netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: David Miller <davem@davemloft.net>
To: bcrl@kvack.org
Cc: netdev@vger.kernel.org
Subject: Re: [PATCH] fib_semantics: prevent long hash chains in access server config
Date: Sun, 13 Jan 2008 18:11:18 -0800 (PST)	[thread overview]
Message-ID: <20080113.181118.08046749.davem@davemloft.net> (raw)
In-Reply-To: <20080113175849.GC12775@kvack.org>

From: Benjamin LaHaise <bcrl@kvack.org>
Date: Sun, 13 Jan 2008 12:58:49 -0500

> Ah, you're right indeed.  It's probably easier for me to change how the 
> daemon adds the local ip address for these point to point interfaces.

I guess you didn't understand, I checked in the following
patch which implements things correctly.

commit d5414fdd4098742a5a9d255c4b9af587318c525f
Author: David S. Miller <davem@davemloft.net>
Date:   Sat Jan 12 21:49:01 2008 -0800

    [IPV4] FIB: Include nexthop device indexes in fib_info hashfn.
    
    Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 3ed920b..0e08df4 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -194,6 +194,15 @@ static __inline__ int nh_comp(const struct fib_info *fi, const struct fib_info *
 	return 0;
 }
 
+static inline unsigned int fib_devindex_hashfn(unsigned int val)
+{
+	unsigned int mask = DEVINDEX_HASHSIZE - 1;
+
+	return (val ^
+		(val >> DEVINDEX_HASHBITS) ^
+		(val >> (DEVINDEX_HASHBITS * 2))) & mask;
+}
+
 static inline unsigned int fib_info_hashfn(const struct fib_info *fi)
 {
 	unsigned int mask = (fib_hash_size - 1);
@@ -202,6 +211,9 @@ static inline unsigned int fib_info_hashfn(const struct fib_info *fi)
 	val ^= fi->fib_protocol;
 	val ^= (__force u32)fi->fib_prefsrc;
 	val ^= fi->fib_priority;
+	for_nexthops(fi) {
+		val ^= fib_devindex_hashfn(nh->nh_oif);
+	} endfor_nexthops(fi)
 
 	return (val ^ (val >> 7) ^ (val >> 12)) & mask;
 }
@@ -232,15 +244,6 @@ static struct fib_info *fib_find_info(const struct fib_info *nfi)
 	return NULL;
 }
 
-static inline unsigned int fib_devindex_hashfn(unsigned int val)
-{
-	unsigned int mask = DEVINDEX_HASHSIZE - 1;
-
-	return (val ^
-		(val >> DEVINDEX_HASHBITS) ^
-		(val >> (DEVINDEX_HASHBITS * 2))) & mask;
-}
-
 /* Check, that the gateway is already configured.
    Used only by redirect accept routine.
  */

      reply	other threads:[~2008-01-14  2:11 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-01-12 18:58 [PATCH] fib_semantics: prevent long hash chains in access server config Benjamin LaHaise
2008-01-13  5:38 ` David Miller
2008-01-13 17:58   ` Benjamin LaHaise
2008-01-14  2:11     ` David Miller [this message]

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=20080113.181118.08046749.davem@davemloft.net \
    --to=davem@davemloft.net \
    --cc=bcrl@kvack.org \
    --cc=netdev@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;
as well as URLs for NNTP newsgroup(s).