mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
To: dtor_core@ameritech.net
Cc: dmitry.torokhov@gmail.com, netdev@oss.sgi.com,
	Greg KH <greg@kroah.com>, Jamal Hadi Salim <hadi@cyberus.ca>,
	Kay Sievers <kay.sievers@vrfy.org>,
	Herbert Xu <herbert@gondor.apana.org.au>,
	James Morris <jmorris@redhat.com>,
	Guillaume Thouvenin <guillaume.thouvenin@bull.net>,
	linux-kernel@vger.kernel.org, Andrew Morton <akpm@osdl.org>,
	Thomas Graf <tgraf@suug.ch>, Jay Lan <jlan@engr.sgi.com>
Subject: Re: [1/1] connector/CBUS: new messaging subsystem. Revision number next.
Date: Tue, 26 Apr 2005 23:28:12 +0400	[thread overview]
Message-ID: <20050426232812.0c7bb3a4@zanzibar.2ka.mipt.ru> (raw)
In-Reply-To: <d120d50005042612069b84ef@mail.gmail.com>

On Tue, 26 Apr 2005 14:06:36 -0500
Dmitry Torokhov <dmitry.torokhov@gmail.com> wrote:

> On 4/26/05, Evgeniy Polyakov <johnpol@2ka.mipt.ru> wrote:
> > On Tue, 26 Apr 2005 13:42:10 -0500
> > Dmitry Torokhov <dmitry.torokhov@gmail.com> wrote:
> > > Yes, that woudl work, although I would urge you to implement a message
> > > queue for callbacks (probably limit it to 1000 messages or so) to
> > > allow bursting.
> > 
> > It already exist, btw, but not exactly in that way -
> > we have skb queue, which can not be filled from userspace
> > if pressure is so strong so work queue can not be scheduled.
> > It is of course different and is influenced by other things
> > but it handles bursts quite well - it was tested on both
> > SMP and UP machines with continuous flows of forks with
> > shape addon of new running tasks [both fith fork bomb and not],
> > so I think it can be called real bursty test.
> > 
> 
> Ok, hear me out and tell me where I am wrong:
> 
> By default a socket can receive at least 128 skbs with 258-byte
> payload, correct? That means that user of cn_netlink_send, if started
> "fresh", 128 average - size connector messages. If sender does not
> want to wait for anything (unlike your fork tests that do schedule)
> that means that 127 of those 128 messages will be dropped, although
> netlink would deliver them in time just fine.
> 
> What am I missing?

Concider netlink_broadcast - it delivers skb to the kernel 
listener directly to the input callback - no queueing actually,
the same is for netlink_unicast - so in-kernel users are always 
called synchronously.
send() [sys_send()] system call(which btw as syscall does schedule too) 
ends up in netlink_sendmsg and thus either netlink_broadcast and 
netlink_unicast, which is called directly as we have seen.

> -- 
> Dmitry


	Evgeniy Polyakov

Only failure makes us experts. -- Theo de Raadt

  reply	other threads:[~2005-04-26 19:29 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-04-11 12:59 Evgeniy Polyakov
2005-04-11 13:32 ` Thomas Graf
2005-04-11 14:49   ` [2/1] " Evgeniy Polyakov
2005-04-26 15:57 ` [1/1] " Dmitry Torokhov
2005-04-26 16:24   ` Evgeniy Polyakov
2005-04-26 16:30     ` Evgeniy Polyakov
2005-04-26 17:34       ` Dmitry Torokhov
2005-04-26 18:07         ` Evgeniy Polyakov
2005-04-26 18:20           ` Dmitry Torokhov
2005-04-26 18:31             ` Evgeniy Polyakov
2005-04-26 18:42               ` Dmitry Torokhov
2005-04-26 18:48                 ` Evgeniy Polyakov
2005-04-26 19:06                   ` Dmitry Torokhov
2005-04-26 19:28                     ` Evgeniy Polyakov [this message]
2005-04-26 20:02                       ` Dmitry Torokhov
2005-04-27  4:06                         ` Evgeniy Polyakov
2005-04-27  5:16                           ` Dmitry Torokhov
2005-04-27  5:32                             ` Evgeniy Polyakov
2005-04-27  5:46                               ` Dmitry Torokhov
2005-04-27  6:08                                 ` Evgeniy Polyakov
2005-04-26 17:31     ` Dmitry Torokhov
2005-04-26 18:03       ` Evgeniy Polyakov
2005-04-26 18:10         ` Evgeniy Polyakov
2005-04-26 18:13           ` Evgeniy Polyakov
2005-04-26 18:25             ` Dmitry Torokhov
2005-04-26 18:35               ` Evgeniy Polyakov
2005-05-10  6:18 ` Dmitry Torokhov
2005-05-10 10:04   ` Evgeniy Polyakov
2005-05-10 14:56     ` Dmitry Torokhov
2005-05-10 15:41       ` Evgeniy Polyakov
2005-05-10 17:50         ` Dmitry Torokhov
2005-05-10 18:24           ` Evgeniy Polyakov
2005-05-11  5:46             ` Dmitry Torokhov
2005-05-11  6:48               ` Evgeniy Polyakov
2005-05-11 14:09   ` Alan Cox

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=20050426232812.0c7bb3a4@zanzibar.2ka.mipt.ru \
    --to=johnpol@2ka.mipt.ru \
    --cc=akpm@osdl.org \
    --cc=dmitry.torokhov@gmail.com \
    --cc=dtor_core@ameritech.net \
    --cc=greg@kroah.com \
    --cc=guillaume.thouvenin@bull.net \
    --cc=hadi@cyberus.ca \
    --cc=herbert@gondor.apana.org.au \
    --cc=jlan@engr.sgi.com \
    --cc=jmorris@redhat.com \
    --cc=kay.sievers@vrfy.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@oss.sgi.com \
    --cc=tgraf@suug.ch \
    /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

all inboxes | Powered by JetHome®