2008-10-27 19:27:55 +03:00
/*
* Copyright ( C ) 2005 , 2006
2009-06-14 18:23:09 +04:00
* Avishay Traeger ( avishay @ gmail . com )
2008-10-27 19:27:55 +03:00
* Copyright ( C ) 2008 , 2009
* Boaz Harrosh < bharrosh @ panasas . com >
*
* Copyrights for code taken from ext2 :
* Copyright ( C ) 1992 , 1993 , 1994 , 1995
* Remy Card ( card @ masi . ibp . fr )
* Laboratoire MASI - Institut Blaise Pascal
* Universite Pierre et Marie Curie ( Paris VI )
* from
* linux / fs / minix / inode . c
* Copyright ( C ) 1991 , 1992 Linus Torvalds
*
* This file is part of exofs .
*
* exofs is free software ; you can redistribute it and / or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation . Since it is based on ext2 , and the only
* valid version of GPL for the Linux kernel is version 2 , the only valid
* version of GPL for exofs is version 2.
*
* exofs 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 exofs ; if not , write to the Free Software
* Foundation , Inc . , 51 Franklin St , Fifth Floor , Boston , MA 02110 - 1301 USA
*/
exofs: Move all operations to an io_engine
In anticipation for multi-device operations, we separate osd operations
into an abstract I/O API. Currently only one device is used but later
when adding more devices, we will drive all devices in parallel according
to a "data_map" that describes how data is arranged on multiple devices.
The file system level operates, like before, as if there is one object
(inode-number) and an i_size. The io engine will split this to the same
object-number but on multiple device.
At first we introduce Mirror (raid 1) layout. But at the final outcome
we intend to fully implement the pNFS-Objects data-map, including
raid 0,4,5,6 over mirrored devices, over multiple device-groups. And
more. See: http://tools.ietf.org/html/draft-ietf-nfsv4-pnfs-obj-12
* Define an io_state based API for accessing osd storage devices
in an abstract way.
Usage:
First a caller allocates an io state with:
exofs_get_io_state(struct exofs_sb_info *sbi,
struct exofs_io_state** ios);
Then calles one of:
exofs_sbi_create(struct exofs_io_state *ios);
exofs_sbi_remove(struct exofs_io_state *ios);
exofs_sbi_write(struct exofs_io_state *ios);
exofs_sbi_read(struct exofs_io_state *ios);
exofs_oi_truncate(struct exofs_i_info *oi, u64 new_len);
And when done
exofs_put_io_state(struct exofs_io_state *ios);
* Convert all source files to use this new API
* Convert from bio_alloc to bio_kmalloc
* In io engine we make use of the now fixed osd_req_decode_sense
There are no functional changes or on disk additions after this patch.
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
2009-11-08 15:54:08 +03:00
# ifndef __EXOFS_H__
# define __EXOFS_H__
2008-10-27 19:27:55 +03:00
# include <linux/fs.h>
# include <linux/time.h>
# include "common.h"
# define EXOFS_ERR(fmt, a...) printk(KERN_ERR "exofs: " fmt, ##a)
# ifdef CONFIG_EXOFS_DEBUG
# define EXOFS_DBGMSG(fmt, a...) \
printk ( KERN_NOTICE " exofs @%s:%d: " fmt , __func__ , __LINE__ , # # a )
# else
# define EXOFS_DBGMSG(fmt, a...) \
do { if ( 0 ) printk ( fmt , # # a ) ; } while ( 0 )
# endif
/* u64 has problems with printk this will cast it to unsigned long long */
# define _LLU(x) (unsigned long long)(x)
/*
* our extension to the in - memory superblock
*/
struct exofs_sb_info {
struct osd_dev * s_dev ; /* returned by get_osd_dev */
exofs: Move all operations to an io_engine
In anticipation for multi-device operations, we separate osd operations
into an abstract I/O API. Currently only one device is used but later
when adding more devices, we will drive all devices in parallel according
to a "data_map" that describes how data is arranged on multiple devices.
The file system level operates, like before, as if there is one object
(inode-number) and an i_size. The io engine will split this to the same
object-number but on multiple device.
At first we introduce Mirror (raid 1) layout. But at the final outcome
we intend to fully implement the pNFS-Objects data-map, including
raid 0,4,5,6 over mirrored devices, over multiple device-groups. And
more. See: http://tools.ietf.org/html/draft-ietf-nfsv4-pnfs-obj-12
* Define an io_state based API for accessing osd storage devices
in an abstract way.
Usage:
First a caller allocates an io state with:
exofs_get_io_state(struct exofs_sb_info *sbi,
struct exofs_io_state** ios);
Then calles one of:
exofs_sbi_create(struct exofs_io_state *ios);
exofs_sbi_remove(struct exofs_io_state *ios);
exofs_sbi_write(struct exofs_io_state *ios);
exofs_sbi_read(struct exofs_io_state *ios);
exofs_oi_truncate(struct exofs_i_info *oi, u64 new_len);
And when done
exofs_put_io_state(struct exofs_io_state *ios);
* Convert all source files to use this new API
* Convert from bio_alloc to bio_kmalloc
* In io engine we make use of the now fixed osd_req_decode_sense
There are no functional changes or on disk additions after this patch.
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
2009-11-08 15:54:08 +03:00
struct exofs_fscb s_fscb ; /* Written often, pre-allocate*/
2008-10-27 19:27:55 +03:00
osd_id s_pid ; /* partition ID of file system*/
int s_timeout ; /* timeout for OSD operations */
uint64_t s_nextid ; /* highest object ID used */
uint32_t s_numfiles ; /* number of files on fs */
spinlock_t s_next_gen_lock ; /* spinlock for gen # update */
u32 s_next_generation ; /* next gen # to use */
atomic_t s_curr_pending ; /* number of pending commands */
uint8_t s_cred [ OSD_CAP_LEN ] ; /* all-powerful credential */
} ;
/*
* our extension to the in - memory inode
*/
struct exofs_i_info {
unsigned long i_flags ; /* various atomic flags */
uint32_t i_data [ EXOFS_IDATA ] ; /*short symlink names and device #s*/
uint32_t i_dir_start_lookup ; /* which page to start lookup */
wait_queue_head_t i_wq ; /* wait queue for inode */
uint64_t i_commit_size ; /* the object's written length */
uint8_t i_cred [ OSD_CAP_LEN ] ; /* all-powerful credential */
struct inode vfs_inode ; /* normal in-memory inode */
} ;
exofs: Move all operations to an io_engine
In anticipation for multi-device operations, we separate osd operations
into an abstract I/O API. Currently only one device is used but later
when adding more devices, we will drive all devices in parallel according
to a "data_map" that describes how data is arranged on multiple devices.
The file system level operates, like before, as if there is one object
(inode-number) and an i_size. The io engine will split this to the same
object-number but on multiple device.
At first we introduce Mirror (raid 1) layout. But at the final outcome
we intend to fully implement the pNFS-Objects data-map, including
raid 0,4,5,6 over mirrored devices, over multiple device-groups. And
more. See: http://tools.ietf.org/html/draft-ietf-nfsv4-pnfs-obj-12
* Define an io_state based API for accessing osd storage devices
in an abstract way.
Usage:
First a caller allocates an io state with:
exofs_get_io_state(struct exofs_sb_info *sbi,
struct exofs_io_state** ios);
Then calles one of:
exofs_sbi_create(struct exofs_io_state *ios);
exofs_sbi_remove(struct exofs_io_state *ios);
exofs_sbi_write(struct exofs_io_state *ios);
exofs_sbi_read(struct exofs_io_state *ios);
exofs_oi_truncate(struct exofs_i_info *oi, u64 new_len);
And when done
exofs_put_io_state(struct exofs_io_state *ios);
* Convert all source files to use this new API
* Convert from bio_alloc to bio_kmalloc
* In io engine we make use of the now fixed osd_req_decode_sense
There are no functional changes or on disk additions after this patch.
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
2009-11-08 15:54:08 +03:00
static inline osd_id exofs_oi_objno ( struct exofs_i_info * oi )
{
return oi - > vfs_inode . i_ino + EXOFS_OBJ_OFF ;
}
struct exofs_io_state ;
typedef void ( * exofs_io_done_fn ) ( struct exofs_io_state * or , void * private ) ;
struct exofs_io_state {
struct kref kref ;
void * private ;
exofs_io_done_fn done ;
struct exofs_sb_info * sbi ;
struct osd_obj_id obj ;
u8 * cred ;
/* Global read/write IO*/
loff_t offset ;
unsigned long length ;
void * kern_buff ;
struct bio * bio ;
/* Attributes */
unsigned in_attr_len ;
struct osd_attr * in_attr ;
unsigned out_attr_len ;
struct osd_attr * out_attr ;
/* Variable array of size numdevs */
unsigned numdevs ;
struct exofs_per_dev_state {
struct osd_request * or ;
struct bio * bio ;
} per_dev [ ] ;
} ;
static inline unsigned exofs_io_state_size ( unsigned numdevs )
{
return sizeof ( struct exofs_io_state ) +
sizeof ( struct exofs_per_dev_state ) * numdevs ;
}
2008-10-27 19:27:55 +03:00
/*
* our inode flags
*/
# define OBJ_2BCREATED 0 /* object will be created soon*/
# define OBJ_CREATED 1 /* object has been created on the osd*/
static inline int obj_2bcreated ( struct exofs_i_info * oi )
{
return test_bit ( OBJ_2BCREATED , & oi - > i_flags ) ;
}
static inline void set_obj_2bcreated ( struct exofs_i_info * oi )
{
set_bit ( OBJ_2BCREATED , & oi - > i_flags ) ;
}
static inline int obj_created ( struct exofs_i_info * oi )
{
return test_bit ( OBJ_CREATED , & oi - > i_flags ) ;
}
static inline void set_obj_created ( struct exofs_i_info * oi )
{
set_bit ( OBJ_CREATED , & oi - > i_flags ) ;
}
int __exofs_wait_obj_created ( struct exofs_i_info * oi ) ;
static inline int wait_obj_created ( struct exofs_i_info * oi )
{
if ( likely ( obj_created ( oi ) ) )
return 0 ;
return __exofs_wait_obj_created ( oi ) ;
}
/*
* get to our inode from the vfs inode
*/
static inline struct exofs_i_info * exofs_i ( struct inode * inode )
{
return container_of ( inode , struct exofs_i_info , vfs_inode ) ;
}
2008-10-28 16:38:12 +03:00
/*
* Maximum count of links to a file
*/
# define EXOFS_LINK_MAX 32000
2008-10-27 19:37:02 +03:00
/*************************
* function declarations *
* * * * * * * * * * * * * * * * * * * * * * * * */
exofs: Move all operations to an io_engine
In anticipation for multi-device operations, we separate osd operations
into an abstract I/O API. Currently only one device is used but later
when adding more devices, we will drive all devices in parallel according
to a "data_map" that describes how data is arranged on multiple devices.
The file system level operates, like before, as if there is one object
(inode-number) and an i_size. The io engine will split this to the same
object-number but on multiple device.
At first we introduce Mirror (raid 1) layout. But at the final outcome
we intend to fully implement the pNFS-Objects data-map, including
raid 0,4,5,6 over mirrored devices, over multiple device-groups. And
more. See: http://tools.ietf.org/html/draft-ietf-nfsv4-pnfs-obj-12
* Define an io_state based API for accessing osd storage devices
in an abstract way.
Usage:
First a caller allocates an io state with:
exofs_get_io_state(struct exofs_sb_info *sbi,
struct exofs_io_state** ios);
Then calles one of:
exofs_sbi_create(struct exofs_io_state *ios);
exofs_sbi_remove(struct exofs_io_state *ios);
exofs_sbi_write(struct exofs_io_state *ios);
exofs_sbi_read(struct exofs_io_state *ios);
exofs_oi_truncate(struct exofs_i_info *oi, u64 new_len);
And when done
exofs_put_io_state(struct exofs_io_state *ios);
* Convert all source files to use this new API
* Convert from bio_alloc to bio_kmalloc
* In io engine we make use of the now fixed osd_req_decode_sense
There are no functional changes or on disk additions after this patch.
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
2009-11-08 15:54:08 +03:00
/* ios.c */
void exofs_make_credential ( u8 cred_a [ OSD_CAP_LEN ] ,
const struct osd_obj_id * obj ) ;
int exofs_read_kern ( struct osd_dev * od , u8 * cred , struct osd_obj_id * obj ,
u64 offset , void * p , unsigned length ) ;
int exofs_get_io_state ( struct exofs_sb_info * sbi , struct exofs_io_state * * ios ) ;
void exofs_put_io_state ( struct exofs_io_state * ios ) ;
int exofs_check_io ( struct exofs_io_state * ios , u64 * resid ) ;
int exofs_sbi_create ( struct exofs_io_state * ios ) ;
int exofs_sbi_remove ( struct exofs_io_state * ios ) ;
int exofs_sbi_write ( struct exofs_io_state * ios ) ;
int exofs_sbi_read ( struct exofs_io_state * ios ) ;
int extract_attr_from_ios ( struct exofs_io_state * ios , struct osd_attr * attr ) ;
int exofs_oi_truncate ( struct exofs_i_info * oi , u64 new_len ) ;
static inline int exofs_oi_write ( struct exofs_i_info * oi ,
struct exofs_io_state * ios )
{
ios - > obj . id = exofs_oi_objno ( oi ) ;
ios - > cred = oi - > i_cred ;
return exofs_sbi_write ( ios ) ;
}
static inline int exofs_oi_read ( struct exofs_i_info * oi ,
struct exofs_io_state * ios )
{
ios - > obj . id = exofs_oi_objno ( oi ) ;
ios - > cred = oi - > i_cred ;
return exofs_sbi_read ( ios ) ;
}
2008-10-27 19:37:02 +03:00
/* inode.c */
void exofs_truncate ( struct inode * inode ) ;
int exofs_setattr ( struct dentry * , struct iattr * ) ;
2008-10-27 20:31:34 +03:00
int exofs_write_begin ( struct file * file , struct address_space * mapping ,
loff_t pos , unsigned len , unsigned flags ,
struct page * * pagep , void * * fsdata ) ;
2008-10-28 16:38:12 +03:00
extern struct inode * exofs_iget ( struct super_block * , unsigned long ) ;
struct inode * exofs_new_inode ( struct inode * , int ) ;
2008-10-28 17:11:41 +03:00
extern int exofs_write_inode ( struct inode * , int ) ;
extern void exofs_delete_inode ( struct inode * ) ;
2008-10-28 16:38:12 +03:00
/* dir.c: */
int exofs_add_link ( struct dentry * , struct inode * ) ;
ino_t exofs_inode_by_name ( struct inode * , struct dentry * ) ;
int exofs_delete_entry ( struct exofs_dir_entry * , struct page * ) ;
int exofs_make_empty ( struct inode * , struct inode * ) ;
struct exofs_dir_entry * exofs_find_entry ( struct inode * , struct dentry * ,
struct page * * ) ;
int exofs_empty_dir ( struct inode * ) ;
struct exofs_dir_entry * exofs_dotdot ( struct inode * , struct page * * ) ;
2009-03-22 13:47:26 +03:00
ino_t exofs_parent_ino ( struct dentry * child ) ;
2008-10-28 16:38:12 +03:00
int exofs_set_link ( struct inode * , struct exofs_dir_entry * , struct page * ,
struct inode * ) ;
2008-10-27 19:37:02 +03:00
2009-06-14 17:52:10 +04:00
/* super.c */
int exofs_sync_fs ( struct super_block * sb , int wait ) ;
2008-10-27 19:37:02 +03:00
/*********************
* operation vectors *
* * * * * * * * * * * * * * * * * * * * */
2008-10-28 16:38:12 +03:00
/* dir.c: */
extern const struct file_operations exofs_dir_operations ;
2008-10-27 19:37:02 +03:00
/* file.c */
extern const struct inode_operations exofs_file_inode_operations ;
extern const struct file_operations exofs_file_operations ;
2008-10-27 20:31:34 +03:00
/* inode.c */
extern const struct address_space_operations exofs_aops ;
exofs: Move all operations to an io_engine
In anticipation for multi-device operations, we separate osd operations
into an abstract I/O API. Currently only one device is used but later
when adding more devices, we will drive all devices in parallel according
to a "data_map" that describes how data is arranged on multiple devices.
The file system level operates, like before, as if there is one object
(inode-number) and an i_size. The io engine will split this to the same
object-number but on multiple device.
At first we introduce Mirror (raid 1) layout. But at the final outcome
we intend to fully implement the pNFS-Objects data-map, including
raid 0,4,5,6 over mirrored devices, over multiple device-groups. And
more. See: http://tools.ietf.org/html/draft-ietf-nfsv4-pnfs-obj-12
* Define an io_state based API for accessing osd storage devices
in an abstract way.
Usage:
First a caller allocates an io state with:
exofs_get_io_state(struct exofs_sb_info *sbi,
struct exofs_io_state** ios);
Then calles one of:
exofs_sbi_create(struct exofs_io_state *ios);
exofs_sbi_remove(struct exofs_io_state *ios);
exofs_sbi_write(struct exofs_io_state *ios);
exofs_sbi_read(struct exofs_io_state *ios);
exofs_oi_truncate(struct exofs_i_info *oi, u64 new_len);
And when done
exofs_put_io_state(struct exofs_io_state *ios);
* Convert all source files to use this new API
* Convert from bio_alloc to bio_kmalloc
* In io engine we make use of the now fixed osd_req_decode_sense
There are no functional changes or on disk additions after this patch.
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
2009-11-08 15:54:08 +03:00
extern const struct osd_attr g_attr_logical_length ;
2008-10-27 20:31:34 +03:00
2008-10-28 16:38:12 +03:00
/* namei.c */
extern const struct inode_operations exofs_dir_inode_operations ;
extern const struct inode_operations exofs_special_inode_operations ;
2008-10-27 20:04:34 +03:00
/* symlink.c */
extern const struct inode_operations exofs_symlink_inode_operations ;
extern const struct inode_operations exofs_fast_symlink_inode_operations ;
2008-10-27 19:27:55 +03:00
# endif