LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
To: Florian Weimer <fweimer@redhat.com>
Cc: Pavel Machek <pavel@ucw.cz>, carlos <carlos@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
	Boqun Feng <boqun.feng@gmail.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	libc-alpha <libc-alpha@sourceware.org>
Subject: Re: Restartable Sequences system call merged into Linux
Date: Thu, 14 Jun 2018 11:09:42 -0400 (EDT)	[thread overview]
Message-ID: <1898936021.13117.1528988982788.JavaMail.zimbra@efficios.com> (raw)
In-Reply-To: <21eae301-50ff-e95a-f3c7-dedcf2f66842@redhat.com>

----- On Jun 14, 2018, at 10:41 AM, Florian Weimer fweimer@redhat.com wrote:

> On 06/14/2018 04:36 PM, Mathieu Desnoyers wrote:
>> ----- On Jun 14, 2018, at 10:00 AM, Florian Weimer fweimer@redhat.com wrote:
>> 
>>> On 06/14/2018 03:49 PM, Pavel Machek wrote:
>>>> Hi!
>>>>
>>>>>>> - rseq_preempt(): on preemption, the scheduler sets the TIF_NOTIFY_RESUME thread
>>>>>>>     flag, so rseq_handle_notify_resume() can check whether it's in a rseq critical
>>>>>>>     section when returning to user-space,
>>>>>>> - rseq_signal_deliver(): on signal delivery, rseq_handle_notify_resume() checks
>>>>>>>     whether it's in a rseq critical section,
>>>>>>> - rseq_migrate: on migration, the scheduler sets TIF_NOTIFY_RESUME as well,
>>>>>>
>>>>>> Yes, this is not likely to be noticeable.
>>>>>>
>>>>>> But the proposal wanted to add a syscall to thread creation, right?
>>>>>> And I believe that may be noticeable.
>>>>>
>>>>> Fair point! Do we have a standard benchmark that would stress this ?
>>>>
>>>> Web server performance benchmarks basically test clone() performance
>>>> in many cases.
>>>
>>> Isn't that fork?  I expect that the rseq arena is inherited on fork and
>>> fork-type clone, otherwise it's going to be painful.
>> 
>> On fork or clone creating a new process, the rseq tls area is inherited
>> from the thread that does the fork syscall.
>> 
>> On creation of a new thread with clone, there is no such inheritance.
> 
> Makes sense.  So fork-based (web) servers will not be impacted by the
> additional system call, and thread-based servers likely use a thread
> pool anyway.  I'm not really concerned about the additional system call
> here.

Just for the sake of completeness, there is (of course) no inheritance
on exec(). So glibc would also have to register the rseq TLS in its
constructors.

Thanks,

Mathieu

-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com

  reply	other threads:[~2018-06-14 15:09 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-11 19:49 Mathieu Desnoyers
2018-06-11 19:55 ` Florian Weimer
2018-06-11 20:04   ` Mathieu Desnoyers
2018-06-12 13:11     ` Florian Weimer
2018-06-12 16:31       ` Mathieu Desnoyers
2018-06-13  8:21         ` Florian Weimer
2018-06-14 12:27         ` Pavel Machek
2018-06-14 13:01           ` Mathieu Desnoyers
2018-06-14 13:25             ` Pavel Machek
2018-06-14 13:32               ` Florian Weimer
2018-06-14 13:46                 ` Mathieu Desnoyers
2018-06-15  5:10                   ` Florian Weimer
2018-06-15 17:44                     ` Mathieu Desnoyers
2018-06-14 13:38               ` Mathieu Desnoyers
2018-06-14 13:49                 ` Pavel Machek
2018-06-14 14:00                   ` Florian Weimer
2018-06-14 14:36                     ` Mathieu Desnoyers
2018-06-14 14:41                       ` Florian Weimer
2018-06-14 15:09                         ` Mathieu Desnoyers [this message]
2018-06-15  5:09             ` Florian Weimer
2018-06-15 17:50               ` Mathieu Desnoyers
2018-06-15  5:07           ` Florian Weimer
2018-06-13 11:48 ` Heiko Carstens
2018-06-13 16:14   ` Mathieu Desnoyers
2018-06-13 19:53     ` 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=1898936021.13117.1528988982788.JavaMail.zimbra@efficios.com \
    --to=mathieu.desnoyers@efficios.com \
    --cc=boqun.feng@gmail.com \
    --cc=carlos@redhat.com \
    --cc=fweimer@redhat.com \
    --cc=libc-alpha@sourceware.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=pavel@ucw.cz \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --subject='Re: Restartable Sequences system call merged into Linux' \
    /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

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).