Re: [PATCH 1/3] timekeeping: NMI safe converter from a given time to monotonic

[Thomas Gleixner]

On Tue, Jan 24 2023 at 01:10, Stephane Eranian wrote:
> On Tue, Jan 24, 2023 at 12:52 AM Thomas Gleixner <tglx@xxxxxxxxxxxxx> wrote:
>> > It's useful to provide a NMI safe function to convert a given time to
>> > monotonic. For example, the perf_event subsystem wants to convert a TSC
>> > of a PEBS record to a monotonic clock in a NMI handler.
>>
>> Why? That's a postprocessing problem, really.
>>
> Because you want to correlate samples captured by PEBS with samples
> from applications timestamped with a user available clock such as
> CLOCK_MONOTONIC, for instance.

Sure. Postprocessing can do that if the kernel provides the relevant
conversion information. Absolutely no need for doing this in the kernel.

> When I create a perf_event event and I stipulate that
> event_attr.clockid=MONOTONIC, I expect all the samples from that event
> to be timestamped using the same clock source, regardless of PEBS or
> IBS.

Just because the kernel can do the conversion, there is no requirement
to actually do so. Instrumentation has to be as lightweight as possible
and a lot of instrumentation goes the extra mile of doing postprocessing
exactly for this reason.

So if we implement this conversion in the kernel then there needs to be
a compelling technical reason to do so.

So far the provided arguments are in the 'I want a pony' realm.

Thanks,

        tglx