Written-By: Andy Whitcroft Signed-off-by: Dave Hansen --- memhotplug-dave/arch/ppc64/Kconfig | 11 ++++++- memhotplug-dave/arch/ppc64/kernel/setup.c | 1 memhotplug-dave/arch/ppc64/mm/Makefile | 2 - memhotplug-dave/arch/ppc64/mm/init.c | 24 +++++++++++++--- memhotplug-dave/include/asm-ppc64/mmzone.h | 41 ++++++++++++++++++++++++----- memhotplug-dave/include/asm-ppc64/page.h | 4 ++ 6 files changed, 68 insertions(+), 15 deletions(-) diff -puN arch/ppc64/Kconfig~B-sparse-170-sparsemem-ppc64 arch/ppc64/Kconfig --- memhotplug/arch/ppc64/Kconfig~B-sparse-170-sparsemem-ppc64 2005-04-13 14:20:20.000000000 -0700 +++ memhotplug-dave/arch/ppc64/Kconfig 2005-04-13 14:20:21.000000000 -0700 @@ -198,10 +198,17 @@ config HMT This option enables hardware multithreading on RS64 cpus. pSeries systems p620 and p660 have such a cpu type. +config ARCH_SELECT_MEMORY_MODEL + def_bool y + config ARCH_DISCONTIGMEM_ENABLE - bool "Discontiguous Memory Support" + def_bool y depends on SMP && PPC_PSERIES +config ARCH_SPARSEMEM_ENABLE + def_bool y + depends on ARCH_DISCONTIGMEM_ENABLE + source "mm/Kconfig" # Some NUMA nodes have memory ranges that span @@ -218,7 +225,7 @@ config NODES_SPAN_OTHER_NODES config NUMA bool "NUMA support" - depends on DISCONTIGMEM + default y if DISCONTIGMEM || SPARSEMEM config SCHED_SMT bool "SMT (Hyperthreading) scheduler support" diff -puN arch/ppc64/kernel/setup.c~B-sparse-170-sparsemem-ppc64 arch/ppc64/kernel/setup.c --- memhotplug/arch/ppc64/kernel/setup.c~B-sparse-170-sparsemem-ppc64 2005-04-13 14:20:20.000000000 -0700 +++ memhotplug-dave/arch/ppc64/kernel/setup.c 2005-04-13 14:20:21.000000000 -0700 @@ -1059,6 +1059,7 @@ void __init setup_arch(char **cmdline_p) /* set up the bootmem stuff with available memory */ do_init_bootmem(); + sparse_init(); /* initialize the syscall map in systemcfg */ setup_syscall_map(); diff -puN arch/ppc64/mm/Makefile~B-sparse-170-sparsemem-ppc64 arch/ppc64/mm/Makefile --- memhotplug/arch/ppc64/mm/Makefile~B-sparse-170-sparsemem-ppc64 2005-04-13 14:20:21.000000000 -0700 +++ memhotplug-dave/arch/ppc64/mm/Makefile 2005-04-13 14:20:21.000000000 -0700 @@ -6,6 +6,6 @@ EXTRA_CFLAGS += -mno-minimal-toc obj-y := fault.o init.o imalloc.o hash_utils.o hash_low.o tlb.o \ slb_low.o slb.o stab.o mmap.o -obj-$(CONFIG_DISCONTIGMEM) += numa.o +obj-$(CONFIG_NUMA) += numa.o obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o obj-$(CONFIG_PPC_MULTIPLATFORM) += hash_native.o diff -puN arch/ppc64/mm/init.c~B-sparse-170-sparsemem-ppc64 arch/ppc64/mm/init.c --- memhotplug/arch/ppc64/mm/init.c~B-sparse-170-sparsemem-ppc64 2005-04-13 14:20:21.000000000 -0700 +++ memhotplug-dave/arch/ppc64/mm/init.c 2005-04-13 14:20:21.000000000 -0700 @@ -601,13 +601,21 @@ EXPORT_SYMBOL(page_is_ram); * Initialize the bootmem system and give it all the memory we * have available. */ -#ifndef CONFIG_DISCONTIGMEM +#ifndef CONFIG_NUMA void __init do_init_bootmem(void) { unsigned long i; unsigned long start, bootmap_pages; unsigned long total_pages = lmb_end_of_DRAM() >> PAGE_SHIFT; int boot_mapsize; + unsigned long start_pfn, end_pfn; + /* + * Note presence of first (logical/coalasced) LMB which will + * contain RMO region + */ + start_pfn = lmb.memory.region[0].physbase >> PAGE_SHIFT; + end_pfn = start_pfn + (lmb.memory.region[0].size >> PAGE_SHIFT); + memory_present(0, start_pfn, end_pfn); /* * Find an area to use for the bootmem bitmap. Calculate the size of @@ -623,12 +631,18 @@ void __init do_init_bootmem(void) max_pfn = max_low_pfn; - /* add all physical memory to the bootmem map. Also find the first */ + /* add all physical memory to the bootmem map. Also, find the first + * presence of all LMBs*/ for (i=0; i < lmb.memory.cnt; i++) { unsigned long physbase, size; physbase = lmb.memory.region[i].physbase; size = lmb.memory.region[i].size; + if (i) { /* already created mappings for first LMB */ + start_pfn = physbase >> PAGE_SHIFT; + end_pfn = start_pfn + (size >> PAGE_SHIFT); + } + memory_present(0, start_pfn, end_pfn); free_bootmem(physbase, size); } @@ -667,7 +681,7 @@ void __init paging_init(void) free_area_init_node(0, &contig_page_data, zones_size, __pa(PAGE_OFFSET) >> PAGE_SHIFT, zholes_size); } -#endif /* CONFIG_DISCONTIGMEM */ +#endif /* ! CONFIG_NUMA */ static struct kcore_list kcore_vmem; @@ -698,7 +712,7 @@ module_init(setup_kcore); void __init mem_init(void) { -#ifdef CONFIG_DISCONTIGMEM +#if defined(CONFIG_DISCONTIGMEM) || defined(CONFIG_SPARSEMEM) int nid; #endif pg_data_t *pgdat; @@ -709,7 +723,7 @@ void __init mem_init(void) num_physpages = max_low_pfn; /* RAM is assumed contiguous */ high_memory = (void *) __va(max_low_pfn * PAGE_SIZE); -#ifdef CONFIG_DISCONTIGMEM +#if defined(CONFIG_DISCONTIGMEM) || defined(CONFIG_SPARSEMEM) for_each_online_node(nid) { if (NODE_DATA(nid)->node_spanned_pages != 0) { printk("freeing bootmem node %x\n", nid); diff -puN include/asm-ppc64/mmzone.h~B-sparse-170-sparsemem-ppc64 include/asm-ppc64/mmzone.h --- memhotplug/include/asm-ppc64/mmzone.h~B-sparse-170-sparsemem-ppc64 2005-04-13 14:20:21.000000000 -0700 +++ memhotplug-dave/include/asm-ppc64/mmzone.h 2005-04-13 14:20:21.000000000 -0700 @@ -10,9 +10,30 @@ #include #include -#ifdef CONFIG_DISCONTIGMEM +/* generic non-linear memory support: + * + * 1) we will not split memory into more chunks than will fit into the + * flags field of the struct page + */ + +/* + * SECTION_SIZE_BITS 2^N: how big each section will be + * MAX_PHYSADDR_BITS 2^N: how much physical address space we have + * MAX_PHYSMEM_BITS 2^N: how much memory we can have in that space + */ +#define SECTION_SIZE_BITS 24 +#define MAX_PHYSADDR_BITS 38 +#define MAX_PHYSMEM_BITS 36 + +#if defined(CONFIG_NUMA) + +#if defined(CONFIG_DISCONTIGMEM) || defined(CONFIG_SPARSEMEM) extern struct pglist_data *node_data[]; +/* + * Return a pointer to the node data for node n. + */ +#define NODE_DATA(nid) (node_data[nid]) /* * Following are specific to this numa platform. @@ -27,6 +48,10 @@ extern int nr_cpus_in_node[]; #define MEMORY_INCREMENT_SHIFT 24 #define MEMORY_INCREMENT (1UL << MEMORY_INCREMENT_SHIFT) +#endif /* !CONFIG_DISCONTIGMEM || !CONFIG_SPARSEMEM */ + +#ifdef CONFIG_DISCONTIGMEM + /* NUMA debugging, will not work on a DLPAR machine */ #undef DEBUG_NUMA @@ -49,11 +74,6 @@ static inline int pa_to_nid(unsigned lon #define pfn_to_nid(pfn) pa_to_nid((pfn) << PAGE_SHIFT) -/* - * Return a pointer to the node data for node n. - */ -#define NODE_DATA(nid) (node_data[nid]) - #define node_localnr(pfn, nid) ((pfn) - NODE_DATA(nid)->node_start_pfn) /* @@ -91,6 +111,15 @@ static inline int pa_to_nid(unsigned lon #endif /* CONFIG_DISCONTIGMEM */ +#ifdef CONFIG_SPARSEMEM +#define pa_to_nid(pa) \ +({ \ + pfn_to_nid(pa >> PAGE_SHIFT); \ +}) +#endif /* CONFIG_SPARSEMEM */ + +#endif /* CONFIG_NUMA */ + #ifdef CONFIG_NEED_MULTIPLE_NODES #define early_pfn_to_nid(pfn) pa_to_nid(((unsigned long)pfn) << PAGE_SHIFT) #endif diff -puN include/asm-ppc64/page.h~B-sparse-170-sparsemem-ppc64 include/asm-ppc64/page.h --- memhotplug/include/asm-ppc64/page.h~B-sparse-170-sparsemem-ppc64 2005-04-13 14:20:21.000000000 -0700 +++ memhotplug-dave/include/asm-ppc64/page.h 2005-04-13 14:20:21.000000000 -0700 @@ -224,7 +224,9 @@ extern u64 ppc64_pft_size; /* Log 2 of #define page_to_pfn(page) discontigmem_page_to_pfn(page) #define pfn_to_page(pfn) discontigmem_pfn_to_page(pfn) #define pfn_valid(pfn) discontigmem_pfn_valid(pfn) -#else +#endif +/* XXX/APW: why is SPARSEMEM not here */ +#ifdef CONFIG_FLATMEM #define pfn_to_page(pfn) (mem_map + (pfn)) #define page_to_pfn(page) ((unsigned long)((page) - mem_map)) #define pfn_valid(pfn) ((pfn) < max_mapnr) _