From: Feng Tang <feng.tang@intel.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: Bjorn Helgaas <helgaas@kernel.org>,
Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
x86@kernel.org, linux-pci@vger.kernel.org,
linux-kernel@vger.kernel.org, Bjorn Helgaas <bhelgaas@google.com>,
Kai-Heng Feng <kai.heng.feng@canonical.com>,
rui.zhang@intel.com, len.brown@intel.com
Subject: Re: [PATCH] x86/PCI: Convert force_disable_hpet() to standard quirk
Date: Fri, 27 Nov 2020 14:11:31 +0800 [thread overview]
Message-ID: <20201127061131.GB105524@shbuild999.sh.intel.com> (raw)
In-Reply-To: <87eekfk8bd.fsf@nanos.tec.linutronix.de>
Hi Thomas,
On Fri, Nov 27, 2020 at 12:27:34AM +0100, Thomas Gleixner wrote:
> Feng,
>
> On Thu, Nov 26 2020 at 09:24, Feng Tang wrote:
> > On Wed, Nov 25, 2020 at 01:46:23PM +0100, Thomas Gleixner wrote:
> >> Now the more interesting question is why this needs to be a PCI quirk in
> >> the first place. Can't we just disable the HPET based on family/model
> >> quirks?
> >>
> >> e0748539e3d5 ("x86/intel: Disable HPET on Intel Ice Lake platforms")
> >> f8edbde885bb ("x86/intel: Disable HPET on Intel Coffee Lake H platforms")
> >> fc5db58539b4 ("x86/quirks: Disable HPET on Intel Coffe Lake platforms")
> >> 62187910b0fc ("x86/intel: Add quirk to disable HPET for the Baytrail platform")
>
> > I added this commit, and I can explain some for Baytrail. There was
> > some discussion about the way to disable it:
> > https://lore.kernel.org/lkml/20140328073718.GA12762@feng-snb/t/
> >
> > It uses PCI ID early quirk in the hope that later Baytrail stepping
> > doesn't have the problem. And later on, there was official document
> > (section 18.10.1.3 http://www.intel.com/content/dam/www/public/us/en/documents/datasheets/atom-z8000-datasheet-vol-1.pdf)
> > stating Baytrail's HPET halts in deep idle. So I think your way of
> > using CPUID to disable Baytrail HPET makes more sense.
>
> Correct.
>
> >> I might be missing something here, but in general on anything modern
> >> HPET is mostly useless.
> >
> > IIUC, nowdays HPET's main use is as a clocksource watchdog monitor.
>
> Plus for the TSC refined calibration, where it is really better than
> anything else we have _if_ it is functional.
>
> > And in one debug case, we found it still useful. The debug platform has
> > early serial console which prints many messages in early boot phase,
> > when the HPET is disabled, the software 'jiffies' clocksource will
> > be used as the monitor. Early printk will disable interrupt will
> > printing message, and this could be quite long for a slow 115200
> > device, and cause the periodic HW timer interrupt get missed, and
> > make the 'jiffies' clocksource not accurate, which will in turn
> > judge the TSC clocksrouce inaccurate, and disablt it. (Adding Rui,
> > Len for more details)
>
> Yes, that can happen. But OTOH, we should start to think about the
> requirements for using the TSC watchdog.
>
> I'm inclined to lift that requirement when the CPU has:
>
> 1) X86_FEATURE_CONSTANT_TSC
> 2) X86_FEATURE_NONSTOP_TSC
> 3) X86_FEATURE_NONSTOP_TSC_S3
IIUC, this feature exists for several generations of Atom platforms,
and it is always coupled with 1) and 2), so it could be skipped for
the checking.
> 4) X86_FEATURE_TSC_ADJUST
>
> 5) At max. 4 sockets
>
> After two decades of horrors we're finally at a point where TSC seems to
> be halfways reliable and less abused by BIOS tinkerers. TSC_ADJUST was
> really key as we can now detect even small modifications reliably and
> the important point is that we can cure them as well (not pretty but
> better than all other options).
>
> The only nasty one in the list above is #5 because AFAIK there is still
> no architecural guarantee for TSCs being synchronized on machines with
> more than 4 sockets. I might be wrong, but then nobody told me.
>
> The only reason I hate to disable HPET upfront at least during boot is
> that HPET is the best mechanism for the refined TSC calibration. PMTIMER
> sucks because it's slow and wraps around pretty quick.
>
> So we could do the following even on platforms where HPET stops in some
> magic PC? state:
>
> - Register it during early boot as clocksource
>
> - Prevent the enablement as clockevent and the chardev hpet timer muck
>
> - Prevent the magic PC? state up to the point where the refined
> TSC calibration is finished.
>
> - Unregister it once the TSC has taken over as system clocksource and
> enable the magic PC? state in which HPET gets disfunctional.
This looks reasonable to me.
I have thought about lowering the hpet rating to lower than PMTIMER, so it
still contributes in early boot phase, and fades out after PMTIMER is
initialised.
Thanks,
Feng
> Hmm?
>
> Thanks,
>
> tglx
>
>
next prev parent reply other threads:[~2020-11-27 6:11 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-19 18:19 [PATCH] x86/PCI: Convert force_disable_hpet() to standard quirk Bjorn Helgaas
2020-11-24 23:27 ` Bjorn Helgaas
2020-11-25 12:46 ` Thomas Gleixner
2020-11-25 19:13 ` Bjorn Helgaas
2020-11-26 0:50 ` Thomas Gleixner
2020-11-26 1:24 ` Feng Tang
2020-11-26 23:27 ` Thomas Gleixner
2020-11-27 6:11 ` Feng Tang [this message]
2020-11-30 19:21 ` Thomas Gleixner
2020-12-01 8:34 ` Feng Tang
2020-12-02 7:28 ` Zhang Rui
2022-09-29 15:52 ` Yu Liao
2022-09-30 0:38 ` Feng Tang
2022-09-30 1:05 ` Xiongfeng Wang
2022-09-30 1:15 ` Feng Tang
2022-09-30 9:45 ` Yu Liao
2022-09-30 10:13 ` Feng Tang
2022-10-01 5:18 ` Zhang Rui
2022-10-01 12:00 ` Feng Tang
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=20201127061131.GB105524@shbuild999.sh.intel.com \
--to=feng.tang@intel.com \
--cc=bhelgaas@google.com \
--cc=bp@alien8.de \
--cc=helgaas@kernel.org \
--cc=kai.heng.feng@canonical.com \
--cc=len.brown@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=rui.zhang@intel.com \
--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).