From 641c767389b19859a45e6de46d8e18cd935bdb60 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Thu, 23 Jun 2005 00:07:59 -0700 Subject: [PATCH] [PATCH] sparsemem swiss cheese numa layouts The part of the sparsemem patch which modifies memmap_init_zone() has recently become a problem. It changes behavior so that there is a call to pfn_to_page() for each individual page inside of a node's range: node_start_pfn through node_end_pfn. It used to simply do this once, at the beginning of the node, but having sparsemem's non-contiguous mem_map[]s inside of a node made it necessary to change. Mike Kravetz recently wrote a patch which made the NUMA code accept some new kinds of layouts. The system's memory was laid out like this, with node 0's memory in two pieces: one before and one after node 1's memory: Node 0: +++++ +++++ Node 1: +++++ Previous behavior before Mike's patch was to assign nodes like this: Node 0: 00000 XXXXX Node 1: 11111 Where the 'X' areas were simply thrown away. The new behavior was to make the pg_data_t span node 0 across all of its areas, including areas that are really node 1's: Node 0: 000000000000000 Node 1: 11111 This wastes a little bit of mem_map space, but ends up being OK, and more fully utilizes the system's memory. memmap_init_zone() initializes all of the "struct page"s for node 0, even for the "hole", but those never get used, because there is no pfn_to_page() that resolves to those pages. However, only calling pfn_to_page() once, memmap_init_zone() always uses the pages that were allocated for node0->node_mem_map because: struct page *start = pfn_to_page(start_pfn); // effectively start = &node->node_mem_map[0] for (page = start; page < (start + size); page++) { init_page_here();... page++; } Slow, and wasteful, but generally harmless. But, modify that to call pfn_to_page() for each loop iteration (like sparsemem does): for (pfn = start_pfn; pfn < < (start_pfn + size); pfn++++) { page = pfn_to_page(pfn); } And you end up trying to initialize node 1's pages too early, along with bogus data from node 0. This patch checks for those weird layouts and declines to touch the pages, making the more frequent pfn_to_page() calls OK to do. Signed-off-by: Dave Hansen Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ppc64/Kconfig | 12 ++++++++++++ include/linux/mmzone.h | 6 ++++++ mm/page_alloc.c | 2 ++ 3 files changed, 20 insertions(+) diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig index 011b5c0bf1d..85f8fcf44b6 100644 --- a/arch/ppc64/Kconfig +++ b/arch/ppc64/Kconfig @@ -211,6 +211,18 @@ config ARCH_FLATMEM_ENABLE source "mm/Kconfig" +# Some NUMA nodes have memory ranges that span +# other nodes. Even though a pfn is valid and +# between a node's start and end pfns, it may not +# reside on that node. +# +# This is a relatively temporary hack that should +# be able to go away when sparsemem is fully in +# place +config NODES_SPAN_OTHER_NODES + def_bool y + depends on NEED_MULTIPLE_NODES + config NUMA bool "NUMA support" depends on DISCONTIGMEM diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 19860d317ec..746b57e3d37 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -528,6 +528,12 @@ void sparse_init(void); #define sparse_init() do {} while (0) #endif /* CONFIG_SPARSEMEM */ +#ifdef CONFIG_NODES_SPAN_OTHER_NODES +#define early_pfn_in_nid(pfn, nid) (early_pfn_to_nid(pfn) == (nid)) +#else +#define early_pfn_in_nid(pfn, nid) (1) +#endif + #ifndef early_pfn_valid #define early_pfn_valid(pfn) (1) #endif diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 5c1b8982a6d..1eb683f9b3a 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1656,6 +1656,8 @@ void __init memmap_init_zone(unsigned long size, int nid, unsigned long zone, for (pfn = start_pfn; pfn < end_pfn; pfn++, page++) { if (!early_pfn_valid(pfn)) continue; + if (!early_pfn_in_nid(pfn, nid)) + continue; page = pfn_to_page(pfn); set_page_links(page, zone, nid, pfn); set_page_count(page, 0); -- 2.41.1