC and guest lr, and the AllocateMemoryPages and AllocateAddressRegion lines need TOTK_TRACE=host
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 16:42:14 +03:00
Valiantsin DzerakhandClaude Opus 5.5 5b783728e5 Count NVN calls per thread and add them to the shared counter 1024 at a time: the main thread and the job workers record command buffers, and a locked add per call kept the counter's cache line moving between their cores
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 16:29:32 +03:00
Valiantsin DzerakhandClaude Opus 5.5 8e15d70fd4 Scale the performance counter by a fixed-point multiplier set before main: plat_now_ns and rt_ticks divided a 128-bit product by the counter frequency on every call, and __udivti3 took 5% of the game's running samples
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 16:22:11 +03:00
Valiantsin DzerakhandClaude Opus 5.5 acd550e822 Merge the VRAM headroom: a background thread allocates the next VMA block of each memory type before the render thread needs it (vk_headroom.c, TOTK_VRAM_HEADROOM)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 16:09:38 +03:00
Valiantsin DzerakhandClaude Opus 5.5 d1149e5060 Allocate the next VMA block of each memory type on a background thread: a new block made by another thread wakes vk_headroom.c, which allocates half a block every 500 ms and keeps it as the anchor of a block it had to create, so the render thread finds room instead of calling vkAllocateMemory for 8-19 ms; no headroom above 85% of the VRAM budget (TOTK_VRAM_HEADROOM)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 16:05:24 +03:00
Valiantsin DzerakhandClaude Opus 5.5 bac4d6cf91 The shader memo skips its 1 KiB handle compare while no texture or image was bound
memo_match compared the vertex and fragment stages' 64 texture and 64 image handles on every draw. Every OP_BIND_TEX and now every OP_BIND_IMG bumps tex_serial; a key whose serial is current has the same handles, and a key that matched by the compare takes the current serial. memo_match went from 0.95% to 0.27% of the fight replay.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 15:54:00 +03:00
Valiantsin DzerakhandClaude Opus 5.5 c238c341c7 Replay command buffer parts span at least a millisecond of recording
A fight frame of about 2,000 draws made 15 parts on the 7900X at 64 draws a part, each split costing the render thread a submit, a new command buffer and a mirror prologue (vk_sync_split, 7.6% of the replay). maybe_split now also waits until TOTK_SPLIT_US (1000, 0 off) has passed since the part began, so a frame makes 7-9 parts there. A millisecond of recording is well under a millisecond of GPU work, so the GPU still starts early in the frame.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 15:53:59 +03:00
Valiantsin DzerakhandClaude Opus 5.5 c9c06f2501 Render target views of registered textures live as long as their image
vk_exec_pass_begin created every attachment view again in each submission and freed them at its fence: 2% of the fight replay on the 7900X, more on drivers with a slower vkCreateImageView. view_of now takes the view from vk_exec_target_view, which keeps it in the sampled-view tables under the texture lock, and vk_exec_image_gone releases it with the image. Images that are not registered textures keep the per-submission views.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 15:53:59 +03:00
Valiantsin DzerakhandClaude Opus 5.5 3984344635 Spin wait gates: tools/spin_waits.py (the nxnative search) lists the loops that only another thread can end, the patch tables carry them as spin lines (857 in 1.4.3, 864 in 1.2.1), the generator sends their back-edges through rt_spin_gate, and the gate pauses the core once per pass and hands the virtual core to a waiter after 50 us of polling (TOTK_SPIN, TOTK_SPIN_HANDOFF_US)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 15:51:47 +03:00