All of lore.kernel.org
 help / color / mirror / Atom feed
From: "J. Bruce Fields" <bfields@fieldses.org>
To: Jeff Layton <jlayton@redhat.com>
Cc: linux-nfs@vger.kernel.org
Subject: Re: [PATCH v1 01/16] nfsd: fix IPv6 address handling in the DRC
Date: Thu, 31 Jan 2013 15:59:01 -0500	[thread overview]
Message-ID: <20130131205901.GC24252@fieldses.org> (raw)
In-Reply-To: <1359402082-29195-2-git-send-email-jlayton@redhat.com>

On Mon, Jan 28, 2013 at 02:41:07PM -0500, Jeff Layton wrote:
> Currently, it only stores the first 16 bytes of any address. struct
> sockaddr_in6 is 28 bytes however, so we're currently ignoring the last
> 12 bytes of the address.
> 
> Expand the c_addr field to a sockaddr_in6, and cast it to a sockaddr_in
> as necessary. Also fix the comparitor to use the existing RPC
> helpers for this.

Applying.  (Assuming the fix for link-local case will be separate.)

--b.

> 
> Signed-off-by: Jeff Layton <jlayton@redhat.com>
> ---
>  fs/nfsd/cache.h             | 6 +++++-
>  fs/nfsd/nfscache.c          | 7 +++++--
>  include/linux/sunrpc/clnt.h | 4 +++-
>  3 files changed, 13 insertions(+), 4 deletions(-)
> 
> diff --git a/fs/nfsd/cache.h b/fs/nfsd/cache.h
> index 93cc9d3..2cac76c 100644
> --- a/fs/nfsd/cache.h
> +++ b/fs/nfsd/cache.h
> @@ -12,6 +12,10 @@
>  
>  /*
>   * Representation of a reply cache entry.
> + *
> + * Note that we use a sockaddr_in6 to hold the address instead of the more
> + * typical sockaddr_storage. This is for space reasons, since sockaddr_storage
> + * is much larger than a sockaddr_in6.
>   */
>  struct svc_cacherep {
>  	struct hlist_node	c_hash;
> @@ -20,7 +24,7 @@ struct svc_cacherep {
>  	unsigned char		c_state,	/* unused, inprog, done */
>  				c_type,		/* status, buffer */
>  				c_secure : 1;	/* req came from port < 1024 */
> -	struct sockaddr_in	c_addr;
> +	struct sockaddr_in6	c_addr;
>  	__be32			c_xid;
>  	u32			c_prot;
>  	u32			c_proc;
> diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c
> index 2cbac34..5dd9ec2 100644
> --- a/fs/nfsd/nfscache.c
> +++ b/fs/nfsd/nfscache.c
> @@ -9,6 +9,7 @@
>   */
>  
>  #include <linux/slab.h>
> +#include <linux/sunrpc/clnt.h>
>  
>  #include "nfsd.h"
>  #include "cache.h"
> @@ -146,7 +147,8 @@ nfsd_cache_lookup(struct svc_rqst *rqstp)
>  		    xid == rp->c_xid && proc == rp->c_proc &&
>  		    proto == rp->c_prot && vers == rp->c_vers &&
>  		    time_before(jiffies, rp->c_timestamp + 120*HZ) &&
> -		    memcmp((char*)&rqstp->rq_addr, (char*)&rp->c_addr, sizeof(rp->c_addr))==0) {
> +		    rpc_cmp_addr(svc_addr(rqstp), (struct sockaddr *)&rp->c_addr) &&
> +		    rpc_get_port(svc_addr(rqstp)) == rpc_get_port((struct sockaddr *)&rp->c_addr)) {
>  			nfsdstats.rchits++;
>  			goto found_entry;
>  		}
> @@ -183,7 +185,8 @@ nfsd_cache_lookup(struct svc_rqst *rqstp)
>  	rp->c_state = RC_INPROG;
>  	rp->c_xid = xid;
>  	rp->c_proc = proc;
> -	memcpy(&rp->c_addr, svc_addr_in(rqstp), sizeof(rp->c_addr));
> +	rpc_copy_addr((struct sockaddr *)&rp->c_addr, svc_addr(rqstp));
> +	rpc_set_port((struct sockaddr *)&rp->c_addr, rpc_get_port(svc_addr(rqstp)));
>  	rp->c_prot = proto;
>  	rp->c_vers = vers;
>  	rp->c_timestamp = jiffies;
> diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
> index 34206b8..47354a2 100644
> --- a/include/linux/sunrpc/clnt.h
> +++ b/include/linux/sunrpc/clnt.h
> @@ -263,7 +263,9 @@ static inline bool __rpc_copy_addr6(struct sockaddr *dst,
>   * @sap1: first sockaddr
>   * @sap2: second sockaddr
>   *
> - * Just compares the family and address portion. Ignores port, scope, etc.
> + * Just compares the family and address portion. Ignores port, but
> + * compares the scope if it's a link-local address.
> + *
>   * Returns true if the addrs are equal, false if they aren't.
>   */
>  static inline bool rpc_cmp_addr(const struct sockaddr *sap1,
> -- 
> 1.7.11.7
> 

  parent reply	other threads:[~2013-01-31 20:59 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-01-28 19:41 [PATCH v1 00/16] nfsd: duplicate reply cache overhaul Jeff Layton
2013-01-28 19:41 ` [PATCH v1 01/16] nfsd: fix IPv6 address handling in the DRC Jeff Layton
2013-01-28 19:51   ` Chuck Lever
2013-01-28 20:05     ` Jeff Layton
2013-01-28 20:16       ` Chuck Lever
2013-01-28 20:26         ` Jeff Layton
2013-01-28 20:34           ` Chuck Lever
2013-01-28 20:45             ` Jeff Layton
2013-01-28 20:54               ` Chuck Lever
2013-01-28 21:06                 ` Jeff Layton
2013-01-28 21:32                   ` Chuck Lever
2013-01-29 13:33                     ` Jeff Layton
2013-01-31 20:59   ` J. Bruce Fields [this message]
2013-01-31 21:06     ` Jeff Layton
2013-01-28 19:41 ` [PATCH v1 02/16] nfsd: remove unneeded spinlock in nfsd_cache_update Jeff Layton
2013-01-31 21:23   ` J. Bruce Fields
2013-01-28 19:41 ` [PATCH v1 03/16] nfsd: get rid of RC_INTR Jeff Layton
2013-01-28 19:41 ` [PATCH v1 04/16] nfsd: create a dedicated slabcache for DRC entries Jeff Layton
2013-01-28 19:41 ` [PATCH v1 05/16] nfsd: add alloc and free functions " Jeff Layton
2013-01-28 19:41 ` [PATCH v1 06/16] nfsd: remove redundant test from nfsd_reply_cache_free Jeff Layton
2013-01-28 19:41 ` [PATCH v1 07/16] nfsd: clean up and clarify the cache expiration code Jeff Layton
2013-01-28 19:41 ` [PATCH v1 08/16] nfsd: break out hashtable search into separate function Jeff Layton
2013-02-01 13:14   ` J. Bruce Fields
2013-02-01 15:26     ` Jeff Layton
2013-02-01 15:48       ` Jeff Layton
2013-02-01 15:51       ` J. Bruce Fields
2013-02-01 16:01         ` Jeff Layton
2013-02-01 16:05           ` J. Bruce Fields
2013-01-28 19:41 ` [PATCH v1 09/16] nfsd: always move DRC entries to the end of LRU list when updating timestamp Jeff Layton
2013-01-28 19:41 ` [PATCH v1 10/16] nfsd: dynamically allocate DRC entries Jeff Layton
2013-01-30 15:16   ` J. Bruce Fields
2013-01-30 15:20     ` Jeff Layton
2013-01-28 19:41 ` [PATCH v1 11/16] nfsd: remove the cache_disabled flag Jeff Layton
2013-01-28 19:41 ` [PATCH v1 12/16] nfsd: when updating an entry with RC_NOCACHE, just free it Jeff Layton
2013-01-28 19:41 ` [PATCH v1 13/16] nfsd: add recurring workqueue job to clean the cache Jeff Layton
2013-01-28 19:41 ` [PATCH v1 14/16] nfsd: track the number of DRC entries in " Jeff Layton
2013-01-28 19:41 ` [PATCH v1 15/16] nfsd: register a shrinker for DRC cache entries Jeff Layton
2013-01-30 15:24   ` J. Bruce Fields
2013-01-30 15:37     ` Jeff Layton
2013-01-28 19:41 ` [PATCH v1 16/16] nfsd: keep a checksum of the first 256 bytes of request Jeff Layton
2013-01-29 18:20   ` Jeff Layton
2013-01-30 15:14   ` J. Bruce Fields

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=20130131205901.GC24252@fieldses.org \
    --to=bfields@fieldses.org \
    --cc=jlayton@redhat.com \
    --cc=linux-nfs@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.