All of lore.kernel.org
 help / color / mirror / Atom feed
From: Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>
To: linux-kernel@vger.kernel.org, linux-acpi@vger.kernel.org
Cc: guz.fnst@cn.fujitsu.com
Subject: logical cpu number is discontinuity
Date: Wed, 18 Jan 2017 14:55:56 +0900	[thread overview]
Message-ID: <587F036C.7050200@jp.fujitsu.com> (raw)

Hi,

On v4.9 and v4.10 kernel, when I booted my box which has two nodes and
each nodes have 48 logical cpus (Hyper Threading is enabled),
the logical cpu number is discontinuity as follows.

node 0:  0-23, 256-279
node 1: 24-47, 280-303

So the following shell script fail to run.
---
#!/bin/bash

for ((cpu = 0; cpu < `nproc`; cpu++))
do
	taskset -c $cpu ./do_work
done
---

I think the logical cpu number should be continuity like as v4.8 and earlier
because user applications may expect the number is continuity.

I believe this behavior was introduced by the patch series beginning with
the following commit.

f7c2883 x86/acpi: Enable acpi to register all possible cpus at boot time

Do anyone have ideas for fix this behavior...?

FYI.
v4.8 kernel, the logical cpu number is continuity as follows.

node 0:  0-23, 48-71
node 1: 24-47, 72-95

- Masayoshi Mizuma

WARNING: multiple messages have this Message-ID (diff)
From: Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>
To: <linux-kernel@vger.kernel.org>, <linux-acpi@vger.kernel.org>
Cc: <guz.fnst@cn.fujitsu.com>
Subject: logical cpu number is discontinuity
Date: Wed, 18 Jan 2017 14:55:56 +0900	[thread overview]
Message-ID: <587F036C.7050200@jp.fujitsu.com> (raw)

Hi,

On v4.9 and v4.10 kernel, when I booted my box which has two nodes and
each nodes have 48 logical cpus (Hyper Threading is enabled),
the logical cpu number is discontinuity as follows.

node 0:  0-23, 256-279
node 1: 24-47, 280-303

So the following shell script fail to run.
---
#!/bin/bash

for ((cpu = 0; cpu < `nproc`; cpu++))
do
	taskset -c $cpu ./do_work
done
---

I think the logical cpu number should be continuity like as v4.8 and earlier
because user applications may expect the number is continuity.

I believe this behavior was introduced by the patch series beginning with
the following commit.

f7c2883 x86/acpi: Enable acpi to register all possible cpus at boot time

Do anyone have ideas for fix this behavior...?

FYI.
v4.8 kernel, the logical cpu number is continuity as follows.

node 0:  0-23, 48-71
node 1: 24-47, 72-95

- Masayoshi Mizuma

             reply	other threads:[~2017-01-18  6:08 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-01-18  5:55 Masayoshi Mizuma [this message]
2017-01-18  5:55 ` logical cpu number is discontinuity Masayoshi Mizuma

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=587F036C.7050200@jp.fujitsu.com \
    --to=m.mizuma@jp.fujitsu.com \
    --cc=guz.fnst@cn.fujitsu.com \
    --cc=linux-acpi@vger.kernel.org \
    --cc=linux-kernel@vger.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 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.