On Mon, Feb 17, 2003 at 08:41:06PM +0100, Manuel Bouyer wrote:
> 
> OK, this is likely a MP issue then, something blocking interrupts
> for too long (maybe an issue with the kernel big lock ?)
> I suspect a UP kernel won't have this issue

There are certainly MP-related locking issues with com(4) on i386.
Do you have COM_MPLOCK in your kernel? I'm not sure of the symptoms
if you don't, and they may well be something much worse than what
your're seeing, but please check.

--
Dan.