The only time I've seen inspecific EOI (outb 0x20, 0x20) cause a problem
is when multiple interrupts occur nearly simultaneously. The first interrupt
causes entry into an interrupt service routine. If that routine leaves
interrupts disabled and another interrupt gets signalled before the EOI, then
the inspecific EOI clears an interrupt that hasn't been processed.
Could this be Richard Johnson's problem?
P.S.: Inspecific EOIs make me nervous. In what situation does specific EOI
cause a problem?
- --
John Paul Nollman ne' Darren Senn sinster@darkwater.com
POBox 64132, Sunnyvale, CA, USA 94088-4132
Unsolicited commercial email will be archived at $1/byte/day.
Congratulations FBI men: Hoover would be proud of you
-----BEGIN PGP SIGNATURE-----
Version: 2.6.2
Comment: Processed by Mailcrypt 3.4, an Emacs/PGP interface
iQCVAwUBM5hdKsyyNwVEgVztAQEmHQP/aMo6oXiWgYdDSl218CkNSX46F/g7p5Bk
oVQxEhkxwAKVg8jQPAYPwNlwlW2CUNUbVIessb8zbMDXPdUujTjdx5JKGLM+IMgj
azjeKZHKlcO9KXjr6s2gMGYW1A0VjcVjPDHx7MS9ussl02seGAokRjqFUQ/c0nHd
7cThTW7jfKE=
=HG+e
-----END PGP SIGNATURE-----