The touch controller reports (display/touch task 4.1b)

2026-09-21. The GT9895 binds, interrupts, and delivers multitouch events to /dev/input/event0.

The interrupt was the whole problem

The device tree declared IRQ_TYPE_EDGE_FALLING. The controller binds either way, goodix_berlin probes, and /dev/input/event0 appears — so every status check passed while the device reported nothing. Measured on the board, same kernel, same everything else, only the DT interrupts flag changed:

`interrupts = <23 ...>`	`/proc/interrupts` count
`IRQ_TYPE_EDGE_FALLING` (0x02)	0
`IRQ_TYPE_LEVEL_LOW` (0x08)	2173 over one 45 s capture

/proc/interrupts confirms the live setting: 87: 2173 gpio-k230 23 Level goodix-berlin.

What it reports

evtest /dev/input/event0, finger dragged on the glass:

Event: time ..., type 3 (EV_ABS), code 57 (ABS_MT_TRACKING_ID), value 27
Event: time ..., type 3 (EV_ABS), code 53 (ABS_MT_POSITION_X), value 456
Event: time ..., type 3 (EV_ABS), code 54 (ABS_MT_POSITION_Y), value 888
Event: time ..., type 3 (EV_ABS), code 48 (ABS_MT_TOUCH_MAJOR), value 3
Event: time ..., type 1 (EV_KEY), code 330 (BTN_TOUCH), value 1
Event: time ..., type 3 (EV_ABS), code 0 (ABS_X), value 456
Event: time ..., type 3 (EV_ABS), code 1 (ABS_Y), value 888
Event: time ..., -------------- SYN_REPORT ------------

Tracking IDs are allocated and retired, BTN_TOUCH goes 1 and 0, and ABS_MT_TOUCH_MAJOR ramps as the contact patch grows. Slot-0 trajectory from a 60 s capture, decoded with ABS_MT_SLOT honoured, spans X 86..728 and Y 765..1765 continuously — both axes move, and the motion is smooth rather than quantised to a few values.

Axis ranges as the driver advertises them, matching the touchscreen-size-x/y added to the DT:

axis	min	max
`ABS_X`, `ABS_MT_POSITION_X`	0	1023
`ABS_Y`, `ABS_MT_POSITION_Y`	0	2399

Note the digitizer grid is 1024x2400 while the panel is 568x1232, so anything consuming these coordinates has to scale.

Not yet established

Whether the axes are swapped or mirrored relative to the panel. That needs a touch at a known screen position, and the panel is currently sheared by the pixel-clock workaround in dsi-burst-headroom.md, which would make any on-screen target land somewhere other than where it is drawn. Deferred to task 4.3, after the corrected image is flashed.