init.c

来自「Linux Kernel 2.6.9 for OMAP1710」· C语言 代码 · 共 907 行 · 第 1/2 页

		((unsigned long)__init_end - (unsigned long)__init_begin) >> 10);
}

#ifdef CONFIG_BLK_DEV_INITRD
void free_initrd_mem(unsigned long start, unsigned long end)
{
	if (start < end)
		printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10);
	for (; start < end; start += PAGE_SIZE) {
		ClearPageReserved(virt_to_page(start));
		set_page_count(virt_to_page(start), 1);
		free_page(start);
		totalram_pages++;
	}
}
#endif

static spinlock_t mmu_context_lock = SPIN_LOCK_UNLOCKED;
static DEFINE_IDR(mmu_context_idr);

int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
{
	int index;
	int err;

again:
	if (!idr_pre_get(&mmu_context_idr, GFP_KERNEL))
		return -ENOMEM;

	spin_lock(&mmu_context_lock);
	err = idr_get_new(&mmu_context_idr, NULL, &index);
	spin_unlock(&mmu_context_lock);

	if (err == -EAGAIN)
		goto again;
	else if (err)
		return err;

	if (index > MAX_CONTEXT) {
		idr_remove(&mmu_context_idr, index);
		return -ENOMEM;
	}

	mm->context.id = index;

	return 0;
}

void destroy_context(struct mm_struct *mm)
{
	spin_lock(&mmu_context_lock);
	idr_remove(&mmu_context_idr, mm->context.id);
	spin_unlock(&mmu_context_lock);

	mm->context.id = NO_CONTEXT;
}

static int __init mmu_context_init(void)
{
	int index;

	/* Reserve the first (invalid) context*/
	idr_pre_get(&mmu_context_idr, GFP_KERNEL);
	idr_get_new(&mmu_context_idr, NULL, &index);
	BUG_ON(0 != index);

	return 0;
}
arch_initcall(mmu_context_init);

/*
 * Do very early mm setup.
 */
void __init mm_init_ppc64(void)
{
#ifndef CONFIG_PPC_ISERIES
	unsigned long i;
#endif

	ppc64_boot_msg(0x100, "MM Init");

	/* This is the story of the IO hole... please, keep seated,
	 * unfortunately, we are out of oxygen masks at the moment.
	 * So we need some rough way to tell where your big IO hole
	 * is. On pmac, it's between 2G and 4G, on POWER3, it's around
	 * that area as well, on POWER4 we don't have one, etc...
	 * We need that as a "hint" when sizing the TCE table on POWER3
	 * So far, the simplest way that seem work well enough for us it
	 * to just assume that the first discontinuity in our physical
	 * RAM layout is the IO hole. That may not be correct in the future
	 * (and isn't on iSeries but then we don't care ;)
	 */

#ifndef CONFIG_PPC_ISERIES
	for (i = 1; i < lmb.memory.cnt; i++) {
		unsigned long base, prevbase, prevsize;

		prevbase = lmb.memory.region[i-1].physbase;
		prevsize = lmb.memory.region[i-1].size;
		base = lmb.memory.region[i].physbase;
		if (base > (prevbase + prevsize)) {
			io_hole_start = prevbase + prevsize;
			io_hole_size = base  - (prevbase + prevsize);
			break;
		}
	}
#endif /* CONFIG_PPC_ISERIES */
	if (io_hole_start)
		printk("IO Hole assumed to be %lx -> %lx\n",
		       io_hole_start, io_hole_start + io_hole_size - 1);

	ppc64_boot_msg(0x100, "MM Init Done");
}

/*
 * This is called by /dev/mem to know if a given address has to
 * be mapped non-cacheable or not
 */
int page_is_ram(unsigned long pfn)
{
	int i;
	unsigned long paddr = (pfn << PAGE_SHIFT);

	for (i=0; i < lmb.memory.cnt; i++) {
		unsigned long base;

#ifdef CONFIG_MSCHUNKS
		base = lmb.memory.region[i].physbase;
#else
		base = lmb.memory.region[i].base;
#endif
		if ((paddr >= base) &&
			(paddr < (base + lmb.memory.region[i].size))) {
			return 1;
		}
	}

	return 0;
}
EXPORT_SYMBOL(page_is_ram);

/*
 * Initialize the bootmem system and give it all the memory we
 * have available.
 */
#ifndef CONFIG_DISCONTIGMEM
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;

	/*
	 * Find an area to use for the bootmem bitmap.  Calculate the size of
	 * bitmap required as (Total Memory) / PAGE_SIZE / BITS_PER_BYTE.
	 * Add 1 additional page in case the address isn't page-aligned.
	 */
	bootmap_pages = bootmem_bootmap_pages(total_pages);

	start = abs_to_phys(lmb_alloc(bootmap_pages<<PAGE_SHIFT, PAGE_SIZE));
	BUG_ON(!start);

	boot_mapsize = init_bootmem(start >> PAGE_SHIFT, total_pages);

	max_pfn = max_low_pfn;

	/* add all physical memory to the bootmem map. Also find the first */
	for (i=0; i < lmb.memory.cnt; i++) {
		unsigned long physbase, size;

		physbase = lmb.memory.region[i].physbase;
		size = lmb.memory.region[i].size;
		free_bootmem(physbase, size);
	}

	/* reserve the sections we're already using */
	for (i=0; i < lmb.reserved.cnt; i++) {
		unsigned long physbase = lmb.reserved.region[i].physbase;
		unsigned long size = lmb.reserved.region[i].size;

		reserve_bootmem(physbase, size);
	}
}

/*
 * paging_init() sets up the page tables - in fact we've already done this.
 */
void __init paging_init(void)
{
	unsigned long zones_size[MAX_NR_ZONES];
	unsigned long zholes_size[MAX_NR_ZONES];
	unsigned long total_ram = lmb_phys_mem_size();
	unsigned long top_of_ram = lmb_end_of_DRAM();

	printk(KERN_INFO "Top of RAM: 0x%lx, Total RAM: 0x%lx\n",
	       top_of_ram, total_ram);
	printk(KERN_INFO "Memory hole size: %ldMB\n",
	       (top_of_ram - total_ram) >> 20);
	/*
	 * All pages are DMA-able so we put them all in the DMA zone.
	 */
	memset(zones_size, 0, sizeof(zones_size));
	memset(zholes_size, 0, sizeof(zholes_size));

	zones_size[ZONE_DMA] = top_of_ram >> PAGE_SHIFT;
	zholes_size[ZONE_DMA] = (top_of_ram - total_ram) >> PAGE_SHIFT;

	free_area_init_node(0, &contig_page_data, zones_size,
			    __pa(PAGE_OFFSET) >> PAGE_SHIFT, zholes_size);
	mem_map = contig_page_data.node_mem_map;
}
#endif /* CONFIG_DISCONTIGMEM */

static struct kcore_list kcore_vmem;

static int __init setup_kcore(void)
{
	int i;

	for (i=0; i < lmb.memory.cnt; i++) {
		unsigned long physbase, size;
		struct kcore_list *kcore_mem;

		physbase = lmb.memory.region[i].physbase;
		size = lmb.memory.region[i].size;

		/* GFP_ATOMIC to avoid might_sleep warnings during boot */
		kcore_mem = kmalloc(sizeof(struct kcore_list), GFP_ATOMIC);
		if (!kcore_mem)
			panic("mem_init: kmalloc failed\n");

		kclist_add(kcore_mem, __va(physbase), size);
	}

	kclist_add(&kcore_vmem, (void *)VMALLOC_START, VMALLOC_END-VMALLOC_START);

	return 0;
}
module_init(setup_kcore);

void __init mem_init(void)
{
#ifdef CONFIG_DISCONTIGMEM
	int nid;
#endif
	pg_data_t *pgdat;
	unsigned long i;
	struct page *page;
	unsigned long reservedpages = 0, codesize, initsize, datasize, bsssize;

	num_physpages = max_low_pfn;	/* RAM is assumed contiguous */
	high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);

#ifdef CONFIG_DISCONTIGMEM
        for (nid = 0; nid < numnodes; nid++) {
		if (NODE_DATA(nid)->node_spanned_pages != 0) {
			printk("freeing bootmem node %x\n", nid);
			totalram_pages +=
				free_all_bootmem_node(NODE_DATA(nid));
		}
	}
#else
	max_mapnr = num_physpages;
	totalram_pages += free_all_bootmem();
#endif

	for_each_pgdat(pgdat) {
		for (i = 0; i < pgdat->node_spanned_pages; i++) {
			page = pgdat->node_mem_map + i;
			if (PageReserved(page))
				reservedpages++;
		}
	}

	codesize = (unsigned long)&_etext - (unsigned long)&_stext;
	initsize = (unsigned long)&__init_end - (unsigned long)&__init_begin;
	datasize = (unsigned long)&_edata - (unsigned long)&__init_end;
	bsssize = (unsigned long)&__bss_stop - (unsigned long)&__bss_start;

	printk(KERN_INFO "Memory: %luk/%luk available (%luk kernel code, "
	       "%luk reserved, %luk data, %luk bss, %luk init)\n",
		(unsigned long)nr_free_pages() << (PAGE_SHIFT-10),
		num_physpages << (PAGE_SHIFT-10),
		codesize >> 10,
		reservedpages << (PAGE_SHIFT-10),
		datasize >> 10,
		bsssize >> 10,
		initsize >> 10);

	mem_init_done = 1;

#ifdef CONFIG_PPC_ISERIES
	iommu_vio_init();
#endif
}

/*
 * This is called when a page has been modified by the kernel.
 * It just marks the page as not i-cache clean.  We do the i-cache
 * flush later when the page is given to a user process, if necessary.
 */
void flush_dcache_page(struct page *page)
{
	if (cur_cpu_spec->cpu_features & CPU_FTR_COHERENT_ICACHE)
		return;
	/* avoid an atomic op if possible */
	if (test_bit(PG_arch_1, &page->flags))
		clear_bit(PG_arch_1, &page->flags);
}

void clear_user_page(void *page, unsigned long vaddr, struct page *pg)
{
	clear_page(page);

	if (cur_cpu_spec->cpu_features & CPU_FTR_COHERENT_ICACHE)
		return;
	/*
	 * We shouldnt have to do this, but some versions of glibc
	 * require it (ld.so assumes zero filled pages are icache clean)
	 * - Anton
	 */

	/* avoid an atomic op if possible */
	if (test_bit(PG_arch_1, &pg->flags))
		clear_bit(PG_arch_1, &pg->flags);
}

void copy_user_page(void *vto, void *vfrom, unsigned long vaddr,
		    struct page *pg)
{
	copy_page(vto, vfrom);

	/*
	 * We should be able to use the following optimisation, however
	 * there are two problems.
	 * Firstly a bug in some versions of binutils meant PLT sections
	 * were not marked executable.
	 * Secondly the first word in the GOT section is blrl, used
	 * to establish the GOT address. Until recently the GOT was
	 * not marked executable.
	 * - Anton
	 */
#if 0
	if (!vma->vm_file && ((vma->vm_flags & VM_EXEC) == 0))
		return;
#endif

	if (cur_cpu_spec->cpu_features & CPU_FTR_COHERENT_ICACHE)
		return;

	/* avoid an atomic op if possible */
	if (test_bit(PG_arch_1, &pg->flags))
		clear_bit(PG_arch_1, &pg->flags);
}

void flush_icache_user_range(struct vm_area_struct *vma, struct page *page,
			     unsigned long addr, int len)
{
	unsigned long maddr;

	maddr = (unsigned long)page_address(page) + (addr & ~PAGE_MASK);
	flush_icache_range(maddr, maddr + len);
}

/*
 * This is called at the end of handling a user page fault, when the
 * fault has been handled by updating a PTE in the linux page tables.
 * We use it to preload an HPTE into the hash table corresponding to
 * the updated linux PTE.
 * 
 * This must always be called with the mm->page_table_lock held
 */
void update_mmu_cache(struct vm_area_struct *vma, unsigned long ea,
		      pte_t pte)
{
	unsigned long vsid;
	void *pgdir;
	pte_t *ptep;
	int local = 0;
	cpumask_t tmp;
	unsigned long flags;

	/* handle i-cache coherency */
	if (!(cur_cpu_spec->cpu_features & CPU_FTR_COHERENT_ICACHE) &&
	    !(cur_cpu_spec->cpu_features & CPU_FTR_NOEXECUTE)) {
		unsigned long pfn = pte_pfn(pte);
		if (pfn_valid(pfn)) {
			struct page *page = pfn_to_page(pfn);
			if (!PageReserved(page)
			    && !test_bit(PG_arch_1, &page->flags)) {
				__flush_dcache_icache(page_address(page));
				set_bit(PG_arch_1, &page->flags);
			}
		}
	}

	/* We only want HPTEs for linux PTEs that have _PAGE_ACCESSED set */
	if (!pte_young(pte))
		return;

	pgdir = vma->vm_mm->pgd;
	if (pgdir == NULL)
		return;

	ptep = find_linux_pte(pgdir, ea);
	if (!ptep)
		return;

	vsid = get_vsid(vma->vm_mm->context.id, ea);

	local_irq_save(flags);
	tmp = cpumask_of_cpu(smp_processor_id());
	if (cpus_equal(vma->vm_mm->cpu_vm_mask, tmp))
		local = 1;

	__hash_page(ea, pte_val(pte) & (_PAGE_USER|_PAGE_RW), vsid, ptep,
		    0x300, local);
	local_irq_restore(flags);
}

void * reserve_phb_iospace(unsigned long size)
{
	void *virt_addr;
		
	if (phbs_io_bot >= IMALLOC_BASE) 
		panic("reserve_phb_iospace(): phb io space overflow\n");
			
	virt_addr = (void *) phbs_io_bot;
	phbs_io_bot += size;

	return virt_addr;
}

kmem_cache_t *zero_cache;

static void zero_ctor(void *pte, kmem_cache_t *cache, unsigned long flags)
{
	memset(pte, 0, PAGE_SIZE);
}

void pgtable_cache_init(void)
{
	zero_cache = kmem_cache_create("zero",
				PAGE_SIZE,
				0,
				SLAB_HWCACHE_ALIGN | SLAB_MUST_HWCACHE_ALIGN,
				zero_ctor,
				NULL);
	if (!zero_cache)
		panic("pgtable_cache_init(): could not create zero_cache!\n");
}