Linux-ARM-Kernel Archive on lore.kernel.org
 help / color / Atom feed
From: Will Deacon <will.deacon@arm.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-arch@vger.kernel.org, Arnd Bergmann <arnd@arndb.de>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Daniel Lezcano <daniel.lezcano@linaro.org>,
	Russell King <linux@armlinux.org.uk>,
	Ralf Baechle <ralf@linux-mips.org>,
	Mark Salyzyn <salyzyn@android.com>,
	Paul Burton <paul.burton@mips.com>,
	Vincenzo Frascino <vincenzo.frascino@arm.com>,
	Peter Collingbourne <pcc@google.com>,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v2 06/28] kernel: Define gettimeofday vdso common code
Date: Wed, 13 Feb 2019 17:05:44 +0000
Message-ID: <20190213170543.GF6346@brain-police> (raw)
In-Reply-To: <alpine.DEB.2.21.1902081950260.1662@nanos.tec.linutronix.de>

Hi Thomas,

On Fri, Feb 08, 2019 at 08:28:07PM +0100, Thomas Gleixner wrote:
> On Fri, 8 Feb 2019, Will Deacon wrote:
> > On Fri, Dec 07, 2018 at 05:53:21PM +0000, Will Deacon wrote:
> > > Anyway, moving the counter read into the protected region is a little fiddly
> > > because the memory barriers we have in there won't give us the ordering we
> > > need. We'll instead need to do something nasty, like create a dependency
> > > from the counter read to the read of the seqlock:
> > > 
> > > Maybe the untested crufty hack below, although this will be a nightmare to
> > > implement in C.
> > 
> > We discussed this in person this week, but you couldn't recall the details
> > off the top of your head so I'm replying here. Please could you clarify what
> > your concern was with the existing code, and whether or not I've got the
> > wrong end of the stick?
> 
> If you just collect the variables under the seqcount protection and have
> the readout of the timer outside of it then you are not guaranteeing
> consistent state.
> 
> The problem is:
> 
> 	do {
> 		seq = read_seqcount_begin(d->seq);
> 		last = d->cycle_last;
> 		mult = d->mult;
> 		shift = d->shift;
> 		ns = d->ns_base;
> 	while (read_seqcount_retry(d->seq, seq));
> 
> 	ns += ((read_clock() - last) * mult);
> 	ns >>= shift;
> 
> So on the first glance this looks consistent because you collect all data
> and then do the read and calc outside the loop.
> 
> But if 'd->mult' gets updated _before_ read_clock() then you can run into a
> situation where time goes backwards with the next read.
> 
> Here is the flow you need for that:
> 
> t1 = gettime()
>      {
> 	collect_data()
> 
>      ---> Interrupt, updates mult (mult becomes smaller)
> 
>      	  This can expand over a very long time when the task is scheduled
>      	  out here and there are multiple updates in between. The farther
>      	  out the read is delayed, the more likely the problem is going to
>      	  observable.
> 
>      	read_clock_and_calc()
>      }
> 
> t2 = gettime()
>      {
> 	collect_data()
>      	read_clock_and_calc()
>      }
> 
> This second read uses updated data, i.e. the smaller mult. So depending on
> the distance of the two reads and the difference of the mult, the caller
> can observe t2 < t1, which is a NONO. You can observe it on a single
> CPU. No virt, SMP, migration needed at all.

Thank you for this explanation, I agree that we have a bug here (and have
done since day 1 on arm64!). I've replied separately on ktime_get().

Will

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  parent reply index

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-11-29 17:05 [PATCH v2 00/28] Unify vDSOs across more architectures Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 01/28] kernel: Standardize vdso_datapage Vincenzo Frascino
2018-11-29 22:39   ` Thomas Gleixner
2018-12-11 13:22     ` Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 02/28] kernel: Add Monotonic boot time support Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 03/28] kernel: Add International Atomic Time support Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 04/28] kernel: Add masks support for Raw and NTP time Vincenzo Frascino
2018-11-29 22:41   ` Thomas Gleixner
2018-12-11 13:24     ` Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 05/28] kernel: Add clock_mode support Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 06/28] kernel: Define gettimeofday vdso common code Vincenzo Frascino
2018-11-29 20:42   ` Arnd Bergmann
2018-12-11 13:39     ` Vincenzo Frascino
2018-12-11 21:41       ` Arnd Bergmann
2018-12-13  9:46         ` Vincenzo Frascino
2018-11-29 22:11   ` Thomas Gleixner
2018-11-30 14:29     ` Arnd Bergmann
2018-12-11 14:02       ` Vincenzo Frascino
2018-12-07 17:53     ` Will Deacon
2019-02-08 17:35       ` Will Deacon
2019-02-08 19:28         ` Thomas Gleixner
2019-02-08 19:30           ` Thomas Gleixner
2019-02-13 17:04             ` Will Deacon
2019-02-13 19:35               ` Thomas Gleixner
2019-02-13 17:05           ` Will Deacon [this message]
2018-12-11 13:54     ` Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 07/28] arm64: Build vDSO with -ffixed-x18 Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 08/28] arm64: Substitute gettimeofday with C implementation Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 09/28] arm64: compat: Alloc separate pages for vectors and sigpage Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 10/28] arm64: compat: Split kuser32 Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 11/28] arm64: compat: Refactor aarch32_alloc_vdso_pages() Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 12/28] arm64: compat: Add KUSER_HELPERS config option Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 13/28] arm64: compat: Add missing syscall numbers Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 14/28] arm64: compat: Expose signal related structures Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 15/28] arm64: compat: Generate asm offsets for signals Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 16/28] lib: vdso: Add compat support Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 17/28] arm64: compat: Add vDSO Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 18/28] arm64: Refactor vDSO code Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 19/28] arm64: compat: vDSO setup for compat layer Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 20/28] arm64: elf: vDSO code page discovery Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 21/28] arm64: compat: Get sigreturn trampolines from vDSO Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 22/28] arm64: Add vDSO compat support Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 23/28] arm64: Enable compat vDSO support Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 24/28] arm: Add support for generic vDSO Vincenzo Frascino
2018-12-10 22:13   ` Mark Salyzyn
2018-12-11 14:15     ` Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 25/28] mips: Introduce vdso_direct Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 26/28] clock: csrc-4k: Add support for vdso_direct Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 27/28] clock: gic-timer: " Vincenzo Frascino
2018-11-29 17:05 ` [PATCH v2 28/28] mips: Add support for generic vDSO Vincenzo Frascino

Reply instructions:

You may reply publically 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=20190213170543.GF6346@brain-police \
    --to=will.deacon@arm.com \
    --cc=arnd@arndb.de \
    --cc=catalin.marinas@arm.com \
    --cc=daniel.lezcano@linaro.org \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux@armlinux.org.uk \
    --cc=paul.burton@mips.com \
    --cc=pcc@google.com \
    --cc=ralf@linux-mips.org \
    --cc=salyzyn@android.com \
    --cc=tglx@linutronix.de \
    --cc=vincenzo.frascino@arm.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

Linux-ARM-Kernel Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-arm-kernel/0 linux-arm-kernel/git/0.git
	git clone --mirror https://lore.kernel.org/linux-arm-kernel/1 linux-arm-kernel/git/1.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-arm-kernel linux-arm-kernel/ https://lore.kernel.org/linux-arm-kernel \
		linux-arm-kernel@lists.infradead.org infradead-linux-arm-kernel@archiver.kernel.org
	public-inbox-index linux-arm-kernel


Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.infradead.lists.linux-arm-kernel


AGPL code for this site: git clone https://public-inbox.org/ public-inbox