public inbox for libc-hacker@sourceware.org
 help / color / mirror / Atom feed
From: Jakub Jelinek <jakub@redhat.com>
To: Ulrich Drepper <drepper@redhat.com>
Cc: Glibc hackers <libc-hacker@sources.redhat.com>
Subject: [PATCH] getnameinfo fixes
Date: Mon, 28 Aug 2006 10:09:00 -0000	[thread overview]
Message-ID: <20060828100850.GL4556@sunsite.mff.cuni.cz> (raw)

Hi!

http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=204122
complains that getnameinfo suffers similar problem like getaddrinfo used to
6 years ago, namely that EAI_AGAIN is not returned when it should.
While looking at it, I noticed e.g. struct hostent *h = NULL; if (h == NULL)
{ ... } bogosity, AF_INET loop not checking herrno == NETDB_INTERNAL and
when returning EAI_SYSTEM setting errno to the saved errno value from when
getnameinfo was entered (but the standard says that for EAI_SYSTEM errno
should contain the error code).  I have also moved the error checking code
out of the loops, removing the need to duplicate the checks, all that
is needed to check in the loops is the NETDB_INTERNAL/ERANGE condition,
so that we can grow the buffer.

What I'm not 100% sure is what should we do in the herrno == TRY_AGAIN
case, when NI_NAMEREQD is not set in flags.  The patch below will return
EAI_AGAIN for that, while previously it would return inet_ntop result.

2006-08-28  Jakub Jelinek  <jakub@redhat.com>

	* inet/getnameinfo.c (getnameinfo): For AF_INET, check errno
	only if herrno is NETDB_INTERNAL.  Handle errors other than
	ERANGE outside of the loops, handle TRY_AGAIN.

--- libc/inet/getnameinfo.c.jj	2006-06-21 17:36:38.000000000 +0200
+++ libc/inet/getnameinfo.c	2006-08-28 11:47:44.000000000 +0200
@@ -203,48 +203,40 @@ getnameinfo (const struct sockaddr *sa, 
 	if (!(flags & NI_NUMERICHOST))
 	  {
 	    struct hostent *h = NULL;
+	    if (sa->sa_family == AF_INET6)
+	      {
+		while (__gethostbyaddr_r ((const void *) &(((const struct sockaddr_in6 *) sa)->sin6_addr),
+					  sizeof(struct in6_addr),
+					  AF_INET6, &th, tmpbuf, tmpbuflen,
+					  &h, &herrno))
+		  if (herrno == NETDB_INTERNAL && errno == ERANGE)
+		    tmpbuf = extend_alloca (tmpbuf, tmpbuflen, 2 * tmpbuflen);
+		  else
+		    break;
+	      }
+	    else
+	      {
+		while (__gethostbyaddr_r ((const void *) &(((const struct sockaddr_in *)sa)->sin_addr),
+					  sizeof(struct in_addr), AF_INET,
+					  &th, tmpbuf, tmpbuflen,
+					  &h, &herrno))
+		  if (herrno == NETDB_INTERNAL && errno == ERANGE)
+		    tmpbuf = extend_alloca (tmpbuf, tmpbuflen, 2 * tmpbuflen);
+		  else
+		    break;
+	      }
+
 	    if (h == NULL)
 	      {
-		if (sa->sa_family == AF_INET6)
+		if (herrno == NETDB_INTERNAL)
 		  {
-		    while (__gethostbyaddr_r ((const void *) &(((const struct sockaddr_in6 *) sa)->sin6_addr),
-					      sizeof(struct in6_addr),
-					      AF_INET6, &th, tmpbuf, tmpbuflen,
-					      &h, &herrno))
-		      {
-			if (herrno == NETDB_INTERNAL)
-			  {
-			    if (errno == ERANGE)
-			      tmpbuf = extend_alloca (tmpbuf, tmpbuflen,
-						      2 * tmpbuflen);
-			    else
-			      {
-				__set_h_errno (herrno);
-				__set_errno (serrno);
-				return EAI_SYSTEM;
-			      }
-			  }
-			else
-			  {
-			    break;
-			  }
-		      }
+		    __set_h_errno (herrno);
+		    return EAI_SYSTEM;
 		  }
-		else
+		if (herrno == TRY_AGAIN)
 		  {
-		    while (__gethostbyaddr_r ((const void *) &(((const struct sockaddr_in *)sa)->sin_addr),
-					      sizeof(struct in_addr), AF_INET,
-					      &th, tmpbuf, tmpbuflen,
-					      &h, &herrno))
-		      {
-			if (errno == ERANGE)
-			  tmpbuf = extend_alloca (tmpbuf, tmpbuflen,
-						  2 * tmpbuflen);
-			else
-			  {
-			    break;
-			  }
-		      }
+		    __set_h_errno (herrno);
+		    return EAI_AGAIN;
 		  }
 	      }
 
@@ -361,10 +353,7 @@ getnameinfo (const struct sockaddr *sa, 
 				 (const void *) &(((const struct sockaddr_in *) sa)->sin_addr),
 				 host, hostlen);
 		if (c == NULL)
-		  {
-		    __set_errno (serrno);
-		    return EAI_SYSTEM;
-		  }
+		  return EAI_SYSTEM;
 	      }
 	    ok = 1;
 	  }

	Jakub

             reply	other threads:[~2006-08-28 10:09 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-08-28 10:09 Jakub Jelinek [this message]
2006-08-28 16:24 ` Ulrich Drepper

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

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

  git send-email \
    --in-reply-to=20060828100850.GL4556@sunsite.mff.cuni.cz \
    --to=jakub@redhat.com \
    --cc=drepper@redhat.com \
    --cc=libc-hacker@sources.redhat.com \
    /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.
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).