Re: [PATCH 1/2] perf: Add closing sibling events' file descriptors

[Alexander Shishkin]

Andi Kleen <ak@xxxxxxxxxxxxxxx> writes:

>> > This adds an opt-in flag to the perf_event_open() syscall to retain
>> > sibling events after their file descriptors are closed. In this case, the
>> > actual events will be closed with the group leader.
>> 
>> So having the 1:1 relation with filedesc imposes a resource limit on
>> userspace.
>> 
>> This patch breaks that and enables a user to basically DoS the system by
>> creating unbound events.
>
> The idea was to account the events in the locked memory allocation too.
> Not sure that made it into the patch though.

It didn't. I can't figure out what to charge on the locked memory, as
all that memory is in kernel-side objects. It also needs to make sense
as iirc the default MLOCK_LIMIT is quite low, you'd hit it sooner than
the file descriptor limit.

> It has a minor issue that it might break some existing setups that rely
> on the mmap fitting exactly into the mmap allocation, but that could
> be solved by allowing a little slack, since the existing setups
> likely don't have that many events.

I don't see how to make this work in a sane way. Besides, if we have to
have a limit anyway, sticking with the existing one is just easier and
1:1 is kind of more logical.

Regards,
--
Alex