lirent 17ad439b9b feat(input): map the keypad (numpad) keys ...

VMCTL_KEYS carried only NumLock, not the keypad block — so any keypad evdev
code sent in CMD_INPUT{KEY} was dropped on both paths: uinput device A never
declared the keybit, and the QMP qcode lookup returned NULL.

Add the full keypad in sorted evdev positions (the table is bsearch'd):
kp_0..kp_9, kp_add/subtract/multiply/divide, kp_enter, kp_decimal, kp_equals,
kp_comma. Both the uinput keybits and the QMP qcodes derive from this one
table, so both paths pick the keys up.

Add test_keymap pinning the strictly-ascending-by-evdev invariant (bsearch
precondition + no duplicates) and the keypad lookup.

Bump 0.3.13.

2026-06-26 02:10:58 +03:00

.gitea/workflows

ci: reach the package registry over https for authenticated apt; bump 0.3.3

2026-06-22 19:33:43 +03:00

cmake

vgpu in-guest producer in-tree, release CI, flexible vmie discovery

2026-06-22 18:35:12 +03:00

include

feat(input): drop absolute-pointer (ABS) support

2026-06-24 17:14:15 +03:00

packaging

fix(discovery): derive the below-4G split robustly from fragmented mtree

2026-06-24 14:26:50 +03:00

src

feat(input): map the keypad (numpad) keys

2026-06-26 02:10:58 +03:00

third_party

vgpu in-guest producer in-tree, release CI, flexible vmie discovery

2026-06-22 18:35:12 +03:00

.gitignore

vgpu in-guest producer in-tree, release CI, flexible vmie discovery

2026-06-22 18:35:12 +03:00

CMakeLists.txt

feat(input): map the keypad (numpad) keys

2026-06-26 02:10:58 +03:00

Makefile

build: derive the project version from VMSIG_VERSION so one tag drives every package + the .so version

2026-06-22 20:35:10 +03:00

S

Description

No description provided

544 KiB

Languages

C 96.5%

CMake 2.7%

Makefile 0.5%

Shell 0.3%