Re: [patch] brd: fix ramdisk regression

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

On Tue, 20 May 2008 15:41:25 +0200
Nick Piggin <npiggin@suse.de> wrote:

> From: Marcin Krol <hawk@pld-linux.org>
> 
> In 2.6.25, ramdisk devices show up in /proc/partitions, which is a
> behaviour change from the old rd.c. Add GENHD_FL_SUPPRESS_PARTITION_INFO,
> which was present in rd.c.
> 
> Signed-off-by: Marcin Krol <hawk@pld-linux.org>

I added your signed-off-by: to this patch.

I renamed it to "brd: don't show ramdisks in /proc/partitions"

> --
> --- linux-2.6.25/drivers/block/brd.c.orig	2008-04-17 04:49:44.000000000 +0200
> +++ linux-2.6.25/drivers/block/brd.c	2008-05-18 01:18:28.381903343 +0200
> @@ -442,6 +442,7 @@
>  	disk->fops		= &brd_fops;
>  	disk->private_data	= brd;
>  	disk->queue		= brd->brd_queue;
> +	disk->flags |= GENHD_FL_SUPPRESS_PARTITION_INFO;
>  	sprintf(disk->disk_name, "ram%d", i);
>  	set_capacity(disk, rd_size * 2);
>  

Why is it a "regression"?

The change in 2.6.25 was a back-compatible one.

This change is not a back-compatible one and if we're going to now
withdraw the newly-added 2.6.25 feature then we should also withdraw it
from 2.6.26.x and 2.6.25.x (if that is still under maintenance).  To
reduce the incidence of "hey where did my feature go" problems.

Really, life would be simpler if we just left the accidentally-added
feature in place.  What problems does it cause?