mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Wei Yongjun" <weiyj@soft.fujitsu.com>
To: <linux-kernel@vger.kernel.org>
Subject: [PATCH]ip_options_fragment() has no effect on fragmentation
Date: Thu, 26 Jan 2006 09:02:06 -0800	[thread overview]
Message-ID: <0a3101c6229a$3cc7c1b0$cfa0220a@WeiYJ> (raw)

[-- Attachment #1: Type: text/plain, Size: 4301 bytes --]

[1]Summary of the problem:
ip_options_fragment() has no effect on fragmentation

[2]Full description of the problem:
When I send IPv4 packet(contain Record Route Option) which need to be 
fragmented to the router, the router can not fragment it correctly.
After fragmented by router, the second fragmentation still contain Record 
Route Option. Refer to RFC791, Record Route Option must Not be copied on 
fragmentation, goes in first fragment only.
ip_options_fragment() is the implemental function, but there are some BUGs 
in it:

ip_option.c: line 207:
void ip_options_fragment(struct sk_buff * skb)
{
 unsigned char * optptr = skb->nh.raw;
 struct ip_options * opt = &(IPCB(skb)->opt);
...

optptr get a error pointer to the ipv4 options, correct is as following:

unsigned char * optptr = skb->nh.raw + sizeof(struct iphdr);

By the way, ip_options_fragment() just fill options not allowed in fragments 
with NOOPs, does not delete the space of the options, following patch has 
corrected the problem.

--- linux-2.6.15.1/net/ipv4/ip_options.c.orig 2006-01-23 08:55:15.045904912 
+0900
+++ linux-2.6.15.1/net/ipv4/ip_options.c 2006-01-23 09:04:28.885708464 +0900
@@ -207,33 +207,63 @@

 void ip_options_fragment(struct sk_buff * skb)
 {
- unsigned char * optptr = skb->nh.raw;
+ unsigned char * optptr = skb->nh.raw + sizeof(struct iphdr);
  struct ip_options * opt = &(IPCB(skb)->opt);
  int  l = opt->optlen;
  int  optlen;
+ int  optneed = 0;
+ unsigned char * pp_ptr = optptr;

  while (l > 0) {
   switch (*optptr) {
   case IPOPT_END:
-   return;
+   goto end;
   case IPOPT_NOOP:
    l--;
+   if(optptr != pp_ptr)
+    memcpy(pp_ptr, optptr, 1);
    optptr++;
+   pp_ptr++;
    continue;
   }
   optlen = optptr[1];
-  if (optlen<2 || optlen>l)
-    return;
-  if (!IPOPT_COPIED(*optptr))
-   memset(optptr, IPOPT_NOOP, optlen);
+  if (optlen<2 || optlen>l) {
+   if(optptr != pp_ptr)
+    memcpy(pp_ptr, optptr, l);
+   optptr += l;
+   pp_ptr += l;
+   optneed = 1;
+   goto error;
+  }
+  if (IPOPT_COPIED(*optptr)) {
+   if(optptr != pp_ptr)
+    memcpy(pp_ptr, optptr, optlen);
+   pp_ptr += optlen;
+   optneed = 1;
+  }
   l -= optlen;
   optptr += optlen;
  }
+end:
  opt->ts = 0;
  opt->rr = 0;
  opt->rr_needaddr = 0;
  opt->ts_needaddr = 0;
  opt->ts_needtime = 0;
+error:
+ if (pp_ptr != optptr) {
+  if (optneed == 1) {
+   opt->optlen -= optptr - pp_ptr;
+   if (opt->optlen & 0x03) {
+    for (l = 0; l < 4 - (opt->optlen & 0x03); l++)
+     *pp_ptr++ = IPOPT_END;
+    opt->optlen = (opt->optlen + 3) & ~3;
+   }
+  } else {
+   opt->optlen = 0;
+  }
+  skb->nh.iph->ihl = 5 + (opt->optlen >> 2);
+ }
  return;
 }

--- linux-2.6.15.1/net/ipv4/ip_output.c.orig 2006-01-23 08:54:57.516569776 
+0900
+++ linux-2.6.15.1/net/ipv4/ip_output.c 2006-01-23 09:09:13.531435736 +0900
@@ -503,12 +503,19 @@
     frag->h.raw = frag->data;
     frag->nh.raw = __skb_push(frag, hlen);
     memcpy(frag->nh.raw, iph, hlen);
+    offset += skb->len - hlen;
+    if (offset == skb->len - hlen) {
+     ip_options_fragment(frag);
+     len = frag->nh.iph->ihl * 4;
+     if (hlen != len) {
+      memmove(frag->nh.raw, frag->h.raw - len, len);
+      frag->nh.raw = __skb_pull(frag, hlen - len);
+      hlen = len;
+     }
+    }
     iph = frag->nh.iph;
     iph->tot_len = htons(frag->len);
     ip_copy_metadata(frag, skb);
-    if (offset == 0)
-     ip_options_fragment(frag);
-    offset += skb->len - hlen;
     iph->frag_off = htons(offset>>3);
     if (frag->next != NULL)
      iph->frag_off |= htons(IP_MF);
@@ -619,6 +626,7 @@
    */
   iph = skb2->nh.iph;
   iph->frag_off = htons((offset >> 3));
+  iph->tot_len = htons(len + hlen);

   /* ANK: dirty, but effective trick. Upgrade options only if
    * the segment to be fragmented was THE FIRST (otherwise,
@@ -626,8 +634,11 @@
    * on the initial skb, so that all the following fragments
    * will inherit fixed options.
    */
-  if (offset == 0)
+  if (offset == 0) {
    ip_options_fragment(skb);
+   hlen = skb->nh.iph->ihl * 4;
+   mtu = dst_pmtu(&rt->u.dst) - hlen;
+  }

   /*
    * Added AC : If we are fragmenting a fragment that's not the
@@ -644,8 +655,6 @@

   IP_INC_STATS(IPSTATS_MIB_FRAGCREATES);

-  iph->tot_len = htons(len + hlen);
-
   ip_send_check(iph);

   err = output(skb2);

Best Regard
Wei Yongjun

[-- Attachment #2: fix-a-bug-in-ip_options_fragment.patch --]
[-- Type: application/octet-stream, Size: 3402 bytes --]

--- linux-2.6.15.1/net/ipv4/ip_options.c.orig	2006-01-23 08:55:15.045904912 +0900
+++ linux-2.6.15.1/net/ipv4/ip_options.c	2006-01-23 09:04:28.885708464 +0900
@@ -207,33 +207,63 @@
 
 void ip_options_fragment(struct sk_buff * skb) 
 {
-	unsigned char * optptr = skb->nh.raw;
+	unsigned char * optptr = skb->nh.raw + sizeof(struct iphdr);
 	struct ip_options * opt = &(IPCB(skb)->opt);
 	int  l = opt->optlen;
 	int  optlen;
+	int  optneed = 0;
+	unsigned char * pp_ptr = optptr;
 
 	while (l > 0) {
 		switch (*optptr) {
 		case IPOPT_END:
-			return;
+			goto end;
 		case IPOPT_NOOP:
 			l--;
+			if(optptr != pp_ptr)
+				memcpy(pp_ptr, optptr, 1);
 			optptr++;
+			pp_ptr++;
 			continue;
 		}
 		optlen = optptr[1];
-		if (optlen<2 || optlen>l)
-		  return;
-		if (!IPOPT_COPIED(*optptr))
-			memset(optptr, IPOPT_NOOP, optlen);
+		if (optlen<2 || optlen>l) {
+			if(optptr != pp_ptr)
+				memcpy(pp_ptr, optptr, l);
+			optptr += l;
+			pp_ptr += l;
+			optneed = 1;
+			goto error;
+		}
+		if (IPOPT_COPIED(*optptr)) {
+			if(optptr != pp_ptr)
+				memcpy(pp_ptr, optptr, optlen);
+			pp_ptr += optlen;
+			optneed = 1;
+		}
 		l -= optlen;
 		optptr += optlen;
 	}
+end:
 	opt->ts = 0;
 	opt->rr = 0;
 	opt->rr_needaddr = 0;
 	opt->ts_needaddr = 0;
 	opt->ts_needtime = 0;
+error:
+	if (pp_ptr != optptr) {
+		if (optneed == 1) {
+			opt->optlen -= optptr - pp_ptr;
+			if (opt->optlen & 0x03) {
+				for (l = 0; l < 4 - (opt->optlen & 0x03); l++)
+					*pp_ptr++ = IPOPT_END;
+				opt->optlen = (opt->optlen + 3) & ~3;
+			}
+		} else {
+			opt->optlen = 0;
+		}
+		skb->nh.iph->ihl = 5 + (opt->optlen >> 2);
+	}
 	return;
 }
 
--- linux-2.6.15.1/net/ipv4/ip_output.c.orig	2006-01-23 08:54:57.516569776 +0900
+++ linux-2.6.15.1/net/ipv4/ip_output.c	2006-01-23 09:09:13.531435736 +0900
@@ -503,12 +503,19 @@
 				frag->h.raw = frag->data;
 				frag->nh.raw = __skb_push(frag, hlen);
 				memcpy(frag->nh.raw, iph, hlen);
+				offset += skb->len - hlen;
+				if (offset == skb->len - hlen) {
+					ip_options_fragment(frag);
+					len = frag->nh.iph->ihl * 4;
+					if (hlen != len) {
+						memmove(frag->nh.raw, frag->h.raw - len, len);
+						frag->nh.raw = __skb_pull(frag, hlen - len);
+						hlen = len;
+					}
+				}
 				iph = frag->nh.iph;
 				iph->tot_len = htons(frag->len);
 				ip_copy_metadata(frag, skb);
-				if (offset == 0)
-					ip_options_fragment(frag);
-				offset += skb->len - hlen;
 				iph->frag_off = htons(offset>>3);
 				if (frag->next != NULL)
 					iph->frag_off |= htons(IP_MF);
@@ -619,6 +626,7 @@
 		 */
 		iph = skb2->nh.iph;
 		iph->frag_off = htons((offset >> 3));
+		iph->tot_len = htons(len + hlen);
 
 		/* ANK: dirty, but effective trick. Upgrade options only if
 		 * the segment to be fragmented was THE FIRST (otherwise,
@@ -626,8 +634,11 @@
 		 * on the initial skb, so that all the following fragments
 		 * will inherit fixed options.
 		 */
-		if (offset == 0)
+		if (offset == 0) {
 			ip_options_fragment(skb);
+			hlen = skb->nh.iph->ihl * 4;
+			mtu = dst_pmtu(&rt->u.dst) - hlen;
+		}
 
 		/*
 		 *	Added AC : If we are fragmenting a fragment that's not the
@@ -644,8 +655,6 @@
 
 		IP_INC_STATS(IPSTATS_MIB_FRAGCREATES);
 
-		iph->tot_len = htons(len + hlen);
-
 		ip_send_check(iph);
 
 		err = output(skb2);

             reply	other threads:[~2006-01-26  0:00 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-01-26 17:02 Wei Yongjun [this message]
2006-01-26  5:59 ` Eric Dumazet
2006-01-26  8:46 ` Patrick McHardy

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='0a3101c6229a$3cc7c1b0$cfa0220a@WeiYJ' \
    --to=weiyj@soft.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    /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

all inboxes | Powered by JetHome®