public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* [patch 0/2] memory hotplug prototype for ia64
@ 2004-04-20  6:03 Fumihiro Tersawa
  0 siblings, 0 replies; only message in thread
From: Fumihiro Tersawa @ 2004-04-20  6:03 UTC (permalink / raw)
  To: linux-kernel, lhms-devel

Hello,

This set of 2 patches make Iwamoto-san's memory hotplug
prototype correspond to an IPF NUMA machines.
The patch is against linux 2.6.5 and works with 
Iwamoto-san's memory hotplug patch.

I don't have an IPF NUMA machine. So, I tested on 
Tiger4 (intel's IPF machine) with NUMA node
in emulation.


Changes from Iwamoto-san's patches are:

- Using SRAT for NUMA node.
- Structures of all nodes are initialized at boot.
  In Iwamoto-san's patch, node0 is initialized
  at boot, and after boot, all nodes except node0 are
  initialized by hot-add operation.

Patch Details:

- Patch(1/2) memory hotplug prototype for ia64
  The main patch.
- Patch(2/2) NUMA node emulation
  To emulate NUMA node on non-NUMA machines.


Remainder items:

The following items are being made.
- Support of HotAdd function.
- Correspond to ACPI's SLIT.
  So, it does not use find_next_best_node() function
  (mm/page_alloc.c) which added to linux 2.6.5.
- Support of hugetlbpage.


How to apply:

 1) First of all, apply Iwamoto-san's patches.
 2) Apply this set of patches.


Thank you,
Fumihiro Terasawa.


^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2004-04-20  6:03 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-04-20  6:03 [patch 0/2] memory hotplug prototype for ia64 Fumihiro Tersawa

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox