All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Brendan Gregg <bgregg@netflix.com>,
	Alexei Starovoitov <alexei.starovoitov@gmail.com>,
	John Stultz <john.stultz@linaro.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Steven Rostedt <rostedt@goodmis.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Subject: [PATCH 4.8 02/37] timekeeping: Fix __ktime_get_fast_ns() regression
Date: Fri, 14 Oct 2016 14:26:48 +0200	[thread overview]
Message-ID: <20161014122549.877055542@linuxfoundation.org> (raw)
In-Reply-To: <20161014122549.411962735@linuxfoundation.org>

4.8-stable review patch.  If anyone has any objections, please let me know.

------------------

From: John Stultz <john.stultz@linaro.org>

commit 58bfea9532552d422bde7afa207e1a0f08dffa7d upstream.

In commit 27727df240c7 ("Avoid taking lock in NMI path with
CONFIG_DEBUG_TIMEKEEPING"), I changed the logic to open-code
the timekeeping_get_ns() function, but I forgot to include
the unit conversion from cycles to nanoseconds, breaking the
function's output, which impacts users like perf.

This results in bogus perf timestamps like:
 swapper     0 [000]   253.427536:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]   254.426573:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]   254.426687:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]   254.426800:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]   254.426905:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]   254.427022:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]   254.427127:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]   254.427239:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]   254.427346:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]   254.427463:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]   255.426572:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])

Instead of more reasonable expected timestamps like:
 swapper     0 [000]    39.953768:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]    40.064839:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]    40.175956:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]    40.287103:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]    40.398217:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]    40.509324:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]    40.620437:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]    40.731546:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]    40.842654:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]    40.953772:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])
 swapper     0 [000]    41.064881:  111111111 cpu-clock:  ffffffff810a0de6 native_safe_halt+0x6 ([kernel.kallsyms])

Add the proper use of timekeeping_delta_to_ns() to convert
the cycle delta to nanoseconds as needed.

Thanks to Brendan and Alexei for finding this quickly after
the v4.8 release. Unfortunately the problematic commit has
landed in some -stable trees so they'll need this fix as
well.

Many apologies for this mistake. I'll be looking to add a
perf-clock sanity test to the kselftest timers tests soon.

Fixes: 27727df240c7 "timekeeping: Avoid taking lock in NMI path with CONFIG_DEBUG_TIMEKEEPING"
Reported-by: Brendan Gregg <bgregg@netflix.com>
Reported-by: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Tested-and-reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/1475636148-26539-1-git-send-email-john.stultz@linaro.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 kernel/time/timekeeping.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -403,8 +403,11 @@ static __always_inline u64 __ktime_get_f
 		tkr = tkf->base + (seq & 0x01);
 		now = ktime_to_ns(tkr->base);
 
-		now += clocksource_delta(tkr->read(tkr->clock),
-					 tkr->cycle_last, tkr->mask);
+		now += timekeeping_delta_to_ns(tkr,
+				clocksource_delta(
+					tkr->read(tkr->clock),
+					tkr->cycle_last,
+					tkr->mask));
 	} while (read_seqcount_retry(&tkf->seq, seq));
 
 	return now;

  parent reply	other threads:[~2016-10-14 12:28 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20161014122720uscas1p185a2db67245495161c918bca0cd90f23@uscas1p1.samsung.com>
2016-10-14 12:26 ` [PATCH 4.8 00/37] 4.8.2-stable review Greg Kroah-Hartman
2016-10-14 12:26   ` [PATCH 4.8 01/37] usb: storage: fix runtime pm issue in usb_stor_probe2 Greg Kroah-Hartman
2016-10-14 12:26   ` Greg Kroah-Hartman [this message]
2016-10-14 12:26   ` [PATCH 4.8 03/37] usb: dwc3: fix Clear Stall EP command failure Greg Kroah-Hartman
2016-10-14 12:26   ` [PATCH 4.8 04/37] phy: sun4i-usb: Use spinlock to guard phyctl register access Greg Kroah-Hartman
2016-10-14 12:26   ` [PATCH 4.8 05/37] ALSA: ali5451: Fix out-of-bound position reporting Greg Kroah-Hartman
2016-10-14 12:26   ` [PATCH 4.8 06/37] ALSA: usb-audio: Extend DragonFly dB scale quirk to cover other variants Greg Kroah-Hartman
2016-10-14 12:26   ` [PATCH 4.8 07/37] ALSA: usb-line6: use the same declaration as definition in header for MIDI manufacturer ID Greg Kroah-Hartman
2016-10-14 12:26   ` [PATCH 4.8 08/37] mfd: rtsx_usb: Avoid setting ucr->current_sg.status Greg Kroah-Hartman
2016-10-14 12:26   ` [PATCH 4.8 09/37] mfd: atmel-hlcdc: Do not sleep in atomic context Greg Kroah-Hartman
2016-10-14 12:26   ` [PATCH 4.8 10/37] mfd: 88pm80x: Double shifting bug in suspend/resume Greg Kroah-Hartman
2016-10-14 12:26   ` [PATCH 4.8 12/37] xen/x86: Update topology map for PV VCPUs Greg Kroah-Hartman
2016-10-14 12:26   ` [PATCH 4.8 13/37] KVM: PPC: Book3s PR: Allow access to unprivileged MMCR2 register Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 14/37] KVM: MIPS: Drop other CPU ASIDs on guest MMU changes Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 15/37] KVM: arm64: Require in-kernel irqchip for PMU support Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 16/37] KVM: arm/arm64: vgic: Dont flush/sync without a working vgic Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 17/37] KVM: PPC: BookE: Fix a sanity check Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 18/37] arm64: fix dump_backtrace/unwind_frame with NULL tsk Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 19/37] x86/boot: Fix kdump, cleanup aborted E820_PRAM max_pfn manipulation Greg Kroah-Hartman
2016-10-14 12:27     ` Greg Kroah-Hartman
2016-10-14 12:27     ` Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 20/37] x86/irq: Prevent force migration of irqs which are not in the vector domain Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 21/37] x86/pkeys: Make protection keys an "eager" feature Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 22/37] x86/cpu: Rename Merrifield2 to Moorefield Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 23/37] x86/platform/intel-mid: Add Intel Penwell to ID table Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 24/37] x86/platform/intel-mid: Keep SRAM powered on at boot Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 25/37] x86/apic: Get rid of apic_version[] array Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 26/37] arch/x86: Handle non enumerated CPU after physical hotplug Greg Kroah-Hartman
2016-10-14 12:27     ` Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 27/37] x86/mm/pkeys: Do not skip PKRU register if debug registers are not used Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 28/37] x86/dumpstack: Fix x86_32 kernel_stack_pointer() previous stack access Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 29/37] ARM: fix delays Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 30/37] ARM: dts: mvebu: armada-390: add missing compatibility string and bracket Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 33/37] ARM: cpuidle: Fix error return code Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 34/37] Bluetooth: Add a new 04ca:3011 QCA_ROME device Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 35/37] ima: use file_dentry() Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 36/37] tpm: fix a race condition in tpm2_unseal_trusted() Greg Kroah-Hartman
2016-10-14 12:27   ` [PATCH 4.8 37/37] tpm_crb: fix crb_req_canceled behavior Greg Kroah-Hartman
2016-10-14 18:53   ` [PATCH 4.8 00/37] 4.8.2-stable review Shuah Khan
2016-10-15 11:50     ` Greg Kroah-Hartman
2016-10-14 19:16   ` Guenter Roeck
2016-10-15 11:51     ` Greg Kroah-Hartman
     [not found]   ` <58013af0.6106c30a.958bb.adf8@mx.google.com>
2016-10-15 11:38     ` Greg Kroah-Hartman
2016-10-17 16:43       ` Kevin Hilman

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=20161014122549.877055542@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=alexei.starovoitov@gmail.com \
    --cc=bgregg@netflix.com \
    --cc=john.stultz@linaro.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=stable@vger.kernel.org \
    --cc=tglx@linutronix.de \
    /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.