[PATCH] sparc32: enabling of the 2nd CPU in 2.6.18-rc4

[PATCH] sparc32: enabling of the 2nd CPU in 2.6.18-rc4

[Krzysztof Helt]

[-- Attachment #1: Type: text/plain, Size: 534 bytes --]

From: Krzysztof Helt (krzysztof.h1@wp.pl)

This is smp_setup_cpu_possible_map() move after the patch did by
davem to sparc64.

Signed-off-by: Krzysztof Helt (krzysztof.h1@wp.pl)
---
This boots up the second CPU on my SS20. It is diff-ed agains
2.6.18-rc4 kernel, so maybe it was already posted by someone.

Regards,
Krzysztof


----------------------------------------------------
Games Convention Lipsk 23-27.08.2006 - Nie możesz tam być?
Zobacz relacje na żywo:
http://klik.wp.pl/?adr=www.gc2006.gry.wp.pl&sid=852

[-- Attachment #2: sparc32-smp-fix.diff --]
[-- Type: application/octet-stream, Size: 456 bytes --]

--- linux-2.6.17/arch/sparc/kernel/setup.c	2006-08-09 23:12:53.000000000 +0200
+++ linux-2.6.17smp/arch/sparc/kernel/setup.c	2006-08-24 22:13:56.000000000 +0200
@@ -348,9 +348,9 @@ void __init setup_arch(char **cmdline_p)
 	init_mm.context = (unsigned long) NO_CONTEXT;
 	init_task.thread.kregs = &fake_swapper_regs;
 
-	smp_setup_cpu_possible_map();
-
 	paging_init();
+
+	smp_setup_cpu_possible_map();
 }
 
 static int __init set_preferred_console(void)

Re: [PATCH] sparc32: enabling of the 2nd CPU in 2.6.18-rc4

[David Miller]

From: "Krzysztof Helt" <krzysztof.h1@wp.pl>
Date: Thu, 24 Aug 2006 22:44:51 +0200

> From: Krzysztof Helt (krzysztof.h1@wp.pl)
> 
> This is smp_setup_cpu_possible_map() move after the patch did by
> davem to sparc64.
> 
> Signed-off-by: Krzysztof Helt (krzysztof.h1@wp.pl)

It took me a while to figure out what this patch was
all about, please provide context and reasons why the
change needs to be made.

I changed the changelog to what you should have said,
to give you an idea:

--------------------
[SPARC]: enabling of the 2nd CPU in 2.6.18-rc4

smp_setup_cpu_possible_map() needs to run after paging_init()
so that the in-kernel device tree is setup.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
--------------------

See?  If you had simply mentioned "paging_init() needs to
run first so that the device tree is built and ready" I
wouldn't have even needed to look at your patch :-)