[PATCH] ppc64: fix prom_init calculation for alloc_bottom

[Arnd Bergmann]

X-BeenThere: linuxppc64-dev at ozlabs.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: 64-bit Linux on PowerPC Developers Mail List
	<linuxppc64-dev.ozlabs.org>
List-Unsubscribe: <https://ozlabs.org/cgi-bin/mailman/listinfo/linuxppc64-dev>, 
	<mailto:linuxppc64-dev-request at ozlabs.org?subject=unsubscribe>
List-Archive: <http://ozlabs.org/pipermail/linuxppc64-dev>
List-Post: <mailto:linuxppc64-dev at ozlabs.org>
List-Help: <mailto:linuxppc64-dev-request at ozlabs.org?subject=help>
List-Subscribe: <https://ozlabs.org/cgi-bin/mailman/listinfo/linuxppc64-dev>, 
	<mailto:linuxppc64-dev-request at ozlabs.org?subject=subscribe>
X-List-Received-Date: Fri, 18 Feb 2005 16:12:44 -0000

I haven't tested this, but (prom_initrd_start + prom_initrd_end) just sounds
wrong. The code prior to the monster cleanup was using initrd_len instead of
initrd_end in some places, perhaps this is a leftover of the conversion.

Signed-off-by: Arnd Bergmann <arndb at de.ibm.com>

--- linux-2.6-ppc.orig/arch/ppc64/kernel/prom_init.c	2005-02-18 10:51:02.518955168 -0500
+++ linux-2.6-ppc/arch/ppc64/kernel/prom_init.c	2005-02-18 10:54:23.570006016 -0500
@@ -679,8 +679,7 @@
 	 * point to after it
 	 */
 	if (RELOC(prom_initrd_start)) {
-		if ((RELOC(prom_initrd_start) + RELOC(prom_initrd_end))
-		    > RELOC(alloc_bottom))
+		if (RELOC(prom_initrd_end) > RELOC(alloc_bottom))
 			RELOC(alloc_bottom) = PAGE_ALIGN(RELOC(prom_initrd_end));
 	}