Re: segfault in stat

[Nathan Straz]

On Fri, Sep 29, 2000 at 05:24:11PM +0000, Petr Vandrovec wrote:
> On 29 Sep 00 at 16:44, Andreas Jaeger wrote:
> >  > It appears to me that when vers != _STAT_VER_KERNEL in __xstat() that
> >  > buf is not checked to be valid before xstat_conv uses it.  I realize
> >  > this is probably nit-picky, but I thought I would bring it up anyway.  I
> >  > would expect a -1 return with errno set to EFAULT as the man page
> >  > states, but the info file definately says otherwise.  
> > The authoritative source for glibc is the info documentation, if the
> > man pages are wrong, please tell Andries (CC'ed).
> 
> If my memory serves correctly, about year and half ago there was some
> talking about this inconsistency on linux-kernel - that kernel returns
> EFAULT, while faults in libraries are returned to app as SIGSEGV/SIGBUS.
> 
> General conclusion was that if apps relies on EFAULT and does not install
> SIGSEGV handler, it is buggy app. No standard requires returning EFAULT
> by kernel and there is even patch for kernel somewhere, which translates 
> all EFAULT return codes to sigsegv...

Ah, now this is the kind of answer I was looking for. I'll adjust the
test case accordingly. 

Thanks,
-- 
Nate Straz                                              address@hidden
sgi, inc                                           http://www.sgi.com/
Linux Test Project                    http://oss.sgi.com/projects/ltp/