Harald Welte has submitted this change and it was merged.

Change subject: fsm: allow graceful exit on FSM termination
......................................................................


fsm: allow graceful exit on FSM termination

The function _osmo_fsm_inst_term() terminates all child FSMs befor
it calls fi->fsm_cleanup(). This prevents the cleanup callback to
perform last actions on the child FSMs (e.g.
osmo_fsm_inst_unlink_parent()).

- Since moving the cleanup callack to the beginning of the function
  would alter the termination behavior and possibly cause malfunction
  in already existing implementation that use OSMO fsm, a new
  optional callback that is called immediately at the beginning of
  the terminatopn process is added.

Change-Id: I0fdda9fe994753f975a658c0f3fb3615949cc8bb
Closes: OS#2915
---
M TODO-RELEASE
M include/osmocom/core/fsm.h
M src/fsm.c
3 files changed, 16 insertions(+), 2 deletions(-)

Approvals:
  Harald Welte: Looks good to me, approved
  Jenkins Builder: Verified



diff --git a/TODO-RELEASE b/TODO-RELEASE
index 782ba19..928b18d 100644
--- a/TODO-RELEASE
+++ b/TODO-RELEASE
@@ -10,3 +10,4 @@
 core           msgb_queue_free()       add inline func to msgb.h
 coding         gsm0503_rach_ext-encode()       add func to gsm0503_coding.h
 codec          ecu.c / ecu.h                   implement ECU for FR (Error 
Concealment Unit)
+fsm            fsmc / fsm.h                    added callback for graceful 
exit => ABI changed
\ No newline at end of file
diff --git a/include/osmocom/core/fsm.h b/include/osmocom/core/fsm.h
index bbfe312..2c2a996 100644
--- a/include/osmocom/core/fsm.h
+++ b/include/osmocom/core/fsm.h
@@ -78,6 +78,8 @@
        int log_subsys;
        /*! human-readable names of events */
        const struct value_string *event_names;
+       /*! graceful exit function, called at the beginning of termination */
+       void (*pre_term)(struct osmo_fsm_inst *fi, enum osmo_fsm_term_cause 
cause);
 };
 
 /*! a single instanceof an osmocom finite state machine */
diff --git a/src/fsm.c b/src/fsm.c
index a127362..176aa8a 100644
--- a/src/fsm.c
+++ b/src/fsm.c
@@ -298,7 +298,11 @@
 
 /*! unlink child FSM from its parent FSM.
  *  \param[in] fi Descriptor of the child FSM to unlink.
- *  \param[in] ctx New talloc context */
+ *  \param[in] ctx New talloc context
+ *
+ * Never call this function from the cleanup callback, because at that time
+ * the child FSMs will already be terminated. If unlinking should be performed
+ * on FSM termination, use the grace callback instead. */
 void osmo_fsm_inst_unlink_parent(struct osmo_fsm_inst *fi, void *ctx)
 {
        if (fi->proc.parent) {
@@ -312,7 +316,10 @@
 /*! change parent instance of an FSM.
  *  \param[in] fi Descriptor of the to-be-allocated FSM.
  *  \param[in] new_parent New parent FSM instance.
- *  \param[in] new_parent_term_event Event to be sent to parent when 
terminating. */
+ *  \param[in] new_parent_term_event Event to be sent to parent when 
terminating.
+ *
+ * Never call this function from the cleanup callback!
+ * (see also osmo_fsm_inst_unlink_parent()).*/
 void osmo_fsm_inst_change_parent(struct osmo_fsm_inst *fi,
                                 struct osmo_fsm_inst *new_parent,
                                 uint32_t new_parent_term_event)
@@ -528,6 +535,10 @@
        LOGPFSMSRC(fi, file, line, "Terminating (cause = %s)\n",
                   osmo_fsm_term_cause_name(cause));
 
+       /* graceful exit (optional) */
+       if (fi->fsm->pre_term)
+               fi->fsm->pre_term(fi, cause);
+
        _osmo_fsm_inst_term_children(fi, OSMO_FSM_TERM_PARENT, NULL,
                                     file, line);
 

-- 
To view, visit https://gerrit.osmocom.org/6452
To unsubscribe, visit https://gerrit.osmocom.org/settings

Gerrit-MessageType: merged
Gerrit-Change-Id: I0fdda9fe994753f975a658c0f3fb3615949cc8bb
Gerrit-PatchSet: 4
Gerrit-Project: libosmocore
Gerrit-Branch: master
Gerrit-Owner: dexter <pma...@sysmocom.de>
Gerrit-Reviewer: Harald Welte <lafo...@gnumonks.org>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: dexter <pma...@sysmocom.de>

Reply via email to