[deliverable/linux.git] / drivers / staging / lustre / lustre / include / lustre_lib.h

/*
 * GPL HEADER START
 *
 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 only,
 * as published by the Free Software Foundation.
 *
 * This program 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 version 2 for more details (a copy is included
 * in the LICENSE file that accompanied this code).
 *
 * You should have received a copy of the GNU General Public License
 * version 2 along with this program; If not, see
 * http://www.gnu.org/licenses/gpl-2.0.html
 *
 * GPL HEADER END
 */
/*
 * Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
 * Use is subject to license terms.
 *
 * Copyright (c) 2011, 2012, Intel Corporation.
 */
/*
 * This file is part of Lustre, http://www.lustre.org/
 * Lustre is a trademark of Sun Microsystems, Inc.
 *
 * lustre/include/lustre_lib.h
 *
 * Basic Lustre library routines.
 */

#ifndef _LUSTRE_LIB_H
#define _LUSTRE_LIB_H

/** \defgroup lib lib
 *
 * @{
 */

#include <linux/sched.h>
#include <linux/signal.h>
#include <linux/types.h>
#include "../../include/linux/libcfs/libcfs.h"
#include "lustre/lustre_idl.h"
#include "lustre_ver.h"
#include "lustre_cfg.h"

/* target.c */
struct kstatfs;
struct ptlrpc_request;
struct obd_export;
struct lu_target;
struct l_wait_info;
#include "lustre_ha.h"
#include "lustre_net.h"

#define LI_POISON 0x5a5a5a5a
#if BITS_PER_LONG > 32
# define LL_POISON 0x5a5a5a5a5a5a5a5aL
#else
# define LL_POISON 0x5a5a5a5aL
#endif
#define LP_POISON ((void *)LL_POISON)

int target_pack_pool_reply(struct ptlrpc_request *req);
int do_set_info_async(struct obd_import *imp,
		      int opcode, int version,
		      u32 keylen, void *key,
		      u32 vallen, void *val,
		      struct ptlrpc_request_set *set);

#define OBD_RECOVERY_MAX_TIME (obd_timeout * 18) /* b13079 */

void target_send_reply(struct ptlrpc_request *req, int rc, int fail_id);

/* client.c */

int client_sanobd_setup(struct obd_device *obddev, struct lustre_cfg *lcfg);
struct client_obd *client_conn2cli(struct lustre_handle *conn);

struct md_open_data;
struct obd_client_handle {
	struct lustre_handle	 och_fh;
	struct lu_fid		 och_fid;
	struct md_open_data	*och_mod;
	struct lustre_handle	 och_lease_handle; /* open lock for lease */
	__u32			 och_magic;
	fmode_t			 och_flags;
};

#define OBD_CLIENT_HANDLE_MAGIC 0xd15ea5ed

/* statfs_pack.c */
void statfs_unpack(struct kstatfs *sfs, struct obd_statfs *osfs);

/* Until such time as we get_info the per-stripe maximum from the OST,
 * we define this to be 2T - 4k, which is the ext3 maxbytes.
 */
#define LUSTRE_STRIPE_MAXBYTES 0x1fffffff000ULL

/* Special values for remove LOV EA from disk */
#define LOVEA_DELETE_VALUES(size, count, offset) (size == 0 && count == 0 && \
						 offset == (typeof(offset))(-1))

#define LMVEA_DELETE_VALUES(count, offset) ((count) == 0 && \
					    (offset) == (typeof(offset))(-1))
/* #define POISON_BULK 0 */

/*
 * l_wait_event is a flexible sleeping function, permitting simple caller
 * configuration of interrupt and timeout sensitivity along with actions to
 * be performed in the event of either exception.
 *
 * The first form of usage looks like this:
 *
 * struct l_wait_info lwi = LWI_TIMEOUT_INTR(timeout, timeout_handler,
 *					   intr_handler, callback_data);
 * rc = l_wait_event(waitq, condition, &lwi);
 *
 * l_wait_event() makes the current process wait on 'waitq' until 'condition'
 * is TRUE or a "killable" signal (SIGTERM, SIKGILL, SIGINT) is pending.  It
 * returns 0 to signify 'condition' is TRUE, but if a signal wakes it before
 * 'condition' becomes true, it optionally calls the specified 'intr_handler'
 * if not NULL, and returns -EINTR.
 *
 * If a non-zero timeout is specified, signals are ignored until the timeout
 * has expired.  At this time, if 'timeout_handler' is not NULL it is called.
 * If it returns FALSE l_wait_event() continues to wait as described above with
 * signals enabled.  Otherwise it returns -ETIMEDOUT.
 *
 * LWI_INTR(intr_handler, callback_data) is shorthand for
 * LWI_TIMEOUT_INTR(0, NULL, intr_handler, callback_data)
 *
 * The second form of usage looks like this:
 *
 * struct l_wait_info lwi = LWI_TIMEOUT(timeout, timeout_handler);
 * rc = l_wait_event(waitq, condition, &lwi);
 *
 * This form is the same as the first except that it COMPLETELY IGNORES
 * SIGNALS.  The caller must therefore beware that if 'timeout' is zero, or if
 * 'timeout_handler' is not NULL and returns FALSE, then the ONLY thing that
 * can unblock the current process is 'condition' becoming TRUE.
 *
 * Another form of usage is:
 * struct l_wait_info lwi = LWI_TIMEOUT_INTERVAL(timeout, interval,
 *					       timeout_handler);
 * rc = l_wait_event(waitq, condition, &lwi);
 * This is the same as previous case, but condition is checked once every
 * 'interval' jiffies (if non-zero).
 *
 * Subtle synchronization point: this macro does *not* necessary takes
 * wait-queue spin-lock before returning, and, hence, following idiom is safe
 * ONLY when caller provides some external locking:
 *
 *	     Thread1			    Thread2
 *
 *   l_wait_event(&obj->wq, ....);				       (1)
 *
 *				    wake_up(&obj->wq):		 (2)
 *					 spin_lock(&q->lock);	  (2.1)
 *					 __wake_up_common(q, ...);     (2.2)
 *					 spin_unlock(&q->lock, flags); (2.3)
 *
 *   kfree(obj);						  (3)
 *
 * As l_wait_event() may "short-cut" execution and return without taking
 * wait-queue spin-lock, some additional synchronization is necessary to
 * guarantee that step (3) can begin only after (2.3) finishes.
 *
 * XXX nikita: some ptlrpc daemon threads have races of that sort.
 *
 */
static inline int back_to_sleep(void *arg)
{
	return 0;
}

#define LWI_ON_SIGNAL_NOOP ((void (*)(void *))(-1))

struct l_wait_info {
	long lwi_timeout;
	long lwi_interval;
	int	    lwi_allow_intr;
	int  (*lwi_on_timeout)(void *);
	void (*lwi_on_signal)(void *);
	void  *lwi_cb_data;
};

/* NB: LWI_TIMEOUT ignores signals completely */
#define LWI_TIMEOUT(time, cb, data)	     \
((struct l_wait_info) {			 \
	.lwi_timeout    = time,		 \
	.lwi_on_timeout = cb,		   \
	.lwi_cb_data    = data,		 \
	.lwi_interval   = 0,		    \
	.lwi_allow_intr = 0		     \
})

#define LWI_TIMEOUT_INTERVAL(time, interval, cb, data)  \
((struct l_wait_info) {				 \
	.lwi_timeout    = time,			 \
	.lwi_on_timeout = cb,			   \
	.lwi_cb_data    = data,			 \
	.lwi_interval   = interval,		     \
	.lwi_allow_intr = 0			     \
})

#define LWI_TIMEOUT_INTR(time, time_cb, sig_cb, data)   \
((struct l_wait_info) {				 \
	.lwi_timeout    = time,			 \
	.lwi_on_timeout = time_cb,		      \
	.lwi_on_signal  = sig_cb,		       \
	.lwi_cb_data    = data,			 \
	.lwi_interval   = 0,			    \
	.lwi_allow_intr = 0			     \
})

#define LWI_TIMEOUT_INTR_ALL(time, time_cb, sig_cb, data)       \
((struct l_wait_info) {					 \
	.lwi_timeout    = time,				 \
	.lwi_on_timeout = time_cb,			      \
	.lwi_on_signal  = sig_cb,			       \
	.lwi_cb_data    = data,				 \
	.lwi_interval   = 0,				    \
	.lwi_allow_intr = 1				     \
})

#define LWI_INTR(cb, data)  LWI_TIMEOUT_INTR(0, NULL, cb, data)

#define LUSTRE_FATAL_SIGS (sigmask(SIGKILL) | sigmask(SIGINT) |		\
			   sigmask(SIGTERM) | sigmask(SIGQUIT) |	\
			   sigmask(SIGALRM))

/**
 * wait_queue_t of Linux (version < 2.6.34) is a FIFO list for exclusively
 * waiting threads, which is not always desirable because all threads will
 * be waken up again and again, even user only needs a few of them to be
 * active most time. This is not good for performance because cache can
 * be polluted by different threads.
 *
 * LIFO list can resolve this problem because we always wakeup the most
 * recent active thread by default.
 *
 * NB: please don't call non-exclusive & exclusive wait on the same
 * waitq if add_wait_queue_exclusive_head is used.
 */
#define add_wait_queue_exclusive_head(waitq, link)		\
{								\
	unsigned long flags;					\
								\
	spin_lock_irqsave(&((waitq)->lock), flags);		\
	__add_wait_queue_exclusive(waitq, link);		\
	spin_unlock_irqrestore(&((waitq)->lock), flags);	\
}

/*
 * wait for @condition to become true, but no longer than timeout, specified
 * by @info.
 */
#define __l_wait_event(wq, condition, info, ret, l_add_wait)		   \
do {									   \
	wait_queue_t __wait;						 \
	long __timeout = info->lwi_timeout;			  \
	sigset_t   __blocked;					      \
	int   __allow_intr = info->lwi_allow_intr;			     \
									       \
	ret = 0;							       \
	if (condition)							 \
		break;							 \
									       \
	init_waitqueue_entry(&__wait, current);					    \
	l_add_wait(&wq, &__wait);					      \
									       \
	/* Block all signals (just the non-fatal ones if no timeout). */       \
	if (info->lwi_on_signal && (__timeout == 0 || __allow_intr))   \
		__blocked = cfs_block_sigsinv(LUSTRE_FATAL_SIGS);	      \
	else								   \
		__blocked = cfs_block_sigsinv(0);			      \
									       \
	for (;;) {							     \
		if (condition)						 \
			break;						 \
									       \
		set_current_state(TASK_INTERRUPTIBLE);			       \
									       \
		if (__timeout == 0) {					  \
			schedule();					       \
		} else {						       \
			long interval = info->lwi_interval ?	  \
					     min_t(long,	     \
						 info->lwi_interval, __timeout) : \
					     __timeout;			\
			long remaining = schedule_timeout(interval);\
			__timeout = cfs_time_sub(__timeout,		    \
					    cfs_time_sub(interval, remaining));\
			if (__timeout == 0) {				  \
				if (!info->lwi_on_timeout ||		      \
				    info->lwi_on_timeout(info->lwi_cb_data)) { \
					ret = -ETIMEDOUT;		      \
					break;				 \
				}					      \
				/* Take signals after the timeout expires. */  \
				if (info->lwi_on_signal)		       \
				    (void)cfs_block_sigsinv(LUSTRE_FATAL_SIGS);\
			}						      \
		}							      \
									       \
		set_current_state(TASK_RUNNING);			       \
									       \
		if (condition)						 \
			break;						 \
		if (signal_pending(current)) {				    \
			if (info->lwi_on_signal &&		     \
			    (__timeout == 0 || __allow_intr)) {		\
				if (info->lwi_on_signal != LWI_ON_SIGNAL_NOOP) \
					info->lwi_on_signal(info->lwi_cb_data);\
				ret = -EINTR;				  \
				break;					 \
			}						      \
			/* We have to do this here because some signals */     \
			/* are not blockable - ie from strace(1).       */     \
			/* In these cases we want to schedule_timeout() */     \
			/* again, because we don't want that to return  */     \
			/* -EINTR when the RPC actually succeeded.      */     \
			/* the recalc_sigpending() below will deliver the */     \
			/* signal properly.			     */     \
			cfs_clear_sigpending();				\
		}							      \
	}								      \
									       \
	cfs_restore_sigs(__blocked);					   \
									       \
	remove_wait_queue(&wq, &__wait);					   \
} while (0)

#define l_wait_event(wq, condition, info)		       \
({							      \
	int		 __ret;			      \
	struct l_wait_info *__info = (info);		    \
								\
	__l_wait_event(wq, condition, __info,		   \
		       __ret, add_wait_queue);		   \
	__ret;						  \
})

#define l_wait_event_exclusive(wq, condition, info)	     \
({							      \
	int		 __ret;			      \
	struct l_wait_info *__info = (info);		    \
								\
	__l_wait_event(wq, condition, __info,		   \
		       __ret, add_wait_queue_exclusive);	 \
	__ret;						  \
})

#define l_wait_event_exclusive_head(wq, condition, info)	\
({							      \
	int		 __ret;			      \
	struct l_wait_info *__info = (info);		    \
								\
	__l_wait_event(wq, condition, __info,		   \
		       __ret, add_wait_queue_exclusive_head);    \
	__ret;						  \
})

#define l_wait_condition(wq, condition)			 \
({							      \
	struct l_wait_info lwi = { 0 };			 \
	l_wait_event(wq, condition, &lwi);		      \
})

#define l_wait_condition_exclusive(wq, condition)	       \
({							      \
	struct l_wait_info lwi = { 0 };			 \
	l_wait_event_exclusive(wq, condition, &lwi);	    \
})

#define l_wait_condition_exclusive_head(wq, condition)	  \
({							      \
	struct l_wait_info lwi = { 0 };			 \
	l_wait_event_exclusive_head(wq, condition, &lwi);       \
})

#define LIBLUSTRE_CLIENT (0)

/** @} lib */

#endif /* _LUSTRE_LIB_H */
Commit	Line	Data
d7e09d03 PT	1	/*
	2	* GPL HEADER START
	3	*
	4	* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
	5	*
	6	* This program is free software; you can redistribute it and/or modify
	7	* it under the terms of the GNU General Public License version 2 only,
	8	* as published by the Free Software Foundation.
	9	*
	10	* This program is distributed in the hope that it will be useful, but
	11	* WITHOUT ANY WARRANTY; without even the implied warranty of
	12	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
	13	* General Public License version 2 for more details (a copy is included
	14	* in the LICENSE file that accompanied this code).
	15	*
	16	* You should have received a copy of the GNU General Public License
	17	* version 2 along with this program; If not, see
6a5b99a4	18	* http://www.gnu.org/licenses/gpl-2.0.html
d7e09d03	19	*
d7e09d03 PT	20	* GPL HEADER END
	21	*/
	22	/*
	23	* Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
	24	* Use is subject to license terms.
	25	*
	26	* Copyright (c) 2011, 2012, Intel Corporation.
	27	*/
	28	/*
	29	* This file is part of Lustre, http://www.lustre.org/
	30	* Lustre is a trademark of Sun Microsystems, Inc.
	31	*
	32	* lustre/include/lustre_lib.h
	33	*
	34	* Basic Lustre library routines.
	35	*/
	36
	37	#ifndef _LUSTRE_LIB_H
	38	#define _LUSTRE_LIB_H
	39
	40	/** \defgroup lib lib
	41	*
	42	* @{
	43	*/
	44
376ef86b JH	45	#include <linux/sched.h>
	46	#include <linux/signal.h>
	47	#include <linux/types.h>
9fdaf8c0	48	#include "../../include/linux/libcfs/libcfs.h"
1accaadf GKH	49	#include "lustre/lustre_idl.h"
	50	#include "lustre_ver.h"
	51	#include "lustre_cfg.h"
d7e09d03 PT	52
d7e09d03 PT	53	/* target.c */
a9c7db39	54	struct kstatfs;
d7e09d03 PT	55	struct ptlrpc_request;
	56	struct obd_export;
	57	struct lu_target;
	58	struct l_wait_info;
1accaadf GKH	59	#include "lustre_ha.h"
1accaadf GKH	60	#include "lustre_net.h"
d7e09d03	61
376ef86b JH	62	#define LI_POISON 0x5a5a5a5a
	63	#if BITS_PER_LONG > 32
	64	# define LL_POISON 0x5a5a5a5a5a5a5a5aL
	65	#else
	66	# define LL_POISON 0x5a5a5a5aL
	67	#endif
	68	#define LP_POISON ((void *)LL_POISON)
d7e09d03 PT	69
	70	int target_pack_pool_reply(struct ptlrpc_request *req);
	71	int do_set_info_async(struct obd_import *imp,
	72	int opcode, int version,
21aef7d9 OD	73	u32 keylen, void *key,
21aef7d9 OD	74	u32 vallen, void *val,
d7e09d03 PT	75	struct ptlrpc_request_set *set);
	76
	77	#define OBD_RECOVERY_MAX_TIME (obd_timeout * 18) /* b13079 */
d7e09d03 PT	78
	79	void target_send_reply(struct ptlrpc_request *req, int rc, int fail_id);
	80
	81	/* client.c */
	82
9c234f6c	83	int client_sanobd_setup(struct obd_device obddev, struct lustre_cfg lcfg);
d7e09d03 PT	84	struct client_obd client_conn2cli(struct lustre_handle conn);
	85
	86	struct md_open_data;
	87	struct obd_client_handle {
d3a8a4e2 JX	88	struct lustre_handle och_fh;
	89	struct lu_fid och_fid;
	90	struct md_open_data *och_mod;
	91	struct lustre_handle och_lease_handle; /* open lock for lease */
	92	__u32 och_magic;
	93	fmode_t och_flags;
d7e09d03	94	};
c9f6bb96	95
d7e09d03 PT	96	#define OBD_CLIENT_HANDLE_MAGIC 0xd15ea5ed
	97
	98	/* statfs_pack.c */
d7e09d03 PT	99	void statfs_unpack(struct kstatfs sfs, struct obd_statfs osfs);
d7e09d03 PT	100
d7e09d03	101	/* Until such time as we get_info the per-stripe maximum from the OST,
c56e256d OD	102	* we define this to be 2T - 4k, which is the ext3 maxbytes.
c56e256d OD	103	*/
d7e09d03 PT	104	#define LUSTRE_STRIPE_MAXBYTES 0x1fffffff000ULL
	105
	106	/* Special values for remove LOV EA from disk */
	107	#define LOVEA_DELETE_VALUES(size, count, offset) (size == 0 && count == 0 && \
	108	offset == (typeof(offset))(-1))
	109
2de35386	110	#define LMVEA_DELETE_VALUES(count, offset) ((count) == 0 && \
2de35386	111	(offset) == (typeof(offset))(-1))
d7e09d03 PT	112	/* #define POISON_BULK 0 */
	113
	114	/*
	115	* l_wait_event is a flexible sleeping function, permitting simple caller
	116	* configuration of interrupt and timeout sensitivity along with actions to
	117	* be performed in the event of either exception.
	118	*
	119	* The first form of usage looks like this:
	120	*
	121	* struct l_wait_info lwi = LWI_TIMEOUT_INTR(timeout, timeout_handler,
	122	* intr_handler, callback_data);
	123	* rc = l_wait_event(waitq, condition, &lwi);
	124	*
	125	* l_wait_event() makes the current process wait on 'waitq' until 'condition'
	126	* is TRUE or a "killable" signal (SIGTERM, SIKGILL, SIGINT) is pending. It
	127	* returns 0 to signify 'condition' is TRUE, but if a signal wakes it before
	128	* 'condition' becomes true, it optionally calls the specified 'intr_handler'
	129	* if not NULL, and returns -EINTR.
	130	*
	131	* If a non-zero timeout is specified, signals are ignored until the timeout
	132	* has expired. At this time, if 'timeout_handler' is not NULL it is called.
	133	* If it returns FALSE l_wait_event() continues to wait as described above with
	134	* signals enabled. Otherwise it returns -ETIMEDOUT.
	135	*
	136	* LWI_INTR(intr_handler, callback_data) is shorthand for
	137	* LWI_TIMEOUT_INTR(0, NULL, intr_handler, callback_data)
	138	*
	139	* The second form of usage looks like this:
	140	*
	141	* struct l_wait_info lwi = LWI_TIMEOUT(timeout, timeout_handler);
	142	* rc = l_wait_event(waitq, condition, &lwi);
	143	*
	144	* This form is the same as the first except that it COMPLETELY IGNORES
	145	* SIGNALS. The caller must therefore beware that if 'timeout' is zero, or if
	146	* 'timeout_handler' is not NULL and returns FALSE, then the ONLY thing that
	147	* can unblock the current process is 'condition' becoming TRUE.
	148	*
	149	* Another form of usage is:
	150	* struct l_wait_info lwi = LWI_TIMEOUT_INTERVAL(timeout, interval,
	151	* timeout_handler);
	152	* rc = l_wait_event(waitq, condition, &lwi);
	153	* This is the same as previous case, but condition is checked once every
	154	* 'interval' jiffies (if non-zero).
	155	*
	156	* Subtle synchronization point: this macro does not necessary takes
	157	* wait-queue spin-lock before returning, and, hence, following idiom is safe
	158	* ONLY when caller provides some external locking:
	159	*
	160	* Thread1 Thread2
	161	*
	162	* l_wait_event(&obj->wq, ....); (1)
	163	*
	164	* wake_up(&obj->wq): (2)
	165	* spin_lock(&q->lock); (2.1)
	166	* __wake_up_common(q, ...); (2.2)
	167	* spin_unlock(&q->lock, flags); (2.3)
	168	*
c0a2472f	169	* kfree(obj); (3)
d7e09d03 PT	170	*
	171	* As l_wait_event() may "short-cut" execution and return without taking
	172	* wait-queue spin-lock, some additional synchronization is necessary to
	173	* guarantee that step (3) can begin only after (2.3) finishes.
	174	*
	175	* XXX nikita: some ptlrpc daemon threads have races of that sort.
	176	*
	177	*/
	178	static inline int back_to_sleep(void *arg)
	179	{
	180	return 0;
	181	}
	182
	183	#define LWI_ON_SIGNAL_NOOP ((void ()(void ))(-1))
	184
	185	struct l_wait_info {
b2d201bd GKH	186	long lwi_timeout;
b2d201bd GKH	187	long lwi_interval;
d7e09d03 PT	188	int lwi_allow_intr;
	189	int (lwi_on_timeout)(void );
	190	void (lwi_on_signal)(void );
	191	void *lwi_cb_data;
	192	};
	193
	194	/* NB: LWI_TIMEOUT ignores signals completely */
	195	#define LWI_TIMEOUT(time, cb, data) \
	196	((struct l_wait_info) { \
	197	.lwi_timeout = time, \
	198	.lwi_on_timeout = cb, \
	199	.lwi_cb_data = data, \
	200	.lwi_interval = 0, \
	201	.lwi_allow_intr = 0 \
	202	})
	203
	204	#define LWI_TIMEOUT_INTERVAL(time, interval, cb, data) \
	205	((struct l_wait_info) { \
	206	.lwi_timeout = time, \
	207	.lwi_on_timeout = cb, \
	208	.lwi_cb_data = data, \
	209	.lwi_interval = interval, \
	210	.lwi_allow_intr = 0 \
	211	})
	212
	213	#define LWI_TIMEOUT_INTR(time, time_cb, sig_cb, data) \
	214	((struct l_wait_info) { \
	215	.lwi_timeout = time, \
	216	.lwi_on_timeout = time_cb, \
	217	.lwi_on_signal = sig_cb, \
	218	.lwi_cb_data = data, \
	219	.lwi_interval = 0, \
	220	.lwi_allow_intr = 0 \
	221	})
	222
	223	#define LWI_TIMEOUT_INTR_ALL(time, time_cb, sig_cb, data) \
	224	((struct l_wait_info) { \
	225	.lwi_timeout = time, \
	226	.lwi_on_timeout = time_cb, \
	227	.lwi_on_signal = sig_cb, \
	228	.lwi_cb_data = data, \
	229	.lwi_interval = 0, \
	230	.lwi_allow_intr = 1 \
	231	})
	232
	233	#define LWI_INTR(cb, data) LWI_TIMEOUT_INTR(0, NULL, cb, data)
	234
376ef86b JH	235	#define LUSTRE_FATAL_SIGS (sigmask(SIGKILL) \| sigmask(SIGINT) \| \
	236	sigmask(SIGTERM) \| sigmask(SIGQUIT) \| \
	237	sigmask(SIGALRM))
	238
564f5d6e JS	239	/**
	240	* wait_queue_t of Linux (version < 2.6.34) is a FIFO list for exclusively
	241	* waiting threads, which is not always desirable because all threads will
	242	* be waken up again and again, even user only needs a few of them to be
	243	* active most time. This is not good for performance because cache can
	244	* be polluted by different threads.
	245	*
	246	* LIFO list can resolve this problem because we always wakeup the most
	247	* recent active thread by default.
	248	*
	249	* NB: please don't call non-exclusive & exclusive wait on the same
	250	* waitq if add_wait_queue_exclusive_head is used.
	251	*/
	252	#define add_wait_queue_exclusive_head(waitq, link) \
	253	{ \
	254	unsigned long flags; \
	255	\
	256	spin_lock_irqsave(&((waitq)->lock), flags); \
	257	__add_wait_queue_exclusive(waitq, link); \
	258	spin_unlock_irqrestore(&((waitq)->lock), flags); \
	259	}
	260
d7e09d03 PT	261	/*
	262	* wait for @condition to become true, but no longer than timeout, specified
	263	* by @info.
	264	*/
	265	#define __l_wait_event(wq, condition, info, ret, l_add_wait) \
	266	do { \
	267	wait_queue_t __wait; \
b2d201bd	268	long __timeout = info->lwi_timeout; \
d7e09d03 PT	269	sigset_t __blocked; \
	270	int __allow_intr = info->lwi_allow_intr; \
	271	\
	272	ret = 0; \
	273	if (condition) \
	274	break; \
	275	\
9e795d35	276	init_waitqueue_entry(&__wait, current); \
d7e09d03 PT	277	l_add_wait(&wq, &__wait); \
	278	\
	279	/* Block all signals (just the non-fatal ones if no timeout). */ \
d2a13989	280	if (info->lwi_on_signal && (__timeout == 0 \|\| __allow_intr)) \
d7e09d03 PT	281	__blocked = cfs_block_sigsinv(LUSTRE_FATAL_SIGS); \
	282	else \
	283	__blocked = cfs_block_sigsinv(0); \
	284	\
	285	for (;;) { \
d7e09d03 PT	286	if (condition) \
	287	break; \
	288	\
65609bd6 OD	289	set_current_state(TASK_INTERRUPTIBLE); \
65609bd6 OD	290	\
d7e09d03	291	if (__timeout == 0) { \
65609bd6	292	schedule(); \
d7e09d03	293	} else { \
b2952d62	294	long interval = info->lwi_interval ? \
b2d201bd	295	min_t(long, \
b2952d62	296	info->lwi_interval, __timeout) : \
d7e09d03	297	__timeout; \
b2d201bd	298	long remaining = schedule_timeout(interval);\
d7e09d03 PT	299	__timeout = cfs_time_sub(__timeout, \
	300	cfs_time_sub(interval, remaining));\
	301	if (__timeout == 0) { \
d2a13989	302	if (!info->lwi_on_timeout \|\| \
d7e09d03 PT	303	info->lwi_on_timeout(info->lwi_cb_data)) { \
	304	ret = -ETIMEDOUT; \
	305	break; \
	306	} \
	307	/* Take signals after the timeout expires. */ \
d2a13989	308	if (info->lwi_on_signal) \
d7e09d03 PT	309	(void)cfs_block_sigsinv(LUSTRE_FATAL_SIGS);\
	310	} \
	311	} \
	312	\
65609bd6 OD	313	set_current_state(TASK_RUNNING); \
65609bd6 OD	314	\
d7e09d03 PT	315	if (condition) \
d7e09d03 PT	316	break; \
9af4826a	317	if (signal_pending(current)) { \
d2a13989	318	if (info->lwi_on_signal && \
d7e09d03 PT	319	(__timeout == 0 \|\| __allow_intr)) { \
	320	if (info->lwi_on_signal != LWI_ON_SIGNAL_NOOP) \
	321	info->lwi_on_signal(info->lwi_cb_data);\
	322	ret = -EINTR; \
	323	break; \
	324	} \
	325	/* We have to do this here because some signals */ \
	326	/* are not blockable - ie from strace(1). */ \
	327	/* In these cases we want to schedule_timeout() */ \
	328	/* again, because we don't want that to return */ \
	329	/* -EINTR when the RPC actually succeeded. */ \
	330	/* the recalc_sigpending() below will deliver the */ \
	331	/* signal properly. */ \
	332	cfs_clear_sigpending(); \
	333	} \
	334	} \
	335	\
	336	cfs_restore_sigs(__blocked); \
	337	\
d7e09d03 PT	338	remove_wait_queue(&wq, &__wait); \
	339	} while (0)
	340
d7e09d03 PT	341	#define l_wait_event(wq, condition, info) \
	342	({ \
	343	int __ret; \
	344	struct l_wait_info *__info = (info); \
	345	\
	346	__l_wait_event(wq, condition, __info, \
	347	__ret, add_wait_queue); \
	348	__ret; \
	349	})
	350
	351	#define l_wait_event_exclusive(wq, condition, info) \
	352	({ \
	353	int __ret; \
	354	struct l_wait_info *__info = (info); \
	355	\
	356	__l_wait_event(wq, condition, __info, \
	357	__ret, add_wait_queue_exclusive); \
	358	__ret; \
	359	})
	360
	361	#define l_wait_event_exclusive_head(wq, condition, info) \
	362	({ \
	363	int __ret; \
	364	struct l_wait_info *__info = (info); \
	365	\
	366	__l_wait_event(wq, condition, __info, \
	367	__ret, add_wait_queue_exclusive_head); \
	368	__ret; \
	369	})
	370
	371	#define l_wait_condition(wq, condition) \
	372	({ \
	373	struct l_wait_info lwi = { 0 }; \
	374	l_wait_event(wq, condition, &lwi); \
	375	})
	376
	377	#define l_wait_condition_exclusive(wq, condition) \
	378	({ \
	379	struct l_wait_info lwi = { 0 }; \
	380	l_wait_event_exclusive(wq, condition, &lwi); \
	381	})
	382
	383	#define l_wait_condition_exclusive_head(wq, condition) \
	384	({ \
	385	struct l_wait_info lwi = { 0 }; \
	386	l_wait_event_exclusive_head(wq, condition, &lwi); \
	387	})
	388
	389	#define LIBLUSTRE_CLIENT (0)
	390
	391	/** @} lib */
	392
	393	#endif /* _LUSTRE_LIB_H */