Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 26 Jan 2000 09:39:20 -0800
From:      "Scott Hess" <scott@avantgo.com>
To:        <freebsd-hackers@freebsd.org>
Subject:   aio_read crashing certain kernels.
Message-ID:  <01b301bf6824$46e928a0$1e80000a@avantgo.com>

next in thread | raw e-mail | index | archive | help
Since nobody here seems to be able to figure it out... I wrote a program to
play around with aio_read(), and it ran stellarly well on my
FreeBSD3.4-release system using EIDE hard drives.  But, when I tried to run
it on a system using our production kernel with Adaptec SCSI controllers,
it locked the kernel up HARD.  No console messages, no Ctrl-Alt-Del, had to
push the hard reset switch.

The kernel in question is a FreeBSD3.4-release kernel compiled with a
different set of options (lots of mbufs, etc, etc).  I can't find anything
in it that seems at all related to async I/O.  The appropriate sysctl
values all seem the same between the systems.  And the fact that it locks
the system up instead of saying ENOTSUPP or "async I/O not compiled in"
concerns me greatly.

Any ideas on why this might be happening?

Thanks much,
scott




To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?01b301bf6824$46e928a0$1e80000a>