Upstream dropped the Emscripten paths, the pak checks are unchanged.
The old pin already sat between v2.86.0 and v2.86.1, so core_version
was behind; HEAD contains v2.86.2.
VR builds mount base_vr/ over base/ and d3xp_vr/ over d3xp/ when
boom3_vr_enable is on, and warn when base_vr/pak399.pk4 is absent
(libretro.cpp:4313-4335). Pin moved to efe140c, notes recaled from
bare basenames to their repository paths.
Upstream added tape autoload and PCG-8100 audio, which read no ROM.
load_system_file moved to libretro.c:977, the ROM table is the same.
Comment citations now carry their repository path.
A disc image runs the configured System Card ROM and refuses to load
without one (mainloop/pce.rs:29-50). The default Arcade Card wants a
v3.0 ROM. Pin moved to d41e2d9, core_version 0.14.1 as Cargo.toml
already read at the old pin.
Port changes are VR, DSU and audio options, no file loading. The
SegaBoot refs now cite the libretro lookup (Sys, then User/Triforce)
at AMMediaboard.cpp:500-514; upstream to e1d35f6, CommonPaths.h same.
bennugd, eden, play, scummvm and uzem cited files their repositories have
since moved (libretro.c into libretro/, fs_paths.h into src/common/fs/,
uzem.cpp into oldsrcs/). Lines unchanged. scummvm's upstream_commit is
its fork base e0ddc60.
For a port that is a GitHub fork of its upstream, the upstream code it
carries is the merge base of the two, not upstream's HEAD. The bump moved
upstream_commit to upstream HEAD on snes9x, kronos, np2kai and others
whose pin was that base. The upstream view's head is now the merge base
GitHub's compare names, the way _follow_submodule already reads a
vendored submodule. Outside a fork network the HEAD stays.
The note citations hold under three source pins; read again at 583f0e9 they land on the acgame loader, the BIOS scan and the resource reads, recorded as prose_written_at. GameDatabase.cpp:39, Patch.cpp:117, SDLInputSource.cpp:24 and linux_build_qt.yml:129 read the same in the fork, the tree RetroBat ships.
libretro/Omicron master is msx80/Omicron aaaa2f1 (2020-09-29) plus .gitlab-ci.yml and the core Makefile. Upstream HEAD 49e4e05 is 168 commits later and moved the libretro frontend files the notes cite.
upstream_commit 3353efd3 is the higan v106r25 release commit nSide synced with (Emulator::Version 106.25). gitlab higan HEAD 00684d2 is a later higan and stays out.
Boot.cpp:408-442, WiimoteEmu.cpp:156-165, NWC24Config.cpp:17 and TitleDatabase.cpp:54-58 land on their code in Dolphin-MMJR2 at 4d0fdb8 and miss it in dolphin-emu at 4ff214c: every citation reads the fork.
Profile the rommgr.cpp table at 376b22c: 208 entries for CPU boards,
freezer cartridges, SCSI and IDE controllers, Picasso IV, x86
bridgeboards, CD32 FMV and the CD32/CDTV images the scan identifies
by SHA-1. A CPU board without its ROM does not start, so the A1200
and A4000 accelerator quickstarts need these files.
A sentence written in two fields was located over the whole file, found
twice and left in both: the writer now searches the lines of its own
scalar. A structured ref was judged pending when its old text appeared
anywhere, so a chain of shifts (tr1x cut2 508 to 531, cut3 531 to 552)
held the pin on the values the previous entries took: it is now looked
up at its own field.
The recale skipped runs in YAML comments and runs wrapped over two
lines of a literal block, while the pin check counted them: the run
wrote nothing and said nothing. Both are now rewritten at their located
position. A citation that still cannot move is named on stderr and the
profile is left untouched.
The bulk advance moved upstream_commit to upstream HEAD. On freej2me_plus,
hakux, kronos, mesen-s, np2kai and snes9x it was, and still is, the
merge base of upstream with the fork's pin (GitHub compare), the revision
the port carries. ax360e, nes-emu and snes9x-explus vendor an upstream
tree whose commit their notes name.
The zn.c rewrite filed the Time Warner and Raizing BIOS sets under new
ids zn1-atari and zn1-raizing, which have no name and stop
generate_site. Every other MAME profile files them as atari-zn1 and
eighting-zn1 (Raizing is Eighting's label).
hiscore.dat and artwork/dotron.zip as mame2003-plus ships them at
ffc68e3 (sha1 d41b1c32, c6d6579f, both as its profile declares), and
Plus.j2d from the JJ2+ installer of 2026-08-31 (get.jj2.plus, sha1
fcec91aa), which jazz2-native converts textured layers from.
A stop by SIGTERM (Start+Select under gptokeyb) reads the summary like
Ctrl-C and prints the count on disk. The systems filter keeps the ticks
and the focus. The bar names what A does on the focused row. Options
check the folder again, a missing network fails in 15 s with plain words,
an unwritable folder offers another one, a folder inside a card offers
its root, and only a failure keeps its log.
New event kind: checking, with files_done and files_total, while the
files in place are hashed. The exact event lists of the check and
install runs now carry two of them, for the two files in place.
The 2.43 tag sits on main2 (b5ae1af, changelog ending at 1.38) while
the 2.43.apk asset carries main4's changelog and RPCS3 20071-1f080485.
The two branches share merge base d4622157. The pin c6ca182 already
names the released tree.
Music tracks are now looked up through the audio language chain,
music/<language>/ first and then music/, and a catalog row may list
alternate names (backend_files.c:108-134). Same level, FMV and track
names.
Upstream split the sound code into crates (soundblaster, sc55, mt32,
gus). The AWE32 and Sound Canvas tables are unchanged. Adds the
optional 128 KB Verite board ROM that verite_rom names, which the
emulated RISC reads in place of the image the emulator builds
(src/bus/verite.rs:45-71).
Upstream changes at the cited lines are the sprite atlas type passed to
loadSff and a branch input for the screenpack clone in the release
workflow. Every file is still read from the same place.
rom_load now maps Scorpion ZS-256, ZS-256 Turbo+, Scorpion GMX and
Didaktik Gama to four new bundled ROMs (src/cpu.c:3549-3564), sized by
get_rom_size at 64K, 64K, 512K and the 16K default. The TK90X Ferranti
models reuse tk90x.rom and tk90xs.rom. Both ZX Evolution ROM images
were replaced upstream, so their sha1 changed.
The game database now merges resources/emucorex_overrides.yaml over
GameIndex.yaml (GameDatabase.cpp:1274-1278), and a libvulkan_malisx2.so
driver is kept only on the Mali GPUs GpuDriverCompatibility.kt:204-208
lists. tfx.glsl changed upstream. Both convert.glsl entries now carry
the size and sha1 of the vendored tree.
The installer page gains a section on copying from an old card, the
found event, the error stage for a --from folder, and the reasons a
check gives for a Mac ._ file and a name in capitals. The first
install page says in a few lines how to copy the BIOS files from an
old card.
--from FOLDER (repeatable, a ZIP accepted) reads an old SD card, a
backup or a Downloads folder: files indexed by size, hashed only when
a size matches an entry, ZIP members included. Each file is copied,
checked again once written, and the source is never modified. Mac ._
files, system folders and links leading out of the folder are skipped.
An install also puts in place the files it finds misplaced in the BIOS
folder (bios/bios/, a pack extracted one level too deep, a name in
capitals, a ZIP never extracted, one folder up), leaving the user's
copy where it is. --check and --dry-run say what a run would copy.
Events gain one kind, found, sent before the copy, which a downloaded
event then reports. The end of the run says where each file came from.
Each manifest carries system_titles, the name the site prints for every system its files list (Sony PlayStation, not sony-playstation or psx), so the installer can show it. Older installers ignore the field.
Abdessamad Derraz