Issues in other projects
The bugs and fixes MoonLight waits on in the projects it builds on: what each one is, what our tree does meanwhile, where it stands, and what changes once the fix ships.
A workaround's reason stays beside it in the code, which links here; the tables below hold its status, so a workaround goes in the release its reason goes.
Prior art the source cites is credited in Attributions instead.
The Status column holds what each tracker shows on 2026-10-26.
ESP-IDF
| Issue |
Filed by |
Submitted |
What |
Our side |
Status |
Once it ships |
| #19308 |
us |
2026-10-25 |
pxPortGetCoprocArea sits in flash. An interrupt entry calls it to save the hardware-loop registers of the task it interrupted, so one landing during a flash write panics with a Cache error, on the P4 and the S31 |
CONFIG_FREERTOS_IN_IRAM=y in the P4 fragment and the S31's, 14 KB of internal RAM |
Open. Espressif moves the function to IRAM in an internal MR. Without the workaround our code survives 60 restores and 186 stressed flash writes on a P4, so the fix is unconfirmed here |
Drop the line, which returns the 14 KB |
| #19025 |
us |
2026-08-28 |
In every release the P4's hardware-loop state erratum workaround is gated on a misnamed macro, so it is always on |
Nothing |
Closed on 2026-10-26, fixed on master (86837ab6, its tests b341f304), in no release |
Nothing on the rev 1 builds; the rev 3 builds lose the workaround, which their silicon does not need |
| #18759 |
another user |
2026-06-24 |
sleep_clock_icg_startup_init runs out of retention memory at boot on the P4 with esp_hosted, a boot loop |
CONFIG_PM_SLEEP_CLK_ICG_ENABLE=n in the P4 WiFi fragment |
Open |
Drop the line |
esp-dsp
| Issue |
Filed by |
Submitted |
What |
Our side |
Status |
Once it ships |
| #119, #102 |
other users |
2026-06-18 and 2025-03-24 |
Some assembly kernels break the P4's hardware-loop constraints (loop boundaries 4-byte aligned, no coprocessor instruction last), so an interrupt on a loop boundary corrupts the loop counter: the FFT crash behind #19025 |
The optimized kernels ship (CONFIG_DSP_OPTIMIZED=y); CONFIG_DSP_ANSI=y in the P4 fragment is the fallback, off |
Open. Espressif commits in #19025 to the alignment fix in an esp-dsp release |
Run the FFT and HLS soak on a rev 1 P4 |
teensy_loader_cli (PJRC)
| Issue |
Filed by |
Submitted |
What |
Our side |
Status |
Once it ships |
| PR #96 |
us |
2026-10-13 |
macOS: match only PJRC's devices in the HID manager, fixing its #60 |
teensy_loader_cli-match-own-devices.patch |
Open, no reply since 2026-10-13 |
Drop the patch |
QNEthernet
| Issue |
Filed by |
Submitted |
What |
Our side |
Status |
Once it ships |
| To file |
us |
|
Six changes: bounded waits in the W5500 driver, an overridable TCP send buffer, the receive window reopened on read, mDNS NSEC answers, each driver's buffers allocated once it finds its hardware, and the Teensy 4.1 driver's per-frame checksum flags following lwIP's checksum options |
The six QNEthernet-*.patch files in teensy/patches |
To file, as the Teensy plan lists |
Drop each patch once the QNEthernet version this tree takes carries its change |