All of lore.kernel.org
 help / color / mirror / Atom feed
* [parisc-linux] Mount segfaults and stuff
@ 2001-10-02  8:13 Jurij Smakov
  2001-10-02  9:09 ` Thomas Marteau
                   ` (2 more replies)
  0 siblings, 3 replies; 7+ messages in thread
From: Jurij Smakov @ 2001-10-02  8:13 UTC (permalink / raw)
  To: parisc-linux

Hi!

I get segfaults from "mount", when I try to mount a non-existent
device, like

># mount /dev/fd0 /floppy
Segmentation fault

The failure is reported to be in strncmp in glibc, but it looks
like it is just trying to dereference a null pointer, passed
from somewhere else (trap 15 with zero IOR). If someone is 
experiencing similar behaviour, I can look further into it.

Also, I've tried looking at todo list to find something to work on...
It seems rather outdated and some tasks in it seem to be already
completed (like unaligned trap handler, etc.). So, where should
I direct my efforts? What is the part of parisc-linux project, which
needs immediate attention? ;-).

Cheers,

Jurij.

^ permalink raw reply	[flat|nested] 7+ messages in thread
* Re: [parisc-linux] Mount segfaults and stuff
@ 2001-10-03  8:42 Jurij Smakov
  2001-10-03 16:04 ` Grant Grundler
  0 siblings, 1 reply; 7+ messages in thread
From: Jurij Smakov @ 2001-10-03  8:42 UTC (permalink / raw)
  To: parisc-linux

> *sigh*. Lots of little stuff like floppy driver and hpux binary support.
> 
> The TODO list is stale intentionally.
> We should be using http://bugs.parisc-linux.org.
> 
> grant

Hi!

The problem is that this bug list is also rather outdated. There are
two more or less recent bugs, others are more than 2 months old. Most
kernel bugs are filed against 2.4.0 kernel. I guess, a lot of things
changed since then. For example, mount bug, that I've mentioned before
just disappeared, once I recompiled it from source with latest CVS gcc.
Also, bug #135 (kernel panic on 712 doing two simultaneous compiles) - 
yesterday I've built the kernel with "make -j 2" without any problems
on a 712/80. Is it enough to close this bug? IMHO, the right strategy
in this situation would be to try and reproduce the bugs with latest
toolchain and drop the bugs, which do not have a recipe to reproduce them.

Anyway, what should I do with mount? Do I have to contact the package
maintainer? I can build the debs myself, but those will be unofficial.

Also, last night I had some weird problems with linux/net/core/netfilter.c
Sometimes its compilation fails, but I could not pinpoint the problem
so far. I'll play with it, but it would be nice to know, if anyone
else has/had any (negative) experiences with this file.

Best regards,

Jurij.

^ permalink raw reply	[flat|nested] 7+ messages in thread
* Re: [parisc-linux] Mount segfaults and stuff
@ 2001-10-04  9:25 Jurij Smakov
  0 siblings, 0 replies; 7+ messages in thread
From: Jurij Smakov @ 2001-10-04  9:25 UTC (permalink / raw)
  To: parisc-linux

Hi!

> Well, sounds like that's something you could do - try to replicate bugs
> and close them if they can't be replicated. I'll take another pass
> at the list this week too.

Ok, let's do that.

> Preferably, if you can, find the bug and send the maintainer a patch.
> Otherwise, file a bug on either bugs.debian.org or bugs.parisc-linux.org
> depending on if it's reproducible on other arches or not.

It turned out, that there is already newer hppa deb available on the
mirrors. Once I've installed it, the problem was gone. No worries :-).

Cheers,

Jurij.

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2001-10-04  9:26 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2001-10-02  8:13 [parisc-linux] Mount segfaults and stuff Jurij Smakov
2001-10-02  9:09 ` Thomas Marteau
2001-10-02 15:18 ` Randolph Chung
2001-10-02 16:41 ` Grant Grundler
  -- strict thread matches above, loose matches on Subject: below --
2001-10-03  8:42 Jurij Smakov
2001-10-03 16:04 ` Grant Grundler
2001-10-04  9:25 Jurij Smakov

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.