From: Andreas Mohr <andi@lisas.de>
To: "Zhang, Yanmin" <yanmin_zhang@linux.intel.com>
Cc: LKML <linux-kernel@vger.kernel.org>, linux-acpi@vger.kernel.org
Subject: Re: Dynamic configure max_cstate
Date: Mon, 27 Jul 2009 09:33:38 +0200 [thread overview]
Message-ID: <20090727073338.GA12669@rhlx01.hs-esslingen.de> (raw)
In-Reply-To: <1248672613.2560.604.camel@ymzhang>
Hi,
> When running a fio workload, I found sometimes cpu C state has
> big impact on the result. Mostly, fio is a disk I/O workload
> which doesn't spend much time with cpu, so cpu switch to C2/C3
> freqently and the latency is big.
Rather than inventing ways to limit ACPI Cx state usefulness, we should
perhaps be thinking of what's wrong here.
And your complaint might just fit into a thought I had recently:
are we actually taking ACPI Cx exit latency into account, for timers???
If we program a timer to fire at some point, then it is quite imaginable
that any ACPI Cx exit latency due to the CPU being idle at that moment
could add to actual timer trigger time significantly.
To combat this, one would need to tweak the timer expiration time
to include the exit latency. But of course once the CPU is running
again, one would need to re-add the latency amount (read: reprogram the
timer hardware, ugh...) to prevent the timer from firing too early.
Given that one would need to reprogram timer hardware quite often,
I don't know whether taking Cx exit latency into account is feasible.
OTOH analysis of the single next timer value and actual hardware reprogramming
would have to be done only once (in ACPI sleep and wake paths each),
thus it might just turn out to be very beneficial after all
(minus prolonging ACPI Cx path activity and thus aggravating CPU power
savings, of course).
Arjan mentioned examples of maybe 10us for C2 and 185us for C3/C4 in an
article.
OTOH even 185us is only 0.185ms, which, when compared to disk seek
latency (around 7ms still, except for SSD), doesn't seem to be all that much.
Or what kind of ballpark figure do you have for percentage of I/O
deterioration?
I'm wondering whether we might have an even bigger problem with disk I/O
related to this than just the raw ACPI exit latency value itself.
Andreas Mohr
next prev parent reply other threads:[~2009-07-27 7:33 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-07-27 5:30 Zhang, Yanmin
2009-07-27 7:33 ` Andreas Mohr [this message]
2009-07-28 2:42 ` Zhang, Yanmin
2009-07-28 7:20 ` Corrado Zoccolo
2009-07-28 9:00 ` Zhang, Yanmin
2009-07-28 10:11 ` Andreas Mohr
2009-07-28 14:03 ` Andreas Mohr
2009-07-28 17:35 ` ok, now would this be useful? (Re: Dynamic configure max_cstate) Andreas Mohr
2009-07-29 8:20 ` Dynamic configure max_cstate Zhang, Yanmin
2009-07-31 3:43 ` Robert Hancock
2009-07-31 7:06 ` Zhang, Yanmin
2009-07-31 8:07 ` Andreas Mohr
2009-07-31 14:40 ` Andi Kleen
2009-07-31 14:56 ` Michael S. Zick
2009-07-31 17:37 ` Pallipadi, Venkatesh
2009-07-31 15:14 ` Len Brown
2009-07-30 6:28 ` Zhang, Yanmin
2009-07-28 19:25 ` Len Brown
2009-07-29 0:17 ` Len Brown
2009-07-29 8:00 ` Andreas Mohr
2009-07-28 19:47 ` Len Brown
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=20090727073338.GA12669@rhlx01.hs-esslingen.de \
--to=andi@lisas.de \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=yanmin_zhang@linux.intel.com \
/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®