Re: Steve's crashing 1.3 machine, cured?

Martin Buck (Martin-2.Buck@student.uni-ulm.de)
Wed, 17 Apr 96 14:51 CDT

Bjorn Ekwall wrote:
> [...]
> There _are_ a couple of things left to do with kerneld, especially with
> handling return messages to requests generated during interrupts,
> so you don't have to hurry to select _what_ hat to use just yet... :-)

There's another thing left to do (which I already reported some time ago):
If the process that called open() and caused kerneld to load a
(character-driver) module gets a signal (SIGALRM in my case) before the
module is loaded, open() will return with ENODEV. kerneld will continue to
load the module, but then it's already too late. I worked around this by
blocking all signals in user-space before the open(), but if I'm correct,
kerneld should handle (un)loading of modules transparently, so the signals
should be blocked automatically in kernel-space.

Martin

-- 
/* Martin Buck                      E-Mail: martin-2.buck@student.uni-ulm.de */
/* Student of electrical engineering   WWW: http://www.uni-ulm.de/~s_mbuck1/ */
/* University of Ulm, Germany  Snail-Mail: Paukengasse 2, 89077 Ulm, Germany */
#include <disclaimer.h>            /* PGP Key available    MIME-Mail welcome */