linux-kernel - Re: [RFC] [PATCH 02/62] mpu401:snd_mpu401_uart_new(): split semantic of irq

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20110908133700.GB2310@zhy>
Date:	Thu, 8 Sep 2011 21:37:00 +0800
From:	Yong Zhang <yong.zhang0@...il.com>
To:	Clemens Ladisch <clemens@...isch.de>
Cc:	linux-kernel@...r.kernel.org, tglx@...utronix.de, mingo@...e.hu,
	Jaroslav Kysela <perex@...ex.cz>, Takashi Iwai <tiwai@...e.de>,
	Uwe Kleine-König 
	<u.kleine-koenig@...gutronix.de>, Jiri Kosina <jkosina@...e.cz>,
	Andreas Mohr <andi@...as.de>,
	Lucas De Marchi <lucas.demarchi@...fusion.mobi>,
	Julia Lawall <julia@...u.dk>,
	Ondrej Zary <linux@...nbow-software.org>,
	Dan Carpenter <error27@...il.com>,
	Joe Perches <joe@...ches.com>, alsa-devel@...a-project.org
Subject: Re: [RFC] [PATCH 02/62] mpu401:snd_mpu401_uart_new(): split
 semantic of irq_flags

On Thu, Sep 08, 2011 at 12:53:14PM +0200, Clemens Ladisch wrote:
> Yong Zhang wrote:
> > Now snd_mpu401_uart_new() parameter 'irq_flags' take two role
> > in it: one is the condition to request_irq and the other is
> > the real irq_flags which will be transfered to request_irq().
> > 
> > So add another parameter 'want_irq' to take the role of the
> > first one, this will make it easy to remove IRQF_DISABLED.
> 
> Please note that the irq number is also intended to pass this
> information:

Yes.

this is a bit subtle:
 * @irq: the irq number, -1 if no interrupt for mpu
    
This semantic of 'irq' is kept by the callers IMHO.

 * @irq_flags: the irq request flags (SA_XXX), 0 if irq was already reserved.

So irq_flags has other meaning--if the irq is already reserved.
Maybe my imprecise description make some kind of misunderstanding.

Seems 'irq_reserved' is more meaningful than 'want_irq', yes?

> 
> >   * @irq: the irq number, -1 if no interrupt for mpu
> > ...
> > -	if (irq >= 0 && irq_flags) {
> >  		if (request_irq(irq, snd_mpu401_uart_interrupt, irq_flags,
> 
> Of course, most of snd_mpu401_uart_new()'s users get this wrong and use
> 0 instead of -1, relying on the irq_flags parameter only.  But if these
> are fixed to use irq == -1, we get the same effect without having to
> introduce another parameter.

Hmm, precisely IRQF_DISABLED imply irq is not reserved in some caller.

BTW, I'm not familiar with mpu401, so maybe I'm missing something here.

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