Re: [PATCH] ARM: Drop support for redundant "keepinitrd" boot-time param.

[Michael Neuling <mikey@neuling.org>]

> NAK.  "keepinitrd" is what people expect on ARM.
> 
> I'm frustrated that whoever invented "retain_initrd" didn't look
> around in the documentation to see if there was any existing
> implementation first.  The generic implementation should be fixed
> IMHO, or at least the option string moved into init/initramfs.c.

Sorry about that.  I'm happy to change retain_initrd to keep keepinitrd.
I don't think many people are using retain_initrd anyway (other than me
and some people I work with).

Otherwise we can support both for a while with something like (and
Robert's patch)....


From: Michael Neuling <mikey@neuling.org>

Make keepinitrd the same as retain_initrd for backwards compatibility
with ARM/AVR32 

Signed-off-by: Michael Neuling <mikey@neuling.org>
---
 init/initramfs.c |    7 +++++++
 1 file changed, 7 insertions(+)

Index: linux-2.6-ozlabs/init/initramfs.c
===================================================================
--- linux-2.6-ozlabs.orig/init/initramfs.c
+++ linux-2.6-ozlabs/init/initramfs.c
@@ -502,6 +502,13 @@ static int __init retain_initrd_param(ch
 }
 __setup("retain_initrd", retain_initrd_param);
 
+/* for backwards compatibility with ARM/AVR32 */
+static int __init keepinitrd_param(char *str)
+{
+	return retain_initrd_param(str);
+}
+__setup("keepinitrd", keepinitrd_param);
+
 extern char __initramfs_start[], __initramfs_end[];
 #ifdef CONFIG_BLK_DEV_INITRD
 #include <linux/initrd.h>