From: Oleg Nesterov <oleg@redhat.com>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Matt Fleming <matt@console-pimps.org>,
Tejun Heo <htejun@gmail.com>,
linux-kernel@vger.kernel.org, Tony Luck <tony.luck@intel.com>,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [RFC][PATCH 0/5] Signal scalability series
Date: Tue, 4 Oct 2011 19:14:37 +0200 [thread overview]
Message-ID: <20111004171437.GA3489@redhat.com> (raw)
In-Reply-To: <1317655320.20367.25.camel@twins>
On 10/03, Peter Zijlstra wrote:
>
> On Mon, 2011-10-03 at 15:07 +0200, Oleg Nesterov wrote:
> > On 10/01, Peter Zijlstra wrote:
>
> > But this series can't help afaics. At least in its current state. It
> > only adds more locking to the sending paths.
>
> Right, so I was hoping Matt had a plan (TM)... :-)
Whatever we do with the locking, this can't remove O(nr_threads),
although read_lock() could help to reduce the contention.
> > I think, the best solution would be: never send the signal from irq
> > context, and ->siglock shouldn't disable irqs.
>
> Bit hard that, posix timers need to deliver signals which pretty much
> mandates we do something from irq context
Of course. We should notify a thread even if it blocks the signal.
> (and the round-trip through
> softirq context really isn't pretty nor good for performance).
No, no. I meant, it would be nice to offload the sending to the target.
The process itself should take care.
> > Probably this is possible too. I was thinking anout this when
> > set_current_blocked() was added. Unfortunately this needs a lot of
> > complications.
>
> Right, so the thing Thomas and I have been promoting for a while now is
> to update a signal target vector on every signal mask update. Mask
> updates should be the slow path. This would leave us with a ready target
> in O(1).
Yes. This is the "obvious" solution ;) Now that we have
set_current_blocked() this is simple. Except, of course, this blows
signal_struct and set_current_blocked() can't rely on TIF_SIGPENDING.
But we can probably add TIF_YOU_ARE_LISTED_IN_CURR_TARGET_ARRAY.
And in fact this was _one_ of the reasons for set_current_blocked().
> > Agreed. But I am not sure how much we should split the locking when
> > it comes to sending/dequeueing/etc signals. 5 locks seems too much.
>
> It doesn't need all 5 locks to send a signal, does it?
Depending on group/private 3 or 4. Plus ->ctrl_lock if the signal
is fatal but I don't really understand this part... Fortunately I
think this is not needed ;)
And mostly (afaics) this tries to help to dequeue the signal, not
to send. At least currently.
Oleg.
next prev parent reply other threads:[~2011-10-04 17:18 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-09-30 15:12 [RFC][PATCH 0/5] Signal scalability series Matt Fleming
2011-09-30 15:12 ` [RFC][PATCH 1/5] signal: Document signal locking rules Matt Fleming
2011-09-30 15:12 ` [RFC][PATCH 2/5] signal: Add rwlock to protect sighand->action Matt Fleming
2011-09-30 15:25 ` Peter Zijlstra
2011-09-30 15:56 ` Matt Fleming
2011-09-30 15:12 ` [RFC][PATCH 3/5] signal: Reduce sighand->siglock hold time in get_signal_to_deliver() Matt Fleming
2011-09-30 15:12 ` [RFC][PATCH 4/5] signal: Add signal->ctrl_lock for job control Matt Fleming
2011-09-30 15:30 ` Peter Zijlstra
2011-09-30 15:36 ` Matt Fleming
2011-09-30 15:12 ` [RFC][PATCH 5/5] signal: Split siglock into shared_siglock and per-thread siglock Matt Fleming
2011-09-30 16:52 ` [RFC][PATCH 0/5] Signal scalability series Oleg Nesterov
2011-09-30 18:54 ` Oleg Nesterov
2011-09-30 20:00 ` Matt Fleming
2011-09-30 23:56 ` Tejun Heo
2011-10-01 10:16 ` Matt Fleming
2011-10-01 13:03 ` Peter Zijlstra
2011-10-03 1:38 ` Tejun Heo
2011-10-03 13:56 ` Thomas Gleixner
2011-10-04 7:37 ` Tejun Heo
2011-10-03 13:07 ` Oleg Nesterov
2011-10-03 15:22 ` Peter Zijlstra
2011-10-04 17:14 ` Oleg Nesterov [this message]
2011-10-04 17:52 ` Oleg Nesterov
2011-10-04 17:54 ` Linus Torvalds
2011-10-04 18:13 ` Oleg Nesterov
2011-10-03 13:16 ` Oleg Nesterov
2011-10-04 8:56 ` Matt Fleming
2011-10-04 17:29 ` Oleg Nesterov
2011-09-30 22:30 ` Andi Kleen
2011-10-01 9:35 ` Matt Fleming
2011-10-03 15:28 ` Linus Torvalds
2011-10-03 15:43 ` Matt Fleming
2011-10-03 16:35 ` Oleg Nesterov
2011-10-03 20:58 ` Matt Fleming
2011-10-03 21:45 ` Linus Torvalds
2011-10-03 22:13 ` Thomas Gleixner
2011-10-04 8:20 ` Matt Fleming
2011-10-04 17:39 ` Oleg Nesterov
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=20111004171437.GA3489@redhat.com \
--to=oleg@redhat.com \
--cc=a.p.zijlstra@chello.nl \
--cc=htejun@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=matt@console-pimps.org \
--cc=tglx@linutronix.de \
--cc=tony.luck@intel.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).