[Cluster-devel] [PATCH 1/2] gfs2: Update file times after grabbing glock

[Cluster-devel] [PATCH 1/2] gfs2: Update file times after grabbing glock

[Andreas Gruenbacher]

In gfs2_page_mkwrite, grab the inode glock in EX mode before calling
file_update_time: grabbing the lock may result in a call to
gfs2_dinode_in, which will reset the file times to their on-disk state.

Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
---
 fs/gfs2/file.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 320e65e..360188f 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -395,9 +395,6 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 	sb_start_pagefault(inode->i_sb);
 
-	/* Update file times before taking page lock */
-	file_update_time(vma->vm_file);
-
 	ret = gfs2_rsqa_alloc(ip);
 	if (ret)
 		goto out;
@@ -409,6 +406,9 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 	if (ret)
 		goto out_uninit;
 
+	/* Update file times before taking page lock */
+	file_update_time(vma->vm_file);
+
 	set_bit(GLF_DIRTY, &ip->i_gl->gl_flags);
 	set_bit(GIF_SW_PAGED, &ip->i_flags);
 
-- 
2.7.4

[Cluster-devel] [PATCH 2/2] gfs2: Initialize atime of I_NEW inodes

[Andreas Gruenbacher]

Fix for commit 719ee344: initialize atime of I_NEW inodes to 0 so that
the timestamps read from disk will always be more recent than the
initial timestamp, and the atime in the I_NEW inode will be set correctly.

Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
---
 fs/gfs2/inode.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index e4da0ec..fb3a810 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -187,6 +187,10 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type,
 		}
 
 		gfs2_set_iop(inode);
+
+		inode->i_atime.tv_sec = 0;
+		inode->i_atime.tv_nsec = 0;
+
 		unlock_new_inode(inode);
 	}
 
-- 
2.7.4

[Cluster-devel] [PATCH 2/2] gfs2: Initialize atime of I_NEW inodes

[Bob Peterson]

----- Original Message -----
| Fix for commit 719ee344: initialize atime of I_NEW inodes to 0 so that
| the timestamps read from disk will always be more recent than the
| initial timestamp, and the atime in the I_NEW inode will be set correctly.
| 
| Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
| ---

Hi,

Thanks. This is now applied to the for-next branch of the linux-gfs2 tree:
https://git.kernel.org/cgit/linux/kernel/git/gfs2/linux-gfs2.git/commit/fs/gfs2?h=for-next&id=332f51d7db13ffb7fcbe2407ed5b3667bc3750a7

Regards,

Bob Peterson
Red Hat File Systems

[Cluster-devel] [PATCH 1/2] gfs2: Update file times after grabbing glock

[Bob Peterson]

----- Original Message -----
| In gfs2_page_mkwrite, grab the inode glock in EX mode before calling
| file_update_time: grabbing the lock may result in a call to
| gfs2_dinode_in, which will reset the file times to their on-disk state.
| 
| Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
| ---
Hi,

Thanks. This is now applied to the for-next branch of the linux-gfs2 tree:
https://git.kernel.org/cgit/linux/kernel/git/gfs2/linux-gfs2.git/commit/fs/gfs2?h=for-next&id=d7c436cd605eb203ecf64c40f9b18aa610ac0248

Regards,

Bob Peterson
Red Hat File Systems