Re: [PATCH] mcheck: Correction to a missing return value from misc_register

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

On Mon, 28 Feb 2011 12:19:04 +0100
Johan Wessfeldt <johan.wessfeldt@gmail.com> wrote:

> Return value from misc_register was not handled correctly
> in mcheck_init_device()
> 
> Signed-off-by: Johan Wessfeldt <johan.wessfeldt@gmail.com>
> ---
>  arch/x86/kernel/cpu/mcheck/mce.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
> index d916183..7274a11 100644
> --- a/arch/x86/kernel/cpu/mcheck/mce.c
> +++ b/arch/x86/kernel/cpu/mcheck/mce.c
> @@ -2140,7 +2140,7 @@ static __init int mcheck_init_device(void)
>  	}
>  
>  	register_hotcpu_notifier(&mce_cpu_notifier);
> -	misc_register(&mce_log_device);
> +	err = misc_register(&mce_log_device);
>  
>  	return err;
>  }

It should be fixed, but not that way.  If misc_register() fails here,
mcheck_init_device() should undo all the things which it just did.

And the memory leak which occurs if sysdev_class_register() fails and
the memory and other resource leaks which occur if mce_create_device()
fails should be fixed too.

mcheck_init_device() a rather optimistic function.