Hi Finn,
I'll have to put the actual input interface call in a tasklet perhaps,
to shorten the time spent in the mouse interrupt handler.
The input interface overhead is cheap enough with ADB so I don't think
that's the problem. The printk(KERN_DEBUG "Keyboard overrun\n") in the
interrupt handler would be costly though. The comments in atakeyb.c could
be taken to mean that overruns are expected from fast mouse movements...
That's always been the case but it's not been quite as bad in old kernels.
Moving input syncing and even reporting of mouse movement into a
software interrupt does not appear to help much. Not generating the
keyboard overun message does help a bit but there's a lot of
'unhandled scancode 7d' messages still happening. These can be
suppressed in the atakbd.c driver.
I'll give the various options a bit more testing - not sure the added
complexity of using a work queue is worth the effort.
Cheers,
Michael
--
To unsubscribe from this list: send the line "unsubscribe linux-m68k" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
