The changed-libraries table of the progress report printed the change in
percentage points (` pp`), while the headline prints it in `%`. Both are
the difference between two percentages, so the table now uses `%` too,
including the small-change path added with the rounding fix (`+0.0006%`,
`+<0.0001%`). Keeps the two decimals, the explicit sign and the `+0.00`
rounding of tiny or negative-zero changes.
The change is moved into `percent_change()`. Tests pin the `%` unit for
the table and the headline, and `docs/dev/PROGRESS.md` describes the
column in the unit of the headline.
Rebased onto main (which now carries #10992's small-change path) and
resolved there: that path also prints `%`.
Co-authored-by: MarcZX10 <[email protected]>
Co-Authored-By: Claude Haiku 5.5 <[email protected]>
sceHttp2SetMinSslVersion returns 0 for versions 2 (SSL 3.0) to 5 (TLS 1.2)
on a template or an unsent request: the range libSceSsl's
sceSslSetMinSslVersion takes on a 13.40 console (docs/dev/debt/libSceSsl.md).
The floor is not kept, as libSceSsl negotiates TLS 1.2 only, which meets
every accepted floor. Versions 0, 1 and above 5 still throw, as do context
and cookie-box ids and sent requests; unknown and deleted ids still return
0x817B1100.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Both call libkernel's pthread_mutex_init and pthread_mutexattr_init and
return 0x80020000 | errno on error, as shadPS4 (ce74386) exports them as
ORBIS(posix_pthread_mutex_init) and ORBIS(posix_pthread_mutexattr_init)
and ChonkyStation4 (c101a23) maps them to its POSIX init functions. A
mutex made without attributes is error-checking. FreeBSD 11.4's
malloc-internal _pthread_mutex_init_calloc_cb makes a normal mutex
instead; the gap is recorded in the libkernel_sys debt file.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
sceBluetoothHidThreadParamInitialize threw NotImplemented. It now fills
the thread parameter block that sceBluetoothHidRegisterCallback reads:
the 32-bit priority at offset 0, the 64-bit affinity mask at offset 8
and 0 in the 32 bits at offset 16, leaving bytes 4 to 7 and 20 to 23
as they were, and returns 0. A null block returns 0x81540001.
This follows BunnyFl's native x86-64 run of firmware 13.42's
libSceBluetoothHid, not a console:
https://github.com/boykopovar/AnyPS5/pull/11124#issuecomment-6416053242
The test checks the run's bytes for (700, 0x3f) and the null answer.
The run does not show the success value, range checks, bytes 20 to 23
or calls before sceBluetoothHidInit; the debt entry records them.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
kldstat was a NotImplemented stub. It now follows FreeBSD 11.4's
sys_kldstat (sys/kern/kern_linker.c): only the version field is read
(EFAULT when that fails), and a version other than 1056
(struct kld_file_stat_1) or 2080 (struct kld_file_stat) is EINVAL.
No linker file is listed here, as for kldnext, so every file id then
gets ENOENT and the structure is never written.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Without a reprojection session the three calls return 0x8a72000b
(reprojection not initialized) for every argument and write nothing.
A retail PS5 answered so from a payload before sceHmd2Initialize,
with every argument null and with every argument pointing at a zeroed
16 KiB buffer (firmware 13.60, T3cm0's hmd2s log on #7798), and in a
PPSA23479 game process (firmware 13.40, frankea on #8645).
sceHmd2ReprojectionInitialize throws on main, so no session can
exist; the in-session answers stay unimplemented and are recorded in
docs/dev/debt/libSceHmd2.md.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Fill the 120-byte SceVideodec2AvcPictureInfo for a single owned H.264
picture queried with thisSize 0x78: SPS fields, VUI, pic_struct from the
picture-timing SEI and per-unit presence flags, parsed per ITU-T H.264
(08/2021) from the access unit that produced the picture. Layout from
SharpProspero Videodec2.cs and shadPS4 videodec2.h. Other accepted
sizes, two records and unsupported syntax still throw after the
validity clears.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Name four Unknown Ot exports by their NIDs and implement them after
FreeType 2.4.12 ftcalc.c and ftutil.c and bundled 2.13.3: truncating
MulDiv, the scaled matrix product and vector transform, and the highest
power of two. Inputs on which the FT_LONG64 and 32-bit builds of
FreeType 2.4.11-2.5.3 and bundled 2.13.3 can differ throw before any
output is written.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Replace the five Unknown stubs whose NIDs hash these ICU names with the
uchar.cpp definitions of ICU 78.1 (049e0d6a), over the Zs, Zl, Zp and
White_Space code points that ICU 65.1 to 78.1 ship. Values outside
U+0000..U+10FFFF return 0, as ICU's trie error value gives.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
cacheDrawStages never admitted a draw an indirect packet runs on the CPU
path, so a title whose draws are mostly DRAW_INDEX_INDIRECT packets with
a patched user word the vertex program reads as a value missed the draw
cache on nearly every lookup whatever its size. On Demon's Souls 97 % of
the draw packets are such draws and the hit rate stayed at 9 % with
APS5_DRAW_CACHE_ENTRIES=32768.
APS5_DRAW_CACHE_INDIRECT admits them. A hit reuses the fragment stage
and skips the first vertex pass; the stage whose user word the record
patches is still re-materialized per record, as the hit path already
provided for.
Tested: agc_driver_draw_cache_reuse_indirect (three DRAW_INDIRECT draws
patching an unfolded user word: one insert and one hit with the knob,
none without), the other draw_cache_reuse, key_miss, register_key and
draw_cache_relocation tests on lavapipe and an RTX 5070 Ti.
The return codes of sceFontRebindRenderer come from our shadPS4-based port and
are unmeasured; the new test pins them, so the debt file now says so.
Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
Late-added ranges and write-watched blocks are read through the same
host-only-aware read as the snapshot, so a CPU store over a guarded
multisample target no longer aborts a capture. The read is a helper
tested for ranges in the middle, at either edge, overlapping and
outside a block, and against a native target and its neighbor.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Omar Elmahy