[deliverable/linux.git] / fs / nfs / blocklayout / blocklayout.h

/*
 *  linux/fs/nfs/blocklayout/blocklayout.h
 *
 *  Module for the NFSv4.1 pNFS block layout driver.
 *
 *  Copyright (c) 2006 The Regents of the University of Michigan.
 *  All rights reserved.
 *
 *  Andy Adamson <andros@citi.umich.edu>
 *  Fred Isaman <iisaman@umich.edu>
 *
 * permission is granted to use, copy, create derivative works and
 * redistribute this software and such derivative works for any purpose,
 * so long as the name of the university of michigan is not used in
 * any advertising or publicity pertaining to the use or distribution
 * of this software without specific, written prior authorization.  if
 * the above copyright notice or any other identification of the
 * university of michigan is included in any copy of any portion of
 * this software, then the disclaimer below must also be included.
 *
 * this software is provided as is, without representation from the
 * university of michigan as to its fitness for any purpose, and without
 * warranty by the university of michigan of any kind, either express
 * or implied, including without limitation the implied warranties of
 * merchantability and fitness for a particular purpose.  the regents
 * of the university of michigan shall not be liable for any damages,
 * including special, indirect, incidental, or consequential damages,
 * with respect to any claim arising out or in connection with the use
 * of the software, even if it has been or is hereafter advised of the
 * possibility of such damages.
 */
#ifndef FS_NFS_NFS4BLOCKLAYOUT_H
#define FS_NFS_NFS4BLOCKLAYOUT_H

#include <linux/device-mapper.h>
#include <linux/nfs_fs.h>
#include <linux/sunrpc/rpc_pipe_fs.h>

#include "../pnfs.h"

#define PAGE_CACHE_SECTORS (PAGE_CACHE_SIZE >> SECTOR_SHIFT)
#define PAGE_CACHE_SECTOR_SHIFT (PAGE_CACHE_SHIFT - SECTOR_SHIFT)

struct block_mount_id {
	spinlock_t			bm_lock;    /* protects list */
	struct list_head		bm_devlist; /* holds pnfs_block_dev */
};

struct pnfs_block_dev {
	struct list_head		bm_node;
	struct nfs4_deviceid		bm_mdevid;    /* associated devid */
	struct block_device		*bm_mdev;     /* meta device itself */
};

enum exstate4 {
	PNFS_BLOCK_READWRITE_DATA	= 0,
	PNFS_BLOCK_READ_DATA		= 1,
	PNFS_BLOCK_INVALID_DATA		= 2, /* mapped, but data is invalid */
	PNFS_BLOCK_NONE_DATA		= 3  /* unmapped, it's a hole */
};

#define MY_MAX_TAGS (15) /* tag bitnums used must be less than this */

struct my_tree {
	sector_t		mtt_step_size;	/* Internal sector alignment */
	struct list_head	mtt_stub; /* Should be a radix tree */
};

struct pnfs_inval_markings {
	spinlock_t	im_lock;
	struct my_tree	im_tree;	/* Sectors that need LAYOUTCOMMIT */
	sector_t	im_block_size;	/* Server blocksize in sectors */
	struct list_head im_extents;	/* Short extents for INVAL->RW conversion */
};

struct pnfs_inval_tracking {
	struct list_head it_link;
	int		 it_sector;
	int		 it_tags;
};

/* sector_t fields are all in 512-byte sectors */
struct pnfs_block_extent {
	struct kref	be_refcnt;
	struct list_head be_node;	/* link into lseg list */
	struct nfs4_deviceid be_devid;  /* FIXME: could use device cache instead */
	struct block_device *be_mdev;
	sector_t	be_f_offset;	/* the starting offset in the file */
	sector_t	be_length;	/* the size of the extent */
	sector_t	be_v_offset;	/* the starting offset in the volume */
	enum exstate4	be_state;	/* the state of this extent */
	struct pnfs_inval_markings *be_inval; /* tracks INVAL->RW transition */
};

/* Shortened extent used by LAYOUTCOMMIT */
struct pnfs_block_short_extent {
	struct list_head bse_node;
	struct nfs4_deviceid bse_devid;
	struct block_device *bse_mdev;
	sector_t	bse_f_offset;	/* the starting offset in the file */
	sector_t	bse_length;	/* the size of the extent */
};

static inline void
BL_INIT_INVAL_MARKS(struct pnfs_inval_markings *marks, sector_t blocksize)
{
	spin_lock_init(&marks->im_lock);
	INIT_LIST_HEAD(&marks->im_tree.mtt_stub);
	INIT_LIST_HEAD(&marks->im_extents);
	marks->im_block_size = blocksize;
	marks->im_tree.mtt_step_size = min((sector_t)PAGE_CACHE_SECTORS,
					   blocksize);
}

enum extentclass4 {
	RW_EXTENT       = 0, /* READWRTE and INVAL */
	RO_EXTENT       = 1, /* READ and NONE */
	EXTENT_LISTS    = 2,
};

static inline int bl_choose_list(enum exstate4 state)
{
	if (state == PNFS_BLOCK_READ_DATA || state == PNFS_BLOCK_NONE_DATA)
		return RO_EXTENT;
	else
		return RW_EXTENT;
}

struct pnfs_block_layout {
	struct pnfs_layout_hdr bl_layout;
	struct pnfs_inval_markings bl_inval; /* tracks INVAL->RW transition */
	spinlock_t		bl_ext_lock;   /* Protects list manipulation */
	struct list_head	bl_extents[EXTENT_LISTS]; /* R and RW extents */
	struct list_head	bl_commit;	/* Needs layout commit */
	struct list_head	bl_committing;	/* Layout committing */
	unsigned int		bl_count;	/* entries in bl_commit */
	sector_t		bl_blocksize;  /* Server blocksize in sectors */
};

#define BLK_ID(lo) ((struct block_mount_id *)(NFS_SERVER(lo->plh_inode)->pnfs_ld_data))

static inline struct pnfs_block_layout *
BLK_LO2EXT(struct pnfs_layout_hdr *lo)
{
	return container_of(lo, struct pnfs_block_layout, bl_layout);
}

static inline struct pnfs_block_layout *
BLK_LSEG2EXT(struct pnfs_layout_segment *lseg)
{
	return BLK_LO2EXT(lseg->pls_layout);
}

struct bl_dev_msg {
	int32_t status;
	uint32_t major, minor;
};

struct bl_msg_hdr {
	u8  type;
	u16 totallen; /* length of entire message, including hdr itself */
};

extern struct dentry *bl_device_pipe;
extern wait_queue_head_t bl_wq;

#define BL_DEVICE_UMOUNT               0x0 /* Umount--delete devices */
#define BL_DEVICE_MOUNT                0x1 /* Mount--create devices*/
#define BL_DEVICE_REQUEST_INIT         0x0 /* Start request */
#define BL_DEVICE_REQUEST_PROC         0x1 /* User level process succeeds */
#define BL_DEVICE_REQUEST_ERR          0x2 /* User level process fails */

/* blocklayoutdev.c */
ssize_t bl_pipe_downcall(struct file *, const char __user *, size_t);
void bl_pipe_destroy_msg(struct rpc_pipe_msg *);
struct block_device *nfs4_blkdev_get(dev_t dev);
int nfs4_blkdev_put(struct block_device *bdev);
struct pnfs_block_dev *nfs4_blk_decode_device(struct nfs_server *server,
						struct pnfs_device *dev);
int nfs4_blk_process_layoutget(struct pnfs_layout_hdr *lo,
				struct nfs4_layoutget_res *lgr, gfp_t gfp_flags);

/* blocklayoutdm.c */
void bl_free_block_dev(struct pnfs_block_dev *bdev);

/* extents.c */
struct pnfs_block_extent *
bl_find_get_extent(struct pnfs_block_layout *bl, sector_t isect,
		struct pnfs_block_extent **cow_read);
int bl_mark_sectors_init(struct pnfs_inval_markings *marks,
			     sector_t offset, sector_t length);
void bl_put_extent(struct pnfs_block_extent *be);
struct pnfs_block_extent *bl_alloc_extent(void);
int bl_is_sector_init(struct pnfs_inval_markings *marks, sector_t isect);
int encode_pnfs_block_layoutupdate(struct pnfs_block_layout *bl,
				   struct xdr_stream *xdr,
				   const struct nfs4_layoutcommit_args *arg);
void clean_pnfs_block_layoutupdate(struct pnfs_block_layout *bl,
				   const struct nfs4_layoutcommit_args *arg,
				   int status);
int bl_add_merge_extent(struct pnfs_block_layout *bl,
			 struct pnfs_block_extent *new);
int bl_mark_for_commit(struct pnfs_block_extent *be,
			sector_t offset, sector_t length,
			struct pnfs_block_short_extent *new);
int bl_push_one_short_extent(struct pnfs_inval_markings *marks);
struct pnfs_block_short_extent *
bl_pop_one_short_extent(struct pnfs_inval_markings *marks);
void bl_free_short_extents(struct pnfs_inval_markings *marks, int num_to_free);

#endif /* FS_NFS_NFS4BLOCKLAYOUT_H */
Commit	Line	Data
155e7524 FI	1	/*
	2	* linux/fs/nfs/blocklayout/blocklayout.h
	3	*
	4	* Module for the NFSv4.1 pNFS block layout driver.
	5	*
	6	* Copyright (c) 2006 The Regents of the University of Michigan.
	7	* All rights reserved.
	8	*
	9	* Andy Adamson <andros@citi.umich.edu>
	10	* Fred Isaman <iisaman@umich.edu>
	11	*
	12	* permission is granted to use, copy, create derivative works and
	13	* redistribute this software and such derivative works for any purpose,
	14	* so long as the name of the university of michigan is not used in
	15	* any advertising or publicity pertaining to the use or distribution
	16	* of this software without specific, written prior authorization. if
	17	* the above copyright notice or any other identification of the
	18	* university of michigan is included in any copy of any portion of
	19	* this software, then the disclaimer below must also be included.
	20	*
	21	* this software is provided as is, without representation from the
	22	* university of michigan as to its fitness for any purpose, and without
	23	* warranty by the university of michigan of any kind, either express
	24	* or implied, including without limitation the implied warranties of
	25	* merchantability and fitness for a particular purpose. the regents
	26	* of the university of michigan shall not be liable for any damages,
	27	* including special, indirect, incidental, or consequential damages,
	28	* with respect to any claim arising out or in connection with the use
	29	* of the software, even if it has been or is hereafter advised of the
	30	* possibility of such damages.
	31	*/
	32	#ifndef FS_NFS_NFS4BLOCKLAYOUT_H
	33	#define FS_NFS_NFS4BLOCKLAYOUT_H
	34
	35	#include <linux/device-mapper.h>
	36	#include <linux/nfs_fs.h>
fe0a9b74 JR	37	#include <linux/sunrpc/rpc_pipe_fs.h>
fe0a9b74 JR	38
155e7524 FI	39	#include "../pnfs.h"
155e7524 FI	40
c1c2a4cd FI	41	#define PAGE_CACHE_SECTORS (PAGE_CACHE_SIZE >> SECTOR_SHIFT)
	42	#define PAGE_CACHE_SECTOR_SHIFT (PAGE_CACHE_SHIFT - SECTOR_SHIFT)
	43
2f9fd182 FI	44	struct block_mount_id {
	45	spinlock_t bm_lock; /* protects list */
	46	struct list_head bm_devlist; /* holds pnfs_block_dev */
	47	};
	48
fe0a9b74 JR	49	struct pnfs_block_dev {
	50	struct list_head bm_node;
	51	struct nfs4_deviceid bm_mdevid; /* associated devid */
	52	struct block_device bm_mdev; / meta device itself */
	53	};
	54
155e7524 FI	55	enum exstate4 {
	56	PNFS_BLOCK_READWRITE_DATA = 0,
	57	PNFS_BLOCK_READ_DATA = 1,
	58	PNFS_BLOCK_INVALID_DATA = 2, /* mapped, but data is invalid */
	59	PNFS_BLOCK_NONE_DATA = 3 /* unmapped, it's a hole */
	60	};
	61
c1c2a4cd FI	62	#define MY_MAX_TAGS (15) /* tag bitnums used must be less than this */
	63
	64	struct my_tree {
	65	sector_t mtt_step_size; /* Internal sector alignment */
	66	struct list_head mtt_stub; /* Should be a radix tree */
	67	};
	68
155e7524	69	struct pnfs_inval_markings {
c1c2a4cd FI	70	spinlock_t im_lock;
	71	struct my_tree im_tree; /* Sectors that need LAYOUTCOMMIT */
	72	sector_t im_block_size; /* Server blocksize in sectors */
7c5465d6	73	struct list_head im_extents; /* Short extents for INVAL->RW conversion */
c1c2a4cd FI	74	};
	75
	76	struct pnfs_inval_tracking {
	77	struct list_head it_link;
	78	int it_sector;
	79	int it_tags;
155e7524 FI	80	};
	81
	82	/* sector_t fields are all in 512-byte sectors */
	83	struct pnfs_block_extent {
	84	struct kref be_refcnt;
	85	struct list_head be_node; /* link into lseg list */
	86	struct nfs4_deviceid be_devid; /* FIXME: could use device cache instead */
	87	struct block_device *be_mdev;
	88	sector_t be_f_offset; /* the starting offset in the file */
	89	sector_t be_length; /* the size of the extent */
	90	sector_t be_v_offset; /* the starting offset in the volume */
	91	enum exstate4 be_state; /* the state of this extent */
	92	struct pnfs_inval_markings be_inval; / tracks INVAL->RW transition */
	93	};
	94
90ace12a FI	95	/* Shortened extent used by LAYOUTCOMMIT */
	96	struct pnfs_block_short_extent {
	97	struct list_head bse_node;
	98	struct nfs4_deviceid bse_devid;
	99	struct block_device *bse_mdev;
	100	sector_t bse_f_offset; /* the starting offset in the file */
	101	sector_t bse_length; /* the size of the extent */
	102	};
	103
155e7524 FI	104	static inline void
	105	BL_INIT_INVAL_MARKS(struct pnfs_inval_markings *marks, sector_t blocksize)
	106	{
c1c2a4cd FI	107	spin_lock_init(&marks->im_lock);
c1c2a4cd FI	108	INIT_LIST_HEAD(&marks->im_tree.mtt_stub);
7c5465d6	109	INIT_LIST_HEAD(&marks->im_extents);
c1c2a4cd FI	110	marks->im_block_size = blocksize;
	111	marks->im_tree.mtt_step_size = min((sector_t)PAGE_CACHE_SECTORS,
	112	blocksize);
155e7524 FI	113	}
	114
	115	enum extentclass4 {
	116	RW_EXTENT = 0, /* READWRTE and INVAL */
	117	RO_EXTENT = 1, /* READ and NONE */
	118	EXTENT_LISTS = 2,
	119	};
	120
03341d2c FI	121	static inline int bl_choose_list(enum exstate4 state)
	122	{
	123	if (state == PNFS_BLOCK_READ_DATA \|\| state == PNFS_BLOCK_NONE_DATA)
	124	return RO_EXTENT;
	125	else
	126	return RW_EXTENT;
	127	}
	128
155e7524 FI	129	struct pnfs_block_layout {
	130	struct pnfs_layout_hdr bl_layout;
	131	struct pnfs_inval_markings bl_inval; /* tracks INVAL->RW transition */
	132	spinlock_t bl_ext_lock; /* Protects list manipulation */
	133	struct list_head bl_extents[EXTENT_LISTS]; /* R and RW extents */
	134	struct list_head bl_commit; /* Needs layout commit */
	135	struct list_head bl_committing; /* Layout committing */
	136	unsigned int bl_count; /* entries in bl_commit */
	137	sector_t bl_blocksize; /* Server blocksize in sectors */
	138	};
	139
2f9fd182 FI	140	#define BLK_ID(lo) ((struct block_mount_id *)(NFS_SERVER(lo->plh_inode)->pnfs_ld_data))
	141
	142	static inline struct pnfs_block_layout *
	143	BLK_LO2EXT(struct pnfs_layout_hdr *lo)
155e7524 FI	144	{
	145	return container_of(lo, struct pnfs_block_layout, bl_layout);
	146	}
	147
a60d2ebd FI	148	static inline struct pnfs_block_layout *
	149	BLK_LSEG2EXT(struct pnfs_layout_segment *lseg)
	150	{
	151	return BLK_LO2EXT(lseg->pls_layout);
	152	}
	153
fe0a9b74	154	struct bl_dev_msg {
fdc17abb	155	int32_t status;
fe0a9b74 JR	156	uint32_t major, minor;
	157	};
	158
	159	struct bl_msg_hdr {
	160	u8 type;
	161	u16 totallen; /* length of entire message, including hdr itself */
	162	};
	163
	164	extern struct dentry *bl_device_pipe;
	165	extern wait_queue_head_t bl_wq;
	166
	167	#define BL_DEVICE_UMOUNT 0x0 /* Umount--delete devices */
	168	#define BL_DEVICE_MOUNT 0x1 /* Mount--create devices*/
	169	#define BL_DEVICE_REQUEST_INIT 0x0 /* Start request */
	170	#define BL_DEVICE_REQUEST_PROC 0x1 /* User level process succeeds */
	171	#define BL_DEVICE_REQUEST_ERR 0x2 /* User level process fails */
	172
	173	/* blocklayoutdev.c */
fe0a9b74 JR	174	ssize_t bl_pipe_downcall(struct file , const char __user , size_t);
	175	void bl_pipe_destroy_msg(struct rpc_pipe_msg *);
	176	struct block_device *nfs4_blkdev_get(dev_t dev);
	177	int nfs4_blkdev_put(struct block_device *bdev);
	178	struct pnfs_block_dev nfs4_blk_decode_device(struct nfs_server server,
2f9fd182	179	struct pnfs_device *dev);
fe0a9b74 JR	180	int nfs4_blk_process_layoutget(struct pnfs_layout_hdr *lo,
	181	struct nfs4_layoutget_res *lgr, gfp_t gfp_flags);
	182
025a70ed JR	183	/* blocklayoutdm.c */
	184	void bl_free_block_dev(struct pnfs_block_dev *bdev);
	185
03341d2c	186	/* extents.c */
6d742ba5 FI	187	struct pnfs_block_extent *
	188	bl_find_get_extent(struct pnfs_block_layout *bl, sector_t isect,
	189	struct pnfs_block_extent **cow_read);
c1c2a4cd	190	int bl_mark_sectors_init(struct pnfs_inval_markings *marks,
60c52e3a	191	sector_t offset, sector_t length);
9e692969	192	void bl_put_extent(struct pnfs_block_extent *be);
03341d2c	193	struct pnfs_block_extent *bl_alloc_extent(void);
c1c2a4cd	194	int bl_is_sector_init(struct pnfs_inval_markings *marks, sector_t isect);
90ace12a FI	195	int encode_pnfs_block_layoutupdate(struct pnfs_block_layout *bl,
	196	struct xdr_stream *xdr,
	197	const struct nfs4_layoutcommit_args *arg);
b2be7811 FI	198	void clean_pnfs_block_layoutupdate(struct pnfs_block_layout *bl,
	199	const struct nfs4_layoutcommit_args *arg,
	200	int status);
03341d2c FI	201	int bl_add_merge_extent(struct pnfs_block_layout *bl,
03341d2c FI	202	struct pnfs_block_extent *new);
31e6306a	203	int bl_mark_for_commit(struct pnfs_block_extent *be,
7c5465d6 PT	204	sector_t offset, sector_t length,
	205	struct pnfs_block_short_extent *new);
	206	int bl_push_one_short_extent(struct pnfs_inval_markings *marks);
	207	struct pnfs_block_short_extent *
	208	bl_pop_one_short_extent(struct pnfs_inval_markings *marks);
	209	void bl_free_short_extents(struct pnfs_inval_markings *marks, int num_to_free);
03341d2c	210
155e7524	211	#endif /* FS_NFS_NFS4BLOCKLAYOUT_H */