exec.c

包括初始化、进程管理、进程通信、内存管理、设备管理、中断、文件系统、系统调用等精选源码分析。

/*
 *  linux/fs/exec.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 */

/*
 * #!-checking implemented by tytso.
 */
/*
 * Demand-loading implemented 01.12.91 - no need to read anything but
 * the header into memory. The inode of the executable is put into
 * "current->executable", and page faults do the actual loading. Clean.
 *
 * Once more I can proudly say that linux stood up to being changed: it
 * was less than 2 hours work to get demand-loading completely implemented.
 *
 * Demand loading changed July 1993 by Eric Youngdale.   Use mmap instead,
 * current->executable is only used by the procfs.  This allows a dispatch
 * table to check for several different types  of binary formats.  We keep
 * trying until we recognize the file or we run out of supported binary
 * formats. 
 */

#include <linux/config.h>
#include <linux/slab.h>
#include <linux/file.h>
#include <linux/mman.h>
#include <linux/a.out.h>
#include <linux/stat.h>
#include <linux/fcntl.h>
#include <linux/smp_lock.h>
#include <linux/init.h>
#include <linux/pagemap.h>
#include <linux/highmem.h>
#include <linux/spinlock.h>
#define __NO_VERSION__
#include <linux/module.h>

#include <asm/uaccess.h>
#include <asm/pgalloc.h>
#include <asm/mmu_context.h>

#ifdef CONFIG_KMOD
#include <linux/kmod.h>
#endif

static struct linux_binfmt *formats;
static rwlock_t binfmt_lock = RW_LOCK_UNLOCKED;

int register_binfmt(struct linux_binfmt * fmt)
{
	struct linux_binfmt ** tmp = &formats;

	if (!fmt)
		return -EINVAL;
	if (fmt->next)
		return -EBUSY;
	write_lock(&binfmt_lock);
	while (*tmp) {
		if (fmt == *tmp) {
			write_unlock(&binfmt_lock);
			return -EBUSY;
		}
		tmp = &(*tmp)->next;
	}
	fmt->next = formats;
	formats = fmt;
	write_unlock(&binfmt_lock);
	return 0;	
}

int unregister_binfmt(struct linux_binfmt * fmt)
{
	struct linux_binfmt ** tmp = &formats;

	write_lock(&binfmt_lock);
	while (*tmp) {
		if (fmt == *tmp) {
			*tmp = fmt->next;
			write_unlock(&binfmt_lock);
			return 0;
		}
		tmp = &(*tmp)->next;
	}
	write_unlock(&binfmt_lock);
	return -EINVAL;
}

static inline void put_binfmt(struct linux_binfmt * fmt)
{
	if (fmt->module)
		__MOD_DEC_USE_COUNT(fmt->module);
}

/*
 * Note that a shared library must be both readable and executable due to
 * security reasons.
 *
 * Also note that we take the address to load from from the file itself.
 */
asmlinkage long sys_uselib(const char * library)
{
	struct file * file;
	struct nameidata nd;
	int error;

	error = user_path_walk(library, &nd);
	if (error)
		goto out;

	error = -EINVAL;
	if (!S_ISREG(nd.dentry->d_inode->i_mode))
		goto exit;

	error = permission(nd.dentry->d_inode, MAY_READ | MAY_EXEC);
	if (error)
		goto exit;

	file = dentry_open(nd.dentry, nd.mnt, O_RDONLY);
	error = PTR_ERR(file);
	if (IS_ERR(file))
		goto out;

	error = -ENOEXEC;
	if(file->f_op && file->f_op->read) {
		struct linux_binfmt * fmt;

		read_lock(&binfmt_lock);
		for (fmt = formats ; fmt ; fmt = fmt->next) {
			if (!fmt->load_shlib)
				continue;
			if (!try_inc_mod_count(fmt->module))
				continue;
			read_unlock(&binfmt_lock);
			error = fmt->load_shlib(file);
			read_lock(&binfmt_lock);
			put_binfmt(fmt);
			if (error != -ENOEXEC)
				break;
		}
		read_unlock(&binfmt_lock);
	}
	fput(file);
out:
  	return error;
exit:
	path_release(&nd);
	goto out;
}

/*
 * count() counts the number of arguments/envelopes
 */
static int count(char ** argv, int max)
{
	int i = 0;

	if (argv != NULL) {
		for (;;) {
			char * p;
			int error;

			error = get_user(p,argv);
			if (error)
				return error;
			if (!p)
				break;
			argv++;
			if(++i > max)
				return -E2BIG;
		}
	}
	return i;
}

/*
 * 'copy_strings()' copies argument/envelope strings from user
 * memory to free pages in kernel mem. These are in a format ready
 * to be put directly into the top of new user memory.
 */
int copy_strings(int argc,char ** argv, struct linux_binprm *bprm) 
{
	while (argc-- > 0) {
		char *str;
		int len;
		unsigned long pos;

		if (get_user(str, argv+argc) || !str || !(len = strnlen_user(str, bprm->p))) 
			return -EFAULT;
		if (bprm->p < len) 
			return -E2BIG; 

		bprm->p -= len;
		/* XXX: add architecture specific overflow check here. */ 

		pos = bprm->p;
		while (len > 0) {
			char *kaddr;
			int i, new, err;
			struct page *page;
			int offset, bytes_to_copy;

			offset = pos % PAGE_SIZE;
			i = pos/PAGE_SIZE;
			page = bprm->page[i];
			new = 0;
			if (!page) {
				page = alloc_page(GFP_HIGHUSER);
				bprm->page[i] = page;
				if (!page)
					return -ENOMEM;
				new = 1;
			}
			kaddr = kmap(page);

			if (new && offset)
				memset(kaddr, 0, offset);
			bytes_to_copy = PAGE_SIZE - offset;
			if (bytes_to_copy > len) {
				bytes_to_copy = len;
				if (new)
					memset(kaddr+offset+len, 0, PAGE_SIZE-offset-len);
			}
			err = copy_from_user(kaddr + offset, str, bytes_to_copy);
			kunmap(page);

			if (err)
				return -EFAULT; 

			pos += bytes_to_copy;
			str += bytes_to_copy;
			len -= bytes_to_copy;
		}
	}
	return 0;
}

/*
 * Like copy_strings, but get argv and its values from kernel memory.
 */
int copy_strings_kernel(int argc,char ** argv, struct linux_binprm *bprm)
{
	int r;
	mm_segment_t oldfs = get_fs();
	set_fs(KERNEL_DS); 
	r = copy_strings(argc, argv, bprm);
	set_fs(oldfs);
	return r; 
}

/*
 * This routine is used to map in a page into an address space: needed by
 * execve() for the initial stack and environment pages.
 */
void put_dirty_page(struct task_struct * tsk, struct page *page, unsigned long address)
{
	pgd_t * pgd;
	pmd_t * pmd;
	pte_t * pte;

	if (page_count(page) != 1)
		printk("mem_map disagrees with %p at %08lx\n", page, address);
	pgd = pgd_offset(tsk->mm, address);
	pmd = pmd_alloc(pgd, address);
	if (!pmd) {
		__free_page(page);
		force_sig(SIGKILL, tsk);
		return;
	}
	pte = pte_alloc(pmd, address);
	if (!pte) {
		__free_page(page);
		force_sig(SIGKILL, tsk);
		return;
	}
	if (!pte_none(*pte)) {
		pte_ERROR(*pte);
		__free_page(page);
		return;
	}
	flush_dcache_page(page);
	flush_page_to_ram(page);
	set_pte(pte, pte_mkdirty(pte_mkwrite(mk_pte(page, PAGE_COPY))));
/* no need for flush_tlb */
}

int setup_arg_pages(struct linux_binprm *bprm)
{
	unsigned long stack_base;
	struct vm_area_struct *mpnt;
	int i;

	stack_base = STACK_TOP - MAX_ARG_PAGES*PAGE_SIZE;

	bprm->p += stack_base;
	if (bprm->loader)
		bprm->loader += stack_base;
	bprm->exec += stack_base;

	mpnt = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL);
	if (!mpnt) 
		return -ENOMEM; 
	
	down(&current->mm->mmap_sem);
	{
		mpnt->vm_mm = current->mm;
		mpnt->vm_start = PAGE_MASK & (unsigned long) bprm->p;
		mpnt->vm_end = STACK_TOP;
		mpnt->vm_page_prot = PAGE_COPY;
		mpnt->vm_flags = VM_STACK_FLAGS;
		mpnt->vm_ops = NULL;
		mpnt->vm_pgoff = 0;
		mpnt->vm_file = NULL;
		mpnt->vm_private_data = (void *) 0;
		insert_vm_struct(current->mm, mpnt);
		current->mm->total_vm = (mpnt->vm_end - mpnt->vm_start) >> PAGE_SHIFT;
	} 

	for (i = 0 ; i < MAX_ARG_PAGES ; i++) {
		struct page *page = bprm->page[i];
		if (page) {
			bprm->page[i] = NULL;
			current->mm->rss++;
			put_dirty_page(current,page,stack_base);
		}
		stack_base += PAGE_SIZE;
	}
	up(&current->mm->mmap_sem);
	
	return 0;
}

struct file *open_exec(const char *name)
{
	struct nameidata nd;
	struct inode *inode;
	struct file *file;
	int err = 0;

	if (path_init(name, LOOKUP_FOLLOW|LOOKUP_POSITIVE, &nd))
		err = path_walk(name, &nd);
	file = ERR_PTR(err);
	if (!err) {
		inode = nd.dentry->d_inode;
		file = ERR_PTR(-EACCES);
		if (!IS_NOEXEC(inode) && S_ISREG(inode->i_mode)) {
			int err = permission(inode, MAY_EXEC);
			file = ERR_PTR(err);
			if (!err) {
				file = dentry_open(nd.dentry, nd.mnt, O_RDONLY);
				if (!IS_ERR(file)) {
					err = deny_write_access(file);
					if (err) {
						fput(file);
						file = ERR_PTR(err);
					}
				}
out:
				return file;
			}
		}
		path_release(&nd);
	}
	goto out;
}

int kernel_read(struct file *file, unsigned long offset,
	char * addr, unsigned long count)
{
	mm_segment_t old_fs;
	loff_t pos = offset;
	int result = -ENOSYS;

	if (!file->f_op->read)
		goto fail;
	old_fs = get_fs();
	set_fs(get_ds());
	result = file->f_op->read(file, addr, count, &pos);
	set_fs(old_fs);
fail:
	return result;
}

static int exec_mmap(void)
{
	struct mm_struct * mm, * old_mm;

	old_mm = current->mm;
	if (old_mm && atomic_read(&old_mm->mm_users) == 1) {
		flush_cache_mm(old_mm);
		mm_release();
		exit_mmap(old_mm);
		flush_tlb_mm(old_mm);
		return 0;
	}

	mm = mm_alloc();
	if (mm) {
		struct mm_struct *active_mm = current->active_mm;

		if (init_new_context(current, mm)) {
			mmdrop(mm);
			return -ENOMEM;
		}

		/* Add it to the list of mm's */
		spin_lock(&mmlist_lock);
		list_add(&mm->mmlist, &init_mm.mmlist);
		spin_unlock(&mmlist_lock);

		task_lock(current);
		current->mm = mm;
		current->active_mm = mm;
		task_unlock(current);
		activate_mm(active_mm, mm);
		mm_release();
		if (old_mm) {
			if (active_mm != old_mm) BUG();
			mmput(old_mm);
			return 0;
		}
		mmdrop(active_mm);
		return 0;
	}
	return -ENOMEM;
}

/*
 * This function makes sure the current process has its own signal table,
 * so that flush_signal_handlers can later reset the handlers without
 * disturbing other processes.  (Other processes might share the signal
 * table via the CLONE_SIGNAL option to clone().)
 */
 
static inline int make_private_signals(void)
{
	struct signal_struct * newsig;

	if (atomic_read(&current->sig->count) <= 1)
		return 0;
	newsig = kmem_cache_alloc(sigact_cachep, GFP_KERNEL);
	if (newsig == NULL)
		return -ENOMEM;
	spin_lock_init(&newsig->siglock);
	atomic_set(&newsig->count, 1);
	memcpy(newsig->action, current->sig->action, sizeof(newsig->action));
	spin_lock_irq(&current->sigmask_lock);
	current->sig = newsig;
	spin_unlock_irq(&current->sigmask_lock);
	return 0;
}
	
/*
 * If make_private_signals() made a copy of the signal table, decrement the
 * refcount of the original table, and free it if necessary.
 * We don't do that in make_private_signals() so that we can back off
 * in flush_old_exec() if an error occurs after calling make_private_signals().
 */

static inline void release_old_signals(struct signal_struct * oldsig)
{
	if (current->sig == oldsig)
		return;
	if (atomic_dec_and_test(&oldsig->count))
		kmem_cache_free(sigact_cachep, oldsig);
}

/*
 * These functions flushes out all traces of the currently running executable
 * so that a new one can be started
 */

static inline void flush_old_files(struct files_struct * files)
{
	long j = -1;

	write_lock(&files->file_lock);
	for (;;) {
		unsigned long set, i;

		j++;
		i = j * __NFDBITS;