git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: David Turner <David.Turner@twosigma.com>
To: 'Junio C Hamano' <gitster@pobox.com>,
	Jonathan Nieder <jrnieder@gmail.com>
Cc: "git@vger.kernel.org" <git@vger.kernel.org>,
	"l.s.r@web.de" <l.s.r@web.de>
Subject: RE: [PATCH v3 2/2] xgethostname: handle long hostnames
Date: Wed, 19 Apr 2017 15:50:34 +0000	[thread overview]
Message-ID: <0701e70b52fe4bdd8e04e4c6918aab7a@exmbdft7.ad.twosigma.com> (raw)
In-Reply-To: <xmqq4lxlcdpf.fsf@gitster.mtv.corp.google.com>

> -----Original Message-----
> From: Junio C Hamano [mailto:gitster@pobox.com]
> Sent: Tuesday, April 18, 2017 10:51 PM
> To: Jonathan Nieder <jrnieder@gmail.com>
> Cc: David Turner <David.Turner@twosigma.com>; git@vger.kernel.org;
> l.s.r@web.de
> Subject: Re: [PATCH v3 2/2] xgethostname: handle long hostnames
> 
> Jonathan Nieder <jrnieder@gmail.com> writes:
> 
> > Hi,
> >
> > David Turner wrote:
> >
> >> If the full hostname doesn't fit in the buffer supplied to
> >> gethostname, POSIX does not specify whether the buffer will be
> >> null-terminated, so to be safe, we should do it ourselves.  Introduce
> >> new function, xgethostname, which ensures that there is always a \0
> >> at the end of the buffer.
> >
> > I think we should detect the error instead of truncating the hostname.
> > That (on top of your patch) would look like the following.
> >
> > Thoughts?
> > Jonathan
> >
> > diff --git i/wrapper.c w/wrapper.c
> > index d837417709..e218bd3bef 100644
> > --- i/wrapper.c
> > +++ w/wrapper.c
> > @@ -660,11 +660,13 @@ int xgethostname(char *buf, size_t len)  {
> >  	/*
> >  	 * If the full hostname doesn't fit in buf, POSIX does not
> > -	 * specify whether the buffer will be null-terminated, so to
> > -	 * be safe, do it ourselves.
> > +	 * guarantee that an error will be returned. Check for ourselves
> > +	 * to be safe.
> >  	 */
> >  	int ret = gethostname(buf, len);
> > -	if (!ret)
> > -		buf[len - 1] = 0;
> > +	if (!ret && !memchr(buf, 0, len)) {
> > +		errno = ENAMETOOLONG;
> > +		return -1;
> > +	}
> 
> Hmmmm.  "Does not specify if the buffer will be NUL-terminated"
> would mean that it is OK for the platform gethostname() to stuff
> sizeof(buf)-1 first bytes of the hostname in the buffer and then truncate by
> placing '\0' at the end of the buf, and we would not notice truncation with the
> above change on such a platform, no?

My read of the docs is that not only is that OK, but it is also permitted
for the platform to put sizeof(buf) bytes into the buffer and *not* 
put \0 at the end.

So in order to do a dynamic approach, we would have to allocate some
buffer, then run gethostname, then check if the penultimate element 
of the buffer was written to, and if so, allocate a larger buffer.  Yucky,
but possible.


  reply	other threads:[~2017-04-19 15:50 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-18 21:57 [PATCH v3 0/2] gethostbyname fixes David Turner
2017-04-18 21:57 ` [PATCH v3 1/2] use HOST_NAME_MAX to size buffers for gethostname(2) David Turner
2017-04-19  1:28   ` Jonathan Nieder
2017-04-19  2:57     ` Junio C Hamano
2017-04-19 14:03     ` René Scharfe
2017-04-19 17:28     ` René Scharfe
2017-04-19 19:08       ` David Turner
2017-04-19 19:09       ` Torsten Bögershausen
2017-04-19 20:02         ` René Scharfe
2017-04-20 18:37           ` Torsten Bögershausen
2017-04-20 19:28             ` René Scharfe
2017-04-21  4:18               ` Torsten Bögershausen
2017-04-18 21:57 ` [PATCH v3 2/2] xgethostname: handle long hostnames David Turner
2017-04-19  1:35   ` Jonathan Nieder
2017-04-19  2:51     ` Junio C Hamano
2017-04-19 15:50       ` David Turner [this message]
2017-04-19 16:43         ` René Scharfe
2017-04-19  2:48   ` Junio C Hamano

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: http://vger.kernel.org/majordomo-info.html

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=0701e70b52fe4bdd8e04e4c6918aab7a@exmbdft7.ad.twosigma.com \
    --to=david.turner@twosigma.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=jrnieder@gmail.com \
    --cc=l.s.r@web.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://80x24.org/mirrors/git.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).