xfs
[Top] [All Lists]

Re: Review: make growing by >2TB work

To: David Chinner <dgc@xxxxxxx>
Subject: Re: Review: make growing by >2TB work
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Mon, 8 Jan 2007 09:12:18 +0000
Cc: xfs-dev@xxxxxxx, xfs@xxxxxxxxxxx
In-reply-to: <20070108044414.GC44411608@melbourne.sgi.com>
References: <20070108044414.GC44411608@melbourne.sgi.com>
Sender: xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.4.2.2i
On Mon, Jan 08, 2007 at 03:44:14PM +1100, David Chinner wrote:
> Growing a filesystem by > 2TB currently causes an overflow
> in the transaction subsystem. Make transaction deltas and associated
> elements explicitly 64 bit types so that we don't get overflows.
> 
> Comments?

Looks good.

>  
> -     AIL_LOCKINIT(&mp->m_ail_lock, "xfs_ail");
>       spinlock_init(&mp->m_sb_lock, "xfs_sb");
>       mutex_init(&mp->m_ilock);
>       initnsema(&mp->m_growlock, 1, "xfs_grow");
> -     /*
> -      * Initialize the AIL.
> -      */
> -     xfs_trans_ail_init(mp);

This seems unrelated (?)

> -xfs_mod_incore_sb(xfs_mount_t *mp, xfs_sb_field_t field, int delta, int rsvd)
> +xfs_mod_incore_sb(xfs_mount_t *mp, xfs_sb_field_t field, int64_t delta, int 
> rsvd)

This seems to be over 80 chars linelength with your patch, just break
the line.


<Prev in Thread] Current Thread [Next in Thread>