init.c

来自「linux 内核源代码」· C语言 代码 · 共 1,976 行 · 第 1/4 页

/* Read OBP translations property into 'prom_trans[]'.  */
static void __init read_obp_translations(void)
{
	int n, node, ents, first, last, i;

	node = prom_finddevice("/virtual-memory");
	n = prom_getproplen(node, "translations");
	if (unlikely(n == 0 || n == -1)) {
		prom_printf("prom_mappings: Couldn't get size.\n");
		prom_halt();
	}
	if (unlikely(n > sizeof(prom_trans))) {
		prom_printf("prom_mappings: Size %Zd is too big.\n", n);
		prom_halt();
	}

	if ((n = prom_getproperty(node, "translations",
				  (char *)&prom_trans[0],
				  sizeof(prom_trans))) == -1) {
		prom_printf("prom_mappings: Couldn't get property.\n");
		prom_halt();
	}

	n = n / sizeof(struct linux_prom_translation);

	ents = n;

	sort(prom_trans, ents, sizeof(struct linux_prom_translation),
	     cmp_ptrans, NULL);

	/* Now kick out all the non-OBP entries.  */
	for (i = 0; i < ents; i++) {
		if (in_obp_range(prom_trans[i].virt))
			break;
	}
	first = i;
	for (; i < ents; i++) {
		if (!in_obp_range(prom_trans[i].virt))
			break;
	}
	last = i;

	for (i = 0; i < (last - first); i++) {
		struct linux_prom_translation *src = &prom_trans[i + first];
		struct linux_prom_translation *dest = &prom_trans[i];

		*dest = *src;
	}
	for (; i < ents; i++) {
		struct linux_prom_translation *dest = &prom_trans[i];
		dest->virt = dest->size = dest->data = 0x0UL;
	}

	prom_trans_ents = last - first;

	if (tlb_type == spitfire) {
		/* Clear diag TTE bits. */
		for (i = 0; i < prom_trans_ents; i++)
			prom_trans[i].data &= ~0x0003fe0000000000UL;
	}
}

static void __init hypervisor_tlb_lock(unsigned long vaddr,
				       unsigned long pte,
				       unsigned long mmu)
{
	unsigned long ret = sun4v_mmu_map_perm_addr(vaddr, 0, pte, mmu);

	if (ret != 0) {
		prom_printf("hypervisor_tlb_lock[%lx:%lx:%lx:%lx]: "
			    "errors with %lx\n", vaddr, 0, pte, mmu, ret);
		prom_halt();
	}
}

static unsigned long kern_large_tte(unsigned long paddr);

static void __init remap_kernel(void)
{
	unsigned long phys_page, tte_vaddr, tte_data;
	int tlb_ent = sparc64_highest_locked_tlbent();

	tte_vaddr = (unsigned long) KERNBASE;
	phys_page = (prom_boot_mapping_phys_low >> 22UL) << 22UL;
	tte_data = kern_large_tte(phys_page);

	kern_locked_tte_data = tte_data;

	/* Now lock us into the TLBs via Hypervisor or OBP. */
	if (tlb_type == hypervisor) {
		hypervisor_tlb_lock(tte_vaddr, tte_data, HV_MMU_DMMU);
		hypervisor_tlb_lock(tte_vaddr, tte_data, HV_MMU_IMMU);
		if (bigkernel) {
			tte_vaddr += 0x400000;
			tte_data += 0x400000;
			hypervisor_tlb_lock(tte_vaddr, tte_data, HV_MMU_DMMU);
			hypervisor_tlb_lock(tte_vaddr, tte_data, HV_MMU_IMMU);
		}
	} else {
		prom_dtlb_load(tlb_ent, tte_data, tte_vaddr);
		prom_itlb_load(tlb_ent, tte_data, tte_vaddr);
		if (bigkernel) {
			tlb_ent -= 1;
			prom_dtlb_load(tlb_ent,
				       tte_data + 0x400000, 
				       tte_vaddr + 0x400000);
			prom_itlb_load(tlb_ent,
				       tte_data + 0x400000, 
				       tte_vaddr + 0x400000);
		}
		sparc64_highest_unlocked_tlb_ent = tlb_ent - 1;
	}
	if (tlb_type == cheetah_plus) {
		sparc64_kern_pri_context = (CTX_CHEETAH_PLUS_CTX0 |
					    CTX_CHEETAH_PLUS_NUC);
		sparc64_kern_pri_nuc_bits = CTX_CHEETAH_PLUS_NUC;
		sparc64_kern_sec_context = CTX_CHEETAH_PLUS_CTX0;
	}
}


static void __init inherit_prom_mappings(void)
{
	read_obp_translations();

	/* Now fixup OBP's idea about where we really are mapped. */
	prom_printf("Remapping the kernel... ");
	remap_kernel();
	prom_printf("done.\n");
}

void prom_world(int enter)
{
	if (!enter)
		set_fs((mm_segment_t) { get_thread_current_ds() });

	__asm__ __volatile__("flushw");
}

void __flush_dcache_range(unsigned long start, unsigned long end)
{
	unsigned long va;

	if (tlb_type == spitfire) {
		int n = 0;

		for (va = start; va < end; va += 32) {
			spitfire_put_dcache_tag(va & 0x3fe0, 0x0);
			if (++n >= 512)
				break;
		}
	} else if (tlb_type == cheetah || tlb_type == cheetah_plus) {
		start = __pa(start);
		end = __pa(end);
		for (va = start; va < end; va += 32)
			__asm__ __volatile__("stxa %%g0, [%0] %1\n\t"
					     "membar #Sync"
					     : /* no outputs */
					     : "r" (va),
					       "i" (ASI_DCACHE_INVALIDATE));
	}
}

/* get_new_mmu_context() uses "cache + 1".  */
DEFINE_SPINLOCK(ctx_alloc_lock);
unsigned long tlb_context_cache = CTX_FIRST_VERSION - 1;
#define MAX_CTX_NR	(1UL << CTX_NR_BITS)
#define CTX_BMAP_SLOTS	BITS_TO_LONGS(MAX_CTX_NR)
DECLARE_BITMAP(mmu_context_bmap, MAX_CTX_NR);

/* Caller does TLB context flushing on local CPU if necessary.
 * The caller also ensures that CTX_VALID(mm->context) is false.
 *
 * We must be careful about boundary cases so that we never
 * let the user have CTX 0 (nucleus) or we ever use a CTX
 * version of zero (and thus NO_CONTEXT would not be caught
 * by version mis-match tests in mmu_context.h).
 *
 * Always invoked with interrupts disabled.
 */
void get_new_mmu_context(struct mm_struct *mm)
{
	unsigned long ctx, new_ctx;
	unsigned long orig_pgsz_bits;
	unsigned long flags;
	int new_version;

	spin_lock_irqsave(&ctx_alloc_lock, flags);
	orig_pgsz_bits = (mm->context.sparc64_ctx_val & CTX_PGSZ_MASK);
	ctx = (tlb_context_cache + 1) & CTX_NR_MASK;
	new_ctx = find_next_zero_bit(mmu_context_bmap, 1 << CTX_NR_BITS, ctx);
	new_version = 0;
	if (new_ctx >= (1 << CTX_NR_BITS)) {
		new_ctx = find_next_zero_bit(mmu_context_bmap, ctx, 1);
		if (new_ctx >= ctx) {
			int i;
			new_ctx = (tlb_context_cache & CTX_VERSION_MASK) +
				CTX_FIRST_VERSION;
			if (new_ctx == 1)
				new_ctx = CTX_FIRST_VERSION;

			/* Don't call memset, for 16 entries that's just
			 * plain silly...
			 */
			mmu_context_bmap[0] = 3;
			mmu_context_bmap[1] = 0;
			mmu_context_bmap[2] = 0;
			mmu_context_bmap[3] = 0;
			for (i = 4; i < CTX_BMAP_SLOTS; i += 4) {
				mmu_context_bmap[i + 0] = 0;
				mmu_context_bmap[i + 1] = 0;
				mmu_context_bmap[i + 2] = 0;
				mmu_context_bmap[i + 3] = 0;
			}
			new_version = 1;
			goto out;
		}
	}
	mmu_context_bmap[new_ctx>>6] |= (1UL << (new_ctx & 63));
	new_ctx |= (tlb_context_cache & CTX_VERSION_MASK);
out:
	tlb_context_cache = new_ctx;
	mm->context.sparc64_ctx_val = new_ctx | orig_pgsz_bits;
	spin_unlock_irqrestore(&ctx_alloc_lock, flags);

	if (unlikely(new_version))
		smp_new_mmu_context_version();
}

/* Find a free area for the bootmem map, avoiding the kernel image
 * and the initial ramdisk.
 */
static unsigned long __init choose_bootmap_pfn(unsigned long start_pfn,
					       unsigned long end_pfn)
{
	unsigned long avoid_start, avoid_end, bootmap_size;
	int i;

	bootmap_size = bootmem_bootmap_pages(end_pfn - start_pfn);
	bootmap_size <<= PAGE_SHIFT;

	avoid_start = avoid_end = 0;
#ifdef CONFIG_BLK_DEV_INITRD
	avoid_start = initrd_start;
	avoid_end = PAGE_ALIGN(initrd_end);
#endif

#ifdef CONFIG_DEBUG_BOOTMEM
	prom_printf("choose_bootmap_pfn: kern[%lx:%lx] avoid[%lx:%lx]\n",
		    kern_base, PAGE_ALIGN(kern_base + kern_size),
		    avoid_start, avoid_end);
#endif
	for (i = 0; i < pavail_ents; i++) {
		unsigned long start, end;

		start = pavail[i].phys_addr;
		end = start + pavail[i].reg_size;

		while (start < end) {
			if (start >= kern_base &&
			    start < PAGE_ALIGN(kern_base + kern_size)) {
				start = PAGE_ALIGN(kern_base + kern_size);
				continue;
			}
			if (start >= avoid_start && start < avoid_end) {
				start = avoid_end;
				continue;
			}

			if ((end - start) < bootmap_size)
				break;

			if (start < kern_base &&
			    (start + bootmap_size) > kern_base) {
				start = PAGE_ALIGN(kern_base + kern_size);
				continue;
			}

			if (start < avoid_start &&
			    (start + bootmap_size) > avoid_start) {
				start = avoid_end;
				continue;
			}

			/* OK, it doesn't overlap anything, use it.  */
#ifdef CONFIG_DEBUG_BOOTMEM
			prom_printf("choose_bootmap_pfn: Using %lx [%lx]\n",
				    start >> PAGE_SHIFT, start);
#endif
			return start >> PAGE_SHIFT;
		}
	}

	prom_printf("Cannot find free area for bootmap, aborting.\n");
	prom_halt();
}

static void __init trim_pavail(unsigned long *cur_size_p,
			       unsigned long *end_of_phys_p)
{
	unsigned long to_trim = *cur_size_p - cmdline_memory_size;
	unsigned long avoid_start, avoid_end;
	int i;

	to_trim = PAGE_ALIGN(to_trim);

	avoid_start = avoid_end = 0;
#ifdef CONFIG_BLK_DEV_INITRD
	avoid_start = initrd_start;
	avoid_end = PAGE_ALIGN(initrd_end);
#endif

	/* Trim some pavail[] entries in order to satisfy the
	 * requested "mem=xxx" kernel command line specification.
	 *
	 * We must not trim off the kernel image area nor the
	 * initial ramdisk range (if any).  Also, we must not trim
	 * any pavail[] entry down to zero in order to preserve
	 * the invariant that all pavail[] entries have a non-zero
	 * size which is assumed by all of the code in here.
	 */
	for (i = 0; i < pavail_ents; i++) {
		unsigned long start, end, kern_end;
		unsigned long trim_low, trim_high, n;

		kern_end = PAGE_ALIGN(kern_base + kern_size);

		trim_low = start = pavail[i].phys_addr;
		trim_high = end = start + pavail[i].reg_size;

		if (kern_base >= start &&
		    kern_base < end) {
			trim_low = kern_base;
			if (kern_end >= end)
				continue;
		}
		if (kern_end >= start &&
		    kern_end < end) {
			trim_high = kern_end;
		}
		if (avoid_start &&
		    avoid_start >= start &&
		    avoid_start < end) {
			if (trim_low > avoid_start)
				trim_low = avoid_start;
			if (avoid_end >= end)
				continue;
		}
		if (avoid_end &&
		    avoid_end >= start &&
		    avoid_end < end) {
			if (trim_high < avoid_end)
				trim_high = avoid_end;
		}

		if (trim_high <= trim_low)
			continue;

		if (trim_low == start && trim_high == end) {
			/* Whole chunk is available for trimming.
			 * Trim all except one page, in order to keep
			 * entry non-empty.
			 */
			n = (end - start) - PAGE_SIZE;
			if (n > to_trim)
				n = to_trim;

			if (n) {
				pavail[i].phys_addr += n;
				pavail[i].reg_size -= n;
				to_trim -= n;
			}
		} else {
			n = (trim_low - start);
			if (n > to_trim)
				n = to_trim;

			if (n) {
				pavail[i].phys_addr += n;
				pavail[i].reg_size -= n;
				to_trim -= n;
			}
			if (to_trim) {
				n = end - trim_high;
				if (n > to_trim)
					n = to_trim;
				if (n) {
					pavail[i].reg_size -= n;
					to_trim -= n;
				}
			}
		}

		if (!to_trim)
			break;
	}

	/* Recalculate.  */
	*cur_size_p = 0UL;
	for (i = 0; i < pavail_ents; i++) {
		*end_of_phys_p = pavail[i].phys_addr +
			pavail[i].reg_size;
		*cur_size_p += pavail[i].reg_size;
	}
}

/* About pages_avail, this is the value we will use to calculate
 * the zholes_size[] argument given to free_area_init_node().  The
 * page allocator uses this to calculate nr_kernel_pages,
 * nr_all_pages and zone->present_pages.  On NUMA it is used
 * to calculate zone->min_unmapped_pages and zone->min_slab_pages.
 *
 * So this number should really be set to what the page allocator
 * actually ends up with.  This means:
 * 1) It should include bootmem map pages, we'll release those.
 * 2) It should not include the kernel image, except for the
 *    __init sections which we will also release.
 * 3) It should include the initrd image, since we'll release
 *    that too.
 */
static unsigned long __init bootmem_init(unsigned long *pages_avail,
					 unsigned long phys_base)
{
	unsigned long bootmap_size, end_pfn;
	unsigned long end_of_phys_memory = 0UL;
	unsigned long bootmap_pfn, bytes_avail, size;
	int i;

#ifdef CONFIG_DEBUG_BOOTMEM
	prom_printf("bootmem_init: Scan pavail, ");
#endif

	bytes_avail = 0UL;
	for (i = 0; i < pavail_ents; i++) {
		end_of_phys_memory = pavail[i].phys_addr +
			pavail[i].reg_size;
		bytes_avail += pavail[i].reg_size;
	}

	/* Determine the location of the initial ramdisk before trying
	 * to honor the "mem=xxx" command line argument.  We must know
	 * where the kernel image and the ramdisk image are so that we
	 * do not trim those two areas from the physical memory map.
	 */

#ifdef CONFIG_BLK_DEV_INITRD
	/* Now have to check initial ramdisk, so that bootmap does not overwrite it */
	if (sparc_ramdisk_image || sparc_ramdisk_image64) {
		unsigned long ramdisk_image = sparc_ramdisk_image ?
			sparc_ramdisk_image : sparc_ramdisk_image64;
		ramdisk_image -= KERNBASE;
		initrd_start = ramdisk_image + phys_base;
		initrd_end = initrd_start + sparc_ramdisk_size;
		if (initrd_end > end_of_phys_memory) {
			printk(KERN_CRIT "initrd extends beyond end of memory "
		                 	 "(0x%016lx > 0x%016lx)\ndisabling initrd\n",
			       initrd_end, end_of_phys_memory);
			initrd_start = 0;
			initrd_end = 0;
		}
	}
#endif	

	if (cmdline_memory_size &&
	    bytes_avail > cmdline_memory_size)
		trim_pavail(&bytes_avail,
			    &end_of_phys_memory);

	*pages_avail = bytes_avail >> PAGE_SHIFT;

	end_pfn = end_of_phys_memory >> PAGE_SHIFT;

	/* Initialize the boot-time allocator. */
	max_pfn = max_low_pfn = end_pfn;
	min_low_pfn = (phys_base >> PAGE_SHIFT);

	bootmap_pfn = choose_bootmap_pfn(min_low_pfn, end_pfn);

#ifdef CONFIG_DEBUG_BOOTMEM
	prom_printf("init_bootmem(min[%lx], bootmap[%lx], max[%lx])\n",
		    min_low_pfn, bootmap_pfn, max_low_pfn);
#endif
	bootmap_size = init_bootmem_node(NODE_DATA(0), bootmap_pfn,
					 min_low_pfn, end_pfn);

	/* Now register the available physical memory with the
	 * allocator.
	 */
	for (i = 0; i < pavail_ents; i++) {
#ifdef CONFIG_DEBUG_BOOTMEM
		prom_printf("free_bootmem(pavail:%d): base[%lx] size[%lx]\n",
			    i, pavail[i].phys_addr, pavail[i].reg_size);
#endif
		free_bootmem(pavail[i].phys_addr, pavail[i].reg_size);