GPU and DRIVER need APS5_PROFILE_GPU or APS5_PROFILE_DRAW, and those switches add timestamp queries to every batch and timers to every packet, which changes the timings the panel shows. A screenshot of such a run could pass for an unprofiled one, so while either switch is on the panel now starts with a PROFILED line.
SHADERS counted the pipelines created between the first and the last present of the last second, so below 1 fps, with one present in that second, it read 0, as FPS did before. It now counts those created during the last frame there, like the other lines. The statistics test checks DRAWS, DISPATCH and SHADERS for a present alone in its second.
USAGE.md explains DRAWS, SHADERS, VRAM, RAM and RES, where each number
comes from, the count and memory formats, and what counting costs while
the overlay is hidden.
Below 1 fps no one-second window holds two presents, so the overlay
showed FPS: 0.0, GPU: 0.0 MS and DRIVER: 0.0 MS, which is the common
case for titles that run slowly today. When the last second holds fewer
than two presents, FPS is now the last frame's rate, 1000 divided by its
time in ms, and GPU and DRIVER are that frame's growth of each total.
With two or more presents in the last second the numbers are the last
second's as before, so the two modes meet at 1 fps.
FPS below 1 is shown with two decimals: a frame every 2 s reads 0.50 and
one every 12 s reads 0.08. Times keep their format and their 9999 cap,
which USAGE.md now states for frames of 10 s or more. Nothing changes
while the overlay is hidden.
Its section moves from after HDR output to before Windows crash diagnostics, next to the other switches for looking into a run. Other open changes add their own sections right after HDR output.
A Vulkan surface can offer HDR10 formats while the display is not in
HDR (KDE Wayland with RADV lists them with the output's HDR disabled),
so on hosts other than Windows the probe now reports an SDR monitor and
HDR output stays off until an output-state signal exists. A probe test
checks that a window without a Windows monitor report reads as SDR with
an HDR10 surface.
The HDR10 present test now also changes the monitor's luminance while
presenting HDR10, and the presenter logs the HDR metadata it sets.
AI-assisted: yes
Titles run natively, so a host CPU without an extension that compiled Zen 2 code uses faults with an illegal instruction deep inside a title (#3719: SHLX, BMI2). libkernel now reads CPUID and XCR0 when it loads on Linux and Windows, logs the host CPU's extensions and throws, naming the CPU and the missing extensions, before any title code runs.
Detect HDR output on the host: on Windows the monitor showing the window
must run in HDR10 (DXGI IDXGIOutput6), and on every host the window's
Vulkan surface must offer a 10-bit VK_COLOR_SPACE_HDR10_ST2084_EXT
swapchain with VK_EXT_swapchain_colorspace and VK_EXT_hdr_metadata. The
monitor's luminance comes from DXGI; hosts that report none get 1000
nits peak and full frame, 0 nits black.
While HDR output is available, sceVideoOutGetOutputStatus reports
dynamic range 2 and flag bit 0, sceSystemServiceGetHdrToneMapLuminance
reports the monitor's luminance, and every open port's output mode
event fires when that changes. PQ display buffers are presented
unchanged through an HDR10 swapchain with HDR metadata; any other
display buffer switches back to the SDR swapchain. Without HDR output
the PQ to SDR conversion uses the peak the title was last told as its
white level.
AI-assisted: yes
On Linux a shared-backed guest range is imported through one udmabuf
covering the whole range, and udmabuf refuses buffers above its
size_limit_mb (64 MiB by default). Larger ranges fell back to copies; an
address-based dispatch then synced its whole mirror, storing back every
pending storage image inside it each time.
Such a range is now one sparse buffer bound to the existing 64 MiB chunk
imports. The bind runs on a sparse binding queue of its own when the
device has one, so imports made at map time (without the GPU mutex) can
use it; otherwise only under the GPU mutex on the main queue.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
A direct memory page mapped at one guest address now lives in private
MEM_WRITE_WATCH memory at that address, and its writes are collected with
GetWriteWatch like flexible memory: no fault is raised. A page mapped at a
second address moves its whole resident run into the section, behind a
read-only fence while it is copied, and every address becomes a view of it,
tracked by write protection. Unmapped resident pages that are not blank are
kept in a private reservation until they are mapped again, not in the
section.
Ported onto main's 4 KiB subpage tracking of shared views: resident runs
keep a per-page collected bit and enter the fresh ranges, so Fresh and
Armed collections select them like shared views; a demoted run carries its
collected bits into the views and reports every page once. MapAlias throws
through refuse() when a resident run extends outside the aliased range, and
the caller keeps main's fallback.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01V1W4jdNSKTV7xwuS6xQPJV
The relinker reads a plain ELF executable and skips SELF containers with a
warning, so an eboot.bin or a *.prx taken from a title cannot be relinked
before it is unwrapped, and the repository has no tool for that.
The inner ELF program headers of a container carry stale offsets: a SELF
inserts its own metadata blocks between the segments, so the data sits at the
offsets of the SELF segment-record table at file offset 0x40. The segments are
also packed without honouring p_align, while ELF requires p_offset % p_align
== p_vaddr % p_align, which the relinker enforces on guest modules. The tool
places the segments from the records, falls back to the vaddr to file mapping
of the containing PT_LOAD for segments the table does not cover, copies every
region to an offset satisfying the alignment, and refuses a container whose
load segments all look encrypted.
Windows writes an exception frame directly below RSP, so a fault resumed on
a guest thread overwrites values a System V leaf function keeps in its red
zone. With --frame-red-zone, a leaf function with a
push rbp; mov rbp, rsp; push ... prologue that reads or writes below its
pushes gets two kinds of trampoline: the prologue lowers RSP by 128 bytes
after its pushes, and each pop ...; pop rbp; ret exit raises it back first.
Sites a branch enters, sites shorter than a jump and sites over a --to-intel
stub are skipped; the conversion reports the framed and skipped counts.
With --trace-imports the entry stub resolves TraceImport_nid_no_patch
in libc.prx and passes each function import through it before writing
the slot: every PLT slot, and a GOT slot only when the original symbol
is a function, so data imports such as __stdoutp keep their address.
Without the option the output is unchanged.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The console's mouse reports relative motion, so the host cursor can never line
up with the cursor a title draws. While a title has a mouse open and the
window is focused the host cursor is hidden and captured, released on focus
loss or while the pad's mouse-to-stick toggle is on. Motion deltas are scaled
from the presented image's size in window units to the title's display
resolution with sub-pixel carry, so the title's cursor travels as far as the
host pointer would, on scaled desktops and letterboxed windows alike.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
Yunus Shaikh