A PS5 (firmware 13.60) reports CPU write 0x2 with CPU read 0x1, 0x80 with 0x40 and 0x200 with 0x100 after a map, sceKernelMprotect, sceKernelMtypeprotect or POSIX mprotect; GPU write 0x20 keeps GPU read 0x10 clear. sceKernelVirtualQuery and sceKernelQueryMemoryProtection now report the protection that way.
NVIDIA's Windows driver reads a host import of guest memory far slower than a buffer it allocated.
Since draw vertex and index ranges are bound in place from host imports (#2253, measured on Linux), Astro
Bot's Crash Site on Windows 11 with an RTX 5070 Ti spends 164 ms of every frame in draws on the GPU, and
45 ms with the ranges copied as before; with the same build, 4.71 fps in place against 7.52 copied. An
RTX 3060 shows the same (draws 32.3 against 24.3 ms per frame), while AMD's Windows driver reads them in
place about as fast as a copy (RX 9070 XT: 69.8 against 73.1 ms), so copying there only costs CPU.
The device decides once (DrawInputsInPlaceUsable, Context::drawInputsInPlace): ranges are copied with
VK_DRIVER_ID_NVIDIA_PROPRIETARY on Windows and bound in place everywhere else. APS5_DRAW_INPUTS_IN_PLACE=1
binds them in place and APS5_NO_DRAW_INPUTS_IN_PLACE=1 copies them on any device, the second winning.
agc_draw_inputs_in_place_driver checks the decision per driver and host; agc_draw_inputs_copied checks
that an aligned imported range is copied and its read not noted when the context copies, whatever the
inherited environment.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01GN8FJFTedayRXLHythXkZ4
Mark described image-table ranges explicitly and apply the eager color-target guard only to ordinary image bindings. Cover native multisample table validation and record the separate table registration gap.
Apply the color-target guard before the precollected texture shortcut can reuse an ordinary sampled image. Distinguish cached and fresh access failures in the regression and unregister tracked memory before unmapping it.
Check described image ranges before either resource reuse proof, preserving the combined color and resident-depth guard for existing callers. Add standalone draw and compute regressions for sampled and storage images, disjoint targets, guarded fresh builds, and target retirement.
--small-copy-sync returns VulkanTestSkipped (77) when OpenVulkanTestDevice
fails, but its add_test had no SKIP_RETURN_CODE, so a runner without a
device counted it as a failure; the other device tests already set 77.
The MinGW libstdc++ reports a dangling symbolic link as missing and clears
the error, so std::filesystem::remove neither deletes it nor fails, and
remove_all loops forever over a tree that contains one. Remove reparse
points with DeleteFileW and RemoveDirectoryW and the fixture through the
shell; record the gap in the libs technical debt.
sceNetUsleep fails with EINVAL for a negative count and otherwise sleeps
through libkernel's sceKernelUsleep, a cancellation point. sceNetInetPtonEx
converts as sceNetInetPton does when its flags are 0 and throws for any
other flags, whose meaning is unknown. The signatures follow shadPS4 and
are recorded as unverified in the libSceNet technical debt.
APS5_NO_GPL opened a device without VK_EXT_graphics_pipeline_library and
the five extensions that go with it. The AMD Windows driver shares its
on-disk pipeline cache between devices with and without those extensions,
so a compute pipeline compiled on one kind of device and loaded on the
other ran with the wrong descriptor layout. The startup f32 flush probe
(#6503) runs the same compute pipelines on every device. In a parallel
test run, every APS5_NO_GPL test therefore page-faulted at 0xbeefdeadbeef,
an address read from the probe's own 0xdeadbeef fill. Each fault reset the
GPU, and twice the reset hung the machine.
The device now enables the extensions and their features whenever the
GPU supports them. APS5_NO_GPL only turns off their use, so every device
on a GPU builds the same pipelines.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The guard rejected every patch below the end of the program header
table, which is correct only while the table sits at the start of the
file. Outputs whose table moved into the extra block refused patches
into their first PT_LOAD, so game_patches failed once synthesis added
a header. Overlap is now checked against the ELF header and the table
range itself.
ProgramLayoutRangeTests.cpp includes this header first, and libc++ does
not provide size_t through <cstdint>, so the macOS relinker jobs failed
to compile the test.
Adrià Franch