/*

 * 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/fm/fs/zfs.h>

#include <sys/spa.h>

#include <sys/txg.h>

#include <sys/spa_impl.h>

#include <sys/vdev_impl.h>

#include <sys/zio_impl.h>

#include <sys/zio_compress.h>

#include <sys/zio_checksum.h>

/*

 * ==========================================================================

 * I/O priority table

 * ==========================================================================

 */

uint8_t zio_priority_table[ZIO_PRIORITY_TABLE_SIZE] = {

	0,	/* ZIO_PRIORITY_NOW		*/

	0,	/* ZIO_PRIORITY_SYNC_READ	*/

	0,	/* ZIO_PRIORITY_SYNC_WRITE	*/

	6,	/* ZIO_PRIORITY_ASYNC_READ	*/

	4,	/* ZIO_PRIORITY_ASYNC_WRITE	*/

	4,	/* ZIO_PRIORITY_FREE		*/

	0,	/* ZIO_PRIORITY_CACHE_FILL	*/

	0,	/* ZIO_PRIORITY_LOG_WRITE	*/

	10,	/* ZIO_PRIORITY_RESILVER	*/

	20,	/* ZIO_PRIORITY_SCRUB		*/

};

/*

 * ==========================================================================

 * I/O type descriptions

 * ==========================================================================

 */

char *zio_type_name[ZIO_TYPES] = {

	"null", "read", "write", "free", "claim", "ioctl" };

/*

 * ==========================================================================

 * I/O kmem caches

 * ==========================================================================

 */

kmem_cache_t *zio_cache;

kmem_cache_t *zio_buf_cache[SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT];

kmem_cache_t *zio_data_buf_cache[SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT];

#ifdef _KERNEL

extern vmem_t *zio_alloc_arena;

#endif

/*

 */

#define	IO_IS_ALLOCATING(zio) \

	((zio)->io_orig_pipeline & (1U << ZIO_STAGE_DVA_ALLOCATE))

void

zio_init(void)

{

	size_t c;

	vmem_t *data_alloc_arena = NULL;

#ifdef _KERNEL

	data_alloc_arena = zio_alloc_arena;

#endif

	zio_cache = kmem_cache_create("zio_cache", sizeof (zio_t), 0,

	    NULL, NULL, NULL, NULL, NULL, 0);

	/*

	 * For small buffers, we want a cache for each multiple of

	 * SPA_MINBLOCKSIZE.  For medium-size buffers, we want a cache

	 * for each quarter-power of 2.  For large buffers, we want

	 * a cache for each multiple of PAGESIZE.

	 */

	for (c = 0; c < SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT; c++) {

		size_t size = (c + 1) << SPA_MINBLOCKSHIFT;

		size_t p2 = size;

		size_t align = 0;

		while (p2 & (p2 - 1))

			p2 &= p2 - 1;

		if (size <= 4 * SPA_MINBLOCKSIZE) {

			align = SPA_MINBLOCKSIZE;

		} else if (P2PHASE(size, PAGESIZE) == 0) {

			align = PAGESIZE;

		} else if (P2PHASE(size, p2 >> 2) == 0) {

			align = p2 >> 2;

		}

		if (align != 0) {

			char name[36];

			(void) sprintf(name, "zio_buf_%lu", (ulong_t)size);

			zio_buf_cache[c] = kmem_cache_create(name, size,

			    align, NULL, NULL, NULL, NULL, NULL, KMC_NODEBUG);

			(void) sprintf(name, "zio_data_buf_%lu", (ulong_t)size);

			zio_data_buf_cache[c] = kmem_cache_create(name, size,

			    align, NULL, NULL, NULL, NULL, data_alloc_arena,

			    KMC_NODEBUG);

		}

	}

	while (--c != 0) {

		ASSERT(zio_buf_cache[c] != NULL);

		if (zio_buf_cache[c - 1] == NULL)

			zio_buf_cache[c - 1] = zio_buf_cache[c];

		ASSERT(zio_data_buf_cache[c] != NULL);

		if (zio_data_buf_cache[c - 1] == NULL)

			zio_data_buf_cache[c - 1] = zio_data_buf_cache[c];

	}

	zio_inject_init();

}

void

zio_fini(void)

{

	size_t c;

	kmem_cache_t *last_cache = NULL;

	kmem_cache_t *last_data_cache = NULL;

	for (c = 0; c < SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT; c++) {

		if (zio_buf_cache[c] != last_cache) {

			last_cache = zio_buf_cache[c];

			kmem_cache_destroy(zio_buf_cache[c]);

		}

		zio_buf_cache[c] = NULL;

		if (zio_data_buf_cache[c] != last_data_cache) {

			last_data_cache = zio_data_buf_cache[c];

			kmem_cache_destroy(zio_data_buf_cache[c]);

		}

		zio_data_buf_cache[c] = NULL;

	}

	kmem_cache_destroy(zio_cache);

	zio_inject_fini();

}

/*

 * ==========================================================================

 * Allocate and free I/O buffers

 * ==========================================================================

 */

/*

 * Use zio_buf_alloc to allocate ZFS metadata.  This data will appear in a

 * crashdump if the kernel panics, so use it judiciously.  Obviously, it's

 * useful to inspect ZFS metadata, but if possible, we should avoid keeping

 * excess / transient data in-core during a crashdump.

 */

void *

zio_buf_alloc(size_t size)

{

	size_t c = (size - 1) >> SPA_MINBLOCKSHIFT;

	ASSERT(c < SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT);

	return (kmem_cache_alloc(zio_buf_cache[c], KM_PUSHPAGE));

}

/*

 * Use zio_data_buf_alloc to allocate data.  The data will not appear in a

 * crashdump if the kernel panics.  This exists so that we will limit the amount

 * of ZFS data that shows up in a kernel crashdump.  (Thus reducing the amount

 * of kernel heap dumped to disk when the kernel panics)

 */

void *

zio_data_buf_alloc(size_t size)

{

	size_t c = (size - 1) >> SPA_MINBLOCKSHIFT;

	ASSERT(c < SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT);

	return (kmem_cache_alloc(zio_data_buf_cache[c], KM_PUSHPAGE));

}

void

zio_buf_free(void *buf, size_t size)

{

	size_t c = (size - 1) >> SPA_MINBLOCKSHIFT;

	ASSERT(c < SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT);

	kmem_cache_free(zio_buf_cache[c], buf);

}

void

zio_data_buf_free(void *buf, size_t size)

{

	size_t c = (size - 1) >> SPA_MINBLOCKSHIFT;

	ASSERT(c < SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT);

	kmem_cache_free(zio_data_buf_cache[c], buf);

}

/*

 * ==========================================================================

 * Push and pop I/O transform buffers

 * ==========================================================================

 */

static void

{

	zio_transform_t *zt = kmem_alloc(sizeof (zio_transform_t), KM_SLEEP);

	zt->zt_bufsize = bufsize;

	zt->zt_next = zio->io_transform_stack;

	zio->io_transform_stack = zt;

	zio->io_data = data;

	zio->io_size = size;

}

static void

{

	}

}

/*

 * ==========================================================================

 * ==========================================================================

 */

static zio_t *

zio_create(zio_t *pio, spa_t *spa, uint64_t txg, blkptr_t *bp,

    void *data, uint64_t size, zio_done_func_t *done, void *private,

{

	zio_t *zio;

	ASSERT3U(size, <=, SPA_MAXBLOCKSIZE);

	ASSERT(P2PHASE(size, SPA_MINBLOCKSIZE) == 0);

	zio = kmem_cache_alloc(zio_cache, KM_SLEEP);

	bzero(zio, sizeof (zio_t));

	if (bp != NULL) {

		zio->io_bp = bp;

		zio->io_bp_copy = *bp;

		zio->io_bp_orig = *bp;

	}

	zio->io_done = done;

	zio->io_private = private;

	zio->io_type = type;

	zio->io_priority = priority;

			zio->io_logical = pio->io_logical;

	}

	return (zio);

}

static void

{

}

zio_t *

zio_null(zio_t *pio, spa_t *spa, zio_done_func_t *done, void *private,

	int flags)

{

	zio_t *zio;

	zio = zio_create(pio, spa, 0, NULL, NULL, 0, done, private,

	return (zio);

}

zio_t *

zio_root(spa_t *spa, zio_done_func_t *done, void *private, int flags)

{

	return (zio_null(NULL, spa, done, private, flags));

}

zio_t *