DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=message-id:date:from:sender:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references:x-google-sender-auth;
        b=nGVe+iwxfn2Y5i960INEUUOgD/TWtUwa7Tu92Kffm9e327o++h0gW82/cyD2hvQMig2llMG+Ah3xqoc7Orfipiss/iUD1f6YSiYmEVppG0gxobq6NlJllXhQ2a1hPviPydDrf9STNZ8W3DUt9Ho70DBlsP3gdw20oYHeMgn90qs=
Message-ID: <3efb10970712061240l3371bd9bj1050f69e5dacf316@mail.gmail.com>
Date: Thu, 6 Dec 2007 21:40:31 +0100
From: "Remy Bohmer" <linux@bohmer.net>
To: "Daniel Walker" <dwalker@mvista.com>
Subject: Re: [PATCH 3/3] printer port driver: semaphore to mutex
Cc: "Ingo Molnar" <mingo@elte.hu>, linux-kernel@vger.kernel.org
In-Reply-To: <1196971945.1568.107.camel@jnielson-xp.ddns.mvista.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
References: <20071206021857.826386004@mvista.com>
	 <20071206021909.131993721@mvista.com> <20071206102334.GE17299@elte.hu>
	 <1196958847.1568.98.camel@jnielson-xp.ddns.mvista.com>
	 <3efb10970712061201l343ad865mae920c3e6d43bfd9@mail.gmail.com>
	 <1196971945.1568.107.camel@jnielson-xp.ddns.mvista.com>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1411
Lines: 37

Hello Daniel,

> I looked at the console_sem , but i was going to leave that as last..
>
> The problem with the console_sem is that it can get locked from
> interrupt context, which is discouraged with mutex types.. I think it
> will be complicated to convert..

At first it looked simple, but after I sent my mail, I saw also the
interrupt issue...
But nevertheless, I am going to look at it anyway... (I think not today)

Further, I looked also at include/linux/device.h where in the 'struct
device', a semaphore is listed that is used at several places as
mutex. I was wondering what was wise here:
* replacing the semaphore with a mutex at once, and replace it
everywhere in the kernel also, or
* just add a new mutex to the struct, and change the the semaphore to
a mutex per driver, and eventually remove the semaphore from the
struct.

The first option creates the biggest patch, the 2nd leaves the risk
for never removing the semaphore from the struct device...

What is your opinion?


BTW: I also will look at the dvb_frontend.c semaphore->mutex.
I also want to look at the mutex in the PS3 code, but I cannot test it
before I receive my new PS3 ;-)

Remy
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/