Date: Sun, 15 Sep 2002 13:13:03 -0700 (PDT) From: Matthew Dillon <dillon@apollo.backplane.com> To: Poul-Henning Kamp <phk@critter.freebsd.dk> Cc: Bruce Evans <bde@zeta.org.au>, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/sys/kern vfs_bio.c src/sys/sys bio.h Message-ID: <200209152013.g8FKD3Ed057840@apollo.backplane.com> References: <2157.1032119476@critter.freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
:I don't see any swear words in the above, only a not entirely
:polished request that you examine facts before going totally Terry.
:
:--
:Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
I'm not sure why you want to believe that, or use it as a justification
for blowing off the rather serious issues I've brought up instead of
tackling or answering or even discussing it. I have looked at the
code quite closely, and there are most certainly problems with it.
In an interrupt/SMP environment the race you have created can be lost
quite easily when Giant is no longer there to govern access
and potentially result in seriously non-deterministic (terrible)
performance for certain situations. Simply changing the timeout from
one second to 1/10 second in your biowait() function is not going to
solve any of the problems, it's just going to hide them a little more
then before. Just put in a spin mutex and be done with it.
-Matt
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200209152013.g8FKD3Ed057840>