Credits are down to SplashDown, whose code the virtual-device layer is
ported from; everything else Control4Free uses, with its license, stays
listed in THIRD_PARTY.md. The AI models named are the ones actually used:
Claude Opus 5.5 and Astra GPT-6.
- README: what Control4Free is and why it reaches menus and sign-in, features,
a quick start, screenshots, requirements, limitations, credits.
- docs/: installation, playing, troubleshooting (every message the page and the
app can show), how it works, and the WebSocket protocol.
- CONTRIBUTING.md: building, testing, code style, releasing.
- Screenshots of the page on phones and a PC and of the app, made against the
real service built for the host, at the example address 192.168.1.20.
- Issue forms, a contact list pointing at the guides and private security
reporting, and FUNDING.yml.
- CHANGELOG.md dated. CI now publishes this version's section of it as the
release notes rather than the whole file.
- THIRD_PARTY.md: the DualShock 4 diagram the buttons are drawn after (CC BY
3.0), and Apollo Save Tool for the package settings.
- Page: the Vibration setting no longer promises to buzz when a game rumbles,
which 1.0.0 cannot do yet.
From a last pass over the code and a set of screenshots on phones, an iPad and
a PC:
- The menu button sat on top of R2 on a phone in landscape: R2 starts about 68px
from the right edge, room for one corner button, not two. The menu and full
screen buttons now stack down the edge instead.
- On a narrow phone the connection status was cut off mid-address ("Connected
to 192.16"). The address is now a detail that narrow screens leave out.
- The controller screen said "Controller 1 · Controller 1": the page's own label
and the console's name for it had become the same words. The console's name
is only added now when it says something more.
- Free controllers still offered "a controller on this device", and the app
"controllers connected to the phone or PC"; both now say gamepad, as the rest
of the page does.
- SECURITY.md said no website could drive the console. A page opened from a
saved file connects with Origin: null, and that has to be accepted because the
saved file is how gamepads work in browsers that keep the Gamepad API to secure
pages; a website can send the same Origin from a sandboxed frame. It now says
exactly that, and what such a page still cannot do.
- THIRD_PARTY.md credited the GoldHEN call to autorun.c; it is in sandbox.c.
- Two comments pointed at a tag that only exists locally.
The package declared itself a game (CATEGORY gd), so the PS4 filed it under
Games in the library.
It now uses the values Apollo Save Tool ships with: CATEGORY gde ("Non-Game Mini
App"), ATTRIBUTE 32, APP_TYPE 1. Apollo is built with the same OpenOrbis tools and
lists under Applications. APP_TYPE 4 ("Freemium"), which is reported to move apps
there too, made the PS4 refuse to start ours (CE-39929-2), so it stays 1.
Confirmed on the console (firmware 10.01): the app installs, starts, and lists
under Library > Applications. A test reads the fields back out of the built
package.
The menu button sat at a fixed 70px from the right while the full-screen button
followed the safe area. On an iPhone in landscape the safe area is 47-59px, so
the full-screen button slid underneath the menu; on Android, with no safe area,
the two stood 24px apart; and wherever full screen is hidden, the menu was left
floating away from the corner.
Both now sit in one row that follows the safe area, 8px apart, and a hidden
button takes its gap with it.
Safari on iPhone lets only video go full screen, so the button did nothing
there. The check for it looked for requestFullscreen, which exists on iPhone even
though every call is refused; it now asks fullscreenEnabled instead.
On iPhone the button now says how to get the same thing: add the page to the Home
Screen and open it from there, where it runs without Safari's bars. When the page
already runs that way, or is installed anywhere else, the button goes away.
Elsewhere it goes full screen as before, prefixed or not.
The app's controller count, and everything else on its screen, only changed when
the app was closed and opened again.
OpenOrbis's time.h gives CLOCK_MONOTONIC the Linux value, 1. The PS4's kernel is
FreeBSD's, where clock 1 is CLOCK_VIRTUAL: the process's own CPU time. The worker
sleeps between checks and uses almost none, so its clock barely moved and the
check due 2 seconds later never came round. Only the first check, on start-up,
ever ran. The payload is unaffected: its SDK has FreeBSD's value, 4.
The launcher now uses 4 when it is built for the console. Checked the other
constants the app uses against FreeBSD's: everything matches except this one,
MSG_NOSIGNAL and SIGSYS, which were already worked around. A test now fails if
any of the three is used directly in the launcher again.
On the console, every new controller was removed straight away with "That
controller was disconnected after sitting unused".
The idle reaper compared unsigned millisecond stamps directly: now - lastInput.
`now` is read at the top of a main-loop iteration, but since creation became a
state machine the claim finishes inside that iteration and stamps lastInput from
the clock afterwards. Once the two fell on different milliseconds the subtraction
wrapped to about 49 days, and every controller in the claim was deleted. On the
console the gap is InsertData and klog work, so it happened nearly every time;
on a fast host it almost never did, which is why the tests missed it.
Every elapsed-time comparison in web.c now goes through c4fSince, which cannot
wrap, and the clock is read again before the reaper runs. The reaper also logs
what it removes and why, so this would have shown up in the log at once.
The test stub can now spend time inside an iteration the way the console does
(C4F_TEST_SLOW_MS), and the new test fails on the old code in its first round.
CHANGELOG.md says what 1.0.0 is. The CI workflow builds both toolchain images,
the payload and the package, runs every host test, and on a vX.Y.Z tag checks the
tag against VERSION before publishing the ELF, the package and their SHA256 sums.
`make C4F_PROBE=1` adds a poll of scePadVirtualDeviceGetRemoteSetting for every
live controller, logging the return code and only the bytes that change. That is
the one call that looks like it could carry a game's rumble and light-bar state
back to a virtual pad, and nothing is known about its buffer, so this is how to
find out on hardware. It builds to control4free-probe.elf and is never in a
release.
Three bits of polish for people who did not write this.
The payload serves /manifest.webmanifest and /icon-192.png, and the page points
at both, so a phone can keep the controller on its home screen instead of typing
an IP address again. The icon is the app's own icon, rendered by the launcher's
drawing code and committed, so the payload build needs no renderer; a test
re-renders it and fails if the two drift apart. One size only: the gradient does
not compress, and a 512 copy would add 150 KB for a splash screen that a plain
HTTP shortcut never shows. write_png now picks the best filter per row, which
also shrinks the package icon.
When the service starts it works out which address the console is actually
reachable on, from the route to the outside world without sending anything, and
puts it in the start-up notification on the TV.
Every error the page can show now says what happened and what to do about it,
rather than "Invalid axis", and the README has a troubleshooting section covering
the ones people will actually hit.
Input was queued and handed to the virtual device on a 16 ms tick, so a press
could wait most of a frame before the console saw it, and the page only sent
state once per animation frame.
The payload now reports a sample as soon as it arrives, unless one went out
within the last 4 ms, and then keeps a 4 ms cadence while the player is doing
something, as a real DualShock 4 does. Half a second after the input stops it
falls back to the 16 ms keepalive, and the main loop's wait follows: 2 ms while
input is moving, 8 ms with a controller sitting idle, 50 ms with no controller at
all, so an unused service costs the console almost nothing.
On the page, a button, touch or key edge is flushed where it happens instead of
waiting for the next frame; stick and trigger movement still rides the frame
loop, where coalescing is what you want. A `ping` method and a round-trip
reading next to the version make a laggy network visible rather than guessed at.
Connecting a controller took a second or two in the middle of the request
handler: drain the kernel log, prove it delivers, call AddDevice, then read the
log until the device turns up. Everyone else waited. A second player adding a
controller froze the first player's input for as long as it took.
The service now owns one kernel-log reader and reads it a line at a time from
its main loop, and creating a controller is a state machine that loop advances:
drain, verify with a marker, AddDevice, adopt the DeviceId. Nothing blocks, so
the other players keep playing and keep being answered while it runs. Their
slots show "Connecting" until the claim is complete. One controller is created
at a time, because two AddDevice calls at once produce two log lines with no way
to tell which device is which; a second request is refused rather than queued.
Nothing in vda.c waits any more, so the callback that kept pads reporting
through those waits is gone, and the line parsing moved to src/klog_line.c,
which the host tests build as it is.
Only the login manager's own line is accepted as our new device now:
SCE_MBUS_EVENT_DEVICE_ADDED with subType 2. The old code fell back to any
"device added" line it saw, so a real pad being plugged in while a controller
was being created could hand us a DeviceId we do not own, and every input would
go to someone else's pad.
On the page: a refused claim re-reads the console's view and gives up only the
controllers that are in use elsewhere, instead of disconnecting everything this
device was playing with; and when getGamepads() starts throwing, every gamepad
source is blanked, so a held button cannot be flushed for ever with nothing left
to release it.
The test stub is now a fake kernel log: it writes the lines firmware 10.01
writes, our own mirrored output included, so the real reading, marker check and
DeviceId matching are what the tests exercise. SECURITY.md writes down what the
open-LAN design does and does not defend.
The staged spike that worked out the VDA call order is no longer part of the
build. It is kept at tag spike-final.
- src/main.c is the service only: instance lock, credentials, MBus and pad
init, then the web service. src/server.c and include/c4f_server.h are gone
with stage 8, and so are the stage-only helpers in src/vda.c: the MBus bind
and holds, the assignment and button-map probes, and the GoldHEN klog stream
path. The service reads /dev/klog, which is what works.
- VERSION holds the only version number. Both Makefiles pass it in, the
packager reads it for param.sfo and the package name, and the build and
deploy scripts read it for the file they look for.
- The payload's log is /data/control4free/control4free.log, not spike.log.
- The page drops the protocol 1 branches: the payload has always answered 2.
A console slot is a "controller" and a pad plugged into the phone or PC is a
"gamepad", so the two no longer read as the same word.
- The host tests move to tests/, with tests/run.py to run all four suites. They
build the real sources with the PS4 calls stubbed, so they need no console.
klog_test now covers the device path only, and asserts that nothing ever
connects to GoldHEN's klog server.
On the console, 0.2.1 often could not add a controller ("Cannot read
controller sign-in events"). Its log shows why: once Control4Free let go
of GoldHEN's klog stream, every new connection read nothing for 4-5
minutes. GoldHEN's klog server serves one client at a time and is slow to
notice that one has left.
The browser service now opens /dev/klog itself, only while a controller
waits for sign-in, and releases it afterwards, so GoldHEN's klog server
works the rest of the time. It never uses GoldHEN's stream; only the
diagnostic stages still fall back to it.
Version 0.2.2 (APP_VER 00.22).
- The service checks its listener every second and notices long pauses
such as rest mode. It then drops stale connections, keeps controllers
neutral, and rebuilds the listener. Confirmed on the console: Control4Free
survives rest mode.
- A lock in /data/control4free stops a second copy before it touches the
host process's credentials or the log.
- Browser mode reads klog through GoldHEN's klog server first, after a
marker check, and falls back to /dev/klog. It holds a reader only while
controllers wait for sign-in, so GoldHEN's own klog keeps working.
- Log lines carry elapsed time, the previous run is kept as
spike.log.previous, and a heartbeat is logged every minute.
- Version 0.2.1 (APP_VER 00.21).
The mark is a vector DualShock 4 seen from above (launcher/logo.c), filled
with stb_truetype's anti-aliased rasterizer through a small path API in
draw.c. The app's top bar and the page's top bar, toast and favicon all use
it. The package icon shows three controllers with their light bars glowing
in the page's player colours.
On the console the app's connections to 127.0.0.1 and to the console's own
address both failed with EACCES: the app sandbox does not let an app reach
the console itself. So the app reached neither Control4Free nor PayLoader,
while a phone or PC on the network could.
The app now takes itself out of the sandbox at startup with GoldHEN's SDK
call, the same one the auto-start setup used, and stays out. The bundled
payload is read into memory first, since /app0 is only visible from inside,
and starts send it from there. When nothing connects at all, the screen
says the app cannot reach Control4Free instead of blaming another copy.
OpenOrbis's headers give SIGSYS Linux's value (31); the kernel's is 12.
- The first Cross copies the bundled payload to /data/payloads and adds it
to GoldHEN's AutoRun list, then starts it if PayLoader is on. Triangle
turns auto-start on or off, or updates an older copy. The app leaves its
sandbox for this through GoldHEN's SDK call and returns right after.
- The app tries 127.0.0.1 and then the console's own address, for both
Control4Free and PayLoader, and shows why neither answered.
- The launcher API no longer requires a loopback source address, since
the app's sandbox may not appear as 127.0.0.1. The custom header and
the Origin check still keep websites out.
- The screen and messages no longer assume a phone.
An OpenOrbis app that sends the bundled payload to GoldHEN's PayLoader,
shows its address and a QR code, and stops it through the loopback API.
It draws its screen in software in the controller page's style
(stb_truetype with Roboto), and its icon with the same code.
The package (title ID CFRE00001) carries OpenOrbis's stub libc.prx and
libSceFios2.prx in sce_module/, as the toolchain asks.
The README covers GoldHEN AutoRun, the launcher, the page and the build.
A GoldHEN ELF payload that creates virtual DualShock 4 controllers with
scePadVirtualDevice*, the path Remote Play uses, so they work on the home
screen, at sign-in and in games. A phone or PC drives them from a page the
payload serves on port 4264.
- Stage 0 (default) is the browser controller: four slots, explicit
creation, native PS4 user selection, input over a WebSocket.
- Stages 1-8 are the diagnostics that established the VDA call order.
- Loopback-only /api/status and /api/stop for the launcher app.
- The klog connection is reopened when a controller is created, so a
start through GoldHEN AutoRun still works if it beats the klog server.
The VDA code is ported from seregonwar/SplashDown (GPL-3.0).
MoHadiShibli