When a plane's values at both ends of the chunk fit in 32 bits, so do all between, and 32-bit
wrapping arithmetic gets them exactly: then in vector lanes, with the clamp.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The coverage stage was a scalar loop per pixel with the snap edge as a branch. A chunk starts a span,
so relative to it the bounds fit in 32 bits and the snap edge's pixel is a fixed offset per lane.
Also read the depth buffer for the early depth test with one load where SSE isn't used.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The span pixel function and the four-pixel bilinear sampler work in vector lanes, a channel per
vector, but took and returned the colors a pixel at a time, so every span was transposed twice on the
way. Now they take a channel at a time with a stride. The staged rows keep each chunk's colors that
way from the start (their stage walks a channel at a time anyway), and DrawRectangle builds them so.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
DrawSpanVector now covers everything but logic ops and the signed blend factors. Stencil tests pick
the SFail, ZFail or ZPass op per lane, and lanes that only get a new stencil keep their old color.
Pixels are read and written in the framebuffer's own format, converting the whole span at once, and
a span with all four pixels drawn is stored in one go.
Only drawn pixels are written: spans are aligned in screen coordinates, so with a draw offset one can
reach into a tile another thread is drawing.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The last stage of a staged row (texturing, the secondary color and the
pixels) is now a template on what's fixed per triangle: textured or not,
the UV planes fitting 32 bits, the mip level's q and through mode.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
When a triangle's q, s and t plane sums fit 32 bits wherever its spans
are walked, and the planes' exponents keep the GE's UV arithmetic in the
normal float range (GEUVSpanSafe), a span's plane values are one vector
add each and GEUVSpanCore runs without its range checks. Tested bit for
bit against the scalar code under those conditions.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Where the pixel function is the C++ one (and textures have planes), a
row is now drawn in stages over up to 64 pixels at a time: coverage,
depth and the early depth test, then the colors, secondary colors and fog
of all its pixels, then texturing and the pixels span by span. Each stage
keeps little state, where the single span loop spilled most of its.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The four-pixel sampler now runs the texture function (modulate, decal,
blend, replace and add, with color doubling) on all four pixels at once,
one vector per channel, instead of GetTextureFunctionOutput per pixel.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Where the pixel function is the C++ one, triangles and sprites now hand
it a whole span of four pixels. Unless the state has a stencil or color
test, a logic op or a signed blend factor, the span goes through one pixel
per vector lane: clamping, alpha test, fog, depth range, test and write,
blending, dither and the color write mask, as the per-pixel code does
them. Checked against the per-pixel code span by span in 8 games and 263
frame dumps.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Depth, colors, fog and the UV planes are now started once per row and
stepped a span at a time, instead of evaluated from scratch at every
span.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Each row's covered pixels now come from the three edges directly, as a
range from 64-bit integer bounds at the pixel centers, and only the spans
in it are visited, with lane masks from the range. The edge functions
per span (32-bit, with a coarse row narrowing) are gone. The mip level's
q comes from the row's covered range instead of edge tests per pixel. A
tall triangle's long edge, which the GE checks at the other end of a span
for its first or last pixel, is still checked per pixel.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
GEUVSpan does the GE's per-pixel UV arithmetic for four pixels at once:
the plane values to float24s by exponent arithmetic, GERecip with its
table entries gathered, and the truncated 24-bit products from the 16-bit
significands' product. Lanes it doesn't cover (values past 2^24,
denormals, results out of range) send the span to the scalar code.
Tested bit for bit against it.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
When the four pixels share their mip level (always, for a triangle's
span), their texel coordinates, clamping or wrapping and texel offsets
are computed in vector lanes, and only the texel and palette reads go
one by one. Adds Vec4S32::Min.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The GE walks rows in spans of four pixels and picks the mip level's q
once per span, so the rasterizer now does the same instead of 2x2
quads: triangle spans are aligned like the GE's, so the four pixels
share their level, which is computed once. Auto LOD without planes takes
its vertical derivative from the row below. Sprites, which always have
planes, lose their unused incremental UVs.
Without a sampler JIT, the four bilinear samples of a span are now taken
in one call. Same results on 263 frame dumps and 8 games.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A texel's byte offset is a sum of one for its row and one for its column,
swizzled or not, so the four texels of a bilinear sample now take two of
each, and the CLUT offset of a mip level is found once per sample. The
UV planes at a quad's four pixels are one sum plus steps, as for depth.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The rasterizer takes a GERecip per pixel, so it's inline now, with its
table exported. The q for the mip level is the same for the two pixels
of a quad row unless they're in different 4-pixel spans, so it's only
looked up again then.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Where there's no sampler JIT (everything but x86-64), bilinear sampling
switched on the texture format per pixel, on the CLUT format per texel,
and went through swizzle checks and vector-to-stack round trips. Now a
template instance per texture format, swizzling and CLUT format is
picked with the state, and the filter does all four channels at once in
16-bit lanes of a uint64_t. Same results (checked on 263 frame dumps).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The per-pixel texture coordinates called ldexp with a shift that's fixed
per triangle or line. The shift keeps the values in double's normal
range, so multiplying by a precomputed 2^-shift is exact. Lines also
convert their end values to fixed point once instead of per pixel.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The GE doesn't clip rectangles: one with a vertex behind the camera or outside
the screen range is dropped whole (the software renderer's ProcessRect, from
gpu/clipping/guardband). The hardware backends drew them, stretched to the far
off-screen vertex.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
N.L changes sign from vertex to vertex, so the branch on it was mispredicted a lot. A factor that isn't
positive scales the color to 0, so the term can always be added. Vertex reading about 6% faster in
Tekken 6.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Every GEAdd4 and GEMulFloat24x4 checked its lanes for the cases it can't do and branched to a
fallback, a vector-to-scalar move and a branch per operation, about 14% of skinning. The unchecked
variants only collect those lanes; a vertex where any appears is skinned again with the checked ones.
The products use GEMulFloat24x4 (the same as a one-term GERowSum4 here), and the position's and
normal's add chains are interleaved. Skinning is about 20% faster.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The position is read once as (x, y, z, 1) for the clip, fog, texgen and
light vector rows, and lighting works on Vec4F32 throughout, without
going through arrays. Each light's constant term, its position minus
the world translation, is computed once per state instead of per
vertex. The texture matrix rows are now prepared with the transform
state, so loading that matrix dirties it.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
GEDot3 forms one term per lane and sums across the lanes, about 1.5-2x
as fast as the scalar GEDot. Adds Vec4S32::GetLane.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Skinning did each component's weighted bone entries and its chain of
GEAdds separately in scalar code. The components now go in lanes, with
GEMulFloat24x4 for the weighted entries, one-term row sums for the
products and GEAdd4 for the sums, in the same order. Checked against the
scalar version on every vertex of 143 skinned frame dumps.
GEMulFloat24x4 matches ProductToFloat24, so inf and NaN bone entries
behave as before. GEAdd4 handles two zero operands without its fallback.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Adds the exponent instead of calling ldexpf, except for reciprocals that
can be denormal. Identical results for every normal float24 input.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
They're truncated to float24 with one AND, and their exponents, fractions
and signs worked out for all terms at once, each term then broadcasting its
lane (new CrossSIMD Vec4S32::SplatLane). The term count becomes a template
parameter so the lanes are constants. About 75-83 M transforms/s on an M1,
from 68; results unchanged.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The four clip space rows of a vertex, the texture matrix's three, the world
normal's three, the vector to a light, and each row of the combined matrix
now take one GERowSum4 call. Same results (pspautotests, frametests and
the probe regression set unchanged).
GERowSum4 scales its result by adding to the exponent instead of
multiplying, and CrossSIMD gets a Vec4S32::Max for the largest lsb. About
68 M transforms/s against 26 for four scalar GERowSums on an M1 (a pure
NEON version with per-lane shifts did about 79); replaying the frametest
dumps gets only about 1% faster, up to 11% on Taiko no Tatsujin.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Four rows sharing their left operands, like a vertex transform's: with
float24 operands each exact product is a 32-bit integer, (1.a' * 1.b') >> 15
= 32768 + a' + b' + (a' b' >> 15), and the terms are aligned by scaling with
a power of two and truncating. Lanes whose result exponent is out of a
single float scale's reach take the scalar GERowSum. Tested bit for bit
against GERowSum on random operands (arm64 and x86-64). Not used yet.
Also adds an arithmetic Vec4S32::Shr and the float/int bit casts to
CrossSIMD.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The GE reads the source in aligned 16-byte blocks, four at a time, into a
buffer of four destination-aligned 16-byte slots, and a read of a block a
slot holds returns the slot's bytes, stale ones included. So a transfer
shifting a row right by less than 64 bytes reads some of what it has just
written, in a pattern that isn't a forward or backward copy. Fitted to
ppsspp-re geprobe exp219 and exp220 on hardware (562 of 570 cases exact).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The fast matrix loads and the joined PRIMs read their commands themselves, then skipped them with
UpdatePC(pc, pc + n * 4), which counts the commands up to pc and resumes counting after the skipped
ones: those were never counted, as for a jump. Now pc moves first and UpdatePC counts up to it. The
software renderer runs those commands one at a time, so the backends now agree on the timing.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
gpu/vertices/carry, recorded on a PSP: a draw whose format has no UVs uses the last ones read, by any
earlier draw in draw order, textured or not. They're kept as read and get the scale and offset of the
draw that uses them. Normals carry over the same way.
We kept the UVs after scaling, only read them with texturing on, and after a cached indexed draw
carried the highest vertex instead of the last one drawn. The carried values move from statics in
ReadVertex to TransformUnit.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Headless replays frame dumps to compare them with references and with the PSP, which a game's
workarounds skew: DarkStalkers' present hack (16510) replaces the game's final upscale, so the
replay didn't match the PSP's. The app still applies them.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Now that density is in configChanges, a density change no longer recreates
the activity - but it also doesn't resize the surface, so no surfaceChanged
follows it. The DPI is only recalculated in backbufferResize, so the UI kept
the old scale until the next real resize or a trip through the background.
Re-report the surface we already have when the density has moved. The native
side needs no change: Recalculate() already returns true when the dp size
changes, which a density change does on its own.
Also make the log line say which size changed, since the pixel size now stays
put in this case.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01152igCdjTNhT29qvphPyPC
It was passed once at init, and decides whether we can restrict the
orientation on Android 17+ large screens. Now that a fold or unfold no
longer recreates the activity, forward the new value from
onConfigurationChanged. Untested on a large screen device.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Now that rotation doesn't recreate the activity, the class is the
SurfaceHolder.Callback plus the hardware scaling logic, and the rest had
outlived its purpose:
- The deferred-surface-while-paused workaround. Also what started the
render thread twice on resume.
- The "bad orientation" workaround, which called recreate().
- The deprecated system UI visibility listener, which only chose between
getMetrics and getRealMetrics, and the view size overrode that whenever
it mattered anyway.
- The coalescing of display measurement updates, and updateDpi, which
updateDisplayMeasurements does itself.
Before the first layout the display size now comes from getRealMetrics.
It only feeds the first-run config defaults.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Every rotation, fold or window resize destroyed the activity and with it
all of graphics, then built it all up again. Declare those config changes
handled, so they arrive as a surfaceChanged and go through the same resize
path as a desktop window resize.
That path needed three things it didn't have on Android:
- The GPU_DISPLAY_RESIZED message, left out for Android since 2020 when
DeviceRestore covered every size change.
- A fixed (hardware scaled) surface size is now recomputed on each in-place
surface change, since surfaceCreated no longer runs on a rotation.
- A 180 degree flip keeps the size but changes the surface transform, so
the display rotation is passed along and a change in it is a resize too.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
It was a debug aid that asserted the activity callbacks arrive in the order
Android promises. That has served its purpose - nothing has tripped it in a long
time, and the lifecycle work this sits on top of has been exercised hard.
What it also did, though, was log every callback, and that is worth keeping: the
begin/end pairs are what makes a bad startup or a surprise recreate legible in a
logcat. So the tracker goes and the Log.i calls stay, at the same points, under
this class's own tag instead of LifeCycle's.
onStart and onStop now only log, so they lose the no-native-binary guard. There
is nothing left in them to guard, and if the library is missing is exactly when
you want to see the callbacks.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01152igCdjTNhT29qvphPyPC
Moving a mouse logged two lines from Java and one at INFO level from C++
for every motion event.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
renderText and getNativeCrashHistory are only called from native code,
like their neighbours that do have the annotation. Harmless today since
minification is off, but it would strip or rename them the day it's on.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
pressedKeys is there so Disconnect() can release what's held, but keys
were only ever added to it, so it listed every key pressed so far.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
It's written by the thread that brings graphics up and down, and read by
every input function on the UI thread. The comment also claimed it was
only used with OpenGL, but the Vulkan path sets it too.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
TextRenderer.init runs in every onCreate and replaced the static Paint. On
the OpenGL path the emu thread survives the activity and can be in the
middle of measuring or drawing text with it.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Each TextDrawerAndroid allocates its fonts afresh, and one is created
every time graphics come up. allocFont gave every request a new ID and map
entry, and nothing ever calls freeAllFonts, so the map grew by a few
entries per resume.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
When the library fails to load, onCreate shows a dialog and returns before
setting anything up. But onResume, onWindowFocusChanged and the rest still
ran, dereferencing the missing SizeManager or calling into the missing
library, so the app crashed before anyone could read the dialog. Make them
all check.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
