init.c

这个linux源代码是很全面的~基本完整了~使用c编译的~由于时间问题我没有亲自测试~但就算用来做参考资料也是非常好的

	 * VMLPT.  I assume that once we run on machines big enough to warrant 64KB pages,
	 * IMPL_VA_MSB will be significantly bigger, so this is unlikely to become a
	 * problem in practice.  Alternatively, we could truncate the top of the mapped
	 * address space to not permit mappings that would overlap with the VMLPT.
	 * --davidm 00/12/06
	 */
#	define pte_bits			3
#	define mapped_space_bits	(3*(PAGE_SHIFT - pte_bits) + PAGE_SHIFT)
	/*
	 * The virtual page table has to cover the entire implemented address space within
	 * a region even though not all of this space may be mappable.  The reason for
	 * this is that the Access bit and Dirty bit fault handlers perform
	 * non-speculative accesses to the virtual page table, so the address range of the
	 * virtual page table itself needs to be covered by virtual page table.
	 */
#	define vmlpt_bits		(impl_va_bits - PAGE_SHIFT + pte_bits)
#	define POW2(n)			(1ULL << (n))

	impl_va_bits = ffz(~(local_cpu_data->unimpl_va_mask | (7UL << 61)));

	if (impl_va_bits < 51 || impl_va_bits > 61)
		panic("CPU has bogus IMPL_VA_MSB value of %lu!\n", impl_va_bits - 1);

	/* place the VMLPT at the end of each page-table mapped region: */
	pta = POW2(61) - POW2(vmlpt_bits);

	if (POW2(mapped_space_bits) >= pta)
		panic("mm/init: overlap between virtually mapped linear page table and "
		      "mapped kernel space!");
	/*
	 * Set the (virtually mapped linear) page table address.  Bit
	 * 8 selects between the short and long format, bits 2-7 the
	 * size of the table, and bit 0 whether the VHPT walker is
	 * enabled.
	 */
	ia64_set_pta(pta | (0 << 8) | (vmlpt_bits << 2) | VHPT_ENABLE_BIT);

	ia64_tlb_init();
}

static int
create_mem_map_page_table (u64 start, u64 end, void *arg)
{
	unsigned long address, start_page, end_page;
	struct page *map_start, *map_end;
	pgd_t *pgd;
	pmd_t *pmd;
	pte_t *pte;

	/* should we use platform_map_nr here? */

	map_start = vmem_map + MAP_NR_DENSE(start);
	map_end   = vmem_map + MAP_NR_DENSE(end);

	start_page = (unsigned long) map_start & PAGE_MASK;
	end_page = PAGE_ALIGN((unsigned long) map_end);

	for (address = start_page; address < end_page; address += PAGE_SIZE) {
		pgd = pgd_offset_k(address);
		if (pgd_none(*pgd))
			pgd_populate(&init_mm, pgd, alloc_bootmem_pages(PAGE_SIZE));
		pmd = pmd_offset(pgd, address);

		if (pmd_none(*pmd))
			pmd_populate(&init_mm, pmd, alloc_bootmem_pages(PAGE_SIZE));
		pte = pte_offset(pmd, address);

		if (pte_none(*pte))
			set_pte(pte, mk_pte_phys(__pa(alloc_bootmem_pages(PAGE_SIZE)),
						 PAGE_KERNEL));
 	}
 	return 0;
}

struct memmap_init_callback_data {
	memmap_init_callback_t *memmap_init;
	struct page *start;
	struct page *end;
	int zone;
	int highmem;
};

static int
virtual_memmap_init (u64 start, u64 end, void *arg)
{
	struct memmap_init_callback_data *args;
	struct page *map_start, *map_end;

	args = (struct memmap_init_callback_data *) arg;

	/* Should we use platform_map_nr here? */

	map_start = mem_map + MAP_NR_DENSE(start);
	map_end   = mem_map + MAP_NR_DENSE(end);

	if (map_start < args->start)
		map_start = args->start;
	if (map_end > args->end)
		map_end = args->end;

	/*
	 * We have to initialize "out of bounds" struct page elements
	 * that fit completely on the same pages that were allocated
	 * for the "in bounds" elements because they may be referenced
	 * later (and found to be "reserved").
	 */
	map_start -= ((unsigned long) map_start & (PAGE_SIZE - 1))
			/ sizeof(struct page);
	map_end += ((PAGE_ALIGN((unsigned long) map_end) -
				(unsigned long) map_end)
			/ sizeof(struct page));

	if (map_start < map_end)
		(*args->memmap_init)(map_start, map_end, args->zone,
				     page_to_phys(map_start), args->highmem);

	return 0;
}

unsigned long
arch_memmap_init (memmap_init_callback_t *memmap_init, struct page *start,
	struct page *end, int zone, unsigned long start_paddr, int highmem)
{
	if (!vmem_map) 
		memmap_init(start,end,zone,page_to_phys(start),highmem);
	else {
		struct memmap_init_callback_data args;

		args.memmap_init = memmap_init;
		args.start = start;
		args.end = end;
		args.zone = zone;
		args.highmem = highmem;

		efi_memmap_walk(virtual_memmap_init, &args);
	}

	return page_to_phys(end);
}

static int
count_dma_pages (u64 start, u64 end, void *arg)
{
	unsigned long *count = arg;

	if (end <= MAX_DMA_ADDRESS)
		*count += (end - start) >> PAGE_SHIFT;
	return 0;
}

int
ia64_page_valid (struct page *page)
{
	char byte;

	return __get_user(byte, (char *) page) == 0;
}

static int
count_pages (u64 start, u64 end, void *arg)
{
	unsigned long *count = arg;

	*count += (end - start) >> PAGE_SHIFT;
	return 0;
}

#ifndef CONFIG_DISCONTIGMEM
static int
find_largest_hole(u64 start, u64 end, void *arg)
{
	u64 *max_gap = arg;
	static u64 last_end = PAGE_OFFSET;

	/* NOTE: this algorithm assumes efi memmap table is ordered */

	if (*max_gap < (start - last_end))
		*max_gap = start - last_end;
	last_end = end;
	return 0;
}
#endif

/*
 * Set up the page tables.
 */
void
paging_init (void)
{
	unsigned long max_dma;
	unsigned long zones_size[MAX_NR_ZONES];
	unsigned long zholes_size[MAX_NR_ZONES];
#ifndef CONFIG_DISCONTIGMEM
	unsigned long max_gap;
#endif

	/* initialize mem_map[] */

	memset(zones_size, 0, sizeof(zones_size));
	memset(zholes_size, 0, sizeof(zholes_size));

	num_physpages = 0;
	efi_memmap_walk(count_pages, &num_physpages);

	num_dma_physpages = 0;
	efi_memmap_walk(count_dma_pages, &num_dma_physpages);

	max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT;

	if (max_low_pfn < max_dma) {
		zones_size[ZONE_DMA] = max_low_pfn;
		zholes_size[ZONE_DMA] = max_low_pfn - num_dma_physpages;
	} else {
		zones_size[ZONE_DMA] = max_dma;
		zholes_size[ZONE_DMA] = max_dma - num_dma_physpages;
		if (num_physpages > num_dma_physpages) {
			zones_size[ZONE_NORMAL] = max_low_pfn - max_dma;
			zholes_size[ZONE_NORMAL] = (max_low_pfn - max_dma)
					- (num_physpages - num_dma_physpages);
		}
	}

#ifdef CONFIG_DISCONTIGMEM
	free_area_init_node(0, NULL, NULL, zones_size, 0, zholes_size);
#else
	max_gap = 0;
	efi_memmap_walk(find_largest_hole, (u64 *)&max_gap);

	if (max_gap < LARGE_GAP) {
		vmem_map = (struct page *)0;
		free_area_init_node(0, NULL, NULL, zones_size, 0, zholes_size);
	}
	else {
		unsigned long map_size;

		/* allocate virtual mem_map */

		map_size = PAGE_ALIGN(max_low_pfn*sizeof(struct page));
		vmalloc_end -= map_size;
		vmem_map = (struct page *) vmalloc_end;
		efi_memmap_walk(create_mem_map_page_table, 0);

		free_area_init_node(0, NULL, vmem_map, zones_size, 0, zholes_size);
		printk("Virtual mem_map starts at 0x%p\n", mem_map);
	}
#endif
}

static int
count_reserved_pages (u64 start, u64 end, void *arg)
{
	unsigned long num_reserved = 0;
	unsigned long *count = arg;
	struct page *pg;

	for (pg = virt_to_page((void *)start); pg < virt_to_page((void *)end); ++pg)
		if (PageReserved(pg))
			++num_reserved;
	*count += num_reserved;
	return 0;
}

void
mem_init (void)
{
	extern char __start_gate_section[];
	long reserved_pages, codesize, datasize, initsize;
	unsigned long num_pgt_pages;

#ifdef CONFIG_PCI
	/*
	 * This needs to be called _after_ the command line has been parsed but _before_
	 * any drivers that may need the PCI DMA interface are initialized or bootmem has
	 * been freed.
	 */
	platform_pci_dma_init();
#endif

	if (!mem_map)
		BUG();

	max_mapnr = max_low_pfn;
	high_memory = __va(max_low_pfn * PAGE_SIZE);

	totalram_pages += free_all_bootmem();

	reserved_pages = 0;
	efi_memmap_walk(count_reserved_pages, &reserved_pages);

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

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

	/*
	 * Allow for enough (cached) page table pages so that we can map the entire memory
	 * at least once.  Each task also needs a couple of page tables pages, so add in a
	 * fudge factor for that (don't use "threads-max" here; that would be wrong!).
	 * Don't allow the cache to be more than 10% of total memory, though.
	 */
#	define NUM_TASKS	500	/* typical number of tasks */
	num_pgt_pages = nr_free_pages() / PTRS_PER_PGD + NUM_TASKS;
	if (num_pgt_pages > nr_free_pages() / 10)
		num_pgt_pages = nr_free_pages() / 10;
	if (num_pgt_pages > pgt_cache_water[1])
		pgt_cache_water[1] = num_pgt_pages;

	/* install the gate page in the global page table: */
	put_gate_page(virt_to_page(__start_gate_section), GATE_ADDR);

#ifdef CONFIG_IA32_SUPPORT
	ia32_gdt_init();
#endif
}