The guest socket layer relies on a few Linux calls:
- The equeue wait uses poll with the timeout rounded up to a millisecond;
macOS has no ppoll.
- A readable stream socket whose peer closed is reported with POLLHUP by
peeking one byte; macOS has no POLLRDHUP.
- FIONREAD on a datagram socket, and the byte count that fstat and kevent
report for it, is SO_NREAD, the size of the next datagram; macOS adds the
source address to FIONREAD.
- socketpair throws for SOCK_SEQPACKET, which has no AF_UNIX pair on macOS.
- gethostbyname replaces gethostbyname_r, which macOS lacks.
The UDP test sends on its connected socket without an address, as #7475
does, because macOS, like FreeBSD, fails sendto with an address on a
connected socket with EISCONN. The log test uses a datagram pair with
enlarged buffers instead of SOCK_SEQPACKET.
sscanf, vsscanf, sscanf_s and vsscanf_s threw on Linux for %ls, %lc, %l[, %S and %C, because glibc stores a 32-bit wchar_t per character into the guest's 16-bit buffer. ScanGuest, the per-directive scanner the _s forms already use, now scans those fields itself: the input is decoded as UTF-8, mirroring the narrow printf family's UTF-8 output, and each character is stored as one 16-bit unit, the width and the _s size counting characters; %ls skips leading white space and stops at the next, %l[ matches an ASCII scanset with ranges and negation, and a %lc field that meets the end of the input early keeps what it read, as FreeBSD's convert_wchar does. sscanf and vsscanf route a format with a wide conversion through ScanGuest and keep the host path otherwise. The UTF-8 decoding and scanset matching live in ScanWideField.cpp, which throws for input that is not UTF-8, a character outside the BMP, a scanset member outside ASCII and positional arguments next to a wide conversion; wide conversions in fscanf still throw.
Queue 0's draw committer records the draws the packet worker hands
over. When the worker's APS5_SAMPLE_WORKER stacks wait in
DrawPipeline::Enqueue, the committer is the thread that bounds the
frame, and only APS5_SAMPLE_THREADS sees it: 40 addresses, 12 frames
deep, without whole stacks.
APS5_SAMPLE_COMMITTER=<file> attaches the worker's sampler to the
committer when its thread starts and writes the same two files. Both
use one start function, and StopWorkerSampler stops both.
The sampled lookup kept the whole array's layer count: a texture over an
array of more than maxImageArrayLayers layers read and snapshotted the
whole surface and then threw in the Texture constructor, on every lookup,
even when the descriptor viewed a few slices. A menu that samples a
16x16 array of up to 8065 layers through its image tables spent
~55 ms per frame reading those surfaces only to reject them.
cachedTexture now applies the window the storage path uses
(StorageArrayWindow, renamed ArrayLayerWindow since both paths use it):
a view whose slices fit the limit is served as an array of its own
starting at its first slice, and a view whose slices alone exceed the
limit throws before anything is read. A 2D image is one layer whatever
its last array field holds and is no longer refused by the storage path.
agc_driver_image_array_window_past_layer_limit samples and loads three
slices at 8100-8102 of an 8103-layer 1D array (before: "a texture array
with more layers than maxImageArrayLayers is not implemented") and checks
that a view of all its layers is refused with the window's error on a
device whose limit is lower.
AI-assisted: yes
A volume's depthOrLastArray is its depth, bounded by maxImageDimension3D,
not by maxImageArrayLayers: a 2049 to 4096 deep volume is a valid image
and threw since the window was introduced. Reported by Dextroid17.
AI-assisted: yes
A PS5 array may hold up to 8192 layers; Vulkan devices allow fewer (2048 on
NVIDIA). A storage view of such an array created its whole image, with more
layers than the device allows: vkCreateImage is then invalid
(VUID-VkImageCreateInfo-arrayLayers-02256) and a store into a layer the device
did not back faults the GPU. Uncharted: Legacy of Thieves Collection (PPSA05684) views a 16x16 RGBA8
array of up to 7809 layers in slices of a few hundred to 2048 in its
orphanage chapter, which lost the device (Xid 109, write invalid) a few minutes in.
StorageArrayWindow describes the viewed slices as an array of their own
starting at the first viewed slice when the whole array is past the limit;
a non-XOR swizzle does not depend on the slice index, so the slices detile
alike. Views whose slices alone exceed the limit, XOR swizzles that mix in
the slice, and any other image past the limit now throw instead of creating
an invalid image.
AI-assisted: yes
On PS4 firmware 10.71 sceNetInetPtonEx reads IPv4 text the way
inet_aton does, not the way sceNetInetPton does: "1.2.3" is 1.2.0.3,
"127.1" is 127.0.0.1, hex and octal parts are accepted, a part value
wraps at 32 bits ("1.4294967296" is 1.0.0.0) and the text ends at the
first whitespace. Two differences from BSD inet_aton were measured: a
part starting with "09" is decimal ("0989" is 989), and an octal part
containing 8 or 9 fails.
The flags argument changed nothing for 0, every single bit up to
0x100, 0x10000, 0x80000000 and -1, so it is ignored instead of
throwing. IPv6 still converts as sceNetInetPton does, which matches
the console.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Without VK_EXT_shader_stencil_export (NVIDIA), stencil reference exports
are emulated by replaying the draw once per value. An export the prover
could not bound to at most 16 values was refused. The replay shader
compares the export's low eight bits, so an unbounded export now
replays all 256 values, and proven values are reduced to their low
eight bits as the comparison sees them.
Refs #4276
Co-Authored-By: Claude Opus 5.5 <[email protected]>
DecodeState and DrawRejection rejected PA_CL_CLIP_CNTL bit 16
(CLIP_DISABLE), skipping draws that set it. Vulkan handles viewport
guard-band clipping automatically without requiring explicit clip
toggles. Accept bit 16 in ClipControlMask without altering depthClamp,
and record the unmodelled console clipping behavior in technical debt.
Tested with agc_driver_graphics_tests.
An image's own VkDeviceMemory was allocated and freed one at a time, and in a scene that re-creates its storage images at every dynamic-resolution step the dispatching queue worker spent half of its image-table resolution freeing evicted images and most of the rest allocating new ones. The pool now keeps the device memory of a released storage or sampled image under its own budget (APS5_IMAGE_POOL_MIB, a quarter of the device tier's budget by default) and hands it to the next image of the same memory type and size.
Implement sceNetGetIfName, sceNetGetRandom, sceNetMemoryAllocate and
sceNetMemoryFree with the signatures shadPS4 uses, and cover them in
guest_sce_net_tests.
Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
Add sceMoveInit, sceMoveTerm, sceMoveOpen, sceMoveGetDeviceInfo and
sceMoveSetVibration, and replace the throwing stubs of the other six
exports. Behaviour and error codes follow shadPS4; signatures follow
Orbital. A guest test covers the init state, argument checks and the
no-controller results.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01Arbzg9c2DA4DEFKVSvRgYv
Port the firmware filter for every text type, with the two codepage sets derived from the firmware converters, and declare sceImeCheckFilterText as unimplemented.
mugurc