Re: [PATCH] waitid system call

[Bill Davidsen <davidsen@tmr.com>]

Andrew Morton wrote:
> Jakub Jelinek <jakub@redhat.com> wrote:
> 
>>On Mon, Aug 30, 2004 at 11:04:46PM -0700, Roland McGrath wrote:
>> > +			/*
>> > +			 * For a WNOHANG return, clear out all the fields
>> > +			 * we would set so the user can easily tell the
>> > +			 * difference.
>> > +			 */
>> > +			if (!retval)
>> > +				retval = put_user(0, &infop->si_signo);
>> > +			if (!retval)
>> > +				retval = put_user(0, &infop->si_errno);
>> > +			if (!retval)
>> > +				retval = put_user(0, &infop->si_code);
>> > +			if (!retval)
>> > +				retval = put_user(0, &infop->si_pid);
>> > +			if (!retval)
>> > +				retval = put_user(0, &infop->si_uid);
>> > +			if (!retval)
>> > +				retval = put_user(0, &infop->si_status);
>>
>> Is it really necessary to check the exit code after each put_user?
>> 	if (!retval && access_ok(VERIFY_WRITE, infop, sizeof(*infop)))) {
>> 		retval = __put_user(0, &infop->si_signo);
>> 		retval |= __put_user(0, &infop->si_errno);
>> 		retval |= __put_user(0, &infop->si_code);
>> 		retval |= __put_user(0, &infop->si_pid);
>> 		retval |= __put_user(0, &infop->si_uid);
>> 		retval |= __put_user(0, &infop->si_status);
>> 	}
>> is what kernel usually does when filling multiple structure members.
> 
> 
> I don't think it matters much.  Taking seven trips into the fault handler
> where one would do seems a bit dumb though.

If all you need is good/bad you could just separate the put_user calls 
with || and only go through the error handler once.

-- 
    -bill davidsen (davidsen@tmr.com)
"The secret to procrastination is to put things off until the
  last possible moment - but no longer"  -me