Re: [PATCH] perf/core: generate overflow signal when samples are dropped (WAS: Re: [REGRESSION] perf/core: PMU interrupts dropped if we entered the kernel in the "skid" region)

[Mark Rutland]

On Tue, Jul 04, 2017 at 10:33:45AM +0100, Mark Rutland wrote:
> On Tue, Jul 04, 2017 at 11:03:13AM +0200, Peter Zijlstra wrote:
> > Faking data gets a wee bit tricky in how much data we need to clear
> > through, its not only IP, pretty much everything we get from the
> > interrupt context, like the branch stack and registers is also suspect.
> 
> Indeed. I'll take a run through __perf_event_output() and callees, and
> see what we need to drop.

Looking at perf_event_sample_format in uapi/linux/perf_event.h, there
are samples that are obviously sensitive, and should be dropped:

* PERF_SAMPLE_IP
* PERF_SAMPLE_CALLCHAIN
* PERF_SAMPLE_BRANCH_STACK
* PERF_SAMPLE_REGS_INTR

... samples that look benign:

* PERF_SAMPLE_TID
* PERF_SAMPLE_TIME
* PERF_SAMPLE_CPU
* PERF_SAMPLE_PERIOD
* PERF_SAMPLE_REGS_USER
* PERF_SAMPLE_STACK_USER
* PERF_SAMPLE_READ
* PERF_SAMPLE_ID
* PERF_SAMPLE_STREAM_ID
* PERF_SAMPLE_IDENTIFIER

.. and samples that I have no idea about:

* PERF_SAMPLE_ADDR
* PERF_SAMPLE_RAW
* PERF_SAMPLE_WEIGHT
* PERF_SAMPLE_DATA_SRC
* PERF_SAMPLE_TRANSACTION

Should any of those be moved into the "should be dropped" pile?

Thanks,
Mark.