Thread (4 messages) 4 messages, 3 authors, 2009-12-22

Re: [PATCH 5/6 v5] CPU probe/release files

From: Nathan Fontenot <hidden>
Date: 2009-12-22 02:17:55
Also in: lkml

Possibly related (same subject, not in this thread)

Benjamin Herrenschmidt wrote:
On Fri, 2009-12-18 at 15:33 +0100, Andreas Schwab wrote:
quoted
Nathan Fontenot [off-list ref] writes:
quoted
Index: powerpc/arch/powerpc/Kconfig
===================================================================
--- powerpc.orig/arch/powerpc/Kconfig	2009-10-28 15:21:47.000000000 -0500
+++ powerpc/arch/powerpc/Kconfig	2009-10-28 15:21:53.000000000 -0500
@@ -320,6 +320,10 @@
 
 	  Say N if you are unsure.
 
+config ARCH_CPU_PROBE_RELEASE
+	def_bool y
+	depends on HOTPLUG_CPU
+
That does not work.

drivers/built-in.o: In function `.store_online':
cpu.c:(.ref.text+0xf5c): undefined reference to `.cpu_hotplug_driver_lock'
cpu.c:(.ref.text+0xfc8): undefined reference to `.cpu_hotplug_driver_unlock'
make: *** [.tmp_vmlinux1] Error 1

cpu_hotplug_driver_lock is only defined on pseries, but HOTPLUG_CPU is
also defined on pmac.
Well, the stuff was merged ... so we need to fix it.

Nathan, the problem is that the above wlil define ARCH_CPU_PROBE_RELEASE
for any powerpc platform that has HOTPLUG_CPU. So a kernel that doesn't
have pSeries support but has hotplug CPU will have a problem, though for
some reason none of my test configs triggered it (I'll have to check why
next week).

The right approach here is to have cpu_hotplug_driver_lock go through
ppc_md. I suppose, along with some of the other CPU hotplug platform
specific bits.
Sounds good to me,  I'll get a patch together.

-Nathan Fontenot
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help