On Tue, Jul 01, 2003 at 12:42:26PM -0700, Sebastien Boving wrote:
>
> *,
>
> ftpd writing to an xfs filesystem (compiled from 2.4.20-18.9XFS1.3.0pre2
> for RH9 src rpm from oss.sgi.com ftp site, but without kdb) crashed,
> leaving the following in the logs:
>
>
> Unable to handle kernel NULL pointer dereference at virtual address 00000008
> printing eip:
> f8afcbc4
> *pde = 00000000
> Oops: 0000
> nfsd iptable_filter ip_tables autofs nfs lockd sunrpc e1000 xfs keybdev
> mousedev input hid usb-uhci usbcore ext3 jbd raid0 3w-xxxx sd_mod scsi_mod
> CPU: 2
> EIP: 0060:[<f8afcbc4>] Not tainted
> EFLAGS: 00010202
>
> EIP is at linvfs_permission [xfs] 0x14 (2.4.20-18.9XFS1.3.0pre2custom)
Looks alot like a sync/iget interaction bug we were seeing recently -
Steve fixed this a couple of weeks ago, the change was:
--- /usr/tmp/TmpDir.2068614-0/linux/fs/xfs/xfs_inode.c_1.376 Wed Jul 2
08:51:27 2003
+++ /usr/tmp/TmpDir.2068614-0/linux/fs/xfs/xfs_inode.c_1.377 Wed Jul 2
08:51:27 2003
@@ -2629,7 +2629,8 @@
if (vp) {
struct inode *inode = LINVFS_GET_IP(vp);
- mark_inode_dirty_sync(inode);
+ if (!(inode->i_state & I_NEW))
+ mark_inode_dirty_sync(inode);
}
wake_up(&ip->i_ipin_wait);
cheers.
--
Nathan
|