From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1753763Ab0LHXfn (ORCPT <rfc822;w@1wt.eu>);
	Wed, 8 Dec 2010 18:35:43 -0500
Received: from mga09.intel.com ([134.134.136.24]:64003 "EHLO mga09.intel.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1751872Ab0LHXfl (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Wed, 8 Dec 2010 18:35:41 -0500
X-ExtLoop1: 1
X-IronPort-AV: E=Sophos;i="4.59,318,1288594800"; 
   d="scan'208";a="581922840"
From: "Luck, Tony" <tony.luck@intel.com>
To: linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org
Cc: tglx@linutronix.de, mingo@elte.hu, greg@kroah.com,
        akpm@linux-foundation.org, ying.huang@intel.com,
        "Borislav Petkov" <bp@alien8.de>, "David Miller" <davem@davemloft.net>,
        "Alan Cox" <alan@lxorguk.ukuu.org.uk>,
        "Jim Keniston" <jkenisto@linux.vnet.ibm.com>,
        "Kyungmin Park" <kmpark@infradead.org>,
        "Geert Uytterhoeven" <geert@linux-m68k.org>
In-Reply-to: <AANLkTin7xDWO8fU2WA7cxC55TNi2+kAb50ajCySCfg9G@mail.gmail.com>
Subject: Re: [RFC] persistent store (version 3) (part 1 of 2)
Date: Wed, 08 Dec 2010 15:35:40 -0800
Message-Id: <4d00164c296166b682@agluck-desktop.sc.intel.com>
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

> > what happens if it IS_ERR?
> 
> Good point - I need to clean up if things fail (and akpm would
> like to see me re-factor so that there is only one "return"
> statement for better maintainabiliy).  I'll fix up stuff here.

Here's what I'll have in the next version. With all the error
handling the pstore_mkfile() routine was getting a bit big, so
I split out the writing-to-the-file part into another function.
The pair of functions now look like this:

/*
 * Set up a file structure as if we had opened this file and
 * write our data to it.
 */
static int pstore_writefile(struct inode *inode, struct dentry *dentry,
	char *data, size_t size)
{
	struct file f;
	ssize_t n;
	mm_segment_t old_fs = get_fs();

	memset(&f, '0', sizeof f);
	f.f_mapping = inode->i_mapping;
	f.f_path.dentry = dentry;
	f.f_path.mnt = pstore_mnt;
	f.f_pos = 0;
	f.f_op = inode->i_fop;
	set_fs(KERNEL_DS);
	n = do_sync_write(&f, data, size, &f.f_pos);
	set_fs(old_fs);

	return n == size;
}

/*
 * Make a regular file in the root directory of our file system.
 * Load it up with "size" bytes of data from "buf".
 * Set the mtime & ctime to the date that this record was originally stored.
 */
int pstore_mkfile(char *name, char *data, size_t size, struct timespec time,
		  void *private)
{
	struct dentry	*root = pstore_sb->s_root;
	struct dentry	*dentry;
	struct inode	*inode;
	int		rc = 0;

	inode = pstore_get_inode(pstore_sb, root->d_inode, S_IFREG | 0444, 0);
	if (!inode) {
		rc = -ENOMEM;
		goto fail;
	}
	inode->i_private = private;

	mutex_lock(&root->d_inode->i_mutex);

	dentry = d_alloc_name(root, name);
	if (IS_ERR(dentry)) {
		mutex_unlock(&root->d_inode->i_mutex);
		rc = -ENOSPC;
		goto fail1;
	}

	d_add(dentry, inode);

	mutex_unlock(&root->d_inode->i_mutex);

	if (!pstore_writefile(inode, dentry, data, size)) {
		inode->i_nlink--;
		mutex_lock(&root->d_inode->i_mutex);
		d_delete(dentry);
		dput(dentry);
		mutex_unlock(&root->d_inode->i_mutex);
		rc = -ENOSPC;
		goto fail;
	}

	if (time.tv_sec)
		inode->i_mtime = inode->i_ctime = time;

	return 0;

fail1:
	iput(inode);
fail:
	return rc;
}