The game's line width was dropped at the call and every line drawn 1 pixel wide; the game draws two lines a frame at 4.5 pixels into a 144x144 mask. The width is now recorded with the draw state and set as dynamic state for draws of the line topology class, held to the device's range. Only pipelines of that class list the new dynamic state, so the others keep their entries in the pipeline caches and the driver's. line_width = 0 draws every line 1 pixel wide as before.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
The 8 ms and 16 ms tried in their place left fewer draws out in a scripted walk, but in play a frame that takes 10 to 14 ms longer felt worse to the player than one that is left out, in two sessions on the same install. The wait stays counted once for each pipeline, and a frame that has left a draw out already still waits for nothing more.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Such a frame is held back or shown without that draw whatever its other pipelines do, and with hundreds of pipelines new at once every frame of an area's load spent its whole 16 ms budget on waits that could not complete it: 49 frames over 25 ms in a walk with no pipeline recorded, against 17 with the earlier 3 ms waits. With the rule the same walk has 15, the picture stands still 15 times instead of 50, and 1 frame is shown incomplete instead of 22.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
A draw that met a new pipeline waited 3 ms for its fast link, in every frame, which is just under what the link takes with the stage libraries in the driver's cache: in a first-visit session on a new install 66 of 87 such draws gave up at the end of the wait and their frame was held back. The wait is now 8 ms for each pipeline, counted from its first miss and shared by all its draws, and 16 ms for a frame, so a pipeline the driver has to compile costs one frame the wait and the frames after it nothing.
The pipeline list tool also takes the compute variants a game folder dispatched, which no pipeline record names and which were translated in the middle of play, and reads a list made before as one of its inputs.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
The 1x1 images bound where a slot has no texture were only moved to GENERAL, so their content was undefined. A material that read its roughness map through one could see roughness near 0, and on Kakariko's distant cliffs at night the specular term reached 3392 in single pixels of the scene target, which bloom and the temporal upscaler spread into bright white spots. Sampled stand-ins now read as opaque mid grey (a middling roughness, a flat two-channel normal), storage stand-ins as zero.
With NX2NAVI_REPLAY_PIPE=1 (off by default) the submitting thread hands a group's chunks to the workers, takes chunks
of it itself until none is left to take, and goes on with the ops after the group in a new command buffer after the
chunks' places in the part (vk_sync_append_later), also into the next group's front pass.
Up to four groups are handed out and not in their part yet; their chunks wait in one queue in the order of the stream.
The submitting thread waits for them only where the order needs it: for room, before a part goes to the GPU, before
copies and texture or buffer clears, where DLAA's run or the contact shadows' hook acts, and where a lookup would make
a texture entry or take back a retired one (the op goes again afterwards, as a chunk's does in turn).
What the chunks read stays as the stream had it while the submitting thread runs ahead: uniform snapshot drops count
from their place in the stream, a chunk reads only the inline uniform updates before its group, the HDR present's
source goes by stream position, the texture pool set syncs under the chunks' lock, each group keeps its own render
conditions, occlusion segments and closed query slots, and the queries after a group's last report go into the next
segment made after it. A group's first chunk runs its pass hooks after the group before began its pass; turns wait
for every earlier chunk, of earlier groups too. Locks the chunks take, the submitting thread takes while it is behind.
Draws keep the sampled-texture descriptors they resolved per draw context and reuse them without the texture lock
while their slot is unchanged; the workers waited for that lock at nearly every draw. Texture lookups that change
nothing no longer wait for their turn. The perf line says where the submitting thread waited, what the chunks took,
and which texture lookups waited; NX2NAVI_GPU_TIMELINE also shows the chunks and the waits.
Off by default because it does not pay yet: in Hyrule Field with 3 workers it shortens the replay (11.5 against 12.1
ms a frame) but not the frame (49.2 against 49.6 fps in the same batch). The GPU gets a group's commands only once its
last chunk is recorded, so the main thread's wait for the GPU after the frame's first submission grew by 0.8 ms.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Maxwell samples a texture by its own descriptor, so a 2D texture under a 3D sample gives its texel at (u, v). Vulkan
has no 3D view of a 2D image, and get_view found no view for such a binding: the draw sampled a stand-in. Such a
texture (plain colour, one sample, up to 1024 x 1024 texels) now gets a 3D image of one slice with all of its levels,
made in the upload batch and copied again when the texture was uploaded or drawn into since, or another layer is
asked for (vk_texture.c vk_texture_flat3d_locked). NX2NAVI_TEX_FLAT3D=0 gives the stand-in as before.
Seen in Kakariko: two of the game's 4 x 4 RGBA8 textures are sampled as 3D textures, and each got its copy; the runs
through eight reloads at 60 and 144 fps ended without errors.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Mesh draws ended a group and stayed on the submitting thread: 119 a frame in Hyrule Field at about 13 us each, and the
groups around them were cut short. They are recorded in the chunks now: the entries filled ahead for the indirect ones
are the draw context's, the ring index is taken atomically and the ring is made before the chunks record, each chunk
counts the ops that break a run of them on its own, and the lazy lookups of the mesh and shading-rate entry points store
their pointer before the flag that says it is there. With the mesh fallback, and while NX2NAVI_GPU_PASSES copies the
workgroup counts, they stay on the submitting thread.
A chunk's closed occlusion query slots go into a list of its own and into their segments once the group is done, in
chunk order, instead of under a lock per query. The per-object motion check decides alone per draw context (its
variant memo is per context) and takes its lock only to list a draw. The occlusion query ring has 16384 slots: the
chunks' blocks are sized to their draws, and with 4096 the ring came round to slots not reset yet.
The perf line splits the submitting thread's time with workers on: ops outside groups by kind, the front pass, its own
chunks and the wait for the others, and which ops ended the groups.
Hyrule Field, RX 7900 XT, 3440x1440, FSR 4 Native AA, same batch: standing 51.7 fps with 3 workers against 43.6 with
the workers off (replay 11.2 against 15.1 ms); along the field route 48.9 against 44.7. Screenshots standing and along
the route match the replay without workers.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Samples-passed resets and reports ended most groups in Hyrule Field (the game reports twice without a reset there, so
counting stays on): the front pass now replays them in the order of the stream, makes each occlusion segment and
queues its completion callback in that order, and each worker follows the counting from its chunk's first op and adds
the queries it closes to the segment the stream gives them. Per-object motion takes a lock instead of waiting for the
earlier chunks: its list's order does not change the picture. The locks the chunks share are taken only inside a
chunk, as outside groups the submitting thread alone replays; taken always, they cost the replay about 1 ms a frame.
The perf line counts the draws by what ended their group.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Stage 2 of NX2NAVI_REPLAY_WORKERS (default 0, the replay as before). A run of draws is cut into groups: a group ends
where the render targets change after its first draw or clear, and before an inline uniform update, a mesh draw and a
samples-passed counter op, which the submitting thread replays itself in the order of the stream. A group of 64 draws
or more is first walked by a front pass on the submitting thread: it applies the state and binds, settles the CPU side
of the timestamp reports and decides the render conditions in the order of the stream, and copies the replay state at
the first op of each chunk. The workers and the submitting thread then take the chunks in order and record each into a
command buffer of its own, which vk_sync_append puts into the part in the order of the stream.
The frame's pass hooks (contact shadows, DLSS) run in a group's first chunk, which the others start after. What still
depends on the order between chunks waits until every earlier chunk of the group is done: a texture lookup that would
make or take back an entry returns nothing and the chunk replays the op again in turn, and so do the contact shadows'
pass, per-object motion and the HDR pass's source. The tables the chunks share take locks (the shader table, the
descriptor lists, mirror use and GPU write notes, uniform snapshot drops, target frames, the DLAA camera votes and the
previous-frame motion notes); each chunk keeps its own occlusion query open, its own feedback memo and no pass marks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Reading the thread-local pointer in the functions that run per binding and per descriptor cost about 0.3 ms of
replay a frame in Hyrule Field (45.0 against 45.8 fps). VkExec.dc now names the context; the draw and look-ahead
entry points fill it from the calling thread's on first use, the functions that hold the state read it from there and
the few helpers without it take it as a parameter. A replay worker's copy of the state carries the worker's context.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Stage 1 of recording a submission on several threads. A submission's draws come in runs between ops that have to
stay in the replay's order on its own thread (dispatches, copies, texture and buffer clears, events, the game's
barriers, calls). With NX2NAVI_REPLAY_WORKERS=1 a run of 16 draws or more is recorded by a worker thread: it takes a
copy of the replay state, records the run into a command buffer of its own pool with a draw context of its own, ends
it behind a full barrier and hands the state back. vk_sync_append puts that command buffer into the part between the
replay's own pieces before and after it, so the GPU runs everything in the op stream's order. The submitting thread
waits for each run, so every side effect keeps the replay's order; this stage checks that a run recorded apart from
a copy of the state gives the same pictures. The default, 0, replays as before.
walk's handling of the ops that do not steer it moves into op_one, which the worker uses too. NX2NAVI_REPLAY_RUNS=1
logs every 600 frames how the draws fall into such runs and which ops break them. The draw replay's hot functions load
their recording context once instead of at every use.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
vk_exec_draw.c kept what a draw, bind or dispatch carries to the next one in file statics: the state the command
buffer holds, the descriptor signatures and skips, the per-binding reuse, the shader and pipeline memos, the shader
lookup's caches and the submission's transient resources. They now live in one DrawCtx, and every thread starts with
the replay's own, so the replay behaves as before. A later stage can give each recording thread a context of its own.
vk_exec.c's inline commands (vk_exec_inline_begin) find the replay state the calling thread records with through a
thread-local pointer too, instead of the file's single state and a flag.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A staging buffer, a transient buffer of the replay or a descriptor pool that the driver refuses for memory used to end
the game at once. Now the buffers kept for reuse go back to the driver first: the staging cache, the staging arena's
free chunks, the write-back snapshot pool and the headroom's spare blocks (no new spare block for 30 s after that), and
the allocation is tried once more; only a second refusal ends the game. On integrated GPUs the device-local heap is the
machine's RAM and fills before the commit checks see anything: staging buffers of 11 KiB to 16 MiB refused there ended
the game in Hyrule Field and on the first scene of play (#465, #567).
NX2NAVI_STAGING_FAIL_AT=n (diagnostics) takes the n-th staging buffer as refused once, to test the path.
Check: a Kakariko run with NX2NAVI_STAGING_FAIL_AT=300 logged "staging buffer of 524288 bytes refused (-2): 348.2 MiB of kept buffers and spare
blocks given back, made at the second try" and played on to the end of its script. The transient buffer and descriptor
pool paths call the same give-back and were not forced.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A place's task and mesh programs all arrive in its first frames, and each one's draws are left out until its compute
pipelines are built, so with one thread the foliage of a first visit filled in one program at a time.
NX2NAVI_MESH_FALLBACK_WORKERS (auto: one thread for each 4 logical processors, 1 to 4) sets how many threads make
them, and the log says how long the queue took to empty the first time.
Measured with the fallback forced on, no pipeline cache and new exe names (cold driver cache), the adult save's
first 75 s: with 1 worker the first 5 programs were ready after 420 ms, 3,266 draws were skipped while a pipeline or
program was pending, 1% low 16.1 fps and 145 frames over 50 ms; with 4 workers 256 ms, 956 draws, 55.7 fps and 35.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Ocarina's last post-process pass stores RGB9E5 texels through an integer view into its R11G11B10F scene target, and
the stand-in for the untranslatable upscaler decodes those raw words. Now and then the same pass writes them into the
RGB9E5 image at that address instead (1 of 529 frames sampled in Hyrule Field on an RX 9070 XT, #522): the stand-in
noted only R11G11B10F images, took the R11G11B10F image there, which held an older picture, and the frame showed a black
scene under an intact HUD. Writes through an integer view into either image are now noted with their order, the
stand-in takes the one written last, and a native RGB9E5 image is decoded as it is, without the copy that turns raw
words into RGB9E5.
Checked: the runtime compiles; a 60 s run without DLSS (the stand-in in use) logs "RGB9E5 fallback decode on", ran at
142.7 fps with no error. The native write itself did not occur in that run; it is rare and AMD-reported.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Ocarina anti-aliases every frame with its own SMAA in three compute passes after the bloom composite. The new aa
setting (nx2navi.ini aa, the Screen tab's Anti-aliasing row) keeps that by default. off, fxaa and smaa leave out the
game's edge and weight passes and replace its composition with a pass of our own that writes the same square-rooted
RGB9E5 words into the scene target: a plain copy, an edge smoothing written here, or the SMAA reference implementation
with its area and search textures (MIT). taa is DLSS or FSR 3.1 at the scene's size, which the row sets through the
upscaler; while either runs, the game's SMAA stays whatever aa says. The game's passes are known by a hash of their
guest code. The shader build now also rebuilds a header when a file its source includes changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Jitter was off on Ocarina because with it the ground and characters went white within seconds. The cause: the game draws its depth pre-pass into the scene depth, copies it into a second image of the same size, and shades the same geometry into the light target against the copy with the depth test EQUAL. Only the draws into the first image took the jitter, so the shading draws compared unjittered depth with jittered depth, EQUAL failed on nearly every pixel, and the light target kept its clear colour. Frame dumps with and without jitter show the shading passes darkening the light target from a mean of 133 to 13 without it and leaving it at 112 with it.
On Ocarina's frame every draw that projects through the camera (it binds the scene context at c3) into a D32F depth of the scene's size now takes the jitter, and full-screen passes drawn with a scene depth attached do not: the bloom composite DLSS reads would otherwise move the picture by the jitter a second time. The viewport is recorded again whenever a draw's jitter differs from the last one recorded. Jitter is on for Ocarina by default (dlaa_jitter = 0 still turns it off), and the DLAA log line counts jittered frames and draws.
Checked in Hyrule Field with the camera turning: the picture's brightness matches the run without jitter (mean 33 to 63 against 35 to 60; 144 to 188 before the fix), and in 20 motion-vector checks the warp error is lowest with the jitter applied as DLSS is told it (0.032 to 0.077), against 0.076 to 0.141 with its sign flipped and 0.059 to 0.106 left out, still cameras included: the scene moves by the jitter of that frame, in render pixels.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The stand-in for the game's tensor-core upscalers wrote the largest image bound to the dispatch. Image slots keep
their bindings from earlier dispatches, and with a 3840x2160 picture the scene upscaler leaves its 3840x2160 RGB9E5
target in slot 2, larger than the movie network's real 2560x1440 output in slot 1. The output kept stale bits for the
whole movie, which showed as frozen coloured stripes (issue #520). The output is now looked for only among the image
entries the program's NVN2 control section counts (at least slot 0). The binding log of NX2NAVI_LOG_CS_BINDS also
prints those counts.
Checked with Dm_Kokiri_0095 at game_res 2160 with DLAA (stripes before, the movie after), at 1440 with DLSS
performance (unchanged), and in Goron City gameplay at 1440 (same picture, 164.9 against 159.8 fps).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
HDR on is the game's scene through the observer-response mapper, anchored to the game's own mid grey and tone slope, with half of the hue bend of the game's clamp: the defaults the menu had. The tone mapper choice, the mid grey, the tone slope and its match, the hue clip, exposure, highlights, shadows, contrast, saturation, highlight saturation, blowout, flare, hue, grade and gamma are gone from the menu, the settings table and the example ini; a file that still sets them works as before, the keys read by nothing and logging nothing. The menu's HDR row is Off and On, with the peak, game and HUD whites under it. The game's own HDR is still there for a file that sets hdr = native, and the row lists it only then. The luminance mapper and its grading code, which no setting reaches any more, are removed from the tonemap library, and the support buffer carries one vec4 for it.
The dark tones of a dark scene sat far above black: the present showed the game's sRGB-encoded picture as that encode decodes, linear near black, while the game is graded for a display of gamma about 2.2, and since the observer-response mapper became the default the 2.2 correction ran with the luminance mapper only. A scene value of 0.003 came out at 0.29 nits at a white of 350 nits, where a 2.2 display shows the game at 0.015. The present now always shows the picture as a 2.2 display would, and the tonemap aims its peak lower by that correction's inverse, so the brightest light still reaches the display's peak; the same value comes out at 0.017 nits.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The author of pull request #353 closed it on 2026-11-11 with the words that they give no permission to use its code. Its port (3134f524) goes: the mark of a pipeline that declares the table (VkPipeInfo.gpa, vk_pipeline.c pipe_add) and of the part that draws with it (vk_mem_gpa_use in vk_exec_draw.c bind_resources and mo_emit), the sync of the table's pools in every such part's prologue (vk_mem_gpa_sync), the submission with a callback in the replay's place as it stood (vk_sync_probe_submit), and --probe-vkmem's read through the table with its shader (test_gpa_read, runtime/shaders/gpa_read.spvasm).
In its place, after pull request #551, which was written without that code: a shadow pool goes into the table and onto a list of the table's shadows (vk_memory.c shadow_table_add); the record of a binding of the table notes the guest submission (vk_exec_draw.c ssbo_info, vk_mem_shadow_table_read); and vk_mem_shadow_sync brings the CPU's writes into every shadow on the list once for that guest submission, before the part is submitted. Once is enough, because the copy-ins read the guest submission's write-watch snapshots, which a second sync would find unchanged. Measured by #551's author on Linux (RTX 4090, Kokiri Forest): 113 and 76 frames over 25 ms in 75 s against 217 and 429 with the port, 43 against 41 fps, the picture complete in both.
What this line had built on the port stays, on the new sync: the test that passes over a pool with nothing written (shadow_pool_clean), the upload of the same pages into a table pool's VRAM mirror, the write-back of stores through the table (NX2NAVI_SHADOW_TABLE_STORES, table_wb_cb, now asked once per guest submission that binds the table) and the counts of NX2NAVI_SHADOW_CHECK. The probes that record their own commands (the lost-device probe, --probe-vkmem's table parts) call vk_sync_submit_own, written anew. vk_mem_shadow_storage stays: one line that returns the mode. Finalized pools in the table were already kept for vk_mem_shutdown here.
--probe-vkmem has a check of its own for the sync (test_table_sync): the shadow of a pool that is never bound keeps its old word through a CPU write and a guest submission that does not bind the table, and holds the new one after a guest submission that does. It passes with NX2NAVI_SHADOW=1 on Windows (RTX 5090), as do the store check with NX2NAVI_SHADOW_TABLE_STORES=1 and the probe without shadow storage; --probe-devlost ends as before. Not run: the game, with or without shadow storage, and anything on Linux.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
A mesh draw becomes its record (the workgroup count, copied on the GPU for an indirect draw), the task dispatch, a
scan that lists the mesh workgroups the tasks started, the mesh dispatch, and an instanced indirect draw with a
generated vertex stage and the program's fragment stage. The compute modules bind the resources of the task and mesh
stages, and their pipelines are made on a thread of their own while the program's draws are skipped.
The indirect mesh ring and the texture pool marks are released at a probe's shutdown.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Verified by runs in the audit: the Market's periodic stall (a pool slot was listed once per refresh pass instead of once; the list fell from 160,385 entries to at most 3,192 and no frame over 20 ms remained while standing), orderly exits delete the pipeline prebuild's running marker (left behind, the next start wrote a skip file and every later start skipped the prebuild), the log and crash report name an area a save asks for that the game does not have, and the menu atlas hash no longer overflows a signed multiply. The other fixes compile and pass the regression route but their faults were never reproduced: see AUDIT.md section 13 for each.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
A slot that resolved kept its place in the list until the next walk, and when it stopped resolving before that walk it was listed again. The list reached 160,000 entries for the 32,768 slots after a minute in the adult Market, and each walk over it held the texture lock for 15 to 30 ms.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
The HDR tonemap gains a mapper built on a finite cone response in LMS: scene grey 0.18 lands where the game's curve puts it and rises through it with the slope of the game's curve, and the hue bend of the game's clamp is given back to bright colours. It takes the scene before the game's exposure gain and the gain itself, which the splice now finds in the game's shader. The luminance mapper stays selectable and the game's own curve stays the third choice. The mapper fits the Rec. 709 gamut here, as the game's scene target holds no negative component. New settings: the tone mapper's mid grey, tone slope match, tone slope and hue clip emulation. The gamma correction on the present applies with the luminance mapper only.
Not part of this: eye adaptation, the anti-aliasing and reconstruction shaders, bloom and local exposure sliders, the output colour space choice and a 16-bit float scene target. Checked by build and numerically against the source of the port; the picture on an HDR display has not been judged. THIRD_PARTY_NOTICES.md names the source.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Four "mem: sizes" lines give the entries and bytes of what no other line counted: the guest heap, physical maps, kernel region, lazily written ranges and the fixed kernel, setjmp, save-file, RomFS and zstd tables; pool records, mirrors, deferred destructions, upload and staging buffers, the device address table and the texture, sampler and slot caches; image views, descriptor spares, the completion queue, uniform snapshot arenas, recording copies and pins; and pipelines, layouts, stage libraries, job queues, the shader memory cache and the translator processes. Each subsystem reads its own counters under its lock; settings now count the replaced values they keep. Segment and scene lines say "-" for the world clock before it starts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An indirect mesh draw reads its workgroup count through a ring entry that a buffer copy fills, and a copy cannot be
recorded inside a pass: every such draw ended its pass, 45 times a frame in a village. The first draw of a run now
fills the entries of the mesh draws that follow it with only state, draws, counter reports and render conditions
between them, and those draws stay in the pass (mesh_ahead, default 1; 0 is the old way, ab switches in A/B blocks).
An entry is taken only by the draw it was filled for, at the same record address, while no other op was replayed
since. Counter reports count as staying in the pass because the replay closes their query inside it and writes the
value from the completion thread; the game reports one after most of these draws, and without that only 38% of the
draws were served.
Measured in the village: 42 of the 45 draws a frame are served by an entry filled ahead; gaps with these copies
41 to 3 a frame (0.22 to 0.04 ms of GPU time); passes and dispatches 248 to 208 a frame (302 to 254 in Hyrule
Field). With the piece release of the commit before, one run each: village 142.6 fps and GPU busy 3.87 ms (138.0
and 135.4 fps, 4.15 ms without this change), Hyrule Field 136.8 fps and 4.34 ms (131.5 and 131.3, 4.70 ms). In A/B
blocks of one run with the narrower rule (counter reports still ending a run) the GPU's busy time went from 4.81 to
4.64 ms.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
A draw whose pipeline is missing waits up to 3 ms for its fast link before it is skipped, 6 ms a frame in all. A mesh pipeline has no stage libraries and gets no fast link, so for the 0.2 to 1.1 s of its compile every frame that drew with it lost those milliseconds and skipped the draw all the same. Once such a pipeline has been building for longer than the 8 ms a cached one needs, its draws are skipped at once, as the one-shot mesh draws already are.
Not run in the game yet.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Each submission's first sync of the bindless texture pool walked every registered slot, 33,000 in Kokiri Forest, to find the few textures the CPU wrote and the few slots that changed. Three parts of that walk change, each with its own switch (pool_fast, pool_presnap, pool_group; on by default, 0 and ab accepted):
- The freshness check goes through a list of the entries the slots resolved to, with a check by 2 MiB region first; slots that resolved to nothing are looked up again when the game's memory map changes and every eighth submission. The use stamps the idle eviction reads are left from a count of the slots that hold each entry.
- The check takes the write-watch snapshots of its regions first, so unchanged entries pass the region test.
- The GPU-written slots of one address bucket with one descriptor form a group, which the dirty walk leaves alone while its first member is unchanged.
Measured on an RTX 5090 at 2560x1440, standing in Kokiri Forest, in-run A/B over five runs with all three switched (three of them with the barrier change switched too): 93.6 to 100.7, 96.4 to 103.1, 95.8 to 102.6, 99.3 to 107.1 and 91.9 to 99.9 frames a second, the replay thread 0.7 to 0.8 ms a frame shorter; a control's blocks differ by 0.4 %. On a walk with NX2NAVI_POOL_GROUP_CHECK=1 no group that was left alone had missed a write, and screenshots across the A and B blocks show the same picture.
From pull request #390. Not tried: the texture eviction under memory pressure (a 9 GiB budget was never reached on the walk), AMD and Intel drivers, Linux. One of the nine runs with it on ended by itself with exit code 0 and nothing in its log two seconds after the window lost the focus; three repeats of that run did not.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Every pass begin and every compute dispatch recorded a full barrier, which makes the GPU finish everything before each of them. With barrier_game = 1 (NX2NAVI_BARRIER_GAME) a pass after an earlier pass gets a barrier for its attachments only and a dispatch after other dispatches gets none; a change between drawing and compute keeps the full barrier, and so do the game's own barriers and the replay's copies, clears, fills and uniform updates. On an RTX 5090 at 2560x1440 the GPU's time for a frame fell from 5.4 to 4.9 ms standing in Kokiri Forest (two in-run A/B runs, the control's blocks differ by 0.7 %) and from 5.1 to 4.4 ms on a walk through it. The frame rate did not change there, because the replay thread is the limit on that machine; it should rise where the GPU is.
The default is 1 on NVIDIA graphics cards only, and 0 on every other vendor's: the game placed its barriers for an NVIDIA GPU, Vulkan asks for more of them than the game recorded, and no AMD or Intel driver has run this. 34 screenshots across the A and B blocks of one run show the same picture on NVIDIA. NX2NAVI_TRACE=perf logs the barriers a frame by kind.
From pull request #389, which had it on for every vendor; the default by vendor and the settings text are added here. The validation layer's synchronization check could not be run on this machine.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Found in the first run of the previous two commits (a first visit of Kokiri Forest from a state that had only seen Hyrule Field):
- The count of frames a target was rendered in started again whenever the guest's texture object changed, and the game initializes its render target objects anew every frame at the same storage: the log showed `seen 2129/2128 in 1 frames`, so no target ever became steady and the rule did nothing. The owner of an image is now the storage address the texture cache keys it by. An image handed on to a texture elsewhere still starts at none.
- A one-shot mesh draw waited its 8 ms in every frame that met the same pipeline while it compiled for 150 to 1100 ms: 17 waits, 170 ms, every draw skipped at the end, up to four in one frame. A mesh pipeline that has been building for longer than that wait is a compile, and the draws after the first do not wait for it (vk_pipeline_building_ms).
- The last wait of a full-screen draw with no picture to hold was 40 ms a draw, and two such draws made 86 ms in one frame. It is 40 ms a frame now.
- A one-shot vertex draw skipped at the end of its wait gets its log line however short the wait was (seven were skipped with no line, so their targets could not be named).
That run against the same cell of the earlier comparison (same exe path, cache state, route and settings; one run each): the load's worst wait of the draws in one frame 99 ms against 735 ms, no mesh wait over 13 ms against 524 and 206 ms. Two long frames remain in both runs (628 and 862 ms here, 660 and 854 ms there): two compute pipelines built at their first dispatch, about 395 and 397 ms.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
A draw whose pipeline is still building is skipped when its pass is rendered every frame and waits otherwise, so that a lookup table or a volume generated once is not left empty. "Every frame" meant "in the last four frames", so after a loading screen or a full-screen menu every pass of the first frame read as new: each draw with a new pipeline waited on the replay thread, up to 30 s each, and a mesh draw, whose pipeline has no fast link, for its whole compile. A player's 0.7.1 log shows it at three area changes: frames of 954, 1015 and 684 ms with 853, 960 and 663 ms of pipeline wait, six mesh draws waiting 52 to 660 ms in passes whose targets had been rendered in thousands of frames.
The target history (now runtime/vk_target_frames.h) also counts the frames each target was rendered in, for as long as it is the same guest texture. From eight frames on its pass is per-frame however long ago the last one was, and its draws are skipped and built by the workers like any other (pipe_skip_steady, on by default). A target rendered for the first time still waits, now from one pool of 50 ms a presented frame that the full-screen draws share (pipe_frame_wait_ms; oneshot_budget = 0 gives the old wait back), a mesh draw 8 ms at most, and the wait for a one-shot draw's shader comes from the same pool (gfx_shader_await takes a limit). A full-screen draw with no picture left to hold waits 40 ms more instead of 500. One frame so waits under 100 ms for pipelines in all.
Every 600 frames with a miss the log says how many draws found a shader or pipeline missing, how long they held the replay thread in all and in the worst frame, by the kind of wait, and how many were skipped at its end.
tests/target_frames_check.py replays the area change of that log against the header. Not yet run in the game.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Eight commits: a draw that misses its pipeline has the rest of its submission read ahead, the library workers link
the pipelines of skipped draws, a pipeline whose fragment shader reads a target of its own pass is fast-linked like
any other, and diagnostics. The merge was clean. Their before and after matrix had not finished when this was merged.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
A large compute shader is built at its first dispatch, on the replay thread (470 to 720 ms each on a first visit to
Kokiri Forest), so whether its program exists long before that dispatch says whether the build could start earlier.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Shaders are looked up and pipelines built when a draw first needs them. The time between nvnProgramSetShaders and a
program's first draw is the head start a build begun at the program's creation would have; the line gives it as
counts under 17 ms, 100 ms, 1 s, 5 s, 30 s and over, mesh programs apart (vk_exec_draw.c prog_lead,
nvn_api_obj.c nvn_prog_made_ns). Off unless asked for.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
It was a constant beside the per-frame budget (pipe_link_ms). With the stage libraries in the driver's cache a fast
link takes 4 to 17 ms for a third of the pipelines a draw misses, so both values decide how many such draws are left
out of a frame; the setting lets that be measured.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
The next draw with a new pipeline was only found when the replay reached it, so the draws of an area's first frame,
or of a turn that brings several new materials into view, paid for their pipelines one after another while the
workers that could have built them together sat idle. The submission is recorded before it is replayed: the first
draw of a submission that finds a shader or pipeline missing now has the ops after it read (vk_exec.c vk_exec_ahead)
with the binds that decide a draw's shaders and shader libraries applied to a copy of the replay state. Each draw
there looks its shaders up as the real draw will, which also queues a shader that is not in memory, and names its
state to vk_pipeline_hint, which queues the pre-rasterization and fragment shader libraries that are missing. No
pipeline is entered and no key recorded, so a wrong guess costs two libraries at most. The draw's own lookup then
links on its thread when both libraries are there.
On a first visit to Kokiri Forest a scan read 6,900 ops in 0.78 ms on average (the longest 5.5 ms; a scan ends
after 3 ms), and the scans of the load and the walk named 377 states without a pipeline and queued 104 libraries
for them.
pipe_ahead = 0 (NX2NAVI_PIPE_AHEAD) turns it off.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
- Every pipeline a draw missed is timed from the miss to its first usable pipeline, the fast link or, without one,
the build itself (vk_pipeline.c ready_note): the gpl lines of a save give the counts under 4, 8, 17, 50 and 150 ms
and over, the average, the longest, and how many had no fast link. That time is how long the pipeline's draws were
missing from the picture or waited.
- The stage library lines add the libraries of 25 ms or more and their sum, and the pre-rasterization libraries that
had outputs added for their fragment shaders: those are another module than the one the preparation built, so the
driver's cache has none of them.
- "a draw waited" lines name a pipeline with a feedback loop.
- The "draws skipped for a pipeline" counter (overlay, perf lines, flight records) was raised by every lookup that
found the pipeline building, also when the draw then waited for its link and was drawn. It is raised where the draw
is skipped now (vk_exec_draw.c draw_skipped).
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
The textures of a loaded area are what fills a 6 GiB card: with the budget taken as 4,730 MiB a walk through Castle Town held 6,907 MiB of device memory, 4,302 MiB of it textures, 201 pools were refused their mirrors and the game ran at 20 fps. With texture_detail = reduced a mipmapped 2D or 2D-array ASTC texture of 256x256 and more that the game's CPU fills is made without its largest level, with low without its two largest when it is 1024x1024 or more. The same walk then held 5,472 MiB (textures 2,357) and 4,987 MiB (textures 1,997), and the ASTC texels decoded fell from 2,589 M to 826 M and 503 M. auto, the default, picks full from 8,192 MiB of device-local budget, low under 6,144 MiB and reduced between. full changes nothing.
The image holds the game's levels from the one it starts at (VkTexImage.skip), and the upload with its ASTC regions and BC3 encoding, WriteTexels, the write-back, the views and the sampler LOD clamp of combined bindings count from there. Render targets, storage images, 3D and cube textures, small and unmipped textures and textures the device samples as ASTC itself are left whole, as is a texture made over a live one that holds GPU-written content. A shader that fetches texels by position or uses a texture's size gets the texture whole: gfx_shader.c marks such bindings from the SPIR-V (the layer count of an array, which the translator's layer clamp asks for, does not count), and the replay makes a reduced texture bound there again with every level.
Not run on a card with 6 GiB: the figures come from an RTX 5090 with NX2NAVI_VRAM_BUDGET_MB=4730.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Eighteen commits, one per Linux pull request that was not in yet: the build fix for a source checkout, NVIDIA's shader
disk cache only with NVIDIA's driver, tools/run_game.py for a Linux game folder, a memory limit the low-memory handling
can read, the recording thread's core on Linux, shadow storage on Linux and Proton, the Steam Deck profile and prompts,
FSR 3.1 offered only where its library loads, DLSS frame generation through NGX on Linux, the AppImage with its window
and its workflow, and the processor level from /proc/cpuinfo. Two conflicts: the settings table keeps the area-load
settings under the new comment, and the example file keeps the lazy write-back as off by default and adds that it
stays off with shadow storage.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
The texture pool creates 16 entries a refresh and stops after 6 ms a frame, so that textures streaming in do not make
long frames. An area load leaves thousands of slots at once (4,936 loading the test save), and at that rate they came
in over the frames after the load screen: the scene's first frames showed a quarter of the picture as black stand-ins,
for 22 presented frames on a free machine and for about three seconds with the CPU busy.
While a refresh leaves pool_load_backlog slots or more (512; 0 never), the pool is loading: the count limit is off, a
frame's refreshes may take pool_load_ms (50), and frame_hold leaves out the presents of frames drawn meanwhile (up to
2 s), so the picture before the load stays until fewer than 64 slots are left. Measured with the flight recorder's
thumbnails of every present: the 4,847 textures are made in one frame of 345 ms, 30 frames before the scene fades in,
and the black share of the picture goes from the fade straight to its final value. The streaming limits are as before
below the backlog.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
With shadow storage (Linux, and Windows builds under Wine) a CPU-visible pool is a host-visible copy with
no memory object of its own, and only pools that owned one went into the device address table: pointers
the game's shaders read from memory reached nothing. Shadow copies are now in the table. A shader that
reads through the table resolves no pool, so nothing queued the pool it reaches for the copy-in: the part
is marked where its pipeline is drawn or dispatched (VkPipeInfo.gpa, vk_mem_gpa_use), and its prologue
brings the shadow of every pool in the table up to date with the CPU's writes and queues the same pages
for those pools' VRAM mirrors (vk_memory.c vk_mem_gpa_sync). --probe-vkmem reads a word through the table
from a pool that is never bound, before and after a plain CPU write.
Pull request #353 by TophC7, ported onto the base copies of the black-world fix:
- taken as it is: the table registration, the mark in vk_exec_draw.c and vk_pipeline.c, the prologue
call and vk_sync_probe_submit in vk_sync.c (two hunks merged by hand), the probe's test;
- written again: vk_mem_gpa_sync, on top of the snapshot-based shadow_sync_pool; a lookup no longer
queues a pool for being in the table, which every shadow pool now is;
- narrowed: the sync runs only with shadow storage. Without it (Windows) a mirrored pool stays unsynced
for the table's readers as before, so the frame there is unchanged, and the probe makes the source's
mirror only with shadow storage and skips when the device has no address table;
- the probe's shader was an array of SPIR-V words in tex_probe.c; it is now the source
runtime/shaders/gpa_read.spvasm, built like the other runtime shaders;
- the exact re-upload count in the texture probe stays checked on Windows.
Files other sessions are changing: runtime/vk_memory.c (vk_pool_init's registration, mem_resolve's touch,
the new section before vk_mem_use).
The Windows files compile and the shader assembles and validates; no probe was run, and nothing was run on
Linux here.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Shadow storage noted a GPU write with the whole range its binding names: the copy-in skipped that range and
the fence copied all of it back, so CPU writes to other fields of the same records never reached the GPU.
Ocarina's lighting parameter is such a field: the lighting came out NaN and the fog pass drew the scene
black. Each page a noted range covers now keeps a base copy, the copy-in moves only the bytes the CPU
changed and the write-back only those the GPU changed (vk_memory.c merge_page). The mirror's merge shader
marks the 4 KiB pages it changed, and one job per submission copies only those back (shadow_wb_cb); the
copy-in reads the CPU-side generation tables through the shared write-watch snapshots; and a caller that
reads guest memory itself resolves with vk_mem_lookup, which does not queue the pool for the copy-in.
Pull request #394 by jacc. Its changes are taken as they are; one hunk met the lazy write-back that came
later (vk_gpu_written_back) and is merged by hand. Added here: the lazy write-back stays off with shadow
storage, because its fetch copies the mirror into the pool's buffer, which is guest memory only for an
imported pool. Files other sessions are changing: runtime/vk_memory.c (the shadow section, the merge
record, vk_gpu_written_back, wb_lazy) and three one-word call changes in runtime/vk_usnap.c.
Without shadow storage (Windows outside Wine) the merge pipeline gains an unused fifth binding and two
push constants and nothing else changes. The Windows files compile; nothing was run, on Windows, Wine or
Linux.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
A draw whose shader or pipeline is still building is left out of its frame, and the frame was shown without it: a black or unlit picture with the interface intact when it was a lighting pass, a light going out, or ground missing for a frame. In a player's session a full-screen draw was skipped after its 50 ms and the frame came out black as a boulder hit.
The frame's present is now left out when draws were skipped (vk_backend.c frame_hold, NX2NAVI_FRAME_HOLD_MS, 120 by default, 0 off). The game runs on, so nothing waits. The hold comes from an allowance that fills again at an eighth of the clock's rate, so a long build or a run of them shows its frames as before and does not freeze the picture; a full-screen draw that would be skipped with the allowance spent waits on instead, 500 ms at most. Not with DLSS Frame Generation.
On a first walk through a village with no pipeline recorded, 21 of the 23 frames with skipped draws were held back; in the Lost Woods, 7 of 8.
The performance overlay: the VRAM line is the card's dedicated memory in use, which counts every program on it: a player read 14.5 GB as the game's while the game held 7.3 GB. The line's third cell now gives what this process holds on the device-local heaps (the driver's budget figure), in place of the GPU clock, which stays where the driver gives no budget.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
The fifteen headers of SPIR-V words under runtime/ are no longer tracked. tools/build_shaders.py builds all of them:
the GLSL with glslc, the SPIR-V assembly with spirv-as, and the two shaders assembled in Python from their scripts.
build_game.py builds the missing and the stale ones before it compiles the runtime, and the build kit, the launcher
release and the AppImage build them before they copy runtime/, so a player's PC still needs no shader compiler.
git ignores the headers and check_content.py refuses a compiled shader in a commit. The headers built this way are
byte for byte the ones removed here. vk_mvobj_fs.h is now mvobj_fs_spv.h, named like the others.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
The preparation builds every stage library once so the driver's disk cache holds it and a library a draw needs later takes under a millisecond. Turning dynamic vertex input off for AMD's RDNA 3 and newer driver changed the dynamic states every library lists, which gave each one another entry in the driver's cache, and nothing built them again: an update carried the prepared shaders over without the driver step. On an RX 7900 XT (26.10.1) all 42,142 prepared libraries were cold, 93 ms a vertex and 242 ms a fragment library on average. With no pipeline recorded, a walk through a village skipped 11,180 draws while their pipelines built and made 94 draws wait 5.5 s in all, four of them 1.0 to 2.0 s; a player's session log shows up to 114,834 draws skipped in 600 frames and 181 waits of 251 ms on average. Skipped ground and walls appear and disappear, and a skipped lighting pass is a black frame with the interface intact.
- The game checks the driver's cache at every start (vk_pipeline.c lw_start): a sample of the prepared fragment libraries is built, and when their median time says the cache is cold the whole prepared list is built into it again on idle-priority threads while the game runs. shader-warm.txt records a warm in progress, so the next start goes on with one a run ended, and stops the tries on a driver that does not keep the libraries. This also covers a driver update and a cache the driver emptied.
- An update that carries the prepared shaders over runs the game's --warm-shaders (tools/make_game.py warm_driver_cache): 11 s with the libraries there, about 8 minutes with all of them cold on that card.
- A draw whose pipeline is missing waits for its fast link from a budget of 6 ms a presented frame, 3 ms at most each (NX2NAVI_PIPE_LINK_MS), before it is skipped: with warm libraries the link is done within that.
- Full-screen draws wait from their 50 ms budget again instead of being skipped (NX2NAVI_PIPE_SKIP = any is the default; fullscreen stays selectable). With warm libraries the wait is a millisecond.
The same walk afterwards: libraries 0.3 and 0.7 ms on average, 1,605 draws skipped, 1,473 of them in the scene's first 600 frames and 8 to 80 in each 600 after, and 19 waits of 60 ms in all. First visits to four other saves (a mountain pass, the Lost Woods, Kokiri Forest, a river) waited 12 to 84 ms in all except the Lost Woods, where mesh shader pipelines, which have no stage libraries, and vertex libraries that need added outputs still compiled once: 23 waits, 2.1 s in all, the longest 776 ms.
The texture cache counted its waits in presents chosen at 60 a second: at 165 presents a second the 1800 that stood for thirty seconds were 11 s, so a released render target went sooner than meant and a texture out of view for 11 s was retired and showed its stand-in for some frames when it came back. The counts are now taken times the present rate over 60 (presents_for), and a texture is retired after two minutes without a sample instead of thirty seconds.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>