From: Peter Zijlstra <peterz@infradead.org>
To: kan.liang@intel.com
Cc: mingo@redhat.com, eranian@google.com,
linux-kernel@vger.kernel.org, alexander.shishkin@linux.intel.com,
acme@redhat.com, jolsa@redhat.com, torvalds@linux-foundation.org,
tglx@linutronix.de, vincent.weaver@maine.edu, ak@linux.intel.com
Subject: Re: [PATCH 1/2] perf/x86/intel: enable CPU ref_cycles for GP counter
Date: Mon, 22 May 2017 10:30:08 +0200 [thread overview]
Message-ID: <20170522083008.fuofuwlgq6muomjn@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <1495213582-3635-1-git-send-email-kan.liang@intel.com>
On Fri, May 19, 2017 at 10:06:21AM -0700, kan.liang@intel.com wrote:
> From: Kan Liang <Kan.liang@intel.com>
>
> The CPU ref_cycles can only be used by one user at the same time,
> otherwise a "not counted" error will be displaced.
> [kan]$ sudo perf stat -x, -e ref-cycles,ref-cycles -- sleep 1
> 1203264,,ref-cycles,513112,100.00,,,,
> <not counted>,,ref-cycles,0,0.00,,,,
>
> CPU ref_cycles can only be counted by fixed counter 2. It uses
> pseudo-encoding. The GP counter doesn't recognize.
>
> BUS_CYCLES (0x013c) is another event which is not affected by core
> frequency changes. It has a constant ratio with the CPU ref_cycles.
> BUS_CYCLES could be used as an alternative event for ref_cycles on GP
> counter.
> A hook is implemented in x86_schedule_events. If the fixed counter 2 is
> occupied and a GP counter is assigned, BUS_CYCLES is used to replace
> ref_cycles. A new flag PERF_X86_EVENT_REF_CYCLES_REP in
> hw_perf_event is introduced to indicate the replacement.
> To make the switch transparent, counting and sampling are also specially
> handled.
> - For counting, it multiplies the result with the constant ratio after
> reading it.
> - For sampling with fixed period, the BUS_CYCLES period = ref_cycles
> period / the constant ratio.
> - For sampling with fixed frequency, the adaptive frequency algorithm
> will figure it out on its own. Do nothing.
>
> The constant ratio is model specific.
> For the model after NEHALEM but before Skylake, the ratio is defined in
> MSR_PLATFORM_INFO.
> For the model after Skylake, it can be get from CPUID.15H.
> For Knights Landing, Goldmont and later, the ratio is always 1.
>
> The old Silvermont/Airmont, Core2 and Atom machines are not covered by
> the patch. The behavior on those machines will not change.
Maybe I missed it, but *why* are we doing this?
next prev parent reply other threads:[~2017-05-22 8:30 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-05-19 17:06 [PATCH 1/2] perf/x86/intel: enable CPU ref_cycles for GP counter kan.liang
2017-05-19 17:06 ` [PATCH 2/2] perf/x86/intel, watchdog: Switch NMI watchdog to ref cycles on x86 kan.liang
2017-05-22 12:03 ` Peter Zijlstra
2017-05-22 12:04 ` Peter Zijlstra
2017-05-22 16:58 ` Liang, Kan
2017-05-22 19:24 ` Peter Zijlstra
2017-05-22 18:20 ` Stephane Eranian
2017-05-22 20:01 ` Andi Kleen
2017-05-22 8:30 ` Peter Zijlstra [this message]
2017-05-22 18:15 ` [PATCH 1/2] perf/x86/intel: enable CPU ref_cycles for GP counter Stephane Eranian
2017-05-22 9:19 ` Peter Zijlstra
2017-05-22 12:22 ` Peter Zijlstra
2017-05-22 16:59 ` Liang, Kan
2017-05-22 16:55 ` Liang, Kan
2017-05-22 19:23 ` Peter Zijlstra
2017-05-22 19:28 ` Stephane Eranian
2017-05-22 21:51 ` Liang, Kan
2017-05-23 6:39 ` Peter Zijlstra
2017-05-23 6:42 ` Stephane Eranian
2017-05-24 15:45 ` Andi Kleen
2017-05-24 16:01 ` Vince Weaver
2017-05-24 16:55 ` Andi Kleen
2017-05-28 20:31 ` Stephane Eranian
2017-05-30 9:25 ` Peter Zijlstra
2017-05-30 13:51 ` Andi Kleen
2017-05-30 16:28 ` Peter Zijlstra
2017-05-30 16:41 ` Stephane Eranian
2017-05-30 17:22 ` Andi Kleen
2017-05-30 17:40 ` Peter Zijlstra
2017-05-30 17:51 ` Andi Kleen
2017-05-30 18:59 ` Peter Zijlstra
2017-05-30 19:40 ` Andi Kleen
2017-05-30 16:39 ` Stephane Eranian
2017-05-30 16:55 ` Thomas Gleixner
2017-05-30 17:25 ` Peter Zijlstra
2017-05-31 20:57 ` Vince Weaver
2017-05-28 2:56 ` kbuild test robot
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20170522083008.fuofuwlgq6muomjn@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=acme@redhat.com \
--cc=ak@linux.intel.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=eranian@google.com \
--cc=jolsa@redhat.com \
--cc=kan.liang@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=vincent.weaver@maine.edu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).