From: Jacob Pan <jacob.jun.pan@linux.intel.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: Daniel Drake <drake@endlessm.com>, Len Brown <lenb@kernel.org>,
x86@kernel.org, LKML <linux-kernel@vger.kernel.org>,
linux@endlessm.com, rjw@rjwysocki.net,
"Rafael J. Wysocki" <rafael.j.wysocki@intel.com>,
jacob.jun.pan@linux.intel.com
Subject: Re: Detecting x86 LAPIC timer frequency from CPUID data
Date: Fri, 19 Apr 2019 13:50:37 -0700 [thread overview]
Message-ID: <20190419135037.5bf5750e@jacob-builder> (raw)
In-Reply-To: <alpine.DEB.2.21.1904191042180.3174@nanos.tec.linutronix.de>
On Fri, 19 Apr 2019 10:57:10 +0200 (CEST)
Thomas Gleixner <tglx@linutronix.de> wrote:
> On Fri, 19 Apr 2019, Daniel Drake wrote:
> > On Fri, Apr 19, 2019 at 6:30 AM Thomas Gleixner
> > <tglx@linutronix.de> wrote:
> > > Time Stamp Counter/Core Crystal Clock Information (0x15):
> > > TSC/clock ratio = 168/2
> > > nominal core crystal clock = 0 Hz
> > >
> > > Processor Frequency Information (0x16):
> > > Core Base Frequency (MHz) = 0x834 (2100)
> > > Core Maximum Frequency (MHz) = 0xed8 (3800)
> > > Bus (Reference) Frequency (MHz) = 0x64 (100)
> > >
> > > Assuming that TSC and local APIC timer run from the same
> > > frequency on these modern machines.
> > >
> > > 2100MHz * 2 / 168 = 25MHz
> > >
> > > and disabling the tsc deadline timer tells me:
> > >
> > > ..... calibration result: 24999
> > >
> > > Close enough.
> >
> > I tested all the Intel SoC generations we have on hand. The
> > assumption that the core crystal clock feeds the APIC seems to be
> > consistently true.
> >
> > (Please note that all the following results are done with
> > CONFIG_HZ=250, which is why the "calibration result" is 4x higher
> > than HZ=1000 as used in previous mails)
> >
> > In the easy case, the low cost platforms do not support CPUID.0x16
> > (so no CPU frequency reporting), but they do tell us the core
> > crystal clock, which is consistent with the APIC calibration
> > result:
>
> ...
>
> > And the 4 higher-end SoCs that we have available for testing all
> > report crystal clock 0Hz from CPUID 0x15, but by combining the
> > CPUID.0x16 base frequency with the CPUID.0x15 TSC/clock ratio, the
> > crystal frequency can be calculated as you describe, and it
> > consistently matches the APIC timer calibration result.
>
> ...
>
> > Is this data convincing enough or should we additionally wait for
> > some comments from Intel?
>
> For me it's pretty convincing, but having some confirmation from Intel
> wouldn't be a bad thing.
>
> > I came up with the patch below. However, upon testing, I realised
> > that, at least for the platforms I have in hand, only the first
> > hunk is really needed. We don't need to use your magic calculation
> > to find the crystal frequency because Intel already told us!
> > native_calibrate_tsc() already hardcodes the crystal frequency for
> > Kabylake, and Amber/Whiskey/Coffee also report the 0x8e/0x9e
> > Kabylake model codes.
>
> I'd rather replace these model checks with math. These tables are
> horrible to maintain.
>
> > Plus ApolloLake/GeminiLake do report the crystal frequency in
> > CPUID.0x15 so that is covered too.
>
> > While looking around this code I also spotted something curious.
> > In calibrate_APIC_clock() for the case where lapic_timer_frequency
> > has been externally provided, we have:
> > lapic_clockevent.max_delta_ns =
> > clockevent_delta2ns(0x7FFFFF,
> > &lapic_clockevent); lapic_clockevent.max_delta_ticks = 0x7FFFFF;
> >
> > But in the case where we calibrate, we have:
> > lapic_clockevent.max_delta_ns =
> > clockevent_delta2ns(0x7FFFFFFF, &lapic_clockevent);
> > lapic_clockevent.max_delta_ticks = 0x7FFFFFFF;
> >
> > 0x7FFFFF vs 0x7FFFFFFF, is that intentional?
>
> I don't think so. Looks like a failed copy and paste. Cc'ed Jacob, he
> might know.
>
At the time of v2.6.35 both places use 0x7FFFFF. But later this patch
increased the latter to 0x7FFFFFFF but forgot the first part. So I
guess it is not exactly a failed copy and paste.
commit 4aed89d6b515b9185351706ca95cd712c9d8d6a3
Author: Pierre Tardy <pierre.tardy@intel.com>
Date: Thu Jan 6 16:23:29 2011 +0100
x86, lapic-timer: Increase the max_delta to 31 bits
> Thanks,
>
> tglx
[Jacob Pan]
next prev parent reply other threads:[~2019-04-19 20:48 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-17 5:28 Detecting x86 LAPIC timer frequency from CPUID data Daniel Drake
2019-04-18 13:12 ` Thomas Gleixner
2019-04-18 22:30 ` Thomas Gleixner
2019-04-19 8:35 ` Daniel Drake
2019-04-19 8:57 ` Thomas Gleixner
2019-04-19 20:50 ` Jacob Pan [this message]
2019-04-19 20:52 ` Thomas Gleixner
2019-04-19 23:09 ` Jacob Pan
2019-05-09 10:34 ` [tip:x86/apic] x86/tsc: Use CPUID.0x16 to calculate missing crystal frequency tip-bot for Daniel Drake
-- strict thread matches above, loose matches on Subject: below --
2019-05-09 5:54 [PATCH v2 1/3] x86/tsc: use " Daniel Drake
2019-05-09 5:54 ` [PATCH v2 2/3] x86/apic: rename lapic_timer_frequency to lapic_timer_period Daniel Drake
2019-05-09 10:34 ` [tip:x86/apic] x86/apic: Rename 'lapic_timer_frequency' to 'lapic_timer_period' tip-bot for Daniel Drake
2019-05-09 5:54 ` [PATCH v2 3/3] x86/tsc: set LAPIC timer period to crystal clock frequency Daniel Drake
2019-05-09 7:25 ` [PATCH v2 1/3] x86/tsc: use CPUID.0x16 to calculate missing crystal frequency Thomas Gleixner
2019-05-09 9:07 ` Ingo Molnar
2019-04-03 7:49 No 8254 PIT & no HPET on new Intel N3350 platforms causes kernel panic during early boot Daniel Drake
2019-04-03 11:21 ` Thomas Gleixner
2019-04-03 12:01 ` Thomas Gleixner
2019-04-09 5:43 ` Daniel Drake
2019-04-10 12:54 ` Thomas Gleixner
2019-04-16 5:21 ` Daniel Drake
2019-05-09 10:35 ` [tip:x86/apic] x86/tsc: Set LAPIC timer period to crystal clock frequency tip-bot for Daniel Drake
2019-06-27 8:54 ` No 8254 PIT & no HPET on new Intel N3350 platforms causes kernel panic during early boot Daniel Drake
2019-06-27 14:06 ` Thomas Gleixner
2019-06-28 3:33 ` Daniel Drake
2019-06-28 5:07 ` Thomas Gleixner
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=20190419135037.5bf5750e@jacob-builder \
--to=jacob.jun.pan@linux.intel.com \
--cc=drake@endlessm.com \
--cc=lenb@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@endlessm.com \
--cc=rafael.j.wysocki@intel.com \
--cc=rjw@rjwysocki.net \
--cc=tglx@linutronix.de \
--cc=x86@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;
as well as URLs for NNTP newsgroup(s).