7 files changed, 92 insertions, 65 deletions

diff --git a/arch/ia64/mm/Makefile b/arch/ia64/mm/Makefile
index d78d20f..bb0a01a 100644
--- a/arch/ia64/mm/Makefile
+++ b/arch/ia64/mm/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the ia64-specific parts of the memory manager.
 #
 
-obj-y := init.o fault.o tlb.o extable.o
+obj-y := init.o fault.o tlb.o extable.o ioremap.o
 
 obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o
 obj-$(CONFIG_NUMA)	   += numa.o
diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c
index acaaec4..84fd1c1 100644
--- a/arch/ia64/mm/contig.c
+++ b/arch/ia64/mm/contig.c
@@ -97,7 +97,7 @@ find_max_pfn (unsigned long start, unsigned long end, void *arg)
  * Find a place to put the bootmap and return its starting address in
  * bootmap_start.  This address must be page-aligned.
  */
-int
+static int __init
 find_bootmap_location (unsigned long start, unsigned long end, void *arg)
 {
 	unsigned long needed = *(unsigned long *)arg;
@@ -141,7 +141,7 @@ find_bootmap_location (unsigned long start, unsigned long end, void *arg)
  * Walk the EFI memory map and find usable memory for the system, taking
  * into account reserved areas.
  */
-void
+void __init
 find_memory (void)
 {
 	unsigned long bootmap_size;
@@ -176,18 +176,20 @@ find_memory (void)
  *
  * Allocate and setup per-cpu data areas.
  */
-void *
+void * __cpuinit
 per_cpu_init (void)
 {
 	void *cpu_data;
 	int cpu;
+	static int first_time=1;
 
 	/*
 	 * get_free_pages() cannot be used before cpu_init() done.  BSP
 	 * allocates "NR_CPUS" pages for all CPUs to avoid that AP calls
 	 * get_zeroed_page().
 	 */
-	if (smp_processor_id() == 0) {
+	if (first_time) {
+		first_time=0;
 		cpu_data = __alloc_bootmem(PERCPU_PAGE_SIZE * NR_CPUS,
 					   PERCPU_PAGE_SIZE, __pa(MAX_DMA_ADDRESS));
 		for (cpu = 0; cpu < NR_CPUS; cpu++) {
@@ -226,7 +228,7 @@ count_dma_pages (u64 start, u64 end, void *arg)
  * Set up the page tables.
  */
 
-void
+void __init
 paging_init (void)
 {
 	unsigned long max_dma;
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index c87d6d1..ec9eeb8 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -379,31 +379,6 @@ static void __init *memory_less_node_alloc(int nid, unsigned long pernodesize)
 }
 
 /**
- * pgdat_insert - insert the pgdat into global pgdat_list
- * @pgdat: the pgdat for a node.
- */
-static void __init pgdat_insert(pg_data_t *pgdat)
-{
-	pg_data_t *prev = NULL, *next;
-
-	for_each_pgdat(next)
-		if (pgdat->node_id < next->node_id)
-			break;
-		else
-			prev = next;
-
-	if (prev) {
-		prev->pgdat_next = pgdat;
-		pgdat->pgdat_next = next;
-	} else {
-		pgdat->pgdat_next = pgdat_list;
-		pgdat_list = pgdat;
-	}
-
-	return;
-}
-
-/**
  * memory_less_nodes - allocate and initialize CPU only nodes pernode
  *	information.
  */
@@ -525,15 +500,20 @@ void __init find_memory(void)
  * find_pernode_space() does most of this already, we just need to set
  * local_per_cpu_offset
  */
-void *per_cpu_init(void)
+void __cpuinit *per_cpu_init(void)
 {
 	int cpu;
+	static int first_time = 1;
+
 
 	if (smp_processor_id() != 0)
 		return __per_cpu_start + __per_cpu_offset[smp_processor_id()];
 
-	for (cpu = 0; cpu < NR_CPUS; cpu++)
-		per_cpu(local_per_cpu_offset, cpu) = __per_cpu_offset[cpu];
+	if (first_time) {
+		first_time = 0;
+		for (cpu = 0; cpu < NR_CPUS; cpu++)
+			per_cpu(local_per_cpu_offset, cpu) = __per_cpu_offset[cpu];
+	}
 
 	return __per_cpu_start + __per_cpu_offset[smp_processor_id()];
 }
@@ -555,7 +535,7 @@ void show_mem(void)
 	printk("Mem-info:\n");
 	show_free_areas();
 	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
-	for_each_pgdat(pgdat) {
+	for_each_online_pgdat(pgdat) {
 		unsigned long present;
 		unsigned long flags;
 		int shared = 0, cached = 0, reserved = 0;
@@ -740,11 +720,5 @@ void __init paging_init(void)
 				    pfn_offset, zholes_size);
 	}
 
-	/*
-	 * Make memory less nodes become a member of the known nodes.
-	 */
-	for_each_node_mask(node, memory_less_mask)
-		pgdat_insert(mem_data[node].pgdat);
-
 	zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page));
 }
diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c
index 2d13889..8d50671 100644
--- a/arch/ia64/mm/hugetlbpage.c
+++ b/arch/ia64/mm/hugetlbpage.c
@@ -68,9 +68,10 @@ huge_pte_offset (struct mm_struct *mm, unsigned long addr)
 #define mk_pte_huge(entry) { pte_val(entry) |= _PAGE_P; }
 
 /*
- * This function checks for proper alignment of input addr and len parameters.
+ * Don't actually need to do any preparation, but need to make sure
+ * the address is in the right region.
  */
-int is_aligned_hugepage_range(unsigned long addr, unsigned long len)
+int prepare_hugepage_range(unsigned long addr, unsigned long len)
 {
 	if (len & ~HPAGE_MASK)
 		return -EINVAL;
@@ -112,8 +113,7 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb,
 			unsigned long floor, unsigned long ceiling)
 {
 	/*
-	 * This is called only when is_hugepage_only_range(addr,),
-	 * and it follows that is_hugepage_only_range(end,) also.
+	 * This is called to free hugetlb page tables.
 	 *
 	 * The offset of these addresses from the base of the hugetlb
 	 * region must be scaled down by HPAGE_SIZE/PAGE_SIZE so that
@@ -125,9 +125,9 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb,
 
 	addr = htlbpage_to_page(addr);
 	end  = htlbpage_to_page(end);
-	if (is_hugepage_only_range(tlb->mm, floor, HPAGE_SIZE))
+	if (REGION_NUMBER(floor) == RGN_HPAGE)
 		floor = htlbpage_to_page(floor);
-	if (is_hugepage_only_range(tlb->mm, ceiling, HPAGE_SIZE))
+	if (REGION_NUMBER(ceiling) == RGN_HPAGE)
 		ceiling = htlbpage_to_page(ceiling);
 
 	free_pgd_range(tlb, addr, end, floor, ceiling);
diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c
index b38b6d2..cafa877 100644
--- a/arch/ia64/mm/init.c
+++ b/arch/ia64/mm/init.c
@@ -109,6 +109,7 @@ lazy_mmu_prot_update (pte_t pte)
 {
 	unsigned long addr;
 	struct page *page;
+	unsigned long order;
 
 	if (!pte_exec(pte))
 		return;				/* not an executable page... */
@@ -119,7 +120,12 @@ lazy_mmu_prot_update (pte_t pte)
 	if (test_bit(PG_arch_1, &page->flags))
 		return;				/* i-cache is already coherent with d-cache */
 
-	flush_icache_range(addr, addr + PAGE_SIZE);
+	if (PageCompound(page)) {
+		order = (unsigned long) (page[1].lru.prev);
+		flush_icache_range(addr, addr + (1UL << order << PAGE_SHIFT));
+	}
+	else
+		flush_icache_range(addr, addr + PAGE_SIZE);
 	set_bit(PG_arch_1, &page->flags);	/* mark page as clean */
 }
 
@@ -197,7 +203,7 @@ free_initmem (void)
 	eaddr = (unsigned long) ia64_imva(__init_end);
 	while (addr < eaddr) {
 		ClearPageReserved(virt_to_page(addr));
-		set_page_count(virt_to_page(addr), 1);
+		init_page_count(virt_to_page(addr));
 		free_page(addr);
 		++totalram_pages;
 		addr += PAGE_SIZE;
@@ -206,7 +212,7 @@ free_initmem (void)
 	       (__init_end - __init_begin) >> 10);
 }
 
-void
+void __init
 free_initrd_mem (unsigned long start, unsigned long end)
 {
 	struct page *page;
@@ -252,7 +258,7 @@ free_initrd_mem (unsigned long start, unsigned long end)
 			continue;
 		page = virt_to_page(start);
 		ClearPageReserved(page);
-		set_page_count(page, 1);
+		init_page_count(page);
 		free_page(start);
 		++totalram_pages;
 	}
@@ -261,7 +267,7 @@ free_initrd_mem (unsigned long start, unsigned long end)
 /*
  * This installs a clean page in the kernel's page table.
  */
-struct page *
+static struct page * __init
 put_kernel_page (struct page *page, unsigned long address, pgprot_t pgprot)
 {
 	pgd_t *pgd;
@@ -294,7 +300,7 @@ put_kernel_page (struct page *page, unsigned long address, pgprot_t pgprot)
 	return page;
 }
 
-static void
+static void __init
 setup_gate (void)
 {
 	struct page *page;
@@ -411,7 +417,7 @@ ia64_mmu_init (void *my_cpu_data)
 
 #ifdef CONFIG_VIRTUAL_MEM_MAP
 
-int
+int __init
 create_mem_map_page_table (u64 start, u64 end, void *arg)
 {
 	unsigned long address, start_page, end_page;
@@ -519,7 +525,7 @@ ia64_pfn_valid (unsigned long pfn)
 }
 EXPORT_SYMBOL(ia64_pfn_valid);
 
-int
+int __init
 find_largest_hole (u64 start, u64 end, void *arg)
 {
 	u64 *max_gap = arg;
@@ -535,7 +541,7 @@ find_largest_hole (u64 start, u64 end, void *arg)
 }
 #endif /* CONFIG_VIRTUAL_MEM_MAP */
 
-static int
+static int __init
 count_reserved_pages (u64 start, u64 end, void *arg)
 {
 	unsigned long num_reserved = 0;
@@ -556,7 +562,7 @@ count_reserved_pages (u64 start, u64 end, void *arg)
  * purposes.
  */
 
-static int nolwsys;
+static int nolwsys __initdata;
 
 static int __init
 nolwsys_setup (char *s)
@@ -567,7 +573,7 @@ nolwsys_setup (char *s)
 
 __setup("nolwsys", nolwsys_setup);
 
-void
+void __init
 mem_init (void)
 {
 	long reserved_pages, codesize, datasize, initsize;
@@ -600,7 +606,7 @@ mem_init (void)
 	kclist_add(&kcore_vmem, (void *)VMALLOC_START, VMALLOC_END-VMALLOC_START);
 	kclist_add(&kcore_kernel, _stext, _end - _stext);
 
-	for_each_pgdat(pgdat)
+	for_each_online_pgdat(pgdat)
 		if (pgdat->bdata->node_bootmem_map)
 			totalram_pages += free_all_bootmem_node(pgdat);
 
@@ -640,7 +646,7 @@ mem_init (void)
 void online_page(struct page *page)
 {
 	ClearPageReserved(page);
-	set_page_count(page, 1);
+	init_page_count(page);
 	__free_page(page);
 	totalram_pages++;
 	num_physpages++;
diff --git a/arch/ia64/mm/ioremap.c b/arch/ia64/mm/ioremap.c
new file mode 100644
index 0000000..643ccc6
--- /dev/null
+++ b/arch/ia64/mm/ioremap.c
@@ -0,0 +1,43 @@
+/*
+ * (c) Copyright 2006 Hewlett-Packard Development Company, L.P.
+ *	Bjorn Helgaas <bjorn.helgaas@hp.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/compiler.h>
+#include <linux/module.h>
+#include <linux/efi.h>
+#include <asm/io.h>
+
+static inline void __iomem *
+__ioremap (unsigned long offset, unsigned long size)
+{
+	return (void __iomem *) (__IA64_UNCACHED_OFFSET | offset);
+}
+
+void __iomem *
+ioremap (unsigned long offset, unsigned long size)
+{
+	if (efi_mem_attribute_range(offset, size, EFI_MEMORY_WB))
+		return phys_to_virt(offset);
+
+	if (efi_mem_attribute_range(offset, size, EFI_MEMORY_UC))
+		return __ioremap(offset, size);
+
+	/*
+	 * Someday this should check ACPI resources so we
+	 * can do the right thing for hot-plugged regions.
+	 */
+	return __ioremap(offset, size);
+}
+EXPORT_SYMBOL(ioremap);
+
+void __iomem *
+ioremap_nocache (unsigned long offset, unsigned long size)
+{
+	return __ioremap(offset, size);
+}
+EXPORT_SYMBOL(ioremap_nocache);
diff --git a/arch/ia64/mm/tlb.c b/arch/ia64/mm/tlb.c
index 6a4eec9..4dbbca0 100644
--- a/arch/ia64/mm/tlb.c
+++ b/arch/ia64/mm/tlb.c
@@ -156,17 +156,19 @@ flush_tlb_range (struct vm_area_struct *vma, unsigned long start,
 		nbits = purge.max_bits;
 	start &= ~((1UL << nbits) - 1);
 
-# ifdef CONFIG_SMP
-	platform_global_tlb_purge(mm, start, end, nbits);
-# else
 	preempt_disable();
+#ifdef CONFIG_SMP
+	if (mm != current->active_mm || cpus_weight(mm->cpu_vm_mask) != 1) {
+		platform_global_tlb_purge(mm, start, end, nbits);
+		preempt_enable();
+		return;
+	}
+#endif
 	do {
 		ia64_ptcl(start, (nbits<<2));
 		start += (1UL << nbits);
 	} while (start < end);
 	preempt_enable();
-# endif
-
 	ia64_srlz_i();			/* srlz.i implies srlz.d */
 }
 EXPORT_SYMBOL(flush_tlb_range);