From 6bc0ed528fc39a6fd2206eebd953358992d63ed2 Mon Sep 17 00:00:00 2001 From: "B. Watson" Date: Sat, 10 May 2025 05:45:00 -0400 Subject: update TODO. --- TODO | 30 ++++++++++++++++++++++++------ 1 file changed, 24 insertions(+), 6 deletions(-) diff --git a/TODO b/TODO index 46b71a2..46222a6 100644 --- a/TODO +++ b/TODO @@ -1,11 +1,29 @@ Definitely: +- Find out why 'marsond virtual keyboard' is showing as both a + keyboard and a pointer in 'xinput --list'. If possible, fix it + so it only appears as a keyboard. + +- Find an actual PS/2 mouse, plug it into the adaptor, make sure + the mouse port still works while marsond is running. It *should* + since it's a separate device under /dev/input, but it's better to + know than to assume. I have an old Kensington PS/2 trackball + in a box somewhere... + Maybe: -- actual realtime scheduling (maybe). sched(7). not sure it's +- If anyone ever finds another adaptor that has the Enter key + issue, with a different USB vendor and product ID, add another + udev rule for it. + +- Actual realtime scheduling (maybe). sched(7). Not sure it's needed (nice -20 seems OK even while compiling mame with -j5), and it - might interfere with stuff like JACK. -- use a log file instead of stderr, so we can have error/warning messages - after detaching from the tty. -- options to set the virtual device name and vendor/product IDs. would anyone - ever need this? + might interfere with stuff like JACK. Plus, it would bloat the + daemon (have to link with -lrt). + +- Use a log file instead of stderr, so we can have error/warning messages + after detaching from the tty. Probably not going to need this, + users can use -vf for troubleshooting. + +- Options to set the virtual device name and vendor/product IDs. Would + anyone ever need this? KISS principle says wait & see. -- cgit v1.2.3