Re: [ORLinux] [PATCH 1/1] Apply transparent_union attribute to union semun

[Jeremy Bennett <jeremy.bennett@embecosm.com>]

On Fri, 2011-05-27 at 00:27 +0200, jonas@southpole.se wrote: 
> From: Jonas Bonn <jonas@southpole.se>
> 
> The syscall handler for semctl is written under the assumption that the
> toolchain will pass "small" unions as function parameters directly instead
> of by reference.  The union semun is "small" and thus fits this description.
> 
> Since it is assumed that the union will be passed directly and not by
> reference, it is safe to access the union members without going via
> get_user.
> 
> The OpenRISC architecture, however, passes all unions by reference, thus
> breaking the above assumption.

Hi Jonas

Joern and I were discussing this - there is a case for changing the ABI.
It would certainly fix a debugging issue.

> The technically correct fix here is to mark the union as being transparent
> so that the ABI of the union's first element determines the parameter
> passing method and thus make explicit what's already implied in the function
> definition.
> 
> Signed-off-by: Jonas Bonn <jonas@southpole.se>
> ---
>  include/linux/sem.h |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/include/linux/sem.h b/include/linux/sem.h
> index f2961af..6985c7d 100644
> --- a/include/linux/sem.h
> +++ b/include/linux/sem.h
> @@ -48,7 +48,7 @@ union semun {
>  	unsigned short __user *array;	/* array for GETALL & SETALL */
>  	struct seminfo __user *__buf;	/* buffer for IPC_INFO */
>  	void __user *__pad;
> -};
> +} __attribute__ ((transparent_union));
>  
>  struct  seminfo {
>  	int semmap;

Excellent. Semaphore test now passes. I look forward to rerunning
regression - we may be down to a single failure now!

Best wishes,


Jeremy

-- 
Tel:      +44 (1590) 610184
Cell:     +44 (7970) 676050
SkypeID: jeremybennett
Email:   jeremy.bennett@embecosm.com
Web:     www.embecosm.com