Re: [v6,1/2] pid: Replace pid bitmap implementation with IDR API

[Gargi Sharma]

On Thu, Oct 19, 2017 at 5:18 PM, Oleg Nesterov <oleg@xxxxxxxxxx> wrote:
> On 10/19, Andrei Vagin wrote:
>>
>> Hi Gargi,
>>
>> This patch breaks CRIU, because it changes a meaning of ns_last_pid.
>
> ...
>
>> > @@ -311,7 +297,7 @@ static int pid_ns_ctl_handler(struct ctl_table *table, int write,
>> >      * it should synchronize its usage with external means.
>> >      */
>> >
>> > -   tmp.data = &pid_ns->last_pid;
>> > +   tmp.data = &pid_ns->idr.idr_next;
>
> Ah, yes, off-by-one error...
>
> Gargi, I don't think you need to make another version, I'd suggest you to send
> the trivial fix to Andrew, afaics you just need to replace these 2 lines with
>
>         unsigned int last;
>         int err;
>
>         tmp.data = &last;
>         err = proc_dointvec_minmax(&tmp, write, buffer, lenp, ppos);
>         if (!err)
>                 idr_set_cursor(&pid_ns->idr, last + 1);
>         return err;
I'm not sure entirely understand how this takes care of rolling over of PIDs?
Can we ignore that? If yes, won't the tests for CRIU still break?

Thanks,
Gargi
>
> Oleg.
>