Compares the software tessellator's positions, normals, UVs and colors
against a longhand double-precision reference (Bernstein polynomials,
Cox-de Boor with clamped knots, finite-difference normals), across
Bezier and spline surfaces, edge types, poles and patch facing, so it
can be optimized with something independent to be wrong against. Also
reports vertices per second.
Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
With a deployment target of 10.13 (10.14 for arm64 libretro), every file
that includes a libc++ header warned "The selected platform is no longer
supported by libc++." Apple Silicon Macs start at 11 anyway.
Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
They use the same quadratic interpolator as rcp and friends, with three
twists. sin indexes the quarter wave from the top, and asin and sin work
in a per-segment exponent whose 4-ulp truncation also applies to results
in a lower binade. log2 truncates exponent + log2(1.m) toward zero to 22
significant bits, and where that step is coarser than 2^-24 the datapath
drops coefficient bits to match; that also covers the region just below
1.0 that needed a special case.
vfpu_sincos now reduces the angle once. With every table gone, so are
the asset folder, the loader, InitVFPU and the fallbacks for tables that
failed to load. All seven functions are bit-exact with the table-based
code over every 32-bit input.
Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
The riscv64/loongarch64 cross builds have no SDL, and Headless.cpp had a
matching "no window" path gated on those two architectures. Gate it on
the option instead (a HEADLESS_NO_SDL define), so any headless-only build
without a usable SDL gets it - an x86_64 build on an arm64 Mac, say, where
Homebrew's SDL3 is arm64 only. That's how the x86 JIT fixes here were
tested, under Rosetta.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
CrossSIMD has four independent implementations and only one is compiled
per machine, so each architecture has to check its own copy against
hand-worked results. Running this under qemu gives us coverage of the
targets we have no hardware for - it's already caught two LSX bugs.
Covers Vec4S32 and Vec4F32 arithmetic, comparisons and the mask helpers,
the lane accessors and shuffles, transpose, the various loads (including
the 24-bit and normalizing ones), NaN/Inf handling, and the matrix
routines the old test already had.
Verified on three of the four implementations: NEON natively, the scalar
fallback via TEST_FALLBACK, and LSX under qemu-loongarch64. SSE2 is left
to CI.
Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
The pieces were nearly all in place already - CMakeLists has detected
riscv64 and set RISCV64 since the backend landed - so this is mostly a
matter of not hardcoding loongarch64 in the cross-build plumbing:
- setup-loongarch64-cross.sh becomes setup-cross.sh <target>, taking the
triple and dynamic linker name from the argument.
- b.sh gains --riscv64, and the GL stub block is keyed off a compiler
variable rather than a loongarch64-only flag.
- New cmake/Toolchains/riscv64-linux-gnu.cmake, mirroring the existing one.
The SDL carve-out moves from LOONGARCH64_DEVICE to a new HEADLESS_CROSS
option set by b.sh. That flag was keyed off the target architecture, which
is also true when building natively on such a machine - harmless so far,
but riscv64 hardware that people actually build PPSSPP on exists, and it
should still get the normal SDL frontend.
Both toolchain files now find qemu rather than assuming the static build:
Debian ships the static binaries in qemu-user-static and the dynamic ones
in qemu-user, and only the latter is available on some releases. Neither
being present is fine too - it just means no compile-check programs run.
Verified by a clean loongarch64 build; riscv64 is untested so far, the
toolchain isn't installed here yet.
Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
sceMpegBaseCscAvc is the top of a profile during video playback, so the loop
that does the work becomes MpegCscRange - a pure function with the HLE plumbing
left behind - and TestMpegCsc measures and checks it.
The measuring half reports megapixels per second for a 480x272 frame in each of
the four pixel formats. The checking half compares against the conversion
written out longhand, over whole frames and over partial ranges with odd offsets
and sizes, plus one-pixel, one-row and one-column ranges and one that reaches
the far edge of the frame. Those are the cases an optimized version gets wrong:
chroma is half resolution, so an odd left edge starts mid-sample, and anything
handling two pixels at a time has to deal with the leftover. The destination is
padded and prefilled, so writing outside the range fails too.
This is only the move - the loop is the same one, so the numbers it gives are
the baseline to improve on. On a Snapdragon X Elite it runs at about 300 MPix/s,
0.44ms for a frame.
Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
Loads a single module standalone - no game, no boot - and writes a report:
module header and segments, the export and import tables with NIDs resolved to
names through the HLE tables, one annotated disassembly file per function, and
a call graph as JSON.
Reuses the emulator's own loader rather than parsing PRXes a second time, so
decryption, decompression, relocation and import resolution can't drift from
what actually runs. Only enough of the system is brought up to load a module:
memory map, timing, HLE tables and the kernel allocators. Nothing executes.
Two things beyond a plain disassembly, both aimed at the questions that come up
when reading unfamiliar MIPS:
- lui/addiu (and lui/load) pairs are folded and reported as the address they
form, which is how every global and constant table gets reached.
- Per function, a register evidence block instead of a guessed signature. A
MIPS function that takes two arguments and passes the second one down often
never reads it, so 'never read but live across a call' is reported as
forwarded rather than quietly dropped from the signature.
SetForceRealModuleLoads() is needed because modules like sceAudiocodec_Driver
have no DisableHLEFlags bit and so can't be turned off the normal way - they'd
fake-load and there would be nothing to look at.
Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
naett has had a complete libcurl backend all along; we just never built it,
so Linux ran with HTTPS_NOT_AVAILABLE. That means no homebrew store over
HTTPS, and RetroAchievements talking to plain http://retroachievements.org.
libcurl is loaded with dlopen rather than linked, the same way we handle the
Vulkan loader, so it stays a soft dependency: we need the curl headers at
build time, but a build made here still starts on a machine without libcurl
installed - it just reports HTTPS as unavailable, exactly like today. Distro
packagers get the behavior they'd expect either way, and certificate
validation comes free from the system CA store.
New net::HTTPSAvailable() answers "did that work", and SDLMain folds it into
SYSPROP_SUPPORTS_HTTPS, which everything downstream already degrades on.
Four fixes to the backend itself, all noted in ext/naett/README-ppsspp.md:
- panic() called exit(1) on a pipe or curl_multi_perform failure. Taking the
emulator down because a download failed isn't acceptable - the backend now
disables itself and requests complete with naettGenericError.
- CURLINFO_RESPONSE_CODE writes a long into res->code, which is an int. Eight
bytes into four, getting away with it only because the next field absorbs
the zeroes.
- curl_easy_setopt is varargs and wants a long for these options; int literals
and int variables are UB on LP64.
- naettPlatformCloseResponse called through a null function pointer when
libcurl was missing. Found by testing that path, which segfaulted.
CI needs libcurl4-openssl-dev (curl-dev on Alpine) or it would quietly keep
building without HTTPS.
C++ homebrew has an unreadable symbol table -
everything is _ZN10PxRenderer7DrawImmE... - which makes the disassembly and
symbol list nearly useless. Add an Itanium C++ ABI demangler and run ELF
symbols through it on load, in both ElfReader::LoadSymbols (unstripped EXECs,
which is what a CMake pspdev EBOOT actually contains) and the companion-ELF
path.
The demangling standard is called Itanium for historical reasons - it
was defined for Itanium but ended up being almost universally
applicable.
Written from scratch rather than using __cxa_demangle, which doesn't exist on
MSVC/UWP, or vendoring LLVM's demangler, whose license doesn't fit. Anything
unrecognized (arbitrary constant expressions, decltype) aborts the parse and
the caller gets the original mangled name back, so a caller never sees a
half-parsed result. Recursion is depth-capped since the input comes from a
file we didn't write.
Checked against c++filt as an oracle: of 1089 mangled symbols in a real C++
homebrew EBOOT, one differs; of 55189 from libstdc++/libLLVM/cc1plus, 22
differ and 413 are declined. Fuzzed with 220k mutated and random inputs under
ASan/UBSan.
Also adds a right-click menu to the ImDebugger symbol list.
Note that SymbolMap stores names in char[128], so the longest STL names get
truncated in the UI. Still far more readable than the mangled form.
Co-Authored-By: Claude Opus 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01X3DbkJ8ShYiXU7q5Tv1LZu
Our Qt backend has long been left behind and doesn't even support Vulkan
currently. There would be a lot of work to make it viable, and I don't
think anyone is really interested.
ImGui on SDL will soon fulfill the need for a more classic user interface
with a menu bar on Linux, and on Mac we already have a native UI.
- AudioEngine.mm/.h: not in the source list at all, only in the ARC
set_source_files_properties list, and written for manual retain/release
([super dealloc]), so it couldn't have compiled under ARC anyway. The actual
audio path is iOSCoreAudio.mm.
- DisplayManager.mm: empty file, referenced nowhere.
- iOSCoreAudioSetDisplayConnected: never called, so g_displayConnected was
always false and the external-display branch in UpdateSession unreachable.
Also dropped an unused local there.
- AppDelegate: the "screen" property was never assigned, so both view
controllers were setting their frame from [nil bounds], i.e. CGRectZero. The
window sizes the root view anyway. Also removed the stale launchPPSSPP:argv:
declaration (it lives on SceneDelegate now), a forward declaration of a
PPSSPPViewController protocol that doesn't exist, and an unreachable return.
- getView: declared on the base class and implemented identically in both
subclasses, called by nobody.
- bindDefaultFBO: moved the empty implementation to the base class so it isn't
missing there, and dropped the Metal no-op override.
- The unused delta/lastTimestamp computation in displayLinkFired.
Also fixed the viewWillAppear/viewWillDisappear log messages in the GL view
controller, which claimed to resume and pause the display link but don't.
The getView and bindDefaultFBO changes clear three -Wincomplete-implementation
warnings, and the AppDelegate.h one clears a fourth.
Co-Authored-By: Claude Opus 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01JZk5y4Fzw811WJoNWZb8Sc
Add unittest/TestTextureReplacer which creates a fictive texture pack
(textures.ini with readable invented hashes plus real PNG files), loads
it via the replacer, and verifies lookups, filtering, hashranges, mip
levels, and missing/ignored entries.
To make the replacer runnable outside the emulator:
- The constructor now accepts a null DrawContext (formats just default
to unsupported).
- FindReplacement/FindFiltering use the replaceEnabled_ member instead
of the global config.
- Added TextureReplacer::LoadPackForTesting() to load an ini from a
path directly.
The LZRC decompressor's only bounds check for output (and input) was a
debug-only _dbg_assert_msg_, which is a no-op in release builds. The
NPDRM demo block device also passed a hardcoded 1 MiB output length
while the real destination buffer (blockBuf_) could be as small as 2048
bytes, allowing a crafted NPDRM image to trigger an unbounded heap
overflow during game load.
Changes:
- rc_putbyte/rc_getbyte now enforce real bounds and set an error flag
instead of relying on debug asserts; decompression aborts with -1 on
overflow or truncated input.
- normalize() reads via rc_getbyte so it stays in bounds.
- Plain-text path clamps the copy size to both the output buffer and the
remaining input (and no longer interprets the size as signed).
- NPDRMDemoBlockDevice::ReadBlock passes blockSize_ (the real buffer
size) instead of 0x00100000 to lzrc_decompress.
- Add unittest/TestLzrc (synthetic input, no test data files): checks the
plain-text clamp, truncated input, and output overflow all fail safely.
- AGENTS.md: note to reuse existing format handlers/decompressors before
writing new ones.
A crafted zip with a parent-directory ("..") entry name could escape the
destination directory during extraction, writing arbitrary files on the
host (e.g. into startup/autostart folders). ExtractZipContents built the
output path by concatenating the raw zip entry name onto the destination
with no traversal check.
Changes:
- Add HasParentDirComponent() utility in Core/Util/PathUtil and use it in
GameManager::ExtractZipContents to reject entries with a ".." component.
Guard both the directory-creation and file-writing passes.
- Expose ExtractZipContents as public for testing.
- Add unittest/TestZipSlip which crafts a zip with a "../evil.txt" entry
and verifies it is not written outside the destination directory.
gason, vma, cityhash, 7zip (ext/lzma-sdk), basis_universal, pugixml,
kirk (ext/libkirk), sfmt19937, xbrz, and xxhash were each defined with
a small add_library() block directly in the root CMakeLists.txt, even
though ext/CMakeLists.txt already exists and handles every other vendor
library (freetype, imgui, naett, discord-rpc, libchdr, zstd, miniupnp,
armips, glew, snappy, ...) via add_subdirectory(). Gave each one its
own ext/<name>/CMakeLists.txt to match that established pattern; xxhash
stays a loose file pair in ext/ (it never had its own directory) so its
add_library() lives directly in ext/CMakeLists.txt instead.
ext/libkirk/CMakeLists.txt already existed but was dead - nothing
add_subdirectory()'d it, and its source list was stale (missing
amctrl.c/.h, which the live inline definition in root had). Replaced
its contents with the current, correct list instead of leaving two
diverging definitions around.
Dropped several target_include_directories()/include_directories()
calls that came along with these (e.g. cityhash, kirk, xbrz, xxhash,
and the lzma-sdk one for 7zip): traced their actual consumers and found
each library's own sources resolve their sibling headers via the
default same-directory quote-include rule, and every external consumer
already uses the full "ext/<name>/..." path resolved through the global
root include - so these were dead weight regardless of position.
Replaced two single-value alias variables (LIB7ZIP_LIBRARY, always
"7zip"; BASISU_LIBRARIES, always "basis_universal") with their target
names directly in Common/CMakeLists.txt, rather than trying to carry
them across the new add_subdirectory boundary - plain set() variables
don't propagate back up out of a child scope without PARENT_SCOPE, so
keeping them as-was would have silently broken Common's link line.
Verified with a fully clean rebuild (removed build/ entirely), a
HEADLESS=ON UNITTEST=ON build, and a LIBRETRO=ON build.
Co-Authored-By: Claude Sonnet 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01PSNaZnHCjmryS3ziVN9gZU
Same treatment as Common/Core/GPU: move the ppsspp_ui target definition
(kept that name rather than renaming to "UI", since it was deliberately
named that in a previous commit) into UI/CMakeLists.txt, pulled in via
add_subdirectory(UI). Source paths are now relative to UI/ instead of
prefixed with "UI/".
PPSSPPHeadless and PPSSPPUnitTest already didn't link ppsspp_ui (they
link Core/Common directly) - that's unchanged, just preserved. The main
per-platform binaries (${TargetBin}) and the Android JNI target keep
linking it as before.
add_compile_definitions(USE_DISCORD=1) has to stay in the root file
rather than move into UI/CMakeLists.txt: it only affects targets created
after the call, and add_subdirectory() snapshots the parent's directory
scope at the point it's invoked, so the definition needs to already be
set before add_subdirectory(UI) runs. Verified UI/DiscordIntegration.cpp
still pulls in Discord_Initialize etc as undefined symbols post-move.
Also removed two now-dead variable inits sitting right next to this
block: "nativeExtra" (never appended to or consumed anywhere anymore,
a leftover from before the old "native" target was removed) and a
duplicate early set(UISource) (the real one now lives inside
UI/CMakeLists.txt's own scope).
Verified with a clean ./b.sh --debug rebuild, a HEADLESS=ON UNITTEST=ON
build (confirmed ppsspp_ui/UI never gets built in that configuration),
and a LIBRETRO=ON build.
Co-Authored-By: Claude Sonnet 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01PSNaZnHCjmryS3ziVN9gZU
Mirrors the earlier Common extraction. The old "core" target folded in
all of GPU/ (~200 files) plus a few ext/ files wholesale; Windows
already treats GPU as its own project (GPU.vcxproj), so GPU/CMakeLists.txt
splits that out too. GPU has a genuine two-way dependency with Core
(Core/System.cpp calls GPU_Init(), GPU/* calls back into Core for
Memory/Config/CoreTiming/etc), so GPU is a CMake OBJECT library: its
object files are always included wherever consumed instead of being
lazily pulled from an archive, avoiding the GNU ld single-pass
archive-ordering problem a two-way STATIC dependency would hit.
Also fixed a few library misattributions discovered while tracing what
each file actually uses:
- GlslangLibs (glslang/spirv-cross) moved from Core to Common, since
it's Common/GPU/ShaderTranslation.cpp and VulkanContext.cpp that
call into it directly. It only worked before because Core happened
to always be linked after Common.
- ZSTD and OPENGL_LIBRARIES/X11_LIBRARIES moved from Core to GPU,
matching where they're actually called (GPU/Debugger/Record.cpp and
Playback.cpp for ZSTD, GPU/GLES for raw gl*() calls).
- GPU also needs Ext::Snappy directly (Playback.cpp calls
snappy_uncompress) and the libretro-common include dir under
LIBRETRO, both previously inherited for free by accident.
Also fixed USE_DISCORD's add_compile_definitions ordering: it was
being defined after ppsspp_ui's add_library call, so the UI target
never actually saw it on non-MSVC platforms.
Co-Authored-By: Claude Sonnet 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01PSNaZnHCjmryS3ziVN9gZU
First step of breaking up the monolithic root CMakeLists.txt: move the
add_library(Common STATIC ...) target definition, and every scattered
target_link_libraries/target_compile_definitions/target_include_directories
call touching it, into Common/CMakeLists.txt, pulled in via
add_subdirectory(Common).
Source paths are now relative to Common/ instead of prefixed with
"Common/". The two files living outside that directory (ext/jpge/*)
use ${CMAKE_SOURCE_DIR}/... instead.
include_directories(Common) stays in the root file rather than moving
into Common/CMakeLists.txt: it's a directory-scope command that needs
to affect targets defined *later* in the root file, and add_subdirectory
scope doesn't propagate upward or sideways, so moving it would have
silently broken unprefixed #includes elsewhere in the project.
add_subdirectory(Common) is placed at the point where the *last*
prerequisite variable it needs (PNG_LIBRARIES, RT_LIB, ATOMIC_LIB, etc.)
is guaranteed already set, not where the old add_library(Common ...)
used to start - some of those are resolved later in the file than
Common's original position was.
Verified with a clean ./b.sh --debug rebuild and a HEADLESS=ON
UNITTEST=ON reconfigure/build.
Co-Authored-By: Claude Sonnet 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01PSNaZnHCjmryS3ziVN9gZU
The Android VR build (ANDROID=true, OPENXR=TRUE) was failing to link
ppsspp_jni with undefined references to xrGetInstanceProcAddr,
xrCreateReferenceSpace, xrEnumerateViewConfigurations, etc. - all called
from Common/VR/*.cpp, which is unconditionally part of Common and always
linked into ppsspp_jni via core.
openxr_loader (built via ext/OpenXR-SDK whenever OPENXR AND NOT
ARMV7_DEVICE) was only ever being added to targetExtraLibs, which just
${TargetBin} - the desktop-style executable - consumes. ppsspp_jni,
the actual Android JNI target Gradle's CMake external native build
uses, never got it. Link it in directly, gated the same way OpenXR is
gated everywhere else in the file.
Co-Authored-By: Claude Sonnet 5 <[email protected]>
Henrik RydgårdandClaude Opus 5.5
Henrik RydgårdandClaude Opus 5
Robert Kirkman
Kevin Reinholz
Craig Carnell