mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Stephane Eranian <eranian@hpl.hp.com>
To: Chuck Ebbert <76306.1226@compuserve.com>
Cc: linux-kernel <linux-kernel@vger.kernel.org>, Andi Kleen <ak@suse.de>
Subject: Re: [PATCH 10/17] 2.6.17.1 perfmon2 patch for review: PMU context switch
Date: Tue, 4 Jul 2006 08:28:57 -0700	[thread overview]
Message-ID: <20060704152857.GA6999@frankl.hpl.hp.com> (raw)
In-Reply-To: <200607011123_MC3-1-C3EB-BFF1@compuserve.com>

Chuck,

On Sat, Jul 01, 2006 at 11:21:22AM -0400, Chuck Ebbert wrote:
> In-Reply-To: <20060630204032.GB22835@frankl.hpl.hp.com>
> 
> On Fri, 30 Jun 2006 13:40:32 -0700, Stephane Eranian wrote:
> 
> > As Andi is suggesting, I think this may depends on how the BIOS implements
> > the low-power state. I have tried the same command on my dual Opteron 250
> > 2.4GHz and I get:
> > $ pfmon --us-c -ecpu_clk_unhalted,interrupts_masked_cycles -k --system-wide -t 10
> > <session to end in 10 seconds>
> > CPU0                     9,520,303 CPU_CLK_UNHALTED
> > CPU0                     3,726,315 INTERRUPTS_MASKED_CYCLES
> > CPU1                    21,268,151 CPU_CLK_UNHALTED
> > CPU1                    14,515,389 INTERRUPTS_MASKED_CYCLES
> 
> That is similar to what I get with idle=halt. Are you not using ACPI
> for idle?
> 
> Try this:
> 
> $ pfmon -ecpu_clk_unhalted,interrupts_masked_cycles_with_interrupt_pending,interrupts_masked_cycles,cycles_no_fpu_ops_retired -k --system-wide -t 10
> <session to end in 10 seconds>
> CPU0     95016828 CPU_CLK_UNHALTED
> CPU0     36472783 INTERRUPTS_MASKED_CYCLES_WITH_INTERRUPT_PENDING
> CPU0     67484408 INTERRUPTS_MASKED_CYCLES
> CPU0    445326968 CYCLES_NO_FPU_OPS_RETIRED
> 
> That's what I get with idle=halt.  Since the kernel doesn't do FP
> the last line should equal clock cycles.  If it were running at full
> speed it would be 16 billion...

Here is what I get on my dual 2.4GHz Opteron 250:

booted with idle=halt
$ pfmon --us-c -ecpu_clk_unhalted,interrupts_masked_cycles_with_interrupt_pending,interrupts_masked_cycles,cycles_no_fpu_ops_retired -k --system-wide -t 10
<session to end in 10 seconds>
CPU0                    11,356,210 CPU_CLK_UNHALTED                               
CPU0                             0 INTERRUPTS_MASKED_CYCLES_WITH_INTERRUPT_PENDING
CPU0                     3,836,107 INTERRUPTS_MASKED_CYCLES                       
CPU0                23,910,784,532 CYCLES_NO_FPU_OPS_RETIRED                      
CPU1                    19,303,632 CPU_CLK_UNHALTED                               
CPU1                             0 INTERRUPTS_MASKED_CYCLES_WITH_INTERRUPT_PENDING
CPU1                    13,942,265 INTERRUPTS_MASKED_CYCLES                       
CPU1                23,911,872,654 CYCLES_NO_FPU_OPS_RETIRED                      

As you can see, CYCLES_NO_FPU_OPS_RETIRED on each CPU is as expected  for 10s.

booted with idle=poll
$ pfmon --us-c -ecpu_clk_unhalted,interrupts_masked_cycles_with_interrupt_pending,interrupts_masked_cycles,cycles_no_fpu_ops_retired -k --system-wide -t 10
<session to end in 10 seconds>
CPU0                23,906,091,982 CPU_CLK_UNHALTED                               
CPU0                             0 INTERRUPTS_MASKED_CYCLES_WITH_INTERRUPT_PENDING
CPU0                     3,771,569 INTERRUPTS_MASKED_CYCLES                       
CPU0                23,906,090,750 CYCLES_NO_FPU_OPS_RETIRED                      
CPU1                23,906,629,241 CPU_CLK_UNHALTED                               
CPU1                             0 INTERRUPTS_MASKED_CYCLES_WITH_INTERRUPT_PENDING
CPU1                    14,805,078 INTERRUPTS_MASKED_CYCLES                       
CPU1                23,906,194,343 CYCLES_NO_FPU_OPS_RETIRED                      

CYCLES_NO_FPU_OPS_RETIRED is as expected and, in this case, it is equal to CPU_CLK_UNHALTED
because we are busy looping.

If I don't specify anything, I get like idle=halt which is expected.

-- 
-Stephane

  reply	other threads:[~2006-07-04 15:37 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-07-01 15:21 Chuck Ebbert
2006-07-04 15:28 ` Stephane Eranian [this message]
  -- strict thread matches above, loose matches on Subject: below --
2006-07-06 17:30 Chuck Ebbert
2006-07-06 20:16 ` Stephane Eranian
2006-06-30 19:17 Chuck Ebbert
2006-06-30 19:37 ` Andi Kleen
2006-06-30 18:33 Chuck Ebbert
2006-06-30 18:42 ` Andi Kleen
2006-06-30 18:43 ` Stephane Eranian
2006-06-30 20:40 ` Stephane Eranian
2006-06-23  9:13 Stephane Eranian
2006-06-30 12:27 ` Andi Kleen
2006-06-30 12:36   ` Stephane Eranian
2006-06-30 12:59     ` Andi Kleen
2006-06-30 13:29       ` Stephane Eranian
2006-06-30 13:41         ` Andi Kleen
2006-06-30 14:12           ` Stephane Eranian
2006-06-30 14:33             ` Andi Kleen
2006-06-30 16:02               ` Stephane Eranian
2006-06-30 17:08                 ` Andi Kleen
2006-06-30 20:47                   ` Stephane Eranian
2006-07-03  9:49       ` Stephane Eranian
2006-07-03 19:25         ` Andi Kleen
2006-07-03 19:22           ` Stephane Eranian
2006-07-03 19:36             ` Andi Kleen

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=20060704152857.GA6999@frankl.hpl.hp.com \
    --to=eranian@hpl.hp.com \
    --cc=76306.1226@compuserve.com \
    --cc=ak@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    /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

all inboxes | Powered by JetHome®