Re: [PATCH] Fix unterminated buffer after readlink() call

[Thomas Jarosch <thomas jarosch intra2net com>]

On Friday, 14. October 2011 17:50:14 Dan Williams wrote:
> > Good catch, I missed the memset() call.
> > 
> > And if readlink() doesn't fill the buffer,
> > 
> > > the remaining bytes will already be set to 0 by the memset, AFAICT. 
> > > Are you seeing a crash here or was this from visual inspection?
> > 
> > Pure visual inspection, I was looking for readlink() usage patterns.
> > 
> > Still the buf[link_len] = '0'; is more explicit and we wouldn't need
> > the memset() call at all. What do you think?
> 
> I try to do the memset() calls specifically to prevent bugs caused by
> forgetting to do NULL termination later on; it's kind of the shotgun
> approach but these paths generally aren't performance sensitive and I
> think the tradeoff of a few cycles to memset (which is a very optimized
> operation these days) is worth the decreased possibility of bugs and
> buffer overruns here.  At least that was my thought :)

Yeah, performance is not an issue here. So let's keep it like it is.

Cheers,
Thomas