Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1592138

Re: [PATCH 02/26] rewrite READ_ONCE/WRITE_ONCE

From Peter Zijlstra <peterz@infradead.org>
Newsgroups linux.kernel
Subject Re: [PATCH 02/26] rewrite READ_ONCE/WRITE_ONCE
Date 2017-03-03 18:00 +0100
Message-ID <tgYIG-3UX-27@gated-at.bofh.it> (permalink)
References (2 earlier) <tgCf8-56J-47@gated-at.bofh.it> <tgDbb-5KR-11@gated-at.bofh.it> <tgGLN-8cb-35@gated-at.bofh.it> <tgGLN-8cb-33@gated-at.bofh.it> <tgVB7-1Rv-7@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw

On Fri, Mar 03, 2017 at 09:26:50AM +0100, Christian Borntraeger wrote:
> Right. The main purpose is to read/write _ONCE_. You can assume a somewhat
> atomic access for sizes <= word size. And there are certainly places that
> rely on that. But the *ONCE thing is mostly used for things where we used
> barrier() 10 years ago.

A lot of code relies on READ/WRITE_ONCE() to generate single
instructions for naturally aligned machined word sized loads/stores
(something GCC used to guarantee, but does no longer IIRC).

So much so that I would say its a bug if READ/WRITE_ONCE() doesn't
generate a single instruction under those conditions.

However, every time I've tried to introduce stricter
semantics/primitives to verify things Linus hated it.

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH 02/26] rewrite READ_ONCE/WRITE_ONCE Arnd Bergmann <arnd@arndb.de> - 2017-03-02 18:00 +0100
  Re: [PATCH 02/26] rewrite READ_ONCE/WRITE_ONCE Christian Borntraeger <borntraeger@de.ibm.com> - 2017-03-02 18:00 +0100
    Re: [PATCH 02/26] rewrite READ_ONCE/WRITE_ONCE Arnd Bergmann <arnd@arndb.de> - 2017-03-02 19:00 +0100
      Re: [PATCH 02/26] rewrite READ_ONCE/WRITE_ONCE Arnd Bergmann <arnd@arndb.de> - 2017-03-02 22:50 +0100
        Re: [PATCH 02/26] rewrite READ_ONCE/WRITE_ONCE Arnd Bergmann <arnd@arndb.de> - 2017-03-03 14:40 +0100
        Re: [PATCH 02/26] rewrite READ_ONCE/WRITE_ONCE Christian Borntraeger <borntraeger@de.ibm.com> - 2017-03-03 14:40 +0100
          Re: [PATCH 02/26] rewrite READ_ONCE/WRITE_ONCE Peter Zijlstra <peterz@infradead.org> - 2017-03-03 18:00 +0100
            Re: [PATCH 02/26] rewrite READ_ONCE/WRITE_ONCE Peter Zijlstra <peterz@infradead.org> - 2017-03-03 18:30 +0100
      Re: [PATCH 02/26] rewrite READ_ONCE/WRITE_ONCE Christian Borntraeger <borntraeger@de.ibm.com> - 2017-03-03 04:20 +0100

csiph-web