diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c
index e725005..d8d6e72 100644
--- a/fs/dlm/lock.c
+++ b/fs/dlm/lock.c
@@ -1,7 +1,7 @@
/******************************************************************************
*******************************************************************************
**
-** Copyright (C) 2005 Red Hat, Inc. All rights reserved.
+** Copyright (C) 2005-2007 Red Hat, Inc. All rights reserved.
**
** This copyrighted material is made available to anyone wishing to use,
** modify, copy, or redistribute it subject to the terms and conditions
@@ -85,6 +85,7 @@ static int _request_lock(struct dlm_rsb *r, struct dlm_lkb *lkb);
static void __receive_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb,
struct dlm_message *ms);
static int receive_extralen(struct dlm_message *ms);
+static void do_purge(struct dlm_ls *ls, int nodeid, int pid);
/*
* Lock compatibilty matrix - thanks Steve
@@ -223,6 +224,16 @@ static inline int is_demoted(struct dlm_lkb *lkb)
return (lkb->lkb_sbflags & DLM_SBF_DEMOTED);
}
+static inline int is_altmode(struct dlm_lkb *lkb)
+{
+ return (lkb->lkb_sbflags & DLM_SBF_ALTMODE);
+}
+
+static inline int is_granted(struct dlm_lkb *lkb)
+{
+ return (lkb->lkb_status == DLM_LKSTS_GRANTED);
+}
+
static inline int is_remote(struct dlm_rsb *r)
{
DLM_ASSERT(r->res_nodeid >= 0, dlm_print_rsb(r););
@@ -254,6 +265,22 @@ static inline int down_conversion(struct dlm_lkb *lkb)
return (!middle_conversion(lkb) && lkb->lkb_rqmode < lkb->lkb_grmode);
}
+static inline int is_overlap_unlock(struct dlm_lkb *lkb)
+{
+ return lkb->lkb_flags & DLM_IFL_OVERLAP_UNLOCK;
+}
+
+static inline int is_overlap_cancel(struct dlm_lkb *lkb)
+{
+ return lkb->lkb_flags & DLM_IFL_OVERLAP_CANCEL;
+}
+
+static inline int is_overlap(struct dlm_lkb *lkb)
+{
+ return (lkb->lkb_flags & (DLM_IFL_OVERLAP_UNLOCK |
+ DLM_IFL_OVERLAP_CANCEL));
+}
+
static void queue_cast(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
{
if (is_master_copy(lkb))
@@ -267,6 +294,12 @@ static void queue_cast(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
dlm_add_ast(lkb, AST_COMP);
}
+static inline void queue_cast_overlap(struct dlm_rsb *r, struct dlm_lkb *lkb)
+{
+ queue_cast(r, lkb,
+ is_overlap_unlock(lkb) ? -DLM_EUNLOCK : -DLM_ECANCEL);
+}
+
static void queue_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int rqmode)
{
if (is_master_copy(lkb))
@@ -547,6 +580,7 @@ static int create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret)
lkb->lkb_grmode = DLM_LOCK_IV;
kref_init(&lkb->lkb_ref);
INIT_LIST_HEAD(&lkb->lkb_ownqueue);
+ INIT_LIST_HEAD(&lkb->lkb_rsb_lookup);
get_random_bytes(&bucket, sizeof(bucket));
bucket &= (ls->ls_lkbtbl_size - 1);
@@ -556,7 +590,7 @@ static int create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret)
/* counter can roll over so we must verify lkid is not in use */
while (lkid == 0) {
- lkid = bucket | (ls->ls_lkbtbl[bucket].counter++ << 16);
+ lkid = (bucket << 16) | ls->ls_lkbtbl[bucket].counter++;
list_for_each_entry(tmp, &ls->ls_lkbtbl[bucket].list,
lkb_idtbl_list) {
@@ -577,8 +611,8 @@ static int create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret)
static struct dlm_lkb *__find_lkb(struct dlm_ls *ls, uint32_t lkid)
{
- uint16_t bucket = lkid & 0xFFFF;
struct dlm_lkb *lkb;
+ uint16_t bucket = (lkid >> 16);
list_for_each_entry(lkb, &ls->ls_lkbtbl[bucket].list, lkb_idtbl_list) {
if (lkb->lkb_id == lkid)
@@ -590,7 +624,7 @@ static struct dlm_lkb *__find_lkb(struct dlm_ls *ls, uint32_t lkid)
static int find_lkb(struct dlm_ls *ls, uint32_t lkid, struct dlm_lkb **lkb_ret)
{
struct dlm_lkb *lkb;
- uint16_t bucket = lkid & 0xFFFF;
+ uint16_t bucket = (lkid >> 16);
if (bucket >= ls->ls_lkbtbl_size)
return -EBADSLT;
@@ -620,7 +654,7 @@ static void kill_lkb(struct kref *kref)
static int __put_lkb(struct dlm_ls *ls, struct dlm_lkb *lkb)
{
- uint16_t bucket = lkb->lkb_id & 0xFFFF;
+ uint16_t bucket = (lkb->lkb_id >> 16);
write_lock(&ls->ls_lkbtbl[bucket].lock);
if (kref_put(&lkb->lkb_ref, kill_lkb)) {
@@ -735,23 +769,75 @@ static void move_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb, int sts)
unhold_lkb(lkb);
}
+static int msg_reply_type(int mstype)
+{
+ switch (mstype) {
+ case DLM_MSG_REQUEST:
+ return DLM_MSG_REQUEST_REPLY;
+ case DLM_MSG_CONVERT:
+ return DLM_MSG_CONVERT_REPLY;
+ case DLM_MSG_UNLOCK:
+ return DLM_MSG_UNLOCK_REPLY;
+ case DLM_MSG_CANCEL:
+ return DLM_MSG_CANCEL_REPLY;
+ case DLM_MSG_LOOKUP:
+ return DLM_MSG_LOOKUP_REPLY;
+ }
+ return -1;
+}
+
/* add/remove lkb from global waiters list of lkb's waiting for
a reply from a remote node */
-static void add_to_waiters(struct dlm_lkb *lkb, int mstype)
+static int add_to_waiters(struct dlm_lkb *lkb, int mstype)
{
struct dlm_ls *ls = lkb->lkb_resource->res_ls;
+ int error = 0;
mutex_lock(&ls->ls_waiters_mutex);
- if (lkb->lkb_wait_type) {
- log_print("add_to_waiters error %d", lkb->lkb_wait_type);
+
+ if (is_overlap_unlock(lkb) ||
+ (is_overlap_cancel(lkb) && (mstype == DLM_MSG_CANCEL))) {
+ error = -EINVAL;
+ goto out;
+ }
+
+ if (lkb->lkb_wait_type || is_overlap_cancel(lkb)) {
+ switch (mstype) {
+ case DLM_MSG_UNLOCK:
+ lkb->lkb_flags |= DLM_IFL_OVERLAP_UNLOCK;
+ break;
+ case DLM_MSG_CANCEL:
+ lkb->lkb_flags |= DLM_IFL_OVERLAP_CANCEL;
+ break;
+ default:
+ error = -EBUSY;
+ goto out;
+ }
+ lkb->lkb_wait_count++;
+ hold_lkb(lkb);
+
+ log_debug(ls, "add overlap %x cur %d new %d count %d flags %x",
+ lkb->lkb_id, lkb->lkb_wait_type, mstype,
+ lkb->lkb_wait_count, lkb->lkb_flags);
goto out;
}
+
+ DLM_ASSERT(!lkb->lkb_wait_count,
+ dlm_print_lkb(lkb);
+ printk("wait_count %d\n", lkb->lkb_wait_count););
+
+ lkb->lkb_wait_count++;
lkb->lkb_wait_type = mstype;
- kref_get(&lkb->lkb_ref);
+ hold_lkb(lkb);
list_add(&lkb->lkb_wait_reply, &ls->ls_waiters);
out:
+ if (error)
+ log_error(ls, "add_to_waiters %x error %d flags %x %d %d %s",
+ lkb->lkb_id, error, lkb->lkb_flags, mstype,
+ lkb->lkb_wait_type, lkb->lkb_resource->res_name);
mutex_unlock(&ls->ls_waiters_mutex);
+ return error;
}
/* We clear the RESEND flag because we might be taking an lkb off the waiters
@@ -759,34 +845,85 @@ static void add_to_waiters(struct dlm_lkb *lkb, int mstype)
request reply on the requestqueue) between dlm_recover_waiters_pre() which
set RESEND and dlm_recover_waiters_post() */
-static int _remove_from_waiters(struct dlm_lkb *lkb)
+static int _remove_from_waiters(struct dlm_lkb *lkb, int mstype)
{
- int error = 0;
+ struct dlm_ls *ls = lkb->lkb_resource->res_ls;
+ int overlap_done = 0;
- if (!lkb->lkb_wait_type) {
- log_print("remove_from_waiters error");
- error = -EINVAL;
- goto out;
+ if (is_overlap_unlock(lkb) && (mstype == DLM_MSG_UNLOCK_REPLY)) {
+ lkb->lkb_flags &= ~DLM_IFL_OVERLAP_UNLOCK;
+ overlap_done = 1;
+ goto out_del;
+ }
+
+ if (is_overlap_cancel(lkb) && (mstype == DLM_MSG_CANCEL_REPLY)) {
+ lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
+ overlap_done = 1;
+ goto out_del;
+ }
+
+ /* N.B. type of reply may not always correspond to type of original
+ msg due to lookup->request optimization, verify others? */
+
+ if (lkb->lkb_wait_type) {
+ lkb->lkb_wait_type = 0;
+ goto out_del;
+ }
+
+ log_error(ls, "remove_from_waiters lkid %x flags %x types %d %d",
+ lkb->lkb_id, lkb->lkb_flags, mstype, lkb->lkb_wait_type);
+ return -1;
+
+ out_del:
+ /* the force-unlock/cancel has completed and we haven't recvd a reply
+ to the op that was in progress prior to the unlock/cancel; we
+ give up on any reply to the earlier op. FIXME: not sure when/how
+ this would happen */
+
+ if (overlap_done && lkb->lkb_wait_type) {
+ log_error(ls, "remove_from_waiters %x reply %d give up on %d",
+ lkb->lkb_id, mstype, lkb->lkb_wait_type);
+ lkb->lkb_wait_count--;
+ lkb->lkb_wait_type = 0;
}
- lkb->lkb_wait_type = 0;
+
+ DLM_ASSERT(lkb->lkb_wait_count, dlm_print_lkb(lkb););
+
lkb->lkb_flags &= ~DLM_IFL_RESEND;
- list_del(&lkb->lkb_wait_reply);
+ lkb->lkb_wait_count--;
+ if (!lkb->lkb_wait_count)
+ list_del_init(&lkb->lkb_wait_reply);
unhold_lkb(lkb);
- out:
- return error;
+ return 0;
}
-static int remove_from_waiters(struct dlm_lkb *lkb)
+static int remove_from_waiters(struct dlm_lkb *lkb, int mstype)
{
struct dlm_ls *ls = lkb->lkb_resource->res_ls;
int error;
mutex_lock(&ls->ls_waiters_mutex);
- error = _remove_from_waiters(lkb);
+ error = _remove_from_waiters(lkb, mstype);
mutex_unlock(&ls->ls_waiters_mutex);
return error;
}
+/* Handles situations where we might be processing a "fake" or "stub" reply in
+ which we can't try to take waiters_mutex again. */
+
+static int remove_from_waiters_ms(struct dlm_lkb *lkb, struct dlm_message *ms)
+{
+ struct dlm_ls *ls = lkb->lkb_resource->res_ls;
+ int error;
+
+ if (ms != &ls->ls_stub_ms)
+ mutex_lock(&ls->ls_waiters_mutex);
+ error = _remove_from_waiters(lkb, ms->m_type);
+ if (ms != &ls->ls_stub_ms)
+ mutex_unlock(&ls->ls_waiters_mutex);
+ return error;
+}
+
static void dir_remove(struct dlm_rsb *r)
{
int to_nodeid;
@@ -988,8 +1125,14 @@ static void remove_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb)
_remove_lock(r, lkb);
}
-static void revert_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
+/* returns: 0 did nothing
+ 1 moved lock to granted
+ -1 removed lock */
+
+static int revert_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
+ int rv = 0;
+
lkb->lkb_rqmode = DLM_LOCK_IV;
switch (lkb->lkb_status) {
@@ -997,6 +1140,7 @@ static void revert_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
break;
case DLM_LKSTS_CONVERT:
move_lkb(r, lkb, DLM_LKSTS_GRANTED);
+ rv = 1;
break;
case DLM_LKSTS_WAITING:
del_lkb(r, lkb);
@@ -1004,15 +1148,17 @@ static void revert_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
/* this unhold undoes the original ref from create_lkb()
so this leads to the lkb being freed */
unhold_lkb(lkb);
+ rv = -1;
break;
default:
log_print("invalid status for revert %d", lkb->lkb_status);
}
+ return rv;
}
-static void revert_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb)
+static int revert_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
- revert_lock(r, lkb);
+ return revert_lock(r, lkb);
}
static void _grant_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
@@ -1055,6 +1201,50 @@ static void grant_lock_pending(struct dlm_rsb *r, struct dlm_lkb *lkb)
queue_cast(r, lkb, 0);
}
+/* The special CONVDEADLK, ALTPR and ALTCW flags allow the master to
+ change the granted/requested modes. We're munging things accordingly in
+ the process copy.
+ CONVDEADLK: our grmode may have been forced down to NL to resolve a
+ conversion deadlock
+ ALTPR/ALTCW: our rqmode may have been changed to PR or CW to become
+ compatible with other granted locks */
+
+static void munge_demoted(struct dlm_lkb *lkb, struct dlm_message *ms)
+{
+ if (ms->m_type != DLM_MSG_CONVERT_REPLY) {
+ log_print("munge_demoted %x invalid reply type %d",
+ lkb->lkb_id, ms->m_type);
+ return;
+ }
+
+ if (lkb->lkb_rqmode == DLM_LOCK_IV || lkb->lkb_grmode == DLM_LOCK_IV) {
+ log_print("munge_demoted %x invalid modes gr %d rq %d",
+ lkb->lkb_id, lkb->lkb_grmode, lkb->lkb_rqmode);
+ return;
+ }
+
+ lkb->lkb_grmode = DLM_LOCK_NL;
+}
+
+static void munge_altmode(struct dlm_lkb *lkb, struct dlm_message *ms)
+{
+ if (ms->m_type != DLM_MSG_REQUEST_REPLY &&
+ ms->m_type != DLM_MSG_GRANT) {
+ log_print("munge_altmode %x invalid reply type %d",
+ lkb->lkb_id, ms->m_type);
+ return;
+ }
+
+ if (lkb->lkb_exflags & DLM_LKF_ALTPR)
+ lkb->lkb_rqmode = DLM_LOCK_PR;
+ else if (lkb->lkb_exflags & DLM_LKF_ALTCW)
+ lkb->lkb_rqmode = DLM_LOCK_CW;
+ else {