From: David Brownell <david-b@pacbell.net>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: Andrew Morton <akpm@linux-foundation.org>,
me@felipebalbi.com, linux-kernel@vger.kernel.org,
linux-input@vger.kernel.org, felipe.balbi@nokia.com,
dmitry.torokhov@gmail.com, sameo@openedhand.com,
a.p.zijlstra@chello.nl
Subject: Re: lockdep and threaded IRQs (was: ...)
Date: Sun, 1 Mar 2009 14:54:21 -0800 [thread overview]
Message-ID: <200903011454.22280.david-b@pacbell.net> (raw)
In-Reply-To: <alpine.LFD.2.00.0903011035360.29264@localhost.localdomain>
On Sunday 01 March 2009, Thomas Gleixner wrote:
> On Sat, 28 Feb 2009, David Brownell wrote:
> > That seems to presume a hardirq-to-taskirq handoff. But the
> > problem case is taskirq-to-taskirq chaining, through e.g.
> > what set_irq_chip_and_handler() provided. (Details not very
> > amenable to brief emails, just UTSL.)
> >
> > Thing is, I'm not sure a per-IRQ thread can work easily with
> > that chaining. The chained IRQs can need to be handled before
> > the top-level IRQ gets re-enabled. That's why the twl4030-irq
> > code uses just one taskirq thread for all incoming events.
>
> This can be solved by a completion as well.
One of many potential solutions; it's probably a better
use case for a counting semaphore though, especially if
they were all going in parallel. And of course there's
the issue of where that synch code lives...
Though I still don't see any real issue with keeping it
simple and serializing them without creating new threads.
In terms of resource consumption, that simple solution is
clearly superior.
> > (Which of course is rarely more than one at a time, so there's
> > little reason not to share that task between the demuxing code
> > and the events being demuxed. Interrupts that need processing
> > via I2C/SPI/etc are more or less by definition not frequent or
> > performance-critical.)
>
> Then all we need to provide in the generic code is a function which
> does not go through the handle_IRQ_event() logic and calls the action
> handler directly.
That is, something to replace handle_simple_irq() and
handle_edge_irq() flow handlers? (irq_desc.handle_irq)
> Not rocket science to do that and better than using
> a facility which is designed to run in hardirq context and expect that
> it works in thread context without complaints.
The main "complaint" is the pre-existing lockdep breakage. :)
The need to call irq_desc.handle_irq() with IRQs disabled is a
bit strange, but not really a problem; and it ensures consistent
locking for the irq_desc statistics and flag updates.
- Dave
next prev parent reply other threads:[~2009-03-01 22:54 UTC|newest]
Thread overview: 106+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-02-27 19:28 [PATCH 1/2] input: misc: add twl4030-pwrbutton driver Felipe Balbi
2009-02-27 19:28 ` [PATCH 2/2] mfd: twl4030: add twl4030-pwrbutton as our child Felipe Balbi
2009-02-27 20:36 ` Andrew Morton
2009-02-27 21:58 ` David Brownell
2009-02-27 22:09 ` Felipe Balbi
2009-02-27 22:12 ` Andrew Morton
2009-02-27 23:20 ` David Brownell
2009-02-27 23:42 ` Andrew Morton
2009-07-22 19:27 ` Trilok Soni
2009-07-22 22:25 ` David Brownell
2009-02-27 20:33 ` [PATCH 1/2] input: misc: add twl4030-pwrbutton driver Andrew Morton
2009-02-27 20:37 ` Felipe Balbi
2009-02-27 21:50 ` lockdep and threaded IRQs (was: [PATCH 1/2] input: misc: add twl4030-pwrbutton driver) David Brownell
2009-02-27 22:09 ` Andrew Morton
2009-02-27 23:18 ` lockdep and threaded IRQs (was: ...) David Brownell
2009-02-27 23:32 ` Andrew Morton
2009-02-28 0:01 ` Andrew Morton
2009-02-28 2:30 ` David Brownell
2009-02-28 2:39 ` Andrew Morton
2009-02-28 4:46 ` David Brownell
2009-02-28 5:12 ` Andrew Morton
2009-02-28 6:17 ` David Brownell
2009-02-28 11:13 ` Thomas Gleixner
2009-02-28 12:16 ` David Brownell
2009-02-28 16:42 ` Thomas Gleixner
2009-02-28 20:02 ` David Brownell
2009-02-28 20:55 ` Thomas Gleixner
2009-02-28 21:13 ` Thomas Gleixner
2009-02-28 22:37 ` David Brownell
2009-02-28 22:05 ` David Brownell
2009-03-01 9:43 ` Thomas Gleixner
2009-03-01 22:54 ` David Brownell [this message]
2009-03-02 13:16 ` Peter Zijlstra
2009-03-02 21:04 ` David Brownell
2009-03-02 21:16 ` Peter Zijlstra
2009-03-02 21:29 ` Andrew Morton
2009-03-02 21:37 ` David Brownell
2009-03-02 21:41 ` Peter Zijlstra
2009-03-02 22:09 ` David Brownell
2009-03-02 22:19 ` Peter Zijlstra
2009-03-02 22:40 ` David Brownell
2009-03-02 22:51 ` Peter Zijlstra
2009-03-02 23:29 ` David Brownell
2009-03-03 7:45 ` Peter Zijlstra
2009-03-02 22:46 ` lockdep and threaded IRQs David Miller
2009-03-02 22:57 ` Andrew Morton
2009-03-02 23:07 ` Peter Zijlstra
2009-03-02 23:02 ` Peter Zijlstra
2009-03-02 23:35 ` lockdep and threaded IRQs (was: ...) Alan Cox
2009-03-01 22:11 ` David Brownell
2009-02-28 11:48 ` lockdep and threaded IRQs Stefan Richter
2009-02-28 20:19 ` David Brownell
2009-02-28 21:10 ` Stefan Richter
2009-03-02 13:16 ` lockdep and threaded IRQs (was: ...) Peter Zijlstra
2009-03-02 22:10 ` David Brownell
2009-03-02 22:25 ` Peter Zijlstra
2009-03-02 23:20 ` David Brownell
2009-03-02 23:26 ` Ingo Molnar
2009-03-02 23:42 ` David Brownell
2009-03-02 23:53 ` Ingo Molnar
2009-03-03 0:33 ` David Brownell
2009-03-03 0:44 ` Ingo Molnar
2009-03-03 2:37 ` David Brownell
2009-03-03 9:27 ` Peter Zijlstra
2009-03-03 9:45 ` Ingo Molnar
2009-03-03 9:47 ` Alan Cox
2009-03-03 10:03 ` Ingo Molnar
2009-03-03 10:30 ` Alan Cox
2009-03-03 10:39 ` Peter Zijlstra
2009-03-03 10:48 ` Ingo Molnar
2009-03-03 11:13 ` Alan Cox
2009-03-03 11:33 ` Ingo Molnar
2009-03-03 11:19 ` Ingo Molnar
2009-03-18 1:04 ` David Brownell
2009-03-18 2:00 ` David Brownell
2009-03-18 2:14 ` [patch/rfc 0/2] handle_threaded_irq() David Brownell
2009-03-18 2:19 ` [patch/rfc 1/2] GENIRQ: add handle_threaded_irq() flow handler David Brownell
2009-03-18 12:00 ` Felipe Balbi
2009-03-18 18:31 ` David Brownell
2009-03-18 18:32 ` Felipe Balbi
2009-03-18 2:22 ` [patch/rfc 2/2] twl4030: use new " David Brownell
2009-03-03 11:53 ` lockdep and threaded IRQs (was: ...) Thomas Gleixner
2009-03-05 2:49 ` David Brownell
2009-03-06 14:40 ` Thomas Gleixner
2009-03-18 3:06 ` David Brownell
2009-03-02 23:48 ` David Brownell
2009-03-02 23:58 ` Ingo Molnar
2009-03-02 15:13 ` [PATCH] genirq: assert that irq handlers are indeed run in hardirq context Peter Zijlstra
2009-03-02 19:48 ` David Brownell
2009-03-02 22:01 ` [tip:irq/genirq] genirq: assert that irq handlers are indeed running " Peter Zijlstra
2009-03-02 23:15 ` Peter Zijlstra
2009-04-10 7:11 ` Eric Miao
2009-04-10 9:57 ` Thomas Gleixner
2009-02-28 11:20 ` lockdep and threaded IRQs Stefan Richter
2009-02-28 20:10 ` David Brownell
2009-02-28 5:51 ` [PATCH 1/2] input: misc: add twl4030-pwrbutton driver Trilok Soni
2009-02-28 12:05 ` [PATCH] mfd: twl4030: add twl4030-pwrbutton as our child Felipe Balbi
2009-02-28 22:23 ` [PATCH 1/2] input: misc: add twl4030-pwrbutton driver Dmitry Torokhov
2009-03-01 0:30 ` Felipe Balbi
2009-03-01 0:58 ` Dmitry Torokhov
2009-03-01 14:40 ` Felipe Balbi
2009-03-04 9:00 ` Dmitry Torokhov
2009-03-04 10:12 ` Felipe Balbi
2009-03-05 1:38 ` David Brownell
2009-03-05 23:54 ` David Brownell
2009-03-06 9:43 ` Dmitry Torokhov
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=200903011454.22280.david-b@pacbell.net \
--to=david-b@pacbell.net \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=dmitry.torokhov@gmail.com \
--cc=felipe.balbi@nokia.com \
--cc=linux-input@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=me@felipebalbi.com \
--cc=sameo@openedhand.com \
--cc=tglx@linutronix.de \
/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).