From: Arnd Bergmann <arnd@arndb.de>
To: Bernd Schmidt <bernds_cb1@t-online.de>
Subject: Re: [PATCH 1/4] Blackfin: arch patch for 2.6.18
Date: Thu, 28 Sep 2006 13:04:30 +0200
User-Agent: KMail/1.9.4
Cc: Robin Getz <rgetz@blackfin.uclinux.org>, luke Yang <luke.adi@gmail.com>,
       Andrew Morton <akpm@osdl.org>, linux-kernel@vger.kernel.org
References: <6.1.1.1.0.20060927121508.01ecea90@ptg1.spd.analog.com> <200609272257.02385.arnd@arndb.de> <451B9675.8070406@t-online.de>
In-Reply-To: <451B9675.8070406@t-online.de>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 8BIT
Content-Disposition: inline
Message-Id: <200609281304.31872.arnd@arndb.de>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 926
Lines: 25

On Thursday 28 September 2006 11:31, Bernd Schmidt wrote:
> > 
> > That completely avoids all the problems you might hit with macro expansion,
> > while still compiling to the same code.
> 
> The operand is an input, not an output. 

ok, my fault.

> We want to restore the proper  
> mask of enabled interrupts with the STI. ?That mask is in the global 
> irq_flags variable (which probably ought to have a different name that 
> doesn't invite clashes).

Shouldn't you just use a constant expression here? A global variable
for it sounds rather strange, especially since the local_irq_disable()
calls are sometimes nested, not to mention the problems you'd hit on
SMP?

	Arnd <><
-
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/