linux-i2c.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jean Delvare <khali-PUYAD+kWke1g9hUCZPvPmw@public.gmane.org>
To: Mika Kuoppala <mika.kuoppala-xNZwKgViW5gAvxtiuMwx3w@public.gmane.org>
Cc: ben-linux-elnMNo+KYs3YtjvyW6yDsg@public.gmane.org,
	linux-i2c-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH 0/1] Better i2c access latencies in high load situations
Date: Wed, 16 Sep 2009 13:49:44 +0200	[thread overview]
Message-ID: <20090916134944.4a329d62@hyperion.delvare> (raw)
In-Reply-To: <1253099829-17655-1-git-send-email-mika.kuoppala-xNZwKgViW5gAvxtiuMwx3w@public.gmane.org>

Hi Mika,

On Wed, 16 Sep 2009 14:17:08 +0300, Mika Kuoppala wrote:
> If low priority thread is holding the bus lock while more high
> priority threads needs the same i2c bus, priority inversion occurs and
> access latency can grow quite large. In my setup i have seen as high as
> 150ms access latencies in some workloads when the actual physical
> transaction is finished in less than 5ms.
> 
> The following patch (in separate email) fixes the priority
> inversion problem described above by converting the i2c bus lock mutex
> to rt_mutex. rt_mutex uses priority inheritance: low priority thread
> holding the mutex will get a kick if high priority thread is trying
> to acquire the lock.

Can you please define "get a kick"? I don't know anything about
rt_mutex.

> 
> Thanks,
> -- Mika
> 
> Mika Kuoppala (1):
>   i2c: Prevent priority inversion on top of bus lock by converting it
>     to rt_mutex
> 
>  drivers/i2c/i2c-core.c |   13 +++++++------
>  include/linux/i2c.h    |    3 +--
>  2 files changed, 8 insertions(+), 8 deletions(-)
> 


-- 
Jean Delvare

  parent reply	other threads:[~2009-09-16 11:49 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-09-16 11:17 [PATCH 0/1] Better i2c access latencies in high load situations Mika Kuoppala
     [not found] ` <1253099829-17655-1-git-send-email-mika.kuoppala-xNZwKgViW5gAvxtiuMwx3w@public.gmane.org>
2009-09-16 11:17   ` [PATCH 1/1] i2c: Prevent priority inversion on top of bus lock Mika Kuoppala
     [not found]     ` <1253099829-17655-2-git-send-email-mika.kuoppala-xNZwKgViW5gAvxtiuMwx3w@public.gmane.org>
2009-09-16 11:51       ` Jean Delvare
     [not found]         ` <20090916135159.0d74f178-ig7AzVSIIG7kN2dkZ6Wm7A@public.gmane.org>
2009-09-16 12:35           ` Mika Kuoppala
2009-09-16 20:32             ` Jean Delvare
2009-09-16 11:49   ` Jean Delvare [this message]
     [not found]     ` <20090916134944.4a329d62-ig7AzVSIIG7kN2dkZ6Wm7A@public.gmane.org>
2009-09-16 12:08       ` [PATCH 0/1] Better i2c access latencies in high load situations Mika Kuoppala
2009-09-16 20:43         ` Jean Delvare
     [not found]           ` <20090916224328.47e349ab-ig7AzVSIIG7kN2dkZ6Wm7A@public.gmane.org>
2009-09-21 13:14             ` Mika Kuoppala
2009-09-21 16:30               ` Jean Delvare

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=20090916134944.4a329d62@hyperion.delvare \
    --to=khali-puyad+kwke1g9huczpvpmw@public.gmane.org \
    --cc=ben-linux-elnMNo+KYs3YtjvyW6yDsg@public.gmane.org \
    --cc=linux-i2c-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=mika.kuoppala-xNZwKgViW5gAvxtiuMwx3w@public.gmane.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).