x86: add memory hotremove config option

I am resending this patch (originally posted by Badari Pulavarty) since the "mm: cleanup to make remove_memory() arch-neutral" patch on which it depends is now in Linus' 2.6.git tree (commit 71088785c6bc68fddb450063d57b1bd1c78e0ea1) and 2.6.28-rc2.

Thanks,
Gary

---
Add memory hotremove config option to x86

Memory hotremove functionality can currently be configured into the ia64, powerpc, and s390 kernels. This patch makes it possible to configure the memory hotremove functionality into the x86 kernel as well.

---
 arch/x86/Kconfig |    4 ++++
 1 file changed, 4 insertions(+)

Index: linux-2.6.28-rc2/arch/x86/Kconfig
===================================================================
@@ -1486,6 +1486,10 @@
 	def_bool y
 	depends on X86_64 || (X86_32 && HIGHMEM)
 
+config ARCH_ENABLE_MEMORY_HOTREMOVE
+	def_bool y
+	depends on MEMORY_HOTPLUG
+
 config HAVE_ARCH_EARLY_PFN_TO_NID
 	def_bool X86_64
 	depends on NUMA

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

applied to tip/x86/mm, thanks!

Ingo

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

You still didn't say how this is actually going to work and what it is good for? See thread last time. The big difference is that the powerpc and s390 kernels have the needed Hypervisor interfaces, x86 has not (not sure about ia64)

iirc the code is useless for hardware based memory hotplug (because it doesn't free full nodes) and not useful for hypervisor based memory hotplug without additional drivers (and actual hypervisor support of course)

Enabling the sysfs interface now is just giving a promise to the user that you cannot hold. Also it makes the kernel bigger without actually giving useful functionality.

If some x86 hypervisor gains support for this I think the interface shouldn't be through sysfs, but controlled through the respective PV drivers which need to be involved anyways.

-Andi

[email protected]

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>