linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jeff Layton <jlayton-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
Cc: linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org,
	linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH 01/11] sunrpc: allocate pipefs inodes using kmalloc
Date: Sun, 1 Dec 2013 09:26:05 -0500	[thread overview]
Message-ID: <20131201092605.0a8cd907@tlielax.poochiereds.net> (raw)
In-Reply-To: <20131201132009.815675002-PfSpb0PWhxZc2C7mugBRk2EX/6BAtgUQ@public.gmane.org>

On Sun, 01 Dec 2013 05:14:42 -0800
Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org> wrote:

> There's not a lot of them, and not needing our own slab makes
> initialization ordering a whole lot simpler.
> 
> Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
> ---
>  net/sunrpc/rpc_pipe.c |   39 +++++++--------------------------------
>  1 file changed, 7 insertions(+), 32 deletions(-)
> 
> diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
> index bf04b30..395eb5f 100644
> --- a/net/sunrpc/rpc_pipe.c
> +++ b/net/sunrpc/rpc_pipe.c
> @@ -40,8 +40,6 @@
>  static struct file_system_type rpc_pipe_fs_type;
>  
>  
> -static struct kmem_cache *rpc_inode_cachep __read_mostly;
> -
>  #define RPC_UPCALL_TIMEOUT (30*HZ)
>  
>  static BLOCKING_NOTIFIER_HEAD(rpc_pipefs_notifier_list);
> @@ -193,24 +191,21 @@ rpc_close_pipes(struct inode *inode)
>  static struct inode *
>  rpc_alloc_inode(struct super_block *sb)
>  {
> -	struct rpc_inode *rpci;
> -	rpci = (struct rpc_inode *)kmem_cache_alloc(rpc_inode_cachep, GFP_KERNEL);
> +	struct rpc_inode *rpci = kmalloc(sizeof(struct rpc_inode), GFP_KERNEL);
>  	if (!rpci)
>  		return NULL;
> -	return &rpci->vfs_inode;
> -}
>  
> -static void
> -rpc_i_callback(struct rcu_head *head)
> -{
> -	struct inode *inode = container_of(head, struct inode, i_rcu);
> -	kmem_cache_free(rpc_inode_cachep, RPC_I(inode));
> +	inode_init_once(&rpci->vfs_inode);
> +	rpci->private = NULL;
> +	rpci->pipe = NULL;
> +	init_waitqueue_head(&rpci->waitq);
> +	return &rpci->vfs_inode;
>  }
>  
>  static void
>  rpc_destroy_inode(struct inode *inode)
>  {
> -	call_rcu(&inode->i_rcu, rpc_i_callback);
> +	kfree_rcu(inode, i_rcu);
>  }
>  
>  static int
> @@ -1327,28 +1322,10 @@ static struct file_system_type rpc_pipe_fs_type = {
>  MODULE_ALIAS_FS("rpc_pipefs");
>  MODULE_ALIAS("rpc_pipefs");
>  
> -static void
> -init_once(void *foo)
> -{
> -	struct rpc_inode *rpci = (struct rpc_inode *) foo;
> -
> -	inode_init_once(&rpci->vfs_inode);
> -	rpci->private = NULL;
> -	rpci->pipe = NULL;
> -	init_waitqueue_head(&rpci->waitq);
> -}
> -
>  int register_rpc_pipefs(void)
>  {
>  	int err;
>  
> -	rpc_inode_cachep = kmem_cache_create("rpc_inode_cache",
> -				sizeof(struct rpc_inode),
> -				0, (SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT|
> -						SLAB_MEM_SPREAD),
> -				init_once);
> -	if (!rpc_inode_cachep)
> -		return -ENOMEM;
>  	err = rpc_clients_notifier_register();
>  	if (err)
>  		goto err_notifier;
> @@ -1360,13 +1337,11 @@ int register_rpc_pipefs(void)
>  err_register:
>  	rpc_clients_notifier_unregister();
>  err_notifier:
> -	kmem_cache_destroy(rpc_inode_cachep);
>  	return err;
>  }
>  
>  void unregister_rpc_pipefs(void)
>  {
>  	rpc_clients_notifier_unregister();
> -	kmem_cache_destroy(rpc_inode_cachep);
>  	unregister_filesystem(&rpc_pipe_fs_type);
>  }

Makes sense...

Acked-by: Jeff Layton <jlayton-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

  parent reply	other threads:[~2013-12-01 14:26 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-12-01 13:14 [PATCH 00/11] [RFC] repair net namespace damage to rpc_pipefs Christoph Hellwig
2013-12-01 13:14 ` [PATCH 01/11] sunrpc: allocate pipefs inodes using kmalloc Christoph Hellwig
     [not found]   ` <20131201132009.815675002-PfSpb0PWhxZc2C7mugBRk2EX/6BAtgUQ@public.gmane.org>
2013-12-01 14:26     ` Jeff Layton [this message]
2013-12-01 13:14 ` [PATCH 02/11] rpc_pipefs: always mount on net namespace initialization Christoph Hellwig
2013-12-01 15:24   ` Jeff Layton
2013-12-01 13:14 ` [PATCH 03/11] sunrpc: remove the rpc_clients_block notifier Christoph Hellwig
2013-12-01 15:25   ` Jeff Layton
2013-12-01 13:14 ` [PATCH 04/11] sunrpc: no need to have a lock or superblock for rpc_unlink Christoph Hellwig
2013-12-01 13:14 ` [PATCH 05/11] sunprc: add sensible pipe creation and removal helpers Christoph Hellwig
2013-12-01 13:14 ` [PATCH 06/11] sunrpc: clean up rpc_pipefs client dir creation Christoph Hellwig
2013-12-01 13:14 ` [PATCH 07/11] sunrpc: make rpc_mkdir_populate net-namespace aware Christoph Hellwig
2013-12-01 13:14 ` [PATCH 08/11] sunrpc: rpc_get_sb_net, die, die, die Christoph Hellwig
2013-12-01 13:14 ` [PATCH 09/11] nfs: convert idmapper to rpc_mkpipe_clnt Christoph Hellwig
2013-12-01 13:14 ` [PATCH 10/11] auth_gss: convert " Christoph Hellwig
2013-12-01 13:14 ` [PATCH 11/11] sunrpc: rpc_pipefs cleanup Christoph Hellwig
     [not found] ` <20131201131441.790963326-PfSpb0PWhxZc2C7mugBRk2EX/6BAtgUQ@public.gmane.org>
2013-12-01 14:36   ` [PATCH 00/11] [RFC] repair net namespace damage to rpc_pipefs Jeff Layton
2013-12-01 15:44   ` Jeff Layton
2013-12-01 15:57     ` Jeff Layton
2013-12-01 18:13   ` Al Viro
     [not found]     ` <20131201181329.GC10323-3bDd1+5oDREiFSDQTTA3OLVCufUGDwFn@public.gmane.org>
2013-12-02  8:12       ` Christoph Hellwig
2013-12-02 13:44         ` Trond Myklebust
2013-12-02 14:24           ` Stanislav Kinsbursky
     [not found]             ` <529C982B.2030906-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-12-02 15:36               ` Christoph Hellwig
2013-12-02 15:58               ` Trond Myklebust
     [not found]                 ` <5872AE74-DF03-42E2-A21B-F8E35634CD1D-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2013-12-03  7:24                   ` Stanislav Kinsbursky
     [not found]           ` <3C65EB4C-6592-44F8-B08D-E5A9EFD6C8C6-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2013-12-02 15:34             ` Christoph Hellwig
     [not found]               ` <20131202153435.GA2804-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2013-12-02 16:00                 ` Trond Myklebust
     [not found]                   ` <BCED1826-0743-4420-AD45-B38D67EF3D99-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2013-12-02 16:27                     ` Christoph Hellwig
     [not found]                       ` <20131202162716.GA17417-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2013-12-02 16:46                         ` Trond Myklebust
2013-12-02 16:33                   ` Jeff Layton
2013-12-02 16:37                     ` J. Bruce Fields
2013-12-02 16:45                       ` Jeff Layton
2013-12-02 15:57             ` Al Viro
2013-12-02 16:04               ` Trond Myklebust
2013-12-03  2:11       ` [RFC] alloc_pid() breakage Al Viro
2013-12-02  7:23   ` [PATCH 00/11] [RFC] repair net namespace damage to rpc_pipefs Stanislav Kinsbursky

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=20131201092605.0a8cd907@tlielax.poochiereds.net \
    --to=jlayton-h+wxahxf7alqt0dzr+alfa@public.gmane.org \
    --cc=hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org \
    --cc=linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.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).