#define ORANGEFS_DEVREQ_MAGIC 0x20030529
#define ORANGEFS_LINK_MAX 0x000000FF
#define ORANGEFS_PURGE_RETRY_COUNT 0x00000005
-#define ORANGEFS_SEEK_END 0x00000002
#define ORANGEFS_MAX_NUM_OPTIONS 0x00000004
#define ORANGEFS_MAX_MOUNT_OPT_LEN 0x00000080
#define ORANGEFS_MAX_FSKEY_LEN 64
-#define MAX_DEV_REQ_UPSIZE (2*sizeof(__s32) + \
+#define MAX_DEV_REQ_UPSIZE (2 * sizeof(__s32) + \
sizeof(__u64) + sizeof(struct orangefs_upcall_s))
-#define MAX_DEV_REQ_DOWNSIZE (2*sizeof(__s32) + \
+#define MAX_DEV_REQ_DOWNSIZE (2 * sizeof(__s32) + \
sizeof(__u64) + sizeof(struct orangefs_downcall_s))
-/* borrowed from irda.h */
-#ifndef MSECS_TO_JIFFIES
-#define MSECS_TO_JIFFIES(ms) (((ms)*HZ+999)/1000)
-#endif
-
/*
* valid orangefs kernel operation states
*
* serviced - op has matching downcall; ok
* purged - op has to start a timer since client-core
* exited uncleanly before servicing op
+ * given up - submitter has given up waiting for it
*/
enum orangefs_vfs_op_states {
OP_VFS_STATE_UNKNOWN = 0,
OP_VFS_STATE_INPROGR = 2,
OP_VFS_STATE_SERVICED = 4,
OP_VFS_STATE_PURGED = 8,
-};
-
-#define set_op_state_waiting(op) ((op)->op_state = OP_VFS_STATE_WAITING)
-#define set_op_state_inprogress(op) ((op)->op_state = OP_VFS_STATE_INPROGR)
-#define set_op_state_serviced(op) ((op)->op_state = OP_VFS_STATE_SERVICED)
-#define set_op_state_purged(op) ((op)->op_state |= OP_VFS_STATE_PURGED)
-
-#define op_state_waiting(op) ((op)->op_state & OP_VFS_STATE_WAITING)
-#define op_state_in_progress(op) ((op)->op_state & OP_VFS_STATE_INPROGR)
-#define op_state_serviced(op) ((op)->op_state & OP_VFS_STATE_SERVICED)
-#define op_state_purged(op) ((op)->op_state & OP_VFS_STATE_PURGED)
-
-#define get_op(op) \
- do { \
- atomic_inc(&(op)->aio_ref_count); \
- gossip_debug(GOSSIP_DEV_DEBUG, \
- "(get) Alloced OP (%p:%llu)\n", \
- op, \
- llu((op)->tag)); \
- } while (0)
-
-#define put_op(op) \
- do { \
- if (atomic_sub_and_test(1, &(op)->aio_ref_count) == 1) { \
- gossip_debug(GOSSIP_DEV_DEBUG, \
- "(put) Releasing OP (%p:%llu)\n", \
- op, \
- llu((op)->tag)); \
- op_release(op); \
- } \
- } while (0)
-
-#define op_wait(op) (atomic_read(&(op)->aio_ref_count) <= 2 ? 0 : 1)
-
-/*
- * Defines for controlling whether I/O upcalls are for async or sync operations
- */
-enum ORANGEFS_async_io_type {
- ORANGEFS_VFS_SYNC_IO = 0,
- ORANGEFS_VFS_ASYNC_IO = 1,
+ OP_VFS_STATE_GIVEN_UP = 16,
};
/*
#define ORANGEFS_CACHE_CREATE_FLAGS 0
#endif /* ((defined ORANGEFS_KERNEL_DEBUG) && (defined CONFIG_DEBUG_SLAB)) */
-#define ORANGEFS_CACHE_ALLOC_FLAGS (GFP_KERNEL)
-#define ORANGEFS_GFP_FLAGS (GFP_KERNEL)
-#define ORANGEFS_BUFMAP_GFP_FLAGS (GFP_KERNEL)
-
/* orangefs xattr and acl related defines */
#define ORANGEFS_XATTR_INDEX_POSIX_ACL_ACCESS 1
#define ORANGEFS_XATTR_INDEX_POSIX_ACL_DEFAULT 2
#define ORANGEFS_XATTR_INDEX_TRUSTED 3
#define ORANGEFS_XATTR_INDEX_DEFAULT 4
-#if 0
-#ifndef POSIX_ACL_XATTR_ACCESS
-#define POSIX_ACL_XATTR_ACCESS "system.posix_acl_access"
-#endif
-#ifndef POSIX_ACL_XATTR_DEFAULT
-#define POSIX_ACL_XATTR_DEFAULT "system.posix_acl_default"
-#endif
-#endif
-
-#define ORANGEFS_XATTR_NAME_ACL_ACCESS POSIX_ACL_XATTR_ACCESS
-#define ORANGEFS_XATTR_NAME_ACL_DEFAULT POSIX_ACL_XATTR_DEFAULT
+#define ORANGEFS_XATTR_NAME_ACL_ACCESS XATTR_NAME_POSIX_ACL_ACCESS
+#define ORANGEFS_XATTR_NAME_ACL_DEFAULT XATTR_NAME_POSIX_ACL_DEFAULT
#define ORANGEFS_XATTR_NAME_TRUSTED_PREFIX "trusted."
#define ORANGEFS_XATTR_NAME_DEFAULT_PREFIX ""
__u64 tag;
/*
- * Set uses_shared_memory to 1 if this operation uses shared memory.
- * If true, then a retry on the op must also get a new shared memory
- * buffer and re-populate it.
+ * Set uses_shared_memory to non zero if this operation uses
+ * shared memory. If true, then a retry on the op must also
+ * get a new shared memory buffer and re-populate it.
+ * Cancels don't care - it only matters for service_operation()
+ * retry logics and cancels don't go through it anymore. It
+ * safely stays non-zero when we use it as slot_to_free.
*/
- int uses_shared_memory;
+ union {
+ int uses_shared_memory;
+ int slot_to_free;
+ };
struct orangefs_upcall_s upcall;
struct orangefs_downcall_s downcall;
- wait_queue_head_t waitq;
+ struct completion waitq;
spinlock_t lock;
- int io_completed;
- wait_queue_head_t io_completion_waitq;
+ int attempts;
- /* VFS aio fields */
+ struct list_head list;
+};
- /* used by the async I/O code to stash the orangefs_kiocb_s structure */
- void *priv;
+#define set_op_state_waiting(op) ((op)->op_state = OP_VFS_STATE_WAITING)
+#define set_op_state_inprogress(op) ((op)->op_state = OP_VFS_STATE_INPROGR)
+#define set_op_state_given_up(op) ((op)->op_state = OP_VFS_STATE_GIVEN_UP)
+static inline void set_op_state_serviced(struct orangefs_kernel_op_s *op)
+{
+ op->op_state = OP_VFS_STATE_SERVICED;
+ complete(&op->waitq);
+}
- /* used again for the async I/O code for deallocation */
- atomic_t aio_ref_count;
+#define op_state_waiting(op) ((op)->op_state & OP_VFS_STATE_WAITING)
+#define op_state_in_progress(op) ((op)->op_state & OP_VFS_STATE_INPROGR)
+#define op_state_serviced(op) ((op)->op_state & OP_VFS_STATE_SERVICED)
+#define op_state_purged(op) ((op)->op_state & OP_VFS_STATE_PURGED)
+#define op_state_given_up(op) ((op)->op_state & OP_VFS_STATE_GIVEN_UP)
+#define op_is_cancel(op) ((op)->upcall.type == ORANGEFS_VFS_OP_CANCEL)
- int attempts;
+void op_release(struct orangefs_kernel_op_s *op);
- struct list_head list;
-};
+extern void orangefs_bufmap_put(int);
+static inline void put_cancel(struct orangefs_kernel_op_s *op)
+{
+ orangefs_bufmap_put(op->slot_to_free);
+ op_release(op);
+}
+
+static inline void set_op_state_purged(struct orangefs_kernel_op_s *op)
+{
+ spin_lock(&op->lock);
+ if (unlikely(op_is_cancel(op))) {
+ list_del_init(&op->list);
+ spin_unlock(&op->lock);
+ put_cancel(op);
+ } else {
+ op->op_state |= OP_VFS_STATE_PURGED;
+ complete(&op->waitq);
+ spin_unlock(&op->lock);
+ }
+}
/* per inode private orangefs info */
struct orangefs_inode_s {
* with this object
*/
unsigned long pinode_flags;
-
- /* All allocated orangefs_inode_s objects are chained to a list */
- struct list_head list;
};
#define P_ATIME_FLAG 0
int op_cache_initialize(void);
int op_cache_finalize(void);
struct orangefs_kernel_op_s *op_alloc(__s32 type);
+void orangefs_new_tag(struct orangefs_kernel_op_s *op);
char *get_opname_string(struct orangefs_kernel_op_s *new_op);
-void op_release(struct orangefs_kernel_op_s *op);
-
-int dev_req_cache_initialize(void);
-int dev_req_cache_finalize(void);
-void *dev_req_alloc(void);
-void dev_req_release(void *);
int orangefs_inode_cache_initialize(void);
int orangefs_inode_cache_finalize(void);
-int kiocb_cache_initialize(void);
-int kiocb_cache_finalize(void);
-struct orangefs_kiocb_s *kiocb_alloc(void);
-void kiocb_release(struct orangefs_kiocb_s *ptr);
-
/*
* defined in orangefs-mod.c
*/
/*
* defined in waitqueue.c
*/
-int wait_for_matching_downcall(struct orangefs_kernel_op_s *op);
-int wait_for_cancellation_downcall(struct orangefs_kernel_op_s *op);
-void orangefs_clean_up_interrupted_operation(struct orangefs_kernel_op_s *op);
void purge_waiting_ops(void);
/*
void *data);
void orangefs_kill_sb(struct super_block *sb);
-int orangefs_remount(struct super_block *sb);
+int orangefs_remount(struct orangefs_sb_info_s *);
int fsid_key_table_initialize(void);
void fsid_key_table_finalize(void);
struct dentry *dentry,
struct kstat *kstat);
+int orangefs_permission(struct inode *inode, int mask);
+
/*
* defined in xattr.c
*/
int orangefs_dev_init(void);
void orangefs_dev_cleanup(void);
int is_daemon_in_service(void);
-int fs_mount_pending(__s32 fsid);
+bool __is_daemon_in_service(void);
/*
* defined in orangefs-utils.c
size_t size,
int flags);
-int orangefs_inode_getattr(struct inode *inode, __u32 mask);
+int orangefs_inode_getattr(struct inode *inode, int new, int size);
-int orangefs_inode_setattr(struct inode *inode, struct iattr *iattr);
+int orangefs_inode_check_changed(struct inode *inode);
-void orangefs_op_initialize(struct orangefs_kernel_op_s *op);
+int orangefs_inode_setattr(struct inode *inode, struct iattr *iattr);
void orangefs_make_bad_inode(struct inode *inode);
-void block_signals(sigset_t *);
-
-void set_signals(sigset_t *);
-
int orangefs_unmount_sb(struct super_block *sb);
-int orangefs_cancel_op_in_progress(__u64 tag);
-
-static inline __u64 orangefs_convert_time_field(const struct timespec *ts)
-{
- return (__u64)ts->tv_sec;
-}
+bool orangefs_cancel_op_in_progress(struct orangefs_kernel_op_s *op);
int orangefs_normalize_to_errno(__s32 error_code);
/*
* misc convenience macros
*/
-#define add_op_to_request_list(op) \
-do { \
- spin_lock(&orangefs_request_list_lock); \
- spin_lock(&op->lock); \
- set_op_state_waiting(op); \
- list_add_tail(&op->list, &orangefs_request_list); \
- spin_unlock(&orangefs_request_list_lock); \
- spin_unlock(&op->lock); \
- wake_up_interruptible(&orangefs_request_list_waitq); \
-} while (0)
-
-#define add_priority_op_to_request_list(op) \
- do { \
- spin_lock(&orangefs_request_list_lock); \
- spin_lock(&op->lock); \
- set_op_state_waiting(op); \
- \
- list_add(&op->list, &orangefs_request_list); \
- spin_unlock(&orangefs_request_list_lock); \
- spin_unlock(&op->lock); \
- wake_up_interruptible(&orangefs_request_list_waitq); \
-} while (0)
-
-#define remove_op_from_request_list(op) \
- do { \
- struct list_head *tmp = NULL; \
- struct list_head *tmp_safe = NULL; \
- struct orangefs_kernel_op_s *tmp_op = NULL; \
- \
- spin_lock(&orangefs_request_list_lock); \
- list_for_each_safe(tmp, tmp_safe, &orangefs_request_list) { \
- tmp_op = list_entry(tmp, \
- struct orangefs_kernel_op_s, \
- list); \
- if (tmp_op && (tmp_op == op)) { \
- list_del(&tmp_op->list); \
- break; \
- } \
- } \
- spin_unlock(&orangefs_request_list_lock); \
- } while (0)
#define ORANGEFS_OP_INTERRUPTIBLE 1 /* service_operation() is interruptible */
#define ORANGEFS_OP_PRIORITY 2 /* service_operation() is high priority */
#define ORANGEFS_OP_CANCELLATION 4 /* this is a cancellation */
-#define ORANGEFS_OP_NO_SEMAPHORE 8 /* don't acquire semaphore */
+#define ORANGEFS_OP_NO_MUTEX 8 /* don't acquire request_mutex */
#define ORANGEFS_OP_ASYNC 16 /* Queue it, but don't wait */
int service_operation(struct orangefs_kernel_op_s *op,
const char *op_name,
int flags);
-/*
- * handles two possible error cases, depending on context.
- *
- * by design, our vfs i/o errors need to be handled in one of two ways,
- * depending on where the error occured.
- *
- * if the error happens in the waitqueue code because we either timed
- * out or a signal was raised while waiting, we need to cancel the
- * userspace i/o operation and free the op manually. this is done to
- * avoid having the device start writing application data to our shared
- * bufmap pages without us expecting it.
- *
- * FIXME: POSSIBLE OPTIMIZATION:
- * However, if we timed out or if we got a signal AND our upcall was never
- * picked off the queue (i.e. we were in OP_VFS_STATE_WAITING), then we don't
- * need to send a cancellation upcall. The way we can handle this is
- * set error_exit to 2 in such cases and 1 whenever cancellation has to be
- * sent and have handle_error
- * take care of this situation as well..
- *
- * if a orangefs sysint level error occured and i/o has been completed,
- * there is no need to cancel the operation, as the user has finished
- * using the bufmap page and so there is no danger in this case. in
- * this case, we wake up the device normally so that it may free the
- * op, as normal.
- *
- * note the only reason this is a macro is because both read and write
- * cases need the exact same handling code.
- */
-#define handle_io_error() \
-do { \
- if (!op_state_serviced(new_op)) { \
- orangefs_cancel_op_in_progress(new_op->tag); \
- op_release(new_op); \
- } else { \
- wake_up_daemon_for_return(new_op); \
- } \
- new_op = NULL; \
- orangefs_bufmap_put(bufmap, buffer_index); \
- buffer_index = -1; \
-} while (0)
-
#define get_interruptible_flag(inode) \
((ORANGEFS_SB(inode->i_sb)->flags & ORANGEFS_OPT_INTR) ? \
ORANGEFS_OP_INTERRUPTIBLE : 0)
-#define add_orangefs_sb(sb) \
-do { \
- gossip_debug(GOSSIP_SUPER_DEBUG, \
- "Adding SB %p to orangefs superblocks\n", \
- ORANGEFS_SB(sb)); \
- spin_lock(&orangefs_superblocks_lock); \
- list_add_tail(&ORANGEFS_SB(sb)->list, &orangefs_superblocks); \
- spin_unlock(&orangefs_superblocks_lock); \
-} while (0)
-
-#define remove_orangefs_sb(sb) \
-do { \
- struct list_head *tmp = NULL; \
- struct list_head *tmp_safe = NULL; \
- struct orangefs_sb_info_s *orangefs_sb = NULL; \
- \
- spin_lock(&orangefs_superblocks_lock); \
- list_for_each_safe(tmp, tmp_safe, &orangefs_superblocks) { \
- orangefs_sb = list_entry(tmp, \
- struct orangefs_sb_info_s, \
- list); \
- if (orangefs_sb && (orangefs_sb->sb == sb)) { \
- gossip_debug(GOSSIP_SUPER_DEBUG, \
- "Removing SB %p from orangefs superblocks\n", \
- orangefs_sb); \
- list_del(&orangefs_sb->list); \
- break; \
- } \
- } \
- spin_unlock(&orangefs_superblocks_lock); \
-} while (0)
-
-#define orangefs_lock_inode(inode) spin_lock(&inode->i_lock)
-#define orangefs_unlock_inode(inode) spin_unlock(&inode->i_lock)
-
#define fill_default_sys_attrs(sys_attr, type, mode) \
do { \
sys_attr.owner = from_kuid(current_user_ns(), current_fsuid()); \
sys_attr.group = from_kgid(current_user_ns(), current_fsgid()); \
- sys_attr.size = 0; \
sys_attr.perms = ORANGEFS_util_translate_mode(mode); \
- sys_attr.objtype = type; \
+ sys_attr.mtime = 0; \
+ sys_attr.atime = 0; \
+ sys_attr.ctime = 0; \
sys_attr.mask = ORANGEFS_ATTR_SYS_ALL_SETABLE; \
} while (0)
-#define orangefs_inode_lock(__i) mutex_lock(&(__i)->i_mutex)
-
-#define orangefs_inode_unlock(__i) mutex_unlock(&(__i)->i_mutex)
-
static inline void orangefs_i_size_write(struct inode *inode, loff_t i_size)
{
#if BITS_PER_LONG == 32 && defined(CONFIG_SMP)
- ornagefs_inode_lock(inode);
+ mutex_lock(&inode->i_mutex);
#endif
i_size_write(inode, i_size);
#if BITS_PER_LONG == 32 && defined(CONFIG_SMP)
- orangefs_inode_unlock(inode);
+ mutex_unlock(&inode->i_mutex);
#endif
}
-static inline unsigned int diff(struct timeval *end, struct timeval *begin)
-{
- if (end->tv_usec < begin->tv_usec) {
- end->tv_usec += 1000000;
- end->tv_sec--;
- }
- end->tv_sec -= begin->tv_sec;
- end->tv_usec -= begin->tv_usec;
- return (end->tv_sec * 1000000) + end->tv_usec;
-}
-
#endif /* __ORANGEFSKERNEL_H */