kern/51982: sio1: interrupt-level buffer overflows

[Bruce Evans]

The following reply was made to PR kern/51982; it has been noted by GNATS.

From: Bruce Evans <bde at zeta.org.au>
To: Ian Freislich <ianf at za.uu.net>
Cc: FreeBSD-gnats-submit at freebsd.org, freebsd-bugs at freebsd.org
Subject: Re: kern/51982: sio1: interrupt-level buffer overflows
Date: Fri, 9 May 2003 10:04:41 +1000 (EST)

 On Fri, 9 May 2003, Bruce Evans wrote:
 
 > ...
 > The original version provides enough buffering for about 4 hardclock
 > ticks (default 40 msec on i386's; much smaller on some other arches)
 > of input at full speed.  The third version provides 400 msec of
 > buffering.
 
 PS: you should also try to find what is holding Giant for more than 40
 msec.  Most syscalls hold it from near their start to near their
 completion, but most syscalls don't take nearly that long.  It is
 easy to write broken ones that do - just spin for a long time, or
 return a huge amount of data via sysctl.
 
 Bruce