cluster-devel.redhat.com archive mirror
 help / color / mirror / Atom feed
From: rpeterso@sourceware.org <rpeterso@sourceware.org>
To: cluster-devel.redhat.com
Subject: [Cluster-devel] cluster/gfs-kernel/src/gfs incore.h inode.c op ...
Date: 10 Oct 2006 18:35:28 -0000	[thread overview]
Message-ID: <20061010183528.27654.qmail@sourceware.org> (raw)

CVSROOT:	/cvs/cluster
Module name:	cluster
Branch: 	RHEL4
Changes by:	rpeterso at sourceware.org	2006-10-10 18:35:28

Modified files:
	gfs-kernel/src/gfs: incore.h inode.c ops_address.c ops_inode.c 

Log message:
	Fix for bugzilla bug 164499: Unable to mount loopback images
	from mounted GFS partition.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/gfs-kernel/src/gfs/incore.h.diff?cvsroot=cluster&only_with_tag=RHEL4&r1=1.17.2.6&r2=1.17.2.7
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/gfs-kernel/src/gfs/inode.c.diff?cvsroot=cluster&only_with_tag=RHEL4&r1=1.20.2.2&r2=1.20.2.3
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/gfs-kernel/src/gfs/ops_address.c.diff?cvsroot=cluster&only_with_tag=RHEL4&r1=1.5.2.5&r2=1.5.2.6
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/gfs-kernel/src/gfs/ops_inode.c.diff?cvsroot=cluster&only_with_tag=RHEL4&r1=1.6.2.3&r2=1.6.2.4

--- cluster/gfs-kernel/src/gfs/incore.h	2006/09/15 21:42:05	1.17.2.6
+++ cluster/gfs-kernel/src/gfs/incore.h	2006/10/10 18:35:27	1.17.2.7
@@ -617,6 +617,7 @@
 
 	unsigned int i_greedy; /* The amount of time to be greedy */
 	unsigned long i_last_pfault; /* The time of the last page fault */
+	struct address_space_operations gfs_file_aops;
 };
 
 /*
--- cluster/gfs-kernel/src/gfs/inode.c	2005/08/16 17:10:41	1.20.2.2
+++ cluster/gfs-kernel/src/gfs/inode.c	2006/10/10 18:35:27	1.20.2.3
@@ -191,7 +191,9 @@
 	if (ip->i_di.di_type == GFS_FILE_REG) {
 		tmp->i_op = &gfs_file_iops;
 		tmp->i_fop = &gfs_file_fops;
-		tmp->i_mapping->a_ops = &gfs_file_aops;
+		memcpy(&ip->gfs_file_aops, &gfs_file_aops,
+			   sizeof(struct address_space_operations));
+		tmp->i_mapping->a_ops = &ip->gfs_file_aops;
 	} else if (ip->i_di.di_type == GFS_FILE_DIR) {
 		tmp->i_op = &gfs_dir_iops;
 		tmp->i_fop = &gfs_dir_fops;
--- cluster/gfs-kernel/src/gfs/ops_address.c	2006/02/20 15:34:05	1.5.2.5
+++ cluster/gfs-kernel/src/gfs/ops_address.c	2006/10/10 18:35:27	1.5.2.6
@@ -32,6 +32,8 @@
 #include "quota.h"
 #include "trans.h"
 
+static int gfs_commit_write(struct file *file, struct page *page,
+							unsigned from, unsigned to);
 /**
  * get_block - Fills in a buffer head with details about a block
  * @inode: The inode
@@ -327,6 +329,13 @@
 
 	atomic_inc(&sdp->sd_ops_address);
 
+	/* We can't set commit_write in the structure in the declare         */
+	/* because if we do, loopback (loop.c) will interpret that to mean   */
+	/* it's okay to do buffered writes without locking through sendfile. */
+	/* This is a kludge to get around the problem with loop.c because    */
+	/* the upstream community rejected my changes to loop.c.             */
+	ip->gfs_file_aops.commit_write = gfs_commit_write;
+
 	if (gfs_assert_warn(sdp, gfs_glock_is_locked_by_me(ip->i_gl)))
 		return -ENOSYS;
 
@@ -395,6 +404,7 @@
 			goto fail;
 	}
 
+	ip->gfs_file_aops.commit_write = NULL;
 	return 0;
 
  fail:
@@ -473,7 +483,6 @@
 	.readpage = gfs_readpage,
 	.sync_page = block_sync_page,
 	.prepare_write = gfs_prepare_write,
-	.commit_write = gfs_commit_write,
 	.bmap = gfs_bmap,
 	.direct_IO = gfs_direct_IO,
 };
--- cluster/gfs-kernel/src/gfs/ops_inode.c	2006/04/20 22:30:14	1.6.2.3
+++ cluster/gfs-kernel/src/gfs/ops_inode.c	2006/10/10 18:35:27	1.6.2.4
@@ -952,7 +952,7 @@
 			return 0;
 	}
 
-	/*  Make sure we aren't trying to move a dirctory into it's subdir  */
+	/*  Make sure we aren't trying to move a directory into its subdir  */
 
 	if (ip->i_di.di_type == GFS_FILE_DIR && odip != ndip) {
 		dir_rename = TRUE;



             reply	other threads:[~2006-10-10 18:35 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-10-10 18:35 rpeterso [this message]
  -- strict thread matches above, loose matches on Subject: below --
2006-08-02 17:21 [Cluster-devel] cluster/gfs-kernel/src/gfs incore.h inode.c op rpeterso

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=20061010183528.27654.qmail@sourceware.org \
    --to=rpeterso@sourceware.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).