Re: [PATCH v4 2/9] perf/core: open access for CAP_SYS_PERFMON privileged process

[Alexey Budankov]

On 14.01.2020 21:06, Alexei Starovoitov wrote:
> On Tue, Jan 14, 2020 at 1:47 AM Alexey Budankov
> <alexey.budankov@xxxxxxxxxxxxxxx> wrote:
>>>>
>>>> As we talked at RFC series of CAP_SYS_TRACING last year, I just expected
>>>> to open it for enabling/disabling kprobes, not for creation.
>>>>
>>>> If we can accept user who has no admin priviledge but the CAP_SYS_PERFMON,
>>>> to shoot their foot by their own risk, I'm OK to allow it. (Even though,
>>>> it should check the max number of probes to be created by something like
>>>> ulimit)
>>>> I think nowadays we have fixed all such kernel crash problems on x86,
>>>> but not sure for other archs, especially on the devices I can not reach.
>>>> I need more help to stabilize it.
>>>
>>> I don't see how enable/disable is any safer than creation.
>>> If there are kernel bugs in kprobes the kernel will crash anyway.
>>> I think such partial CAP_SYS_PERFMON would be very confusing to the users.
>>> CAP_* is about delegation of root privileges to non-root.
>>> Delegating some of it is ok, but disallowing creation makes it useless
>>> for bpf tracing, so we would need to add another CAP later.
>>> Hence I suggest to do it right away instead of breaking
>>> sys_perf_even_open() access into two CAPs.
>>>
>>
>> Alexei, Masami,
>>
>> Thanks for your meaningful input.
>> If we know in advance that it still can crash the system in some cases and on
>> some archs, even though root fully controls delegation thru CAP_SYS_PERFMON,
>> such delegation looks premature until the crashes are avoided. So it looks like
>> access to eBPF for CAP_SYS_PERFMON privileged processes is the subject for
>> a separate patch set.
> 
> perf_event_open is always dangerous. sw cannot guarantee non-bugginess of hw.

Sure, software cannot guarantee, but known software bugs could still be fixed,
that's what I meant.

> imo adding a cap just for pmc is pointless.
> if you add a new cap it should cover all of sys_perf_event_open syscall.
> subdividing it into sw vs hw counters, kprobe create vs enable, etc will
> be the source of ongoing confusion. nack to such cap.
> 

Well, as this patch set already covers complete perf_event_open functionality,
and also eBPF related parts too, could you please review and comment on it?
Does the patches 2/9 and 5/9 already bring all required extentions?

Thanks,
Alexey