Re: [PATCH 2/2] mm: optimize kill_bdev()

[Andrew Morton <akpm@linux-foundation.org>]

On Thu, 12 Apr 2007 17:21:22 +0200 Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:

> 
> Remove duplicate work in kill_bdev().
> 
> It currently invalidates and then truncates the bdev's mapping.
> invalidate_mapping_pages() will opportunistically remove pages from the
> mapping. And truncate_inode_pages() will forcefully remove all pages.
> 
> The only thing truncate doesn't do is flush the bh lrus. So do that explicitly.
> This avoids (very unlikely) but possible invalid lookup results if the
> same bdev is quickyl re-issued.
> 
> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> ---
>  fs/block_dev.c              |    2 +-
>  fs/buffer.c                 |    3 +--
>  include/linux/buffer_head.h |    1 +
>  3 files changed, 3 insertions(+), 3 deletions(-)
> 
> Index: linux-2.6/fs/block_dev.c
> ===================================================================
> --- linux-2.6.orig/fs/block_dev.c	2007-04-12 16:01:13.000000000 +0200
> +++ linux-2.6/fs/block_dev.c	2007-04-12 16:20:14.000000000 +0200
> @@ -61,7 +61,7 @@ static sector_t max_block(struct block_d
>  /* Kill _all_ buffers, dirty or not.. */
>  static void kill_bdev(struct block_device *bdev)
>  {
> -	invalidate_bdev(bdev);
> +	invalidate_bh_lrus();
>  	truncate_inode_pages(bdev->bd_inode->i_mapping, 0);
>  }	

Fair enough, thanks.

The check for mapping->nr_pages != 0 was added to invalidate_bdev() to
avoid unpleasant IPI-induced stalls when large ia64 machines poll their
CDROM drives for media.  I don't know if kill_bdev() gets called on the
probe-a-cdrom path, but we might as well put

	 if (bdev->bd_inode->i_mapping->nr_pages == 0)
		return;

into kill_bdev().  I'll make that change.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/