From: David Vrabel <david.vrabel@citrix.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>, "H. Peter Anvin" <hpa@zytor.com>,
LKML <linux-kernel@vger.kernel.org>, <konrad.wilk@oracle.com>,
<john.stultz@linaro.org>, <xen-devel@lists.xen.org>,
Artem Savkov <artem.savkov@gmail.com>
Subject: Re: [tip:timers/core] hrtimers: Support resuming with two or more CPUs online (but stopped)
Date: Fri, 5 Jul 2013 14:46:01 +0100 [thread overview]
Message-ID: <51D6CE19.1050503@citrix.com> (raw)
In-Reply-To: <alpine.DEB.2.02.1307051225060.11637@ionos.tec.linutronix.de>
On 05/07/13 11:25, Thomas Gleixner wrote:
> On Fri, 5 Jul 2013, David Vrabel wrote:
>
> You failed to CC Artem :(
>
>> On 05/07/13 10:30, Artem Savkov wrote:
>>> This commit brings up a warning about a potential deadlock in
>>> smp_call_function_many() discussed previously:
>>> https://lkml.org/lkml/2013/4/18/546
>>
>> Can we just avoid the wait in clock_was_set()? Something like this?
>>
>> 8<------------------------------------------------------
>> hrtimers: do not wait for other CPUs in clock_was_set()
>>
>> Calling on_each_cpu() and waiting in a softirq causes a WARNing about
>> a potential deadlock.
>>
>> Because hrtimers are per-CPU, it is sufficient to ensure that all
>> other CPUs' timers are reprogrammed as soon as possible and before the
>> next softirq on that CPU. There is no need to wait for this to be
>> complete on all CPUs.
Unfortunately this doesn't look sufficient. on_each_cpu(..., 0) may
still wait for other calls to complete before queuing the calls due to
the use of a single set of per-CPU csd data.
David
next prev parent reply other threads:[~2013-07-05 13:46 UTC|newest]
Thread overview: 71+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-27 10:35 [PATCHv6 0/5] xen: maintain an accurate persistent clock in more cases David Vrabel
2013-06-27 10:35 ` [PATCH 1/5] hrtimers: support resuming with two or more CPUs online (but stopped) David Vrabel
2013-06-27 10:35 ` David Vrabel
2013-06-28 21:18 ` [tip:timers/core] hrtimers: Support " tip-bot for David Vrabel
2013-07-05 9:30 ` Artem Savkov
2013-07-05 10:07 ` David Vrabel
2013-07-05 10:07 ` David Vrabel
2013-07-05 10:10 ` Thomas Gleixner
2013-07-05 10:10 ` Thomas Gleixner
2013-07-05 10:25 ` Thomas Gleixner
2013-07-05 10:25 ` Thomas Gleixner
2013-07-05 10:36 ` Thomas Gleixner
2013-07-05 10:36 ` Thomas Gleixner
2013-07-05 10:45 ` Artem Savkov
2013-07-05 10:45 ` Artem Savkov
2013-07-05 13:26 ` Thomas Gleixner
2013-07-05 13:26 ` Thomas Gleixner
2013-07-05 10:38 ` Artem Savkov
2013-07-05 10:38 ` Artem Savkov
2013-07-05 13:46 ` David Vrabel [this message]
2013-07-05 13:51 ` Thomas Gleixner
2013-07-05 13:51 ` Thomas Gleixner
2013-07-05 13:46 ` David Vrabel
2013-07-05 10:09 ` Thomas Gleixner
2013-07-05 10:09 ` Thomas Gleixner
2013-07-05 9:30 ` Artem Savkov
2013-06-28 21:18 ` tip-bot for David Vrabel
2013-06-28 21:18 ` [tip:timers/core] xen: Remove clock_was_set() call in the resume path tip-bot for David Vrabel
2013-06-28 21:18 ` tip-bot for David Vrabel
2013-06-27 10:35 ` [PATCH 2/5] time: pass flags instead of multiple bools to timekeeping_update() David Vrabel
2013-06-27 17:39 ` John Stultz
2013-06-27 17:39 ` John Stultz
2013-06-28 21:18 ` [tip:timers/core] timekeeping: Pass " tip-bot for David Vrabel
2013-06-28 21:18 ` tip-bot for David Vrabel
2013-06-27 10:35 ` [PATCH 2/5] time: pass " David Vrabel
2013-06-27 10:35 ` [PATCH 3/5] time: indicate that the clock was set in the pvclock gtod notifier chain David Vrabel
2013-06-27 10:35 ` David Vrabel
2013-06-27 17:37 ` John Stultz
2013-06-27 17:37 ` John Stultz
2013-06-28 10:20 ` David Vrabel
2013-06-28 10:20 ` David Vrabel
2013-06-28 21:19 ` [tip:timers/core] timekeeping: Indicate that clock was set in the pvclock gtod notifier tip-bot for David Vrabel
2013-06-28 21:19 ` tip-bot for David Vrabel
2013-06-27 10:35 ` [PATCH 4/5] x86/xen: sync the wallclock when the system time is set David Vrabel
2013-06-28 21:19 ` [tip:timers/core] x86: xen: Sync " tip-bot for David Vrabel
2013-06-28 21:19 ` tip-bot for David Vrabel
2013-06-27 10:35 ` [PATCH 4/5] x86/xen: sync " David Vrabel
2013-06-27 10:35 ` [PATCH 5/5] x86/xen: sync the CMOS RTC as well as the Xen wallclock David Vrabel
2013-06-27 10:35 ` David Vrabel
2013-06-28 15:38 ` Thomas Gleixner
2013-06-28 15:38 ` Thomas Gleixner
2013-06-28 15:49 ` David Vrabel
2013-06-28 15:49 ` David Vrabel
2013-06-28 16:09 ` Thomas Gleixner
2013-06-28 16:09 ` Thomas Gleixner
2013-06-28 16:51 ` David Vrabel
2013-06-28 16:51 ` David Vrabel
2013-06-28 20:41 ` Thomas Gleixner
2013-06-28 20:41 ` Thomas Gleixner
2013-06-28 21:19 ` [tip:timers/core] x86: xen: Sync " tip-bot for David Vrabel
2013-06-28 21:19 ` tip-bot for David Vrabel
2013-06-28 13:14 ` [PATCHv6 0/5] xen: maintain an accurate persistent clock in more cases Thomas Gleixner
2013-06-28 15:01 ` Konrad Rzeszutek Wilk
2013-06-28 15:01 ` Konrad Rzeszutek Wilk
2013-06-28 15:12 ` Thomas Gleixner
2013-06-28 15:12 ` Thomas Gleixner
2013-06-28 16:19 ` Konrad Rzeszutek Wilk
2013-06-28 18:58 ` Thomas Gleixner
2013-06-28 18:58 ` Thomas Gleixner
2013-06-28 16:19 ` Konrad Rzeszutek Wilk
2013-06-28 13:14 ` 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=51D6CE19.1050503@citrix.com \
--to=david.vrabel@citrix.com \
--cc=artem.savkov@gmail.com \
--cc=hpa@zytor.com \
--cc=john.stultz@linaro.org \
--cc=konrad.wilk@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=tglx@linutronix.de \
--cc=xen-devel@lists.xen.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.