From mboxrd@z Thu Jan  1 00:00:00 1970
From: u.kleine-koenig@pengutronix.de (Uwe =?iso-8859-1?Q?Kleine-K=F6nig?=)
Date: Sun, 7 Dec 2014 12:38:14 +0100
Subject: Do I need to invalidate caches before enabling (on ARMv7)?
In-Reply-To: <20141207112719.GB21341@lunn.ch>
References: <20141204100357.GO2129@pengutronix.de>
 <20141206210031.GA1758@afzal-ThinkPad-R50e>
 <20141207100124.GT2129@pengutronix.de>
 <20141207105823.GA21341@lunn.ch>
 <20141207111103.GV2129@pengutronix.de>
 <20141207112719.GB21341@lunn.ch>
Message-ID: <20141207113814.GW2129@pengutronix.de>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Hey Andrew,

On Sun, Dec 07, 2014 at 12:27:19PM +0100, Andrew Lunn wrote:
> > Even handling the cache being on would be possible, that would result in
> > something like:
> 
> We came to the conclusion it is not possible to handle it. Something
> to do with the MMU i think, but i don't remember. There is a thread in
> the archive, and for the platform which i know is broken, the dts file
> points to:
> 
> https://lists.debian.org/debian-arm/2012/08/msg00128.html
> 
> which contains instructions how to append a few bytes of code to the
> front of the image to turn the cache off.
I remember the problem. It's about L2 caches here which are not used by
the decompressor.

It's similar to the L1 caches that are in use: If the bootloader keeps
this cache on, the decompressor must be aware of it (and be able to
determine if the cache is on to decide if it needs to flush or
invalidate).

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-K?nig            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |