Re: error in xfs_logprint?

[Nathan Scott <nathans@sgi.com>]

hi there,

On Wed, Aug 15, 2001 at 11:06:52PM +0800, Tang Lingbo wrote:
> Hi,
> 
> There was a core dump in xfs_logprint when I did as follows:
> 

I didn't see a core dump when I ran this, but I did see
the error message you describe below...

> 1)$ dd if=/dev/zero of=/tmp/xfs bs=1024 count=50000
> 2)$ mkfs.xfs /tmp/xfs
> 3)$ mount -t xfs -o loop /tmp/xfs /xfs
> 4)$ touch /xfs/test
> 5)$ for((i=0;i<1024;i++));do echo -n 1 >> /tmp/dummy;done
> 6)$ value=`cat /tmp/dummy`;attr -s test -V $value /xfs/test
> 7)$ xfs_logprint /tmp/xfs
> 
> output:
> 
> ...
> xlog_print_trans_inode: illegal inode type
> 
> And I think the patch will be as follows:
> ...[snip]...
> Any comments?
> 

This patch looks good - I'll apply it shortly if that hasn't
happened already.

Thanks!

-- 
Nathan