[mirror_zfs.git] / module / zfs / zio_checksum.c

/*
 * CDDL HEADER START
 *
 * The contents of this file are subject to the terms of the
 * Common Development and Distribution License (the "License").
 * You may not use this file except in compliance with the License.
 *
 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
 * or http://www.opensolaris.org/os/licensing.
 * See the License for the specific language governing permissions
 * and limitations under the License.
 *
 * When distributing Covered Code, include this CDDL HEADER in each
 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
 * If applicable, add the following below this CDDL HEADER, with the
 * fields enclosed by brackets "[]" replaced with your own identifying
 * information: Portions Copyright [yyyy] [name of copyright owner]
 *
 * CDDL HEADER END
 */
/*
 * Copyright 2008 Sun Microsystems, Inc.  All rights reserved.
 * Use is subject to license terms.
 */

#include <sys/zfs_context.h>
#include <sys/spa.h>
#include <sys/zio.h>
#include <sys/zio_checksum.h>

/*
 * Checksum vectors.
 *
 * In the SPA, everything is checksummed.  We support checksum vectors
 * for three distinct reasons:
 *
 *   1. Different kinds of data need different levels of protection.
 *	For SPA metadata, we always want a very strong checksum.
 *	For user data, we let users make the trade-off between speed
 *	and checksum strength.
 *
 *   2. Cryptographic hash and MAC algorithms are an area of active research.
 *	It is likely that in future hash functions will be at least as strong
 *	as current best-of-breed, and may be substantially faster as well.
 *	We want the ability to take advantage of these new hashes as soon as
 *	they become available.
 *
 *   3. If someone develops hardware that can compute a strong hash quickly,
 *	we want the ability to take advantage of that hardware.
 *
 * Of course, we don't want a checksum upgrade to invalidate existing
 * data, so we store the checksum *function* in five bits of the DVA.
 * This gives us room for up to 32 different checksum functions.
 *
 * When writing a block, we always checksum it with the latest-and-greatest
 * checksum function of the appropriate strength.  When reading a block,
 * we compare the expected checksum against the actual checksum, which we
 * compute via the checksum function specified in the DVA encoding.
 */

/*ARGSUSED*/
static void
zio_checksum_off(const void *buf, uint64_t size, zio_cksum_t *zcp)
{
	ZIO_SET_CHECKSUM(zcp, 0, 0, 0, 0);
}

zio_checksum_info_t zio_checksum_table[ZIO_CHECKSUM_FUNCTIONS] = {
	{{NULL,			NULL},			0, 0,	"inherit"},
	{{NULL,			NULL},			0, 0,	"on"},
	{{zio_checksum_off,	zio_checksum_off},	0, 0,	"off"},
	{{zio_checksum_SHA256,	zio_checksum_SHA256},	1, 1,	"label"},
	{{zio_checksum_SHA256,	zio_checksum_SHA256},	1, 1,	"gang_header"},
	{{fletcher_2_native,	fletcher_2_byteswap},	0, 1,	"zilog"},
	{{fletcher_2_native,	fletcher_2_byteswap},	0, 0,	"fletcher2"},
	{{fletcher_4_native,	fletcher_4_byteswap},	1, 0,	"fletcher4"},
	{{zio_checksum_SHA256,	zio_checksum_SHA256},	1, 0,	"SHA256"},
};

uint8_t
zio_checksum_select(uint8_t child, uint8_t parent)
{
	ASSERT(child < ZIO_CHECKSUM_FUNCTIONS);
	ASSERT(parent < ZIO_CHECKSUM_FUNCTIONS);
	ASSERT(parent != ZIO_CHECKSUM_INHERIT && parent != ZIO_CHECKSUM_ON);

	if (child == ZIO_CHECKSUM_INHERIT)
		return (parent);

	if (child == ZIO_CHECKSUM_ON)
		return (ZIO_CHECKSUM_ON_VALUE);

	return (child);
}

/*
 * Set the external verifier for a gang block based on <vdev, offset, txg>,
 * a tuple which is guaranteed to be unique for the life of the pool.
 */
static void
zio_checksum_gang_verifier(zio_cksum_t *zcp, blkptr_t *bp)
{
	dva_t *dva = BP_IDENTITY(bp);
	uint64_t txg = bp->blk_birth;

	ASSERT(BP_IS_GANG(bp));

	ZIO_SET_CHECKSUM(zcp, DVA_GET_VDEV(dva), DVA_GET_OFFSET(dva), txg, 0);
}

/*
 * Set the external verifier for a label block based on its offset.
 * The vdev is implicit, and the txg is unknowable at pool open time --
 * hence the logic in vdev_uberblock_load() to find the most recent copy.
 */
static void
zio_checksum_label_verifier(zio_cksum_t *zcp, uint64_t offset)
{
	ZIO_SET_CHECKSUM(zcp, offset, 0, 0, 0);
}

/*
 * Generate the checksum.
 */
void
zio_checksum_compute(zio_t *zio, enum zio_checksum checksum,
	void *data, uint64_t size)
{
	blkptr_t *bp = zio->io_bp;
	uint64_t offset = zio->io_offset;
	zio_block_tail_t *zbt = (zio_block_tail_t *)((char *)data + size) - 1;
	zio_checksum_info_t *ci = &zio_checksum_table[checksum];
	zio_cksum_t zbt_cksum;

	ASSERT((uint_t)checksum < ZIO_CHECKSUM_FUNCTIONS);
	ASSERT(ci->ci_func[0] != NULL);

	if (ci->ci_zbt) {
		if (checksum == ZIO_CHECKSUM_GANG_HEADER)
			zio_checksum_gang_verifier(&zbt->zbt_cksum, bp);
		else if (checksum == ZIO_CHECKSUM_LABEL)
			zio_checksum_label_verifier(&zbt->zbt_cksum, offset);
		else
			bp->blk_cksum = zbt->zbt_cksum;
		zbt->zbt_magic = ZBT_MAGIC;
		ci->ci_func[0](data, size, &zbt_cksum);
		zbt->zbt_cksum = zbt_cksum;
	} else {
		ci->ci_func[0](data, size, &bp->blk_cksum);
	}
}

int
zio_checksum_error(zio_t *zio)
{
	blkptr_t *bp = zio->io_bp;
	uint_t checksum = (bp == NULL ? zio->io_prop.zp_checksum :
	    (BP_IS_GANG(bp) ? ZIO_CHECKSUM_GANG_HEADER : BP_GET_CHECKSUM(bp)));
	int byteswap;
	void *data = zio->io_data;
	uint64_t size = (bp == NULL ? zio->io_size :
	    (BP_IS_GANG(bp) ? SPA_GANGBLOCKSIZE : BP_GET_PSIZE(bp)));
	uint64_t offset = zio->io_offset;
	zio_block_tail_t *zbt = (zio_block_tail_t *)((char *)data + size) - 1;
	zio_checksum_info_t *ci = &zio_checksum_table[checksum];
	zio_cksum_t actual_cksum, expected_cksum, verifier;

	if (checksum >= ZIO_CHECKSUM_FUNCTIONS || ci->ci_func[0] == NULL)
		return (EINVAL);

	if (ci->ci_zbt) {
		if (checksum == ZIO_CHECKSUM_GANG_HEADER)
			zio_checksum_gang_verifier(&verifier, bp);
		else if (checksum == ZIO_CHECKSUM_LABEL)
			zio_checksum_label_verifier(&verifier, offset);
		else
			verifier = bp->blk_cksum;

		byteswap = (zbt->zbt_magic == BSWAP_64(ZBT_MAGIC));

		if (byteswap)
			byteswap_uint64_array(&verifier, sizeof (zio_cksum_t));

		expected_cksum = zbt->zbt_cksum;
		zbt->zbt_cksum = verifier;
		ci->ci_func[byteswap](data, size, &actual_cksum);
		zbt->zbt_cksum = expected_cksum;

		if (byteswap)
			byteswap_uint64_array(&expected_cksum,
			    sizeof (zio_cksum_t));
	} else {
		ASSERT(!BP_IS_GANG(bp));
		byteswap = BP_SHOULD_BYTESWAP(bp);
		expected_cksum = bp->blk_cksum;
		ci->ci_func[byteswap](data, size, &actual_cksum);
	}

	if (!ZIO_CHECKSUM_EQUAL(actual_cksum, expected_cksum))
		return (ECKSUM);

	if (zio_injection_enabled && !zio->io_error)
		return (zio_handle_fault_injection(zio, ECKSUM));

	return (0);
}
Commit	Line	Data
34dc7c2f BB	1	/*
	2	* CDDL HEADER START
	3	*
	4	* The contents of this file are subject to the terms of the
	5	* Common Development and Distribution License (the "License").
	6	* You may not use this file except in compliance with the License.
	7	*
	8	* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
	9	* or http://www.opensolaris.org/os/licensing.
	10	* See the License for the specific language governing permissions
	11	* and limitations under the License.
	12	*
	13	* When distributing Covered Code, include this CDDL HEADER in each
	14	* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
	15	* If applicable, add the following below this CDDL HEADER, with the
	16	* fields enclosed by brackets "[]" replaced with your own identifying
	17	* information: Portions Copyright [yyyy] [name of copyright owner]
	18	*
	19	* CDDL HEADER END
	20	*/
	21	/*
b128c09f	22	* Copyright 2008 Sun Microsystems, Inc. All rights reserved.
34dc7c2f BB	23	* Use is subject to license terms.
	24	*/
	25
34dc7c2f BB	26	#include <sys/zfs_context.h>
	27	#include <sys/spa.h>
	28	#include <sys/zio.h>
	29	#include <sys/zio_checksum.h>
	30
	31	/*
	32	* Checksum vectors.
	33	*
	34	* In the SPA, everything is checksummed. We support checksum vectors
	35	* for three distinct reasons:
	36	*
	37	* 1. Different kinds of data need different levels of protection.
	38	* For SPA metadata, we always want a very strong checksum.
	39	* For user data, we let users make the trade-off between speed
	40	* and checksum strength.
	41	*
	42	* 2. Cryptographic hash and MAC algorithms are an area of active research.
	43	* It is likely that in future hash functions will be at least as strong
	44	* as current best-of-breed, and may be substantially faster as well.
	45	* We want the ability to take advantage of these new hashes as soon as
	46	* they become available.
	47	*
	48	* 3. If someone develops hardware that can compute a strong hash quickly,
	49	* we want the ability to take advantage of that hardware.
	50	*
	51	* Of course, we don't want a checksum upgrade to invalidate existing
	52	* data, so we store the checksum function in five bits of the DVA.
	53	* This gives us room for up to 32 different checksum functions.
	54	*
	55	* When writing a block, we always checksum it with the latest-and-greatest
	56	* checksum function of the appropriate strength. When reading a block,
	57	* we compare the expected checksum against the actual checksum, which we
	58	* compute via the checksum function specified in the DVA encoding.
	59	*/
	60
	61	/ARGSUSED/
	62	static void
	63	zio_checksum_off(const void buf, uint64_t size, zio_cksum_t zcp)
	64	{
	65	ZIO_SET_CHECKSUM(zcp, 0, 0, 0, 0);
	66	}
	67
	68	zio_checksum_info_t zio_checksum_table[ZIO_CHECKSUM_FUNCTIONS] = {
	69	{{NULL, NULL}, 0, 0, "inherit"},
	70	{{NULL, NULL}, 0, 0, "on"},
	71	{{zio_checksum_off, zio_checksum_off}, 0, 0, "off"},
	72	{{zio_checksum_SHA256, zio_checksum_SHA256}, 1, 1, "label"},
	73	{{zio_checksum_SHA256, zio_checksum_SHA256}, 1, 1, "gang_header"},
	74	{{fletcher_2_native, fletcher_2_byteswap}, 0, 1, "zilog"},
	75	{{fletcher_2_native, fletcher_2_byteswap}, 0, 0, "fletcher2"},
	76	{{fletcher_4_native, fletcher_4_byteswap}, 1, 0, "fletcher4"},
	77	{{zio_checksum_SHA256, zio_checksum_SHA256}, 1, 0, "SHA256"},
	78	};
	79
	80	uint8_t
	81	zio_checksum_select(uint8_t child, uint8_t parent)
	82	{
	83	ASSERT(child < ZIO_CHECKSUM_FUNCTIONS);
	84	ASSERT(parent < ZIO_CHECKSUM_FUNCTIONS);
	85	ASSERT(parent != ZIO_CHECKSUM_INHERIT && parent != ZIO_CHECKSUM_ON);
	86
	87	if (child == ZIO_CHECKSUM_INHERIT)
	88	return (parent);
	89
90	if (child == ZIO_CHECKSUM_ON)
91	return (ZIO_CHECKSUM_ON_VALUE);
92
93	return (child);
94	}
95
b128c09f BB	96	/*
	97	* Set the external verifier for a gang block based on <vdev, offset, txg>,
	98	* a tuple which is guaranteed to be unique for the life of the pool.
	99	*/
	100	static void
	101	zio_checksum_gang_verifier(zio_cksum_t zcp, blkptr_t bp)
	102	{
	103	dva_t *dva = BP_IDENTITY(bp);
	104	uint64_t txg = bp->blk_birth;
	105
	106	ASSERT(BP_IS_GANG(bp));
	107
	108	ZIO_SET_CHECKSUM(zcp, DVA_GET_VDEV(dva), DVA_GET_OFFSET(dva), txg, 0);
	109	}
	110
	111	/*
	112	* Set the external verifier for a label block based on its offset.
	113	* The vdev is implicit, and the txg is unknowable at pool open time --
	114	* hence the logic in vdev_uberblock_load() to find the most recent copy.
	115	*/
	116	static void
	117	zio_checksum_label_verifier(zio_cksum_t *zcp, uint64_t offset)
	118	{
	119	ZIO_SET_CHECKSUM(zcp, offset, 0, 0, 0);
	120	}
	121
34dc7c2f BB	122	/*
	123	* Generate the checksum.
	124	*/
	125	void
b128c09f BB	126	zio_checksum_compute(zio_t *zio, enum zio_checksum checksum,
b128c09f BB	127	void *data, uint64_t size)
34dc7c2f	128	{
b128c09f BB	129	blkptr_t *bp = zio->io_bp;
b128c09f BB	130	uint64_t offset = zio->io_offset;
34dc7c2f BB	131	zio_block_tail_t zbt = (zio_block_tail_t )((char *)data + size) - 1;
	132	zio_checksum_info_t *ci = &zio_checksum_table[checksum];
	133	zio_cksum_t zbt_cksum;
	134
b128c09f	135	ASSERT((uint_t)checksum < ZIO_CHECKSUM_FUNCTIONS);
34dc7c2f BB	136	ASSERT(ci->ci_func[0] != NULL);
	137
	138	if (ci->ci_zbt) {
b128c09f BB	139	if (checksum == ZIO_CHECKSUM_GANG_HEADER)
	140	zio_checksum_gang_verifier(&zbt->zbt_cksum, bp);
	141	else if (checksum == ZIO_CHECKSUM_LABEL)
	142	zio_checksum_label_verifier(&zbt->zbt_cksum, offset);
	143	else
	144	bp->blk_cksum = zbt->zbt_cksum;
34dc7c2f BB	145	zbt->zbt_magic = ZBT_MAGIC;
	146	ci->ci_func[0](data, size, &zbt_cksum);
	147	zbt->zbt_cksum = zbt_cksum;
	148	} else {
b128c09f	149	ci->ci_func[0](data, size, &bp->blk_cksum);
34dc7c2f BB	150	}
	151	}
	152
	153	int
	154	zio_checksum_error(zio_t *zio)
	155	{
	156	blkptr_t *bp = zio->io_bp;
b128c09f BB	157	uint_t checksum = (bp == NULL ? zio->io_prop.zp_checksum :
	158	(BP_IS_GANG(bp) ? ZIO_CHECKSUM_GANG_HEADER : BP_GET_CHECKSUM(bp)));
	159	int byteswap;
34dc7c2f	160	void *data = zio->io_data;
b128c09f BB	161	uint64_t size = (bp == NULL ? zio->io_size :
	162	(BP_IS_GANG(bp) ? SPA_GANGBLOCKSIZE : BP_GET_PSIZE(bp)));
	163	uint64_t offset = zio->io_offset;
34dc7c2f BB	164	zio_block_tail_t zbt = (zio_block_tail_t )((char *)data + size) - 1;
34dc7c2f BB	165	zio_checksum_info_t *ci = &zio_checksum_table[checksum];
b128c09f	166	zio_cksum_t actual_cksum, expected_cksum, verifier;
34dc7c2f BB	167
	168	if (checksum >= ZIO_CHECKSUM_FUNCTIONS \|\| ci->ci_func[0] == NULL)
	169	return (EINVAL);
	170
	171	if (ci->ci_zbt) {
	172	if (checksum == ZIO_CHECKSUM_GANG_HEADER)
b128c09f BB	173	zio_checksum_gang_verifier(&verifier, bp);
	174	else if (checksum == ZIO_CHECKSUM_LABEL)
	175	zio_checksum_label_verifier(&verifier, offset);
	176	else
	177	verifier = bp->blk_cksum;
	178
	179	byteswap = (zbt->zbt_magic == BSWAP_64(ZBT_MAGIC));
34dc7c2f	180
b128c09f BB	181	if (byteswap)
	182	byteswap_uint64_array(&verifier, sizeof (zio_cksum_t));
	183
	184	expected_cksum = zbt->zbt_cksum;
	185	zbt->zbt_cksum = verifier;
	186	ci->ci_func[byteswap](data, size, &actual_cksum);
	187	zbt->zbt_cksum = expected_cksum;
	188
	189	if (byteswap)
34dc7c2f BB	190	byteswap_uint64_array(&expected_cksum,
34dc7c2f BB	191	sizeof (zio_cksum_t));
34dc7c2f BB	192	} else {
34dc7c2f BB	193	ASSERT(!BP_IS_GANG(bp));
b128c09f BB	194	byteswap = BP_SHOULD_BYTESWAP(bp);
b128c09f BB	195	expected_cksum = bp->blk_cksum;
34dc7c2f BB	196	ci->ci_func[byteswap](data, size, &actual_cksum);
	197	}
	198
b128c09f	199	if (!ZIO_CHECKSUM_EQUAL(actual_cksum, expected_cksum))
34dc7c2f BB	200	return (ECKSUM);
	201
	202	if (zio_injection_enabled && !zio->io_error)
	203	return (zio_handle_fault_injection(zio, ECKSUM));
	204
	205	return (0);
	206	}