Re: [PATCH 1/7] perf: introduce raw_type attribute to specify the type of a raw sample

From: Stephane Eranian
Date: Thu May 20 2010 - 05:42:49 EST

On Thu, May 20, 2010 at 11:23 AM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> On Thu, 2010-05-20 at 10:10 +0200, Stephane Eranian wrote:
>> I still don't understand why you need all of this to encode IBS.
>> I still believe that with attr.config there is plenty of bits to choose
>> from. I do understand the need for PERF_SAMPLE_RAW. I think
>> there is no other way.
>> You simply need to pick an encoding to mark the config as IBS. You
>> need two bits for this: 00 regular counters, 01 IBS Fetch, 10 IBS op.
>> Regular counters use 43 bits, IBS fetch uses 58, IBS op uses 52.
>> So you could use bits 62-63 for instance. You don't need to encode
>> the sampling period in attr.config for either IBS. You can use
>> attr.sample_period, so you free up 16 bits.
>> I understand that IBS may evolve and thus may use more bits. But
>> you still have at least 16 bits of margin.
>> Users and tools would rely on an library to provide the event encoding.
>> No need to come up with some raw hex number on the cmdline.
> No need for any of that afaict, how about:
> For Instruction-Fetch:
> Â0:15 sample-period (r/w)
> Â16:31 cnt      (r/w)
> Â32:47 latency    (r/w)
>  Â48 enable    Â(r/w)
>  Â49 valid     (r/w)
> Â50:56 Â Â Â Â Â Â Â (ro)
>  Â57 randomized  Â(r/w)
Your are mixing output and input parameters.

The only input parameters you have are:
- sample-period, enable, random
The rest is output only.

So I would say: you don't need to encode anything
sample-period -> attr.sample_period
random -> attr.random_width (once we add that)

> For Instruction-Execution:
> Â0:15 sample-period (r/w)
>  Â17 enable    Â(r/w)
>  Â18 valid     (r/w)
Same thing here, sample-period is the only input parameter.
so sample-period -> attr.sample_period

Both IBFETCHCTL, IBSOP would also be included in
the PERF_SAMPLE_RAW. Because they have the valid
bits and latency.

> So if we add perf_event_attr::latency (can also be used for
> PEBS-load-latency, can Sparc/PowerPC/ARM/SH use this too?), we can

I suspect you can encode the latency with the event code used for PEBS
load-latency (PEBS-LL). It can only be used with one event.

> encode these IBS things as:
> Â0x87 Instruction Fetch Stall -- Ins-Fetch
> Â0xC0 Retired Instructions  Â-- Ins-Exec
I think those events do not map to the behavior of IBS. We have
add that discussion before.

> When we set perf_event_attr::precise > 0
Well, then you'd have to document that when you do that you use IBS and
that the event do not count the same things.

> The Ins-Exec will have to re-construct the actual event->count by adding
> sample-period on each interrupt, as it seems we lack an actual counter
> in hardware.
For what? counting mode?

> Furthermore, these counters will have to deal with sample-period > 2^16
> by 'ignoring' interrupts until we get ->period_left down to 0.
Well, it's not 2^16, it's 2^20 but bottom 4 bits must be zero.
What about simply failing perf_event_open() is sample_period does not fit the

> The extra data could possibly be exposed through attaching non-sampling
> group events and using SAMPLE_READ, like L1-misses, although
> reconstructing the count from just one bit seems 'interesting'.
> The IbsFetchLinAd/IbsOpRip would go straight into PERF_SAMPLE_IP by
> replacing pt_regs->ip I guess.
> IbsDcLinAd goes into PERF_SAMPLE_ADDR
What about the rest, the TLB, alignment, data sources?
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at