linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: Logan Gunthorpe <logang@deltatee.com>,
	Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
	linux-kernel@vger.kernel.org
Cc: Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@kernel.org>, Will Deacon <will@kernel.org>,
	"Paul E . McKenney" <paulmck@kernel.org>,
	Joel Fernandes <joel@joelfernandes.org>,
	Steven Rostedt <rostedt@goodmis.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Kurt Schwemmer <kurt.schwemmer@microsemi.com>,
	Bjorn Helgaas <bhelgaas@google.com>,
	linux-pci@vger.kernel.org
Subject: Re: [PATCH 3/9] pci/switchtec: Don't abuse completion wait queue for poll
Date: Mon, 16 Mar 2020 20:34:34 +0100	[thread overview]
Message-ID: <87v9n4ccvp.fsf@nanos.tec.linutronix.de> (raw)
In-Reply-To: <4d3a997d-ced4-3dbe-d766-0b1e9fc35b29@deltatee.com>

Logan Gunthorpe <logang@deltatee.com> writes:
> On 2020-03-13 11:46 a.m., Sebastian Andrzej Siewior wrote:
>>  1) It cannot work with EPOLLEXCLUSIVE
>
> Why? You don't explain this.

man epoll_ctt(2)

EPOLLEXCLUSIVE (since Linux 4.5)

  Sets an exclusive wakeup mode for the epoll file descriptor that is
  being attached to the target file descriptor, fd.  When a wakeup event
  occurs and multiple epoll file descriptors are attached to the same
  target file using EPOLLEXCLUSIVE, one or more of the epoll file
  descriptors will receive an event with epoll_wait(2).

As this uses complete_all() there is no distinction possible, because
complete_all() wakes up everything.

> And I don't see how this patch would change anything to do with the
> call to poll_wait(). All you've done is open-code the completion.

wake_up_interruptible(x) resolves to: 

     __wake_up(x, TASK_INTERRUPTIBLE, 1, NULL)

which wakes exactly 1 exclusive waiter.

Also the other way round is just working because the waker side uses
complete_all(). Why? Because completion internally defaults to exclusive
mode and complete() wakes exactly one exlusive waiter.

There is a conceptual difference and while it works for that particular
purpose to some extent it's not suitable as a general wait notification
construct.

Thanks,

        tglx

  parent reply	other threads:[~2020-03-16 19:34 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-13 17:46 [PATCH 0/9] Lock ordering documentation and annotation for lockdep Sebastian Andrzej Siewior
2020-03-13 17:46 ` [PATCH 1/9] Documentation: Add lock ordering and nesting documentation Sebastian Andrzej Siewior
2020-03-14 22:57   ` Randy Dunlap
2020-03-16 10:34     ` Sebastian Andrzej Siewior
2020-03-16 15:13       ` Randy Dunlap
2020-03-13 17:46 ` [PATCH 2/9] timekeeping: Split jiffies seqlock Sebastian Andrzej Siewior
2020-03-13 17:46 ` [PATCH 3/9] pci/switchtec: Don't abuse completion wait queue for poll Sebastian Andrzej Siewior
2020-03-13 18:11   ` Logan Gunthorpe
2020-03-14  0:23     ` Thomas Gleixner
2020-03-14  6:01       ` Logan Gunthorpe
2020-03-16 18:52         ` Thomas Gleixner
2020-03-16 19:24           ` Logan Gunthorpe
2020-03-16 19:34     ` Thomas Gleixner [this message]
2020-03-16 21:53       ` Logan Gunthorpe
2020-03-17  0:17         ` Thomas Gleixner
2020-03-17  1:15           ` Logan Gunthorpe
2020-03-13 19:31   ` Peter Zijlstra
2020-03-13 17:46 ` [PATCH 4/9] sched/swait: Prepare usage in completions Sebastian Andrzej Siewior
2020-03-14  0:26   ` Thomas Gleixner
2020-03-13 17:46 ` [PATCH 5/9] completion: Use simple wait queues Sebastian Andrzej Siewior
2020-03-14 15:40   ` Linus Torvalds
2020-03-16 18:55     ` Thomas Gleixner
2020-03-13 17:46 ` [PATCH 6/9] lockdep: Introduce wait-type checks Sebastian Andrzej Siewior
2020-03-31 13:25   ` Geert Uytterhoeven
2020-03-31 13:42     ` Sebastian Andrzej Siewior
2020-03-31 14:55     ` Peter Zijlstra
2020-03-31 15:28       ` Peter Zijlstra
2020-03-31 15:37         ` Frederic Weisbecker
2020-03-13 17:46 ` [PATCH 7/9] lockdep: Add hrtimer context tracing bits Sebastian Andrzej Siewior
2020-03-13 17:47 ` [PATCH 8/9] lockdep: Annotate irq_work Sebastian Andrzej Siewior
2020-03-13 17:47 ` [PATCH 9/9] lockdep: Add posixtimer context tracing bits Sebastian Andrzej Siewior

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=87v9n4ccvp.fsf@nanos.tec.linutronix.de \
    --to=tglx@linutronix.de \
    --cc=bhelgaas@google.com \
    --cc=bigeasy@linutronix.de \
    --cc=joel@joelfernandes.org \
    --cc=kurt.schwemmer@microsemi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=logang@deltatee.com \
    --cc=mingo@kernel.org \
    --cc=paulmck@kernel.org \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=torvalds@linux-foundation.org \
    --cc=will@kernel.org \
    /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).