From: Rasmus Villemoes <linux@rasmusvillemoes.dk>
To: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Thomas Gleixner <tglx@linutronix.de>,
Andrew Morton <akpm@linux-foundation.org>,
Russell King <linux@arm.linux.org.uk>,
Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
linux-kernel@vger.kernel.org, linux-api@vger.kernel.org,
Paul Turner <pjt@google.com>, Andrew Hunter <ahh@google.com>,
Peter Zijlstra <peterz@infradead.org>,
Andy Lutomirski <luto@amacapital.net>,
Andi Kleen <andi@firstfloor.org>,
Dave Watson <davejwatson@fb.com>, Chris Lameter <cl@linux.com>,
Ben Maurer <bmaurer@fb.com>, Steven Rostedt <rostedt@goodmis.org>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
Josh Triplett <josh@joshtriplett.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
Catalin Marinas <catalin.marinas@arm.com>,
Will Deacon <will.deacon@arm.com>,
Michael Kerrisk <mtk.manpages@gmail.com>
Subject: Re: [PATCH v4 1/5] getcpu_cache system call: cache CPU number of running thread
Date: Fri, 26 Feb 2016 00:32:19 +0100 [thread overview]
Message-ID: <87egc0l62k.fsf@rasmusvillemoes.dk> (raw)
In-Reply-To: <alpine.DEB.2.11.1602241206020.3670@nanos> (Thomas Gleixner's message of "Wed, 24 Feb 2016 12:11:15 +0100 (CET)")
On Wed, Feb 24 2016, Mathieu Desnoyers <mathieu.desnoyers@efficios.com> wrote:
>
> Typically, a library or application will keep the cpu number
> cache in a thread-local storage variable, or other memory
> areas belonging to each thread. It is recommended to perform
> a volatile read of the cpu number cache to prevent the com‐
> piler from doing load tearing. An alternative approach is to
> read the cpu number cache from inline assembly in a single
> instruction.
>
> Each thread is responsible for registering its own cpu number
> cache. Only one cpu cache address can be registered per
> thread.
>
> The symbol __getcpu_cache_tls is recommended to be used
> across libraries and applications wishing to register a
> thread-local getcpu_cache. The attribute "weak" is recom‐
> mended when declaring this variable in libraries. Applica‐
> tions can choose to define their own version of this symbol
> without the weak attribute as a performance improvement.
>
> In a typical usage scenario, the thread registering the cpu
> number cache will be performing reads from that cache. It is
> however also allowed to read the cpu number cache from other
> threads. The cpu number cache updates performed by the kernel
> provide single-copy atomicity semantics, which guarantee that
> other threads performing single-copy atomic reads of the cpu
> number cache will always observe a consistent value.
>
> Memory registered as cpu number cache should never be deallo‐
> cated before the thread which registered it exits: specifi‐
> cally, it should not be freed, and the library containing the
> registered thread-local storage should not be dlclose'd.
Maybe spell out the consequence if this is violated - since the SIGSEGV
only happens on migration, it may take a while to strike.
Random thoughts: The current implementation ensures that getcpu_cache is
"idempotent" from within a single thread - once set, it can never get
unset nor set to some other pointer. I think that can be useful, since
it means a library can reliably use the TLS variable itself (initialized
with some negative number) as an indicator of whether
getcpu_cache(GETCPU_CACHE_SET) has been called. So if a single test on a
fast path where the library would need to load __getcpu_cache_tls anyway
is acceptable, it can avoid requiring some library init function to be
called in each thread - which can sometimes be hard to arrange. Is this
something we want to guarantee - that is, will we never implement
GETCPU_CACHE_UNSET or a "force" flag to _SET? Either way, I think we
should spend a few words on it to avoid the current behaviour becoming
accidental ABI.
In another thread:
> However, there are other use-cases for having a fast mechanism for
> reading the current CPU number, besides restartable sequences. For
> instance, it can be used by glibc to implement a faster sched_getcpu.
Will glibc do that? It may be a little contentious for glibc to claim a
unique resource such as task_struct::cpu_cache for itself, even if
everybody is supposed to use the same symbol. Hm, maybe one could say
that if an application does define the symbol __getcpu_cache_tls (which
is techically in the implementation namespace), that gives glibc (and
any other library) license to do getcpu_cache(SET, &&__getcpu_cache_tls)
(pseudo-code, of course). If a library initializes its own weak version
with -2 it can check whether the application defined
__getcpu_cache_tls. Ok, I'm probably overthinking this...
Rasmus
next prev parent reply other threads:[~2016-02-25 23:32 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-02-23 23:28 [PATCH v4 0/5] getcpu_cache system call for 4.6 Mathieu Desnoyers
2016-02-23 23:28 ` [PATCH v4 1/5] getcpu_cache system call: cache CPU number of running thread Mathieu Desnoyers
2016-02-24 11:11 ` Thomas Gleixner
2016-02-24 17:17 ` Mathieu Desnoyers
2016-02-25 23:32 ` Rasmus Villemoes [this message]
2016-02-26 17:47 ` Mathieu Desnoyers
2016-02-25 9:56 ` Peter Zijlstra
2016-02-25 16:55 ` Mathieu Desnoyers
2016-02-25 17:04 ` Peter Zijlstra
2016-02-25 17:17 ` Mathieu Desnoyers
2016-02-26 11:33 ` Peter Zijlstra
2016-02-26 16:29 ` Thomas Gleixner
2016-02-26 17:20 ` Mathieu Desnoyers
2016-02-26 18:01 ` Thomas Gleixner
2016-02-26 20:24 ` Mathieu Desnoyers
2016-02-26 23:04 ` H. Peter Anvin
2016-02-27 0:40 ` Mathieu Desnoyers
2016-02-27 6:24 ` H. Peter Anvin
2016-02-27 14:15 ` Mathieu Desnoyers
2016-02-27 14:58 ` Peter Zijlstra
2016-02-27 18:35 ` Linus Torvalds
2016-02-27 19:01 ` H. Peter Anvin
2016-02-27 23:53 ` Mathieu Desnoyers
[not found] ` <CA+55aFwcgwRxvVBz5kk_3O8dESXAGJ4KHBkf=pSXjiS7Xh4NwA@mail.gmail.com>
[not found] ` <1082926946.10326.1456619994590.JavaMail.zimbra@efficios.com>
2016-02-28 0:57 ` Linus Torvalds
2016-02-28 14:32 ` Mathieu Desnoyers
2016-02-29 10:35 ` Peter Zijlstra
2016-03-01 20:23 ` Mathieu Desnoyers
2016-03-01 21:32 ` Peter Zijlstra
2016-03-01 21:36 ` Peter Zijlstra
2016-03-01 21:47 ` H. Peter Anvin
2016-03-02 10:34 ` Peter Zijlstra
2016-02-29 10:32 ` Peter Zijlstra
2016-02-29 10:39 ` Arnd Bergmann
2016-02-29 12:41 ` Mathieu Desnoyers
2016-02-29 13:08 ` Arnd Bergmann
2016-02-29 18:19 ` H. Peter Anvin
2016-03-02 10:44 ` Geert Uytterhoeven
2016-03-01 18:25 ` H. Peter Anvin
2016-03-01 18:40 ` Mathieu Desnoyers
2016-02-28 13:07 ` Geert Uytterhoeven
2016-02-28 16:21 ` Linus Torvalds
2016-02-29 10:01 ` Peter Zijlstra
2016-02-27 15:04 ` H. Peter Anvin
2016-02-23 23:28 ` [PATCH v4 2/5] getcpu_cache: ARM resume notifier Mathieu Desnoyers
2016-02-23 23:28 ` [PATCH v4 3/5] getcpu_cache: wire up ARM system call Mathieu Desnoyers
2016-02-24 0:54 ` kbuild test robot
2016-02-24 1:05 ` [PATCH v4 (updated)] " Mathieu Desnoyers
2016-02-24 5:28 ` kbuild test robot
2016-02-24 6:54 ` kbuild test robot
2016-02-23 23:28 ` [PATCH v4 4/5] getcpu_cache: x86 32/64 resume notifier Mathieu Desnoyers
2016-02-23 23:28 ` [PATCH v4 5/5] getcpu_cache: wire up x86 32/64 system call Mathieu Desnoyers
2016-02-24 1:36 ` [PATCH v4 0/5] getcpu_cache system call for 4.6 H. Peter Anvin
2016-02-24 4:09 ` Mathieu Desnoyers
2016-02-24 20:07 ` H. Peter Anvin
2016-02-24 22:38 ` Mathieu Desnoyers
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=87egc0l62k.fsf@rasmusvillemoes.dk \
--to=linux@rasmusvillemoes.dk \
--cc=ahh@google.com \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=bmaurer@fb.com \
--cc=catalin.marinas@arm.com \
--cc=cl@linux.com \
--cc=davejwatson@fb.com \
--cc=hpa@zytor.com \
--cc=josh@joshtriplett.org \
--cc=linux-api@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@arm.linux.org.uk \
--cc=luto@amacapital.net \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@redhat.com \
--cc=mtk.manpages@gmail.com \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
--cc=pjt@google.com \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=will.deacon@arm.com \
/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).