public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: John Stultz <john.stultz@linaro.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
	Ingo Molnar <mingo@kernel.org>,
	Steven Rostedt <rostedt@goodmis.org>
Subject: [patch V2 62/64] seqcount: Add raw_write_seqcount_latch()
Date: Wed, 16 Jul 2014 21:05:21 -0000	[thread overview]
Message-ID: <20140716205057.481379409@linutronix.de> (raw)
In-Reply-To: 20140716205018.175419210@linutronix.de

[-- Attachment #1: seqcount-add-raw-write-seqcount-latch.patch --]
[-- Type: text/plain, Size: 1668 bytes --]

From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>

For NMI safe access to clock monotonic we use the seqcount LSB as
index of a timekeeper array. The update sequence looks like this:

      smp_wmb();      <- prior stores to a[1]
      seq++;
      smp_wmb();      <- seq increment before update of a[0]
      update(a[0]);
      smp_wmb();      <- update of a[0]	
      seq++;
      smp_wmb();      <- seq increment before update of a[1]	
      update(a[1]);

To avoid open coded barriers, provide a helper function.

[ tglx: Split out of a combo patch against the first implementation of
  	the NMI safe accessor ]

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/seqlock.h |   11 +++++++++++
 1 file changed, 11 insertions(+)

Index: tip/include/linux/seqlock.h
===================================================================
--- tip.orig/include/linux/seqlock.h
+++ tip/include/linux/seqlock.h
@@ -236,6 +236,17 @@ static inline void raw_write_seqcount_en
 }
 
 /*
+ * raw_write_seqcount_latch - redirect readers to even/odd copy
+ * @s: pointer to seqcount_t
+ */
+static inline void raw_write_seqcount_latch(seqcount_t *s)
+{
+       smp_wmb();      /* prior stores before incrementing "sequence" */
+       s->sequence++;
+       smp_wmb();      /* increment "sequence" before following stores */
+}
+
+/*
  * Sequence counter only version assumes that callers are using their
  * own mutexing.
  */



  parent reply	other threads:[~2014-07-16 21:06 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-16 21:03 [patch V2 00/64] timekeeping: 2038, optimizations, NMI safe accessors Thomas Gleixner
2014-07-16 21:03 ` [patch V2 01/64] tile: Convert VDSO timekeeping to the precise mechanism Thomas Gleixner
2014-07-16 21:03 ` [patch V2 02/64] timekeeping: Simplify arch_gettimeoffset() Thomas Gleixner
2014-07-16 21:03 ` [patch V2 03/64] hrtimer: Cleanup hrtimer accessors to the timekepeing state Thomas Gleixner
2014-07-16 21:03 ` [patch V2 04/64] ktime: Kill non-scalar ktime_t implementation for 2038 Thomas Gleixner
2014-07-16 21:03 ` [patch V2 05/64] ktime: Sanitize ktime_to_us/ms conversion Thomas Gleixner
2014-07-16 21:03 ` [patch V2 06/64] ktime: Change ktime_set() to take 64bit seconds value Thomas Gleixner
2014-07-16 21:03 ` [patch V2 07/64] time64: Add time64.h header and define struct timespec64 Thomas Gleixner
2014-07-16 21:03 ` [patch V2 08/64] time: More core infrastructure for timespec64 Thomas Gleixner
2014-07-16 21:04 ` [patch V2 09/64] timekeeping: Convert timekeeping core to use timespec64s Thomas Gleixner
2014-07-16 21:04 ` [patch V2 10/64] time: Consolidate the time accessor prototypes Thomas Gleixner
2014-07-16 21:04 ` [patch V2 11/64] timekeeping: Provide timespec64 based interfaces Thomas Gleixner
2014-07-16 21:04 ` [patch V2 12/64] timekeeper: Move tk_xtime to core code Thomas Gleixner
2014-07-23 21:15   ` John Stultz
2014-07-23 21:59     ` Thomas Gleixner
2014-07-16 21:04 ` [patch V2 13/64] timekeeping: Cache optimize struct timekeeper Thomas Gleixner
2014-07-16 21:04 ` [patch V2 14/64] timekeeping: Use timekeeping_update() instead of memcpy() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 15/64] timekeeping: Provide internal ktime_t based data Thomas Gleixner
2014-07-16 21:04 ` [patch V2 16/64] timekeeping: Use ktime_t based data for ktime_get() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 17/64] timekeeping: Provide ktime_get_with_offset() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 18/64] timekeeping: Use ktime_t based data for ktime_get_real() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 19/64] timekeeping; Use ktime_t based data for ktime_get_boottime() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 20/64] timekeeping: Use ktime_t based data for ktime_get_clocktai() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 21/64] timekeeping: Use ktime_t data for ktime_get_update_offsets_now() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 22/64] timekeeping; Use ktime based data for ktime_get_update_offsets_tick() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 23/64] timekeeping: Provide ktime_mono_to_any() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 24/64] timerfd: Use ktime_mono_to_real() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 25/64] input: evdev: " Thomas Gleixner
2014-07-16 21:04 ` [patch V2 26/64] drm: " Thomas Gleixner
2014-07-16 21:04 ` [patch V2 27/64] timekeeping: Remove ktime_get_monotonic_offset() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 28/64] timekeeping: Provide ktime_get[*]_ns() helpers Thomas Gleixner
2014-07-16 21:04 ` [patch V2 29/64] time: Export nsecs_to_jiffies() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 30/64] sched: Make task->real_start_time nanoseconds based Thomas Gleixner
2014-07-16 21:04 ` [patch V2 31/64] sched: Make task->start_time " Thomas Gleixner
2014-07-16 21:04 ` [patch V2 32/64] delayacct: Make accounting nanosecond based Thomas Gleixner
2014-07-16 21:04 ` [patch V2 33/64] delayacct: Remove braindamaged type conversions Thomas Gleixner
2014-07-16 21:04 ` [patch V2 34/64] powerpc: cell: Use ktime_get_ns() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 35/64] connector: " Thomas Gleixner
2014-07-16 21:06   ` Evgeniy Polyakov
2014-07-16 21:04 ` [patch V2 36/64] mfd: cros_ec_spi: " Thomas Gleixner
2014-07-16 21:04 ` [patch V2 37/64] misc: ioc4: " Thomas Gleixner
2014-07-16 21:04 ` [patch V2 38/64] net: mlx5: " Thomas Gleixner
2014-07-16 21:04 ` [patch V2 39/64] fs: lockd: " Thomas Gleixner
2014-07-16 21:04 ` [patch V2 40/64] hwmon: ibmaem: " Thomas Gleixner
2014-07-21  8:37   ` Jean Delvare
2014-07-21 21:38     ` Darrick J. Wong
2014-07-16 21:04 ` [patch V2 41/64] iio: Use ktime_get_real_ns() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 42/64] arm: bL_switcher:k " Thomas Gleixner
2014-07-16 21:04 ` [patch V2 43/64] x86: kvm: Use ktime_get_boot_ns() Thomas Gleixner
2014-07-17 10:58   ` Paolo Bonzini
2014-07-16 21:04 ` [patch V2 44/64] x86: kvm: Make kvm_get_time_and_clockread() nanoseconds based Thomas Gleixner
2014-07-17 10:58   ` Paolo Bonzini
2014-07-16 21:04 ` [patch V2 45/64] timekeeping: Remove monotonic_to_bootbased Thomas Gleixner
2014-07-16 21:04 ` [patch V2 46/64] timekeeping: Use ktime_get_boottime() for get_monotonic_boottime() Thomas Gleixner
2014-07-16 21:04 ` [patch V2 47/64] timekeeping: Simplify getboottime() Thomas Gleixner
2014-07-16 21:05 ` [patch V2 48/64] timekeeping: Remove timekeeper.total_sleep_time Thomas Gleixner
2014-07-16 21:05 ` [patch V2 49/64] timekeeping: Simplify timekeeping_clocktai() Thomas Gleixner
2014-07-16 21:05 ` [patch V2 50/64] hangcheck-timer: Use ktime_get_ns() Thomas Gleixner
2014-07-16 22:08   ` Greg Kroah-Hartman
2014-07-16 21:05 ` [patch V2 51/64] timekeeping: Provide ktime_get_raw() Thomas Gleixner
2014-07-16 21:05 ` [patch V2 52/64] drm: i915: Use nsec based interfaces Thomas Gleixner
2014-07-16 21:05 ` [patch V2 53/64] drm: vmwgfx: " Thomas Gleixner
2014-07-16 21:05 ` [patch V2 54/64] wireless: ath9k: Get rid of timespec conversions Thomas Gleixner
2014-07-16 21:05 ` [patch V2 55/64] clocksource: Make delta calculation a function Thomas Gleixner
2014-07-16 21:05 ` [patch V2 56/64] clocksource: Move cycle_last validation to core code Thomas Gleixner
2014-07-16 21:05 ` [patch V2 57/64] clocksource: Get rid of cycle_last Thomas Gleixner
2014-07-16 21:05 ` [patch V2 58/64] timekeeping: Restructure the timekeeper some more Thomas Gleixner
2014-07-16 21:05 ` [patch V2 59/64] timekeeping: Create struct tk_read_base and use it in struct timekeeper Thomas Gleixner
2014-07-16 21:05 ` [patch V2 60/64] timekeeping: Use tk_read_base as argument for timekeeping_get_ns() Thomas Gleixner
2014-07-16 21:05 ` [patch V2 61/64] seqcount: Provide raw_read_seqcount() Thomas Gleixner
2014-07-16 21:05 ` Thomas Gleixner [this message]
2014-07-16 21:05 ` [patch V2 63/64] timekeeping: Provide fast and NMI safe access to CLOCK_MONOTONIC Thomas Gleixner
2014-07-16 21:05 ` [patch V2 64/64] ftrace: Provide trace clocks monotonic Thomas Gleixner
2014-07-17 12:55   ` Steven Rostedt
2014-07-17 13:12     ` Steven Rostedt
2014-07-17 22:32 ` [patch V2 00/64] timekeeping: 2038, optimizations, NMI safe accessors 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=20140716205057.481379409@linutronix.de \
    --to=tglx@linutronix.de \
    --cc=john.stultz@linaro.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.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