linux-kernel - Re: [PATCH] cdrom: support devices that have check_events but not media

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20110131101244.GA7459@htj.dyndns.org>
Date:	Mon, 31 Jan 2011 11:12:44 +0100
From:	Tejun Heo <tj@...nel.org>
To:	Simon Arlott <simon@...e.lp0.eu>
Cc:	axboe@...nel.dk, jeff@...zik.org, linux-ide@...r.kernel.org,
	linux-kernel@...r.kernel.org, linux-scsi@...r.kernel.org,
	kay.sievers@...y.org, jack@...e.cz,
	James.Bottomley@...senPartnership.com
Subject: Re: [PATCH] cdrom: support devices that have check_events but not
 media_changed

Hello,

On Sun, Jan 30, 2011 at 01:31:16AM +0000, Simon Arlott wrote:
> 93aae17af1172c40c6f74b7294e93a90c3cfaa5d replaced the media_changed op
> with the check_events op in drivers/scsi/sr.c
> 
> All users that check for the CDC_MEDIA_CHANGED capability try both
> the check_events op and the media_changed op, but register_cdrom()
> was requiring media_changed.
> 
> This patch fixes the capability checking and removes a redundant
> check that media_changed != NULL.
> 
> The cdrom_select_disc ioctl is also using the two operations, so
> they should be required for CDC_SELECT_DISC too.

First of all, thanks a lot for fixing this.  Just minor nits below.

> diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c
> index 14033a3..4716d76 100644
> --- a/drivers/cdrom/cdrom.c
> +++ b/drivers/cdrom/cdrom.c
> @@ -409,7 +409,9 @@ int register_cdrom(struct cdrom_device_info *cdi)
>  	}
>  
>  	ENSURE(drive_status, CDC_DRIVE_STATUS );
> -	ENSURE(media_changed, CDC_MEDIA_CHANGED);
> +	if (cdo->check_events == NULL && cdo->media_changed == NULL) {
> +		*change_capability = ~(CDC_MEDIA_CHANGED | CDC_SELECT_DISC);
> +	}

Can we lose the unnecessary {}?

>  	ENSURE(tray_move, CDC_CLOSE_TRAY | CDC_OPEN_TRAY);
>  	ENSURE(lock_door, CDC_LOCK);
>  	ENSURE(select_speed, CDC_SELECT_SPEED);
> @@ -1471,7 +1473,7 @@ int cdrom_media_changed(struct cdrom_device_info *cdi)
>  	/* This talks to the VFS, which doesn't like errors - just 1 or 0.  
>  	 * Returning "0" is always safe (media hasn't been changed). Do that 
>  	 * if the low-level cdrom driver dosn't support media changed. */ 
> -	if (cdi == NULL || cdi->ops->media_changed == NULL)
> +	if (cdi == NULL)

This change is probably not necessary as cdrom_media_changed() is only
called from drivers which implement media_changed.  I don't think it's
gonna break anything but the whole media_changed stuff is scheduled
for removal anyway, so I think it would be better to keep the changes
minimal.

Thank you.

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