Re: [RFC] Race condition?

[Keith Owens <kaos@ocs.com.au>]

On Fri, 02 Aug 2002 10:00:13 -0700, 
Dave Hansen <haveblue@us.ibm.com> wrote:
>Kasper Dupont wrote:
>> Is there a race condition in this piece of code from do_fork in
>> linux/kernel/fork.c? I cannot see what prevents two processes
>> from calling this at the same time and both successfully fork
>> even though the user had only one process left.
>> 
>>         if (atomic_read(&p->user->processes) >= p->rlim[RLIMIT_NPROC].rlim_cur
>>                       && !capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RESOURCE))
>>                 goto bad_fork_free;
>> 
>>         atomic_inc(&p->user->__count);
>>         atomic_inc(&p->user->processes);
>
>I don't see any locking in the call chain leading to this function, so 
>I think you're right.  The attached patch fixes this.  It costs an 
>extra 2 atomic ops in the failure case, but otherwise just makes the 
>processes++ operation earlier.

Does this race really justify extra locks?  AFAICT the worst case is
that a user can go slightly over their RLIMIT_NPROC, and that will only
occur if they fork on multiple cpus "at the same time".  Given the
timing constraints on that small window, I would be surprised if this
race could be exploited to gain more than a couple of extra processes.
This looks like a case where close enough is good enough.