cryptsetup/lib/utils.c

/*
 * utils - miscellaneous device utilities for cryptsetup
 *
 * Copyright (C) 2004, Christophe Saout <christophe@saout.de>
 * Copyright (C) 2004-2007, Clemens Fruhwirth <clemens@endorphin.org>
 * Copyright (C) 2009-2011, Red Hat, Inc. All rights reserved.
 *
 * 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.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 */

#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <stddef.h>
#include <stdarg.h>
#include <errno.h>
#include <linux/fs.h>
#include <sys/types.h>
#include <unistd.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <sys/ioctl.h>
#include <fcntl.h>
#include <sys/mman.h>
#include <sys/resource.h>

#include "libcryptsetup.h"
#include "internal.h"

static char *error=NULL;

__attribute__((format(printf, 1, 0)))
void set_error_va(const char *fmt, va_list va)
{
	int r;

	if(error) {
		free(error);
		error = NULL;
	}

	if(!fmt) return;

	r = vasprintf(&error, fmt, va);
	if (r < 0) {
		free(error);
		error = NULL;
		return;
	}

	if (r && error[r - 1] == '\n')
		error[r - 1] = '\0';
}

__attribute__((format(printf, 1, 2)))
void set_error(const char *fmt, ...)
{
	va_list va;

	va_start(va, fmt);
	set_error_va(fmt, va);
	va_end(va);
}

const char *get_error(void)
{
	return error;
}

static int get_alignment(int fd)
{
	int alignment = DEFAULT_MEM_ALIGNMENT;

#ifdef _PC_REC_XFER_ALIGN
	alignment = fpathconf(fd, _PC_REC_XFER_ALIGN);
	if (alignment < 0)
		alignment = DEFAULT_MEM_ALIGNMENT;
#endif
	return alignment;
}

static void *aligned_malloc(void **base, int size, int alignment)
{
#ifdef HAVE_POSIX_MEMALIGN
	return posix_memalign(base, alignment, size) ? NULL : *base;
#else
/* Credits go to Michal's padlock patches for this alignment code */
	char *ptr;

	ptr  = malloc(size + alignment);
	if(ptr == NULL) return NULL;

	*base = ptr;
	if(alignment > 1 && ((long)ptr & (alignment - 1))) {
		ptr += alignment - ((long)(ptr) & (alignment - 1));
	}
	return ptr;
#endif
}

int device_read_ahead(const char *dev, uint32_t *read_ahead)
{
	int fd, r = 0;
	long read_ahead_long;

	if ((fd = open(dev, O_RDONLY)) < 0)
		return 0;

	r = ioctl(fd, BLKRAGET, &read_ahead_long) ? 0 : 1;
	close(fd);

	if (r)
		*read_ahead = (uint32_t) read_ahead_long;

	return r;
}

static int sector_size(int fd) 
{
	int bsize;
	if (ioctl(fd,BLKSSZGET, &bsize) < 0)
		return -EINVAL;
	else
		return bsize;
}

int sector_size_for_device(const char *device)
{
	int fd = open(device, O_RDONLY);
	int r;
	if(fd < 0)
		return -EINVAL;
	r = sector_size(fd);
	close(fd);
	return r;
}

ssize_t write_blockwise(int fd, void *orig_buf, size_t count)
{
	void *hangover_buf, *hangover_buf_base = NULL;
	void *buf, *buf_base = NULL;
	int r, hangover, solid, bsize, alignment;
	ssize_t ret = -1;

	if ((bsize = sector_size(fd)) < 0)
		return bsize;

	hangover = count % bsize;
	solid = count - hangover;
	alignment = get_alignment(fd);

	if ((long)orig_buf & (alignment - 1)) {
		buf = aligned_malloc(&buf_base, count, alignment);
		if (!buf)
			goto out;
		memcpy(buf, orig_buf, count);
	} else
		buf = orig_buf;

	r = write(fd, buf, solid);
	if (r < 0 || r != solid)
		goto out;

	if (hangover) {
		hangover_buf = aligned_malloc(&hangover_buf_base, bsize, alignment);
		if (!hangover_buf)
			goto out;

		r = read(fd, hangover_buf, bsize);
		if (r < 0 || r != bsize)
			goto out;

		r = lseek(fd, -bsize, SEEK_CUR);
		if (r < 0)
			goto out;
		memcpy(hangover_buf, (char*)buf + solid, hangover);

		r = write(fd, hangover_buf, bsize);
		if (r < 0 || r != bsize)
			goto out;
	}
	ret = count;
out:
	free(hangover_buf_base);
	if (buf != orig_buf)
		free(buf_base);
	return ret;
}

ssize_t read_blockwise(int fd, void *orig_buf, size_t count) {
	void *hangover_buf, *hangover_buf_base = NULL;
	void *buf, *buf_base = NULL;
	int r, hangover, solid, bsize, alignment;
	ssize_t ret = -1;

	if ((bsize = sector_size(fd)) < 0)
		return bsize;

	hangover = count % bsize;
	solid = count - hangover;
	alignment = get_alignment(fd);

	if ((long)orig_buf & (alignment - 1)) {
		buf = aligned_malloc(&buf_base, count, alignment);
		if (!buf)
			return -1;
	} else
		buf = orig_buf;

	r = read(fd, buf, solid);
	if(r < 0 || r != solid)
		goto out;

	if (hangover) {
		hangover_buf = aligned_malloc(&hangover_buf_base, bsize, alignment);
		if (!hangover_buf)
			goto out;
		r = read(fd, hangover_buf, bsize);
		if (r <  0 || r != bsize)
			goto out;

		memcpy((char *)buf + solid, hangover_buf, hangover);
	}
	ret = count;
out:
	free(hangover_buf_base);
	if (buf != orig_buf) {
		memcpy(orig_buf, buf, count);
		free(buf_base);
	}
	return ret;
}

/*
 * Combines llseek with blockwise write. write_blockwise can already deal with short writes
 * but we also need a function to deal with short writes at the start. But this information
 * is implicitly included in the read/write offset, which can not be set to non-aligned
 * boundaries. Hence, we combine llseek with write.
 */
ssize_t write_lseek_blockwise(int fd, char *buf, size_t count, off_t offset) {
	char *frontPadBuf;
	void *frontPadBuf_base = NULL;
	int r, bsize, frontHang;
	size_t innerCount = 0;
	ssize_t ret = -1;

	if ((bsize = sector_size(fd)) < 0)
		return bsize;

	frontHang = offset % bsize;

	if (lseek(fd, offset - frontHang, SEEK_SET) < 0)
		goto out;

	if (frontHang) {
		frontPadBuf = aligned_malloc(&frontPadBuf_base,
					     bsize, get_alignment(fd));
		if (!frontPadBuf)
			goto out;

		r = read(fd, frontPadBuf, bsize);
		if (r < 0 || r != bsize)
			goto out;

		innerCount = bsize - frontHang;
		if (innerCount > count)
			innerCount = count;

		memcpy(frontPadBuf + frontHang, buf, innerCount);

		if (lseek(fd, offset - frontHang, SEEK_SET) < 0)
			goto out;

		r = write(fd, frontPadBuf, bsize);
		if (r < 0 || r != bsize)
			goto out;

		buf += innerCount;
		count -= innerCount;
	}

	ret = count ? write_blockwise(fd, buf, count) : 0;
	if (ret >= 0)
		ret += innerCount;
out:
	free(frontPadBuf_base);

	return ret;
}

int device_ready(struct crypt_device *cd, const char *device, int mode)
{
	int devfd, r = 0;
	ssize_t s;
	struct stat st;
	char buf[512];

	if(stat(device, &st) < 0) {
		log_err(cd, _("Device %s doesn't exist or access denied.\n"), device);
		return -EINVAL;
	}

	if (!S_ISBLK(st.st_mode))
		return -ENOTBLK;

	log_dbg("Trying to open and read device %s.", device);
	devfd = open(device, mode | O_DIRECT | O_SYNC);
	if(devfd < 0) {
		log_err(cd, _("Cannot open device %s for %s%s access.\n"), device,
			(mode & O_EXCL) ? _("exclusive ") : "",
			(mode & O_RDWR) ? _("writable") : _("read-only"));
		return -EINVAL;
	}

	 /* Try to read first sector */
	s = read_blockwise(devfd, buf, sizeof(buf));
	if (s < 0 || s != sizeof(buf)) {
		log_verbose(cd, _("Cannot read device %s.\n"), device);
		r = -EIO;
	}

	memset(buf, 0, sizeof(buf));
	close(devfd);

	return r;
}

int device_size(const char *device, uint64_t *size)
{
	int devfd, r = 0;

	devfd = open(device, O_RDONLY);
	if(devfd == -1)
		return -EINVAL;

	if (ioctl(devfd, BLKGETSIZE64, size) < 0)
		r = -EINVAL;

	close(devfd);
	return r;
}

static int get_device_infos(const char *device, enum devcheck device_check,
			    int *readonly, uint64_t *size)
{
	struct stat st;
	unsigned long size_small;
	int fd, r = -1;
	int flags = 0;

	*readonly = 0;
	*size = 0;

	if (stat(device, &st) < 0)
		return -EINVAL;

	/* never wipe header on mounted device */
	if (device_check == DEV_EXCL && S_ISBLK(st.st_mode))
		flags |= O_EXCL;

	/* Try to open read-write to check whether it is a read-only device */
	fd = open(device, O_RDWR | flags);
	if (fd == -1 && errno == EROFS) {
		*readonly = 1;
		fd = open(device, O_RDONLY | flags);
	}

	if (fd == -1 && device_check == DEV_EXCL && errno == EBUSY)
		return -EBUSY;

	if (fd == -1)
		return -EINVAL;

	/* If the device can be opened read-write, i.e. readonly is still 0, then
	 * check whether BKROGET says that it is read-only. E.g. read-only loop
	 * devices may be openend read-write but are read-only according to BLKROGET
	 */
	if (*readonly == 0 && (r = ioctl(fd, BLKROGET, readonly)) < 0)
		goto out;

	if (ioctl(fd, BLKGETSIZE64, size) >= 0) {
		*size >>= SECTOR_SHIFT;
		r = 0;
		goto out;
	}

	if (ioctl(fd, BLKGETSIZE, &size_small) >= 0) {
		*size = (uint64_t)size_small;
		r = 0;
		goto out;
	}

	r = -EINVAL;
out:
	close(fd);
	return r;
}

int device_check_and_adjust(struct crypt_device *cd,
			    const char *device,
			    enum devcheck device_check,
			    uint64_t *size,
			    uint64_t *offset,
			    uint32_t *flags)
{
	int r, real_readonly;
	uint64_t real_size;

	if (!device)
		return -ENOTBLK;

	r = get_device_infos(device, device_check, &real_readonly, &real_size);
	if (r < 0) {
		if (r == -EBUSY)
			log_err(cd, _("Cannot use device %s which is in use "
				      "(already mapped or mounted).\n"),
				      device);
		else
			log_err(cd, _("Cannot get info about device %s.\n"),
				device);
		return r;
	}

	if (*offset >= real_size) {
		log_err(cd, _("Requested offset is beyond real size of device %s.\n"),
			device);
		return -EINVAL;
	}

	if (!*size) {
		*size = real_size;
		if (!*size) {
			log_err(cd, _("Device %s has zero size.\n"), device);
			return -ENOTBLK;
		}
		*size -= *offset;
	}

	/* in case of size is set by parameter */
	if ((real_size - *offset) < *size) {
		log_dbg("Device %s: offset = %" PRIu64 " requested size = %" PRIu64
			", backing device size = %" PRIu64,
			device, *offset, *size, real_size);
		log_err(cd, _("Device %s is too small.\n"), device);
		return -EINVAL;
	}

	if (device_check == DEV_SHARED) {
		log_dbg("Checking crypt segments for device %s.", device);
		r = crypt_sysfs_check_crypt_segment(device, *offset, *size);
		if (r < 0) {
			log_err(cd, _("Cannot use device %s (crypt segments "
				    "overlaps or in use by another device).\n"),
				    device);
			return r;
		}
	}

	if (real_readonly)
		*flags |= CRYPT_ACTIVATE_READONLY;

	log_dbg("Calculated device size is %" PRIu64 " sectors (%s), offset %" PRIu64 ".",
		*size, real_readonly ? "RO" : "RW", *offset);
	return 0;
}

int wipe_device_header(const char *device, int sectors)
{
	struct stat st;
	char *buffer;
	int size = sectors * SECTOR_SIZE;
	int r = -1;
	int devfd;
	int flags = O_RDWR | O_DIRECT | O_SYNC;

	if (stat(device, &st) < 0)
		return -EINVAL;

	/* never wipe header on mounted device */
	if (S_ISBLK(st.st_mode))
		flags |= O_EXCL;

	devfd = open(device, flags);
	if(devfd == -1)
		return errno == EBUSY ? -EBUSY : -EINVAL;

	buffer = malloc(size);
	if (!buffer) {
		close(devfd);
		return -ENOMEM;
	}
	memset(buffer, 0, size);

	r = write_blockwise(devfd, buffer, size) < size ? -EIO : 0;

	free(buffer);
	close(devfd);

	return r;
}

/* MEMLOCK */
#define DEFAULT_PROCESS_PRIORITY -18

static int _priority;
static int _memlock_count = 0;

// return 1 if memory is locked
int crypt_memlock_inc(struct crypt_device *ctx)
{
	if (!_memlock_count++) {
		log_dbg("Locking memory.");
		if (mlockall(MCL_CURRENT | MCL_FUTURE) == -1) {
			log_err(ctx, _("WARNING!!! Possibly insecure memory. Are you root?\n"));
			_memlock_count--;
			return 0;
		}
		errno = 0;
		if (((_priority = getpriority(PRIO_PROCESS, 0)) == -1) && errno)
			log_err(ctx, _("Cannot get process priority.\n"));
		else
			if (setpriority(PRIO_PROCESS, 0, DEFAULT_PROCESS_PRIORITY))
				log_err(ctx, _("setpriority %d failed: %s\n"),
					DEFAULT_PROCESS_PRIORITY, strerror(errno));
	}
	return _memlock_count ? 1 : 0;
}

int crypt_memlock_dec(struct crypt_device *ctx)
{
	if (_memlock_count && (!--_memlock_count)) {
		log_dbg("Unlocking memory.");
		if (munlockall() == -1)
			log_err(ctx, _("Cannot unlock memory.\n"));
		if (setpriority(PRIO_PROCESS, 0, _priority))
			log_err(ctx, _("setpriority %d failed: %s\n"), _priority, strerror(errno));
	}
	return _memlock_count ? 1 : 0;
}

/* DEVICE TOPOLOGY */

/* block device topology ioctls, introduced in 2.6.32 */
#ifndef BLKIOMIN
#define BLKIOMIN    _IO(0x12,120)
#define BLKIOOPT    _IO(0x12,121)
#define BLKALIGNOFF _IO(0x12,122)
#endif

void get_topology_alignment(const char *device,
			    unsigned long *required_alignment, /* bytes */
			    unsigned long *alignment_offset,   /* bytes */
			    unsigned long default_alignment)
{
	int dev_alignment_offset = 0;
	unsigned int min_io_size = 0, opt_io_size = 0;
	unsigned long temp_alignment = 0;
	int fd;

	*required_alignment = default_alignment;
	*alignment_offset = 0;

	fd = open(device, O_RDONLY);
	if (fd == -1)
		return;

	/* minimum io size */
	if (ioctl(fd, BLKIOMIN, &min_io_size) == -1) {
		log_dbg("Topology info for %s not supported, using default offset %lu bytes.",
			device, default_alignment);
		goto out;
	}

	/* optimal io size */
	if (ioctl(fd, BLKIOOPT, &opt_io_size) == -1)
		opt_io_size = min_io_size;

	/* alignment offset, bogus -1 means misaligned/unknown */
	if (ioctl(fd, BLKALIGNOFF, &dev_alignment_offset) == -1 || dev_alignment_offset < 0)
		dev_alignment_offset = 0;
	*alignment_offset = (unsigned long)dev_alignment_offset;

	temp_alignment = (unsigned long)min_io_size;

	if (temp_alignment < (unsigned long)opt_io_size)
		temp_alignment = (unsigned long)opt_io_size;

	/* If calculated alignment is multiple of default, keep default */
	if (temp_alignment && (default_alignment % temp_alignment))
		*required_alignment = temp_alignment;

	log_dbg("Topology: IO (%u/%u), offset = %lu; Required alignment is %lu bytes.",
		min_io_size, opt_io_size, *alignment_offset, *required_alignment);
out:
	(void)close(fd);
}