rajames/LithosAnanke
1
1
Fork 0
Code Issues 61 Pull Requests Actions Packages Projects Releases 5 Activity

feat(usb): a USB keyboard -- a full one -- and no hang on a PC without PS/2

Browse Source 
boards/BRINGUP.md A3 and S3, FABRIC-3 SXLI.  A HID boot keyboard is
configured (interrupt IN endpoint, SET_PROTOCOL boot) and its reports
become the same Linux-keycode events as PS/2 and virtio (usb_kbd.c): the
full layout -- F-keys, navigation block, arrows, Num Lock, keypad.  The
console treats Caps and Num Lock as toggles shown on the keyboard's
lights, and the keypad as digits or navigation.  While a USB keyboard
runs, its event ring is serviced every tick, not once a second.

Found on the way: with no i8042 (port 0x64 reads 0xFF) the kernel hung at
boot in an endless drain -- a legacy-free PC would stop there.  Detected
and skipped now; drains bounded.

QEMU, amd64 with the USB keyboard the only one: 1 2 + . on main keys and
keypad, Caps Lock, Num Lock all right; aarch64/riscv64 configure it too.
Acceptance on three ISAs unchanged; identities hot-plug.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
rajames
2026-10-14 14:04:33 -04:00
co-authored by Claude Opus 5.5
parent 080b7012ba
commit 739cb9c163
30 changed files with 82072 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files
kernel/src/arch/amd64/i8042.c
+24 -6
View File
@@ -62,13 +62,33 @@ static inline uint8_t inb(uint16_t port) {
return val;
}
/* Is there a controller at all? A PC without one -- "legacy-free", common
* now, the keyboard on USB alone -- answers port 0x64 with 0xFF, every bit
* set, "output full" included, for ever: until 2026-10-14 the drains below
* then never ended, and the kernel hung right here (boards/BRINGUP.md S3,
* found rehearsing A3 with QEMU's i8042=off). */
static int g_i8042_present;
/* Read and drop what the controller holds -- a bounded number of bytes:
* a real controller holds at most a few. */
static void i8042_drain(void) {
int i;
for (i = 0; i < 64 && (inb(I8042_STATUS_PORT) & I8042_STATUS_OUTPUT_FULL); i++) {
(void)inb(I8042_DATA_PORT);
}
}
void i8042_init(void) {
if (inb(I8042_STATUS_PORT) == 0xFF) {
console_println("i8042: no PS/2 controller -- keyboard by USB only");
return;
}
g_i8042_present = 1;
/* Drain any stale output byte left by firmware before enabling IRQ1 —
* otherwise the first "interrupt" would really be this leftover byte
* read via polling, not a real IRQ1 delivery. */
while (inb(I8042_STATUS_PORT) & I8042_STATUS_OUTPUT_FULL) {
(void)inb(I8042_DATA_PORT);
}
i8042_drain();
outb(I8042_CMD_PORT, I8042_CMD_READ_CONFIG);
uint8_t config = inb(I8042_DATA_PORT);
@@ -110,9 +130,7 @@ void i8042_drain_stale(void) {
* edge ever again (the textbook "fires once, then never" pattern).
* i8042_init()'s drain happens at M4; VM bootstrap runs for real wall
* time between then and unmask, so re-drain immediately before. */
while (inb(I8042_STATUS_PORT) & I8042_STATUS_OUTPUT_FULL) {
(void)inb(I8042_DATA_PORT);
}
if (g_i8042_present) i8042_drain();
}
/* Ring buffer: single producer (the ISR), single reader (i8042_pop_scancode(),