Lennart Augustsson <lennart@augustsson.net> writes:
> I realize where the code comes from, but we've had PCMCIA support for
> about 2.5 years now.  So you'd think that those delay()s would be gone
> by now...

yeah.  The problel is, when things work normally, e.g. the hard drive
responds quickly in the way that's expected, people don't notice the
delays.


> Perhaps a DIAGNOSTIC in delay() would help us find them?

perhaps rename it something else and make people evaluate every
usage...  8-)  but that's pretty drastic.

certainly, a DIAGNOSTIC check if it's a long delay might be in order.
i dunno.


cgd
-- 
Chris Demetriou - cgd@netbsd.org - http://www.netbsd.org/People/Pages/cgd.html
Disclaimer: Not speaking for NetBSD, just expressing my own opinion.