windows / build (aarch64-pc-windows-msvc) (pull_request) Successful in 1m10s
ci / docs-site (pull_request) Successful in 1m16s
apple / swift (pull_request) Successful in 1m21s
apple / screenshots (pull_request) Skipped
ci / web (pull_request) Successful in 1m37s
ci / rust-arm64 (pull_request) Successful in 2m0s
windows / build (x86_64-pc-windows-msvc) (pull_request) Successful in 2m2s
android / android (pull_request) Successful in 5m3s
ci / rust (pull_request) Successful in 9m44s
Twelve findings from the sweep's DRY/docs/dead-code tail. Most are small; three found real defects hiding behind the duplication. **The UHID event ABI existed five times.** Every UHID gamepad backend — DualSense, DualShock 4, Switch Pro, Steam Controller, Steam Controller 2 — carried its own verbatim copy of the kernel's constants plus its own `put_cstr`, and they had already drifted: `switch_pro` was missing the SET_REPORT pair entirely, and `steam_controller` read a FIXED 16-byte SET_REPORT window instead of the event's own `size`. That last one is a bug in both directions — a longer report was truncated, and a shorter one had the parser reading whatever the reused event buffer still held past the payload, i.e. acting on rumble values the game never wrote. Now one `uhid_abi` module owns the numbers plus the two accessors that are easy to get subtly wrong, with tests on exactly that. **A dead force-feedback id fallback.** ff-core's `input_ff_upload` picks a free effect slot and writes it into the effect BEFORE uinput forwards the request, so the `id == -1` branch could never run — and allocating from a local counter would have been the wrong answer anyway, since the kernel owns that id space. Removed, with a `debug_assert` where it stood. **Apple's HID path silently dropped weak rumble.** `hidByte` took the top byte with no non-zero floor, so every amplitude below 0x0100 rendered as exactly nothing. Android has always floored it at 1; this was the odd one out. That converter also existed twice byte-identically inside one Gradle module — now one `wireAmplitudeToByte`. Also: the DS5 output-report layout gets named offsets (`dualsense_proto::out_report`) documenting all three transport bases — USB 0, SDL payload −1, Bluetooth +2 — since the differing bases are transport-forced, not drift. `pf-client-core` cannot import them (it and `pf-inject` do not depend on each other, and a DualSense layout has no business in `punktfunk-core`, their only shared crate), so its copy now DERIVES its offsets by explicit subtraction and a test pins the relationship. `PUNKTFUNK_HID_EFFECT_MAX` sizes the struct it describes instead of a second literal 11 — the header now emits `uint8_t effect[PUNKTFUNK_HID_EFFECT_MAX]`. The rumble policy engine's `min_pulse_ms` and `keepalive_ms` docs stop naming cases nothing implements: no in-tree caller sets `min_pulse_ms`, and the macOS DualSense-over-BT keepalive the doc cited CANNOT be served by the quirk, because that renderer skips writes whose levels are unchanged and would swallow the engine's re-emit — it keeps its own keepalive instead. `TrackpadHaptic` is marked as staged scaffolding (the tag is on a shipped wire; removing the variant would not reclaim it). Three ×257-vs-`<<8` doc comments corrected — the scaling itself is fine, both round-trip to 255. `backstop_ms.max(160)` deleted as unreachable (the engine floors at 500). New tests for `Ds5Feedback` and for the Android rumble JNI packing on BOTH sides, with `MAX_PADS <= 16` now a compile-time assertion rather than a comment. Closes S1-S9, S11, T2, T3 (design/haptics-sweep-2026-08-03.md M12). S11's second half is NOT a defect and was left alone: `clients/session/src/main.rs` calls `set_forwarding` unconditionally on every params-build (its own comment explains why — browse mode reuses one service across launches), so `Ctl::Forwarding` routinely arrives unchanged and that early-out is what stops a redundant `sync_open` + Valve-HIDAPI cycle each launch. Verified: pf-inject clippy -D warnings 0 / 91 tests; pf-client-core + punktfunk-core clippy 0 / 437 tests (amd64 container); punktfunk-client-android 7 tests; Android :kit: 6 tests; Apple swift build + 189 tests / 0 failures; cargo fmt --all --check clean. Each new test probed by reverting its fix — the fixed SET_REPORT window fails 3, a broken pack shift fails 3, dropping the amplitude floor fails 1, and a wrong DS5 offset either fails the pin or refuses to compile.
243 lines
10 KiB
Rust
243 lines
10 KiB
Rust
//! Host→client gamepad feedback pulls (Option B): blocking JNI shims that forward to the connector's
|
|
//! rumble (0xCA) / HID-output (0xCD) planes and return one decoded event. Kotlin owns the poll
|
|
//! threads + the Android Vibrator/Lights rendering (see `GamepadFeedback.kt`) — no JNI upcalls, no
|
|
//! `JavaVM` attach, no cached method ids. Mirrors the audio plane's one-thread-per-plane contract,
|
|
//! except the thread lives in Kotlin and we just expose the blocking pull.
|
|
//!
|
|
//! Not android-gated: `next_rumble`/`next_hidout` are pure-Rust on the `quic` feature, so these
|
|
//! compile on the host build too (parity with the input shims in [`crate::session`]).
|
|
|
|
use crate::session::{jni_guard, SessionHandle};
|
|
use jni::objects::{JByteBuffer, JObject};
|
|
use jni::sys::{jint, jlong};
|
|
use jni::JNIEnv;
|
|
use punktfunk_core::quic::HidOutput;
|
|
use std::time::Duration;
|
|
|
|
/// Short blocking timeout: long enough not to busy-spin, short enough that the Kotlin poll thread
|
|
/// observes its `running=false` flag promptly on teardown.
|
|
const PULL_TIMEOUT: Duration = Duration::from_millis(100);
|
|
|
|
/// Width of the packed `pad` field in [`pack_rumble`] — 4 bits, i.e. indices 0..15.
|
|
const PAD_BITS: u32 = 4;
|
|
/// The packing is only lossless while every representable pad index fits in [`PAD_BITS`]. This was
|
|
/// a comment before; growing `MAX_PADS` past 16 would have silently aliased pad 16 onto pad 0
|
|
/// rather than failing the build.
|
|
const _: () = assert!(
|
|
punktfunk_core::input::MAX_PADS <= 1usize << PAD_BITS,
|
|
"MAX_PADS no longer fits the 4-bit pad field in the packed rumble long"
|
|
);
|
|
|
|
/// Pack one effective rumble command into the `jlong` `nativeNextRumble` returns.
|
|
///
|
|
/// Layout — mirrored by `unpackRumbleEvent` in `RumbleWire.kt`: bits 49..52 `pad`, 32..47
|
|
/// `backstop_ms`, 16..31 `low`, 0..15 `high`. Always non-negative, so the `-1` timeout/closed
|
|
/// sentinel stays unambiguous. Split out from the JNI entry point purely so it can be tested
|
|
/// without a live session handle — the shift arithmetic is the part worth pinning.
|
|
fn pack_rumble(pad: u16, low: u16, high: u16, backstop_ms: u32) -> jlong {
|
|
(jlong::from(pad & ((1 << PAD_BITS) - 1)) << 49)
|
|
| (jlong::from(backstop_ms.min(0xFFFF) as u16) << 32)
|
|
| (jlong::from(low) << 16)
|
|
| jlong::from(high)
|
|
}
|
|
|
|
// HID-output kind tags written into the returned ByteBuffer (Kotlin reads them back).
|
|
const TAG_LED: u8 = 0x01;
|
|
const TAG_PLAYER_LEDS: u8 = 0x02;
|
|
const TAG_TRIGGER: u8 = 0x03;
|
|
const TAG_HID_RAW: u8 = 0x05;
|
|
|
|
/// `NativeBridge.nativeNextRumble(handle): Long` — block up to ~100 ms for the next EFFECTIVE
|
|
/// rumble command from the core's shared policy engine (`design/rumble-root-fix.md` §D). The
|
|
/// engine owns ALL rumble policy — v2 lease expiry, legacy-host staleness (a uniform 1 s, ending
|
|
/// the old 60 s Android exposure), connection-close drain zeros — so Kotlin applies commands
|
|
/// verbatim: `(0, 0)` = cancel now, non-zero = one-shot at this level.
|
|
///
|
|
/// Returns a packed positive long: bits 49..52 = wire `pad` index (0..15), bits 32..47 = the
|
|
/// command's `backstop_ms` (≤ 5000 — the one-shot duration, i.e. the hardware net under a stalled
|
|
/// poll thread; the engine emits explicit zeros at every policy stop, so it is never the stop
|
|
/// mechanism), bits 16..31 = `low`, bits 0..15 = `high` (0..=0xFFFF). `-1` on timeout / session
|
|
/// closed (all packed values are positive, so `-1` stays unambiguous). Kotlin routes the command
|
|
/// back to the controller holding that wire `pad` index (multi-pad rumble). Run from a Kotlin
|
|
/// poll thread.
|
|
#[no_mangle]
|
|
pub extern "system" fn Java_io_unom_punktfunk_kit_NativeBridge_nativeNextRumble(
|
|
_env: JNIEnv,
|
|
_this: JObject,
|
|
handle: jlong,
|
|
) -> jlong {
|
|
// Runs on a Kotlin poll thread, so a panic here would abort the process; guard the boundary.
|
|
jni_guard(-1, || {
|
|
if handle == 0 {
|
|
return -1;
|
|
}
|
|
// SAFETY: live handle per the nativeConnect/nativeClose contract; next_rumble_command is
|
|
// &self on the Sync connector — safe alongside the decode/audio/input threads. Kotlin
|
|
// stops these poll threads (and joins them — unbounded) before nativeClose frees the
|
|
// handle.
|
|
let h = unsafe { &*(handle as *const SessionHandle) };
|
|
match h.client.next_rumble_command(PULL_TIMEOUT) {
|
|
Ok(cmd) => pack_rumble(cmd.pad, cmd.low, cmd.high, cmd.backstop_ms),
|
|
Err(_) => -1, // NoFrame (timeout) or Closed — Kotlin loops on its running flag
|
|
}
|
|
})
|
|
}
|
|
|
|
/// `NativeBridge.nativeNextHidout(handle, buf): Int` — block up to ~100 ms for the next DualSense
|
|
/// HID-output event, written into the caller's direct ByteBuffer as `[pad][kind][fields…]` (the
|
|
/// leading `pad` is the wire pad index the event is addressed to, so Kotlin routes it to that
|
|
/// controller — multi-pad HID feedback):
|
|
/// Led → `[pad][0x01][r][g][b]` (len 5)
|
|
/// PlayerLeds → `[pad][0x02][bits]` (len 3)
|
|
/// Trigger → `[pad][0x03][which][effect…]` (len 3 + effect.len())
|
|
/// Returns the byte count written, or `-1` on timeout / session closed / buffer too small.
|
|
#[no_mangle]
|
|
pub extern "system" fn Java_io_unom_punktfunk_kit_NativeBridge_nativeNextHidout(
|
|
env: JNIEnv,
|
|
_this: JObject,
|
|
handle: jlong,
|
|
buf: JByteBuffer,
|
|
) -> jint {
|
|
// Runs on a Kotlin poll thread, so a panic here would abort the process; guard the boundary.
|
|
jni_guard(-1, || {
|
|
if handle == 0 {
|
|
return -1;
|
|
}
|
|
// SAFETY: live handle per the contract; next_hidout is &self on the Sync connector.
|
|
let h = unsafe { &*(handle as *const SessionHandle) };
|
|
let ev = match h.client.next_hidout(PULL_TIMEOUT) {
|
|
Ok(ev) => ev,
|
|
Err(_) => return -1, // timeout or closed — Kotlin loops
|
|
};
|
|
|
|
// The caller passes a direct ByteBuffer (allocateDirect) so we write its backing store directly.
|
|
let cap = match env.get_direct_buffer_capacity(&buf) {
|
|
Ok(c) => c,
|
|
Err(_) => return -1,
|
|
};
|
|
let ptr = match env.get_direct_buffer_address(&buf) {
|
|
Ok(p) if !p.is_null() => p,
|
|
_ => return -1,
|
|
};
|
|
// SAFETY: `ptr`/`cap` describe the direct ByteBuffer's backing store, valid for this call.
|
|
let out = unsafe { std::slice::from_raw_parts_mut(ptr, cap) };
|
|
|
|
// out[0] = wire pad index; out[1] = kind tag; the rest is the per-kind payload.
|
|
let n = match ev {
|
|
HidOutput::Led { pad, r, g, b } => {
|
|
if cap < 5 {
|
|
return -1;
|
|
}
|
|
out[0] = pad;
|
|
out[1] = TAG_LED;
|
|
out[2] = r;
|
|
out[3] = g;
|
|
out[4] = b;
|
|
5
|
|
}
|
|
HidOutput::PlayerLeds { pad, bits } => {
|
|
if cap < 3 {
|
|
return -1;
|
|
}
|
|
out[0] = pad;
|
|
out[1] = TAG_PLAYER_LEDS;
|
|
out[2] = bits;
|
|
3
|
|
}
|
|
HidOutput::Trigger { pad, which, effect } => {
|
|
let n = 3 + effect.len();
|
|
if cap < n {
|
|
return -1; // the raw DS5 trigger block is ~11 bytes; Kotlin allocates 64
|
|
}
|
|
out[0] = pad;
|
|
out[1] = TAG_TRIGGER;
|
|
out[2] = which;
|
|
out[3..n].copy_from_slice(&effect);
|
|
n
|
|
}
|
|
HidOutput::TrackpadHaptic { .. } => {
|
|
// Steam Controller trackpad-coil haptics — no Android equivalent; drop it (motor
|
|
// rumble already rides the universal 0xCA plane).
|
|
return -1;
|
|
}
|
|
HidOutput::HidRaw { pad, kind, data } => {
|
|
// As-is SC2 passthrough: the host's hidraw consumer (Steam) wrote this report to
|
|
// the virtual pad; Kotlin replays it verbatim on the physical controller.
|
|
// `[pad][0x05][kind][report…]` — kind 0 = output report, 1 = feature report.
|
|
let n = 3 + data.len();
|
|
if cap < n {
|
|
return -1; // reports are ≤ 64 bytes; Kotlin allocates 128
|
|
}
|
|
out[0] = pad;
|
|
out[1] = TAG_HID_RAW;
|
|
out[2] = kind;
|
|
out[3..n].copy_from_slice(&data);
|
|
n
|
|
}
|
|
};
|
|
n as jint
|
|
})
|
|
}
|
|
|
|
#[cfg(test)]
|
|
mod pack_rumble_tests {
|
|
use super::*;
|
|
use punktfunk_core::input::MAX_PADS;
|
|
|
|
/// Kotlin's `unpackRumbleEvent`, transcribed — if these two ever disagree the boundary is
|
|
/// broken, and nothing else in the build would say so.
|
|
fn unpack(ev: jlong) -> (u16, u16, u16, u32) {
|
|
let pad = ((ev >> 49) & 0xF) as u16;
|
|
let backstop = ((ev >> 32) & 0xFFFF) as u32;
|
|
let low = ((ev >> 16) & 0xFFFF) as u16;
|
|
let high = (ev & 0xFFFF) as u16;
|
|
(pad, low, high, backstop)
|
|
}
|
|
|
|
#[test]
|
|
fn round_trips_every_field_at_its_extremes() {
|
|
for &(pad, low, high, backstop) in &[
|
|
(0u16, 0u16, 0u16, 0u32),
|
|
(15, 0xFFFF, 0xFFFF, 0xFFFF),
|
|
(1, 0x1234, 0x5678, 500),
|
|
(7, 0, 0xFFFF, 2000),
|
|
] {
|
|
let ev = pack_rumble(pad, low, high, backstop);
|
|
assert_eq!(unpack(ev), (pad, low, high, backstop), "pad {pad}");
|
|
}
|
|
}
|
|
|
|
#[test]
|
|
fn every_representable_pad_survives_the_four_bit_field() {
|
|
for pad in 0..MAX_PADS as u16 {
|
|
let (got, ..) = unpack(pack_rumble(pad, 1, 2, 3));
|
|
assert_eq!(got, pad, "pad {pad} aliased in the packed long");
|
|
}
|
|
}
|
|
|
|
#[test]
|
|
fn a_packed_command_is_never_negative() {
|
|
// `-1` is the timeout/closed sentinel; any packed value colliding with it would read as
|
|
// "no command" and the rumble would simply vanish.
|
|
assert!(pack_rumble(15, 0xFFFF, 0xFFFF, 0xFFFF) >= 0);
|
|
assert!(pack_rumble(0, 0, 0, 0) >= 0);
|
|
}
|
|
|
|
#[test]
|
|
fn an_oversized_backstop_saturates_instead_of_corrupting_the_pad_field() {
|
|
let ev = pack_rumble(3, 0, 0, u32::MAX);
|
|
let (pad, _, _, backstop) = unpack(ev);
|
|
assert_eq!(pad, 3, "a huge backstop must not bleed into the pad bits");
|
|
assert_eq!(backstop, 0xFFFF);
|
|
}
|
|
|
|
#[test]
|
|
fn a_stop_is_distinguishable_from_a_hold() {
|
|
let stop = pack_rumble(2, 0, 0, 0);
|
|
let hold = pack_rumble(2, 0x8000, 0x8000, 500);
|
|
assert_ne!(stop, hold);
|
|
assert_eq!(unpack(stop).1, 0);
|
|
assert_eq!(unpack(stop).2, 0);
|
|
}
|
|
}
|