From: Guo Ren <ren_guo@c-sky.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org,
daniel.lezcano@linaro.org, jason@lakedaemon.net, arnd@arndb.de,
c-sky_gcc_upstream@c-sky.com, gnu-csky@mentor.com,
thomas.petazzoni@bootlin.com, wbx@uclibc-ng.org,
green.hu@gmail.com
Subject: Re: [PATCH V2 18/19] clocksource: add C-SKY clocksource drivers
Date: Thu, 5 Jul 2018 13:03:12 +0800 [thread overview]
Message-ID: <20180705050311.GA20088@guoren> (raw)
In-Reply-To: <alpine.DEB.2.21.1807041632140.1752@nanos.tec.linutronix.de>
On Wed, Jul 04, 2018 at 04:35:43PM +0200, Thomas Gleixner wrote:
> On Wed, 4 Jul 2018, Guo Ren wrote:
> > On Tue, Jul 03, 2018 at 11:39:05AM +0200, Thomas Gleixner wrote:
> > > > +static inline u64 get_ccvr(void)
> > > > +{
> > > > + u32 lo, hi, t;
> > > > +
> > > > + do {
> > > > + hi = mfcr(PTIM_CCVR_HI);
> > > > + lo = mfcr(PTIM_CCVR_LO);
> > > > + t = mfcr(PTIM_CCVR_HI);
> > > > + } while(t != hi);
> > >
> > > No idea which frequency this timer ticks at, but if the 32 bit wrap does
> > > not come too fast, then you really should avoid that loop. That function is
> > > called very frequently.
> >
> > 0000006c <clksrc_read>:
> > hi = mfcr(PTIM_CCVR_HI);
> > 6c: c1c26023 mfcr r3, cr<2, 14>
> > lo = mfcr(PTIM_CCVR_LO);
> > 70: c1c36021 mfcr r1, cr<3, 14>
> > t = mfcr(PTIM_CCVR_HI);
> > 74: c1c26022 mfcr r2, cr<2, 14>
> > } while(t != hi);
> > 78: 648e cmpne r3, r2
> > 7a: 0bf9 bt 0x6c // 6c <clksrc_read>
> >
> > When two read cr<2, 14> is not equal, we'll retry. So only when
> > CCVR_LO is at 0xffffffff between the two read of CCVR_HI. That's very
> > very small probability event for "bt 0x6c".
> >
> > Don't worry about the "do {...} whie(t != hi)", it's no performance issue.
>
> But _three_ mfcr plus a conditional jump which _cannot_ be predicted are a
> performance issue. When you can replace that with a single mfcr, then you
> win a lot, really. The time keeping and the sched clock code can handle
> that nicely unless you really have fast wrap arounds on the LO word.
Timer's frequency is about 50Mhz-100Mhz and LO word wrap arounds timer is
about 42s ~ 84s.
Our Branch prediction buffer will let the CPU speculative execute
continually. So the "bt" won't be the performance issue.
And I could modify it like this:
static inline u64 get_ccvr(void)
{
u32 lo, hi, t;
t = mfcr(PTIM_CCVR_LO);
hi = mfcr(PTIM_CCVR_HI);
lo = mfcr(PTIM_CCVR_LO);
if (lo < t) hi++;
return ((u64)hi << 32) | lo;
}
t = mfcr(PTIM_CCVR_LO);
50: c1c36023 mfcr r3, cr<3, 14>
hi = mfcr(PTIM_CCVR_HI);
54: c1c26021 mfcr r1, cr<2, 14>
lo = mfcr(PTIM_CCVR_LO);
58: c1c3602c mfcr r12, cr<3, 14>
if (lo < t) hi++;
5c: 64f0 cmphs r12, r3
5e: c4210c21 incf r1, r1, 1
return ((u64)hi << 32) | lo;
62: 3200 movi r2, 0
Hmm? (No jump at all)
Guo Ren
next prev parent reply other threads:[~2018-07-05 5:03 UTC|newest]
Thread overview: 87+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-07-01 17:30 [PATCH V2 00/19] C-SKY(csky) Linux Kernel Port Guo Ren
2018-07-01 17:30 ` [PATCH V2 01/19] csky: Build infrastructure Guo Ren
2018-07-01 21:01 ` Randy Dunlap
2018-07-02 1:13 ` Guo Ren
2018-07-03 3:33 ` Rob Herring
2018-07-03 9:14 ` Guo Ren
2018-07-03 16:03 ` Arnd Bergmann
2018-07-04 11:40 ` Guo Ren
2018-07-01 17:30 ` [PATCH V2 02/19] csky: defconfig Guo Ren
2018-07-03 3:16 ` Rob Herring
2018-07-03 8:31 ` Guo Ren
2018-07-01 17:30 ` [PATCH V2 03/19] csky: Kernel booting Guo Ren
2018-07-01 17:30 ` [PATCH V2 04/19] csky: Exception handling Guo Ren
2018-07-01 17:30 ` [PATCH V2 05/19] csky: System Call Guo Ren
2018-07-03 19:53 ` Arnd Bergmann
2018-07-04 11:49 ` Guo Ren
2018-07-04 21:04 ` Arnd Bergmann
2018-07-05 5:38 ` Guo Ren
2018-07-01 17:30 ` [PATCH V2 06/19] csky: Cache and TLB routines Guo Ren
2018-07-05 17:40 ` Peter Zijlstra
2018-07-07 11:51 ` Guo Ren
2018-07-01 17:30 ` [PATCH V2 07/19] csky: MMU and page table management Guo Ren
2018-07-02 13:29 ` Christoph Hellwig
2018-07-03 2:53 ` Guo Ren
2018-07-01 17:30 ` [PATCH V2 08/19] csky: Process management and Signal Guo Ren
2018-07-01 17:30 ` [PATCH V2 09/19] csky: VDSO and rt_sigreturn Guo Ren
2018-07-01 17:30 ` [PATCH V2 10/19] csky: IRQ handling Guo Ren
2018-07-01 17:30 ` [PATCH V2 11/19] csky: Atomic operations Guo Ren
2018-07-05 17:50 ` Peter Zijlstra
2018-07-06 11:01 ` Guo Ren
2018-07-06 11:56 ` Peter Zijlstra
2018-07-06 12:17 ` Peter Zijlstra
2018-07-07 8:08 ` Guo Ren
2018-07-07 20:10 ` Andrea Parri
2018-07-08 1:05 ` Guo Ren
2018-07-07 7:42 ` Guo Ren
2018-07-07 19:54 ` Andrea Parri
2018-07-08 0:39 ` Guo Ren
2018-07-05 17:59 ` Peter Zijlstra
2018-07-06 11:44 ` Guo Ren
2018-07-06 12:03 ` Peter Zijlstra
2018-07-06 13:01 ` Peter Zijlstra
2018-07-06 14:06 ` Guo Ren
2018-07-05 18:00 ` Peter Zijlstra
2018-07-06 11:48 ` Guo Ren
2018-07-06 12:05 ` Peter Zijlstra
2018-07-06 13:46 ` Guo Ren
2018-07-01 17:30 ` [PATCH V2 12/19] csky: ELF and module probe Guo Ren
2018-07-01 17:30 ` [PATCH V2 13/19] csky: Library functions Guo Ren
2018-07-03 20:04 ` Arnd Bergmann
2018-07-04 10:51 ` Guo Ren
2018-07-01 17:30 ` [PATCH V2 14/19] csky: User access Guo Ren
2018-07-01 17:30 ` [PATCH V2 15/19] csky: Debug and Ptrace GDB Guo Ren
2018-07-01 17:30 ` [PATCH V2 16/19] csky: SMP support Guo Ren
2018-07-05 18:05 ` Peter Zijlstra
2018-07-06 6:07 ` Guo Ren
2018-07-06 9:39 ` Peter Zijlstra
2018-07-06 13:22 ` Guo Ren
2018-07-06 5:24 ` Mark Rutland
2018-07-06 11:32 ` Guo Ren
2018-07-06 11:43 ` Mark Rutland
2018-07-06 12:26 ` Guo Ren
2018-07-06 16:21 ` Mark Rutland
2018-07-07 6:16 ` Guo Ren
2018-07-01 17:30 ` [PATCH V2 17/19] csky: Misc headers Guo Ren
2018-07-01 17:30 ` [PATCH V2 18/19] clocksource: add C-SKY clocksource drivers Guo Ren
2018-07-01 17:34 ` Guo Ren
2018-07-03 9:39 ` Thomas Gleixner
2018-07-04 10:49 ` Guo Ren
2018-07-04 14:35 ` Thomas Gleixner
2018-07-05 5:03 ` Guo Ren [this message]
2018-07-04 17:05 ` Daniel Lezcano
2018-07-05 3:30 ` Guo Ren
2018-07-05 9:23 ` Daniel Lezcano
2018-07-06 5:57 ` Guo Ren
2018-07-01 17:30 ` [PATCH V2 19/19] irqchip: add C-SKY irqchip drivers Guo Ren
2018-07-03 3:27 ` Rob Herring
2018-07-03 7:38 ` Guo Ren
2018-07-03 9:28 ` Thomas Gleixner
2018-07-04 5:08 ` Guo Ren
2018-07-04 6:43 ` Thomas Gleixner
2018-07-04 11:58 ` Guo Ren
2018-07-11 9:51 ` [PATCH V2 00/19] C-SKY(csky) Linux Kernel Port David Howells
2018-07-12 12:51 ` Guo Ren
2018-07-12 16:04 ` Sandra Loosemore
2018-07-13 1:30 ` Guo Ren
2018-07-13 10:23 ` David Howells
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=20180705050311.GA20088@guoren \
--to=ren_guo@c-sky.com \
--cc=arnd@arndb.de \
--cc=c-sky_gcc_upstream@c-sky.com \
--cc=daniel.lezcano@linaro.org \
--cc=gnu-csky@mentor.com \
--cc=green.hu@gmail.com \
--cc=jason@lakedaemon.net \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=tglx@linutronix.de \
--cc=thomas.petazzoni@bootlin.com \
--cc=wbx@uclibc-ng.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).