Linux Power Management development
 help / color / mirror / Atom feed
From: "Rafael J. Wysocki" <rafael.j.wysocki@intel.com>
To: WANG Chao <chao.wang@ucloud.cn>,
	Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
	Vikas Shivappa <vikas.shivappa@linux.intel.com>,
	Kate Stewart <kstewart@linuxfoundation.org>,
	Len Brown <len.brown@intel.com>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Philippe Ombredanne <pombredanne@nexb.com>,
	Mathias Krause <minipli@googlemail.com>,
	x86@kernel.org, "Rafael J. Wysocki" <rafael@kernel.org>,
	Linux PM <linux-pm@vger.kernel.org>
Subject: Re: [PATCH] x86: use cpufreq_quick_get() for /proc/cpuinfo "cpu MHz" again
Date: Thu, 9 Nov 2017 17:06:41 +0100	[thread overview]
Message-ID: <6583ed1f-3ea3-c7fd-7e69-1430c8387e54@intel.com> (raw)
In-Reply-To: <20171109103814.70688-1-chao.wang@ucloud.cn>

Hi Linus,

On 11/9/2017 11:38 AM, WANG Chao wrote:
> Commit 941f5f0f6ef5 (x86: CPU: Fix up "cpu MHz" in /proc/cpuinfo) caused
> a serious performance issue when reading from /proc/cpuinfo on system
> with aperfmperf.
>
> For each cpu, arch_freq_get_on_cpu() sleeps 20ms to get its frequency.
> On a system with 64 cpus, it takes 1.5s to finish running `cat
> /proc/cpuinfo`, while it previously was done in 15ms.

Honestly, I'm not sure what to do to address this ATM.

The last requested frequency is only available in the non-HWP case, so 
it cannot be used universally.

> Some programs use /proc/cpuinfo during startup or runtime, eg. QEMU.
> This patch will restore the previous behavior and let user space happy.
>
> Fixes: 941f5f0f6ef5 (x86: CPU: Fix up "cpu MHz" in /proc/cpuinfo)
> Cc: stable@kernel.org   # 4.13
> Signed-off-by: WANG Chao <chao.wang@ucloud.cn>
> ---
>   arch/x86/kernel/cpu/Makefile | 2 +-
>   arch/x86/kernel/cpu/proc.c   | 4 +---
>   2 files changed, 2 insertions(+), 4 deletions(-)
>
> diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
> index 236999c54edc..c60922a66385 100644
> --- a/arch/x86/kernel/cpu/Makefile
> +++ b/arch/x86/kernel/cpu/Makefile
> @@ -22,7 +22,7 @@ obj-y			+= common.o
>   obj-y			+= rdrand.o
>   obj-y			+= match.o
>   obj-y			+= bugs.o
> -obj-y			+= aperfmperf.o
> +obj-$(CONFIG_CPU_FREQ)	+= aperfmperf.o
>   
>   obj-$(CONFIG_PROC_FS)	+= proc.o
>   obj-$(CONFIG_X86_FEATURE_NAMES) += capflags.o powerflags.o
> diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
> index 4378a729b933..6b7e17bf0b71 100644
> --- a/arch/x86/kernel/cpu/proc.c
> +++ b/arch/x86/kernel/cpu/proc.c
> @@ -78,10 +78,8 @@ static int show_cpuinfo(struct seq_file *m, void *v)
>   		seq_printf(m, "microcode\t: 0x%x\n", c->microcode);
>   
>   	if (cpu_has(c, X86_FEATURE_TSC)) {
> -		unsigned int freq = arch_freq_get_on_cpu(cpu);
> +		unsigned int freq = cpufreq_quick_get(cpu);
>   
> -		if (!freq)
> -			freq = cpufreq_quick_get(cpu);
>   		if (!freq)
>   			freq = cpu_khz;
>   		seq_printf(m, "cpu MHz\t\t: %u.%03u\n",

       reply	other threads:[~2017-11-09 16:06 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20171109103814.70688-1-chao.wang@ucloud.cn>
2017-11-09 16:06 ` Rafael J. Wysocki [this message]
2017-11-09 22:30   ` [PATCH] x86: use cpufreq_quick_get() for /proc/cpuinfo "cpu MHz" again Rafael J. Wysocki
2017-11-10  0:06     ` Rafael J. Wysocki
2017-11-10  4:04       ` WANG Chao
2017-11-10  4:11         ` WANG Chao
2017-11-10 19:11     ` Linus Torvalds
2017-11-10 23:09       ` Rafael J. Wysocki
2017-11-14 22:47         ` Rafael J. Wysocki
2017-11-14 23:02           ` Linus Torvalds
2017-11-14 23:53             ` Thomas Gleixner
2017-11-15  0:04               ` Linus Torvalds
2017-11-15  0:06                 ` Linus Torvalds
2017-11-15  0:30                   ` Rafael J. Wysocki
2017-11-15  0:34                     ` Linus Torvalds
2017-11-15  1:13                       ` [PATCH] x86 / CPU: Always show current CPU frequency in /proc/cpuinfo Rafael J. Wysocki
2017-11-15  8:47                         ` Thomas Gleixner
2017-11-15  9:33                         ` WANG Chao
2017-11-16  0:24                           ` Rafael J. Wysocki
2017-11-16  9:50                             ` WANG Chao
2017-11-16 13:54                               ` Rafael J. Wysocki
2017-11-17  4:27                                 ` WANG Chao
2017-11-17 13:33                                   ` Rafael J. Wysocki
2017-11-15  7:43                     ` [PATCH] x86: use cpufreq_quick_get() for /proc/cpuinfo "cpu MHz" again Ingo Molnar
2017-11-15  7:54                       ` Greg Kroah-Hartman
2017-11-15 17:27                       ` Linus Torvalds
2017-11-15 18:05                         ` Thomas Gleixner
2017-11-15  8:47                 ` Thomas Gleixner
2017-11-15  0:06               ` Rafael J. Wysocki
2017-11-10  7:25   ` Ingo Molnar
2017-11-10  9:21     ` WANG Chao

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=6583ed1f-3ea3-c7fd-7e69-1430c8387e54@intel.com \
    --to=rafael.j.wysocki@intel.com \
    --cc=chao.wang@ucloud.cn \
    --cc=gregkh@linuxfoundation.org \
    --cc=hpa@zytor.com \
    --cc=kstewart@linuxfoundation.org \
    --cc=len.brown@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=minipli@googlemail.com \
    --cc=pombredanne@nexb.com \
    --cc=rafael@kernel.org \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    --cc=vikas.shivappa@linux.intel.com \
    --cc=x86@kernel.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