[BIKESHED] Giving abort(2) a reason

[M. Warner Losh]

In message: <61286.1095024313 at critter.freebsd.dk>
            "Poul-Henning Kamp" <phk at phk.freebsd.dk> writes:
: In message <20040912.152047.16265436.imp at bsdimp.com>, "M. Warner Losh" writes:
: >In message: <61109.1095023635 at critter.freebsd.dk>
: >            "Poul-Henning Kamp" <phk at phk.freebsd.dk> writes:
: >: In message <20040912.142552.83283958.imp at bsdimp.com>, "M. Warner Losh" writes:
: >: 
: >: >: Given that we are usually pretty stumped when we get to call abort(2)
: >: >: it needs to work without malloc or anything like it and varargs into
: >: >: the kernel is not at all in my future.
: >: >
: >: >Only in malloc.  Everywhere else, people have enough state to cope.
: >: >Do we really want to have another kernel API just to support malloc
: >: >failures?
: >: 
: >: Well, the problem is that practically nothing else works once malloc
: >: fails, and people seem to find the lack of visible explanation a
: >: problem.
: >: 
: >: syslog() or anything else using varargs is not going to work...
: >
: >Wouldn't it be better to have a more generic 'Put this into dmesg'
: >thing that doesn't require malloc to work?  It seems silly to bloat
: >the kernel for only a malloc failure case...
: 
: That is what I thought I proposed...

You proposed 'put this into dmesg, with a lot of other stuff and then
abort the program.'  I was wanting 'put this into dmesg' and nothing
else...

Warner