gprs: deattach driver when gprs status indicates unregistered
PATCH 1/1 gprs: deattach driver when gprs status indicates unregistered
From: Richard Röjfors <[email protected]>
In the same way as we consider the driver attached when the gprs status indicates we are registered, we should consider it deattached when the status indicates unregistration.
If we don't, then we would not always recover from the case when deattaching the driver fails. We would just revert back the driver attached status to true, and "ignore" if the status indicates the opposite when we check the registration status afterwards.
--- src/gprs.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
src/gprs.c +6 -6
@@ -2568,12 +2568,6 @@ void ofono_gprs_status_notify(struct ofono_gprs *gprs, int status) gprs->status = status; -if (status !=NETWORK_REGISTRATION_STATUS_REGISTERED && - status !=NETWORK_REGISTRATION_STATUS_ROAMING ) { -gprs_attached_update (gprs); -return ; - } -/* * If we're already taking action, e.g. attaching or detaching, then * ignore this notification for now, we will take appropriate action
@@ -2582,6 +2576,12 @@ void ofono_gprs_status_notify(struct ofono_gprs *gprs, int status)if (gprs->flags &GPRS_FLAG_ATTACHING )return ; +if (status !=NETWORK_REGISTRATION_STATUS_REGISTERED && + status !=NETWORK_REGISTRATION_STATUS_ROAMING ) { +ofono_gprs_detached_notify (gprs); +return ; + } +/* We registered without being powered */ if (gprs->powered == FALSE) goto detach;
Signature · 1 line
2.20.1
...and a log for this:
Got deregistered
Sep 17 20:02:22 daemon.info: Aux: < \r\n+CREG: 0\r\n Sep 17 20:02:22 daemon.debug: ../git/src/network.c:ofono_netreg_status_notify() /ublox_0 status 0 tech -1 lac -1 ci -1 Sep 17 20:02:22 daemon.debug: ../git/src/network.c:current_operator_callback() 0x1bb4360, 0x1bb6ae0 Sep 17 20:02:22 daemon.debug: ../git/src/gprs.c:netreg_status_changed() 0 (unregistered) Sep 17 20:02:22 daemon.debug: ../git/src/gprs.c:gprs_netreg_update() attach: 0, driver_attached: 1 Sep 17 20:02:22 daemon.debug: ../git/src/gprs.c:netreg_status_changed() 0 (unregistered) Sep 17 20:02:22 daemon.debug: ../git/src/gprs.c:gprs_netreg_update() attach: 0, driver_attached: 0 Sep 17 20:02:22 daemon.info: Aux: > AT\r Sep 17 20:02:22 daemon.info: Aux: < \r\n+CIEV: 7,0\r\n Sep 17 20:02:22 daemon.info: Aux: < \r\nOK\r\n Sep 17 20:02:22 daemon.info: Aux: Finally woke up the modem Sep 17 20:02:22 daemon.info: Aux: > AT+CGATT=0\r Sep 17 20:02:22 daemon.info: Aux: < \r\n+CME ERROR: 148\r\n
Detach failed - driver_attached is set to 1 again
Sep 17 20:02:22 daemon.debug: ../git/src/gprs.c:gprs_attach_callback() /ublox_0 error = 1 Sep 17 20:02:22 daemon.info: Aux: > AT+CGREG?\r Sep 17 20:02:22 daemon.info: Aux: < \r\n+CGREG: 4\r\n Sep 17 20:02:22 daemon.info: Aux: < \r\n+UREG: 0\r\n Sep 17 20:02:22 daemon.info: Aux: < \r\n+CIEV: 9,1\r\n Sep 17 20:02:22 daemon.info: Aux: < \r\n+CGREG: 2,4\r\n Sep 17 20:02:22 daemon.info: Aux: < \r\nOK\r\n Sep 17 20:02:22 daemon.debug: ../git/src/gprs.c:registration_status_cb()
/ublox_0 error 0 status 4
We read back the registration status, but does not set the driver_attached to 0
Sep 17 20:02:22 daemon.debug: ../git/src/gprs.c:ofono_gprs_status_notify() /ublox_0 status unknown (4) Sep 17 20:02:22 daemon.debug: ../git/src/gprs.c:gprs_set_attached_property() 0x1bb1500 1 -> 0 Sep 17 20:02:33 daemon.info: Aux: < \r\n+CREG: 0\r\n Sep 17 20:02:33 daemon.debug: ../git/src/network.c:ofono_netreg_status_notify() /ublox_0 status 0 tech -1 lac -1 ci -1 Sep 17 20:02:33 daemon.debug: ../git/src/network.c:current_operator_callback() 0x1bb4360, (nil) Sep 17 20:02:33 daemon.debug: ../git/src/gprs.c:netreg_status_changed() 0 (unregistered) Sep 17 20:02:33 daemon.debug: ../git/src/gprs.c:gprs_netreg_update()
attach: 0, driver_attached: 1
We still believe the driver is attached, trying to detach again
Sep 17 20:02:33 daemon.info: Aux: > AT\r Sep 17 20:02:33 daemon.info: Aux: < \r\nOK\r\n Sep 17 20:02:33 daemon.info: Aux: Finally woke up the modem Sep 17 20:02:33 daemon.info: Aux: > AT+CGATT=0\r Sep 17 20:05:33 daemon.info: Aux: < \r\n+CME ERROR: 2000\r\n
Detach failed again, setting back driver_attached to 1
Sep 17 20:05:33 daemon.debug: ../git/src/gprs.c:gprs_attach_callback() /ublox_0 error = 1 Sep 17 20:05:33 daemon.info: Aux: > AT\r Sep 17 20:05:33 daemon.info: Aux: < \r\nOK\r\n Sep 17 20:05:33 daemon.info: Aux: Finally woke up the modem Sep 17 20:05:33 daemon.info: Aux: > AT+CGREG?\r Sep 17 20:05:33 daemon.info: Aux: < \r\n+CGREG: 2,4\r\n Sep 17 20:05:33 daemon.info: Aux: < \r\nOK\r\n Sep 17 20:05:33 daemon.debug: ../git/src/gprs.c:registration_status_cb()
/ublox_0 error 0 status 4
We still do not set driver_attached to 0, even CREG says unknown status
Sep 17 20:05:33 daemon.debug: ../git/src/gprs.c:ofono_gprs_status_notify() /ublox_0 status unknown (4) Sep 17 20:05:33 daemon.info: Aux: < \r\n+CTZE: +08,1,"19/09/17,22:02:36"\r\n Sep 17 20:05:33 daemon.info: Aux: < \r\n+CREG: 5,"0000","00000010",2\r\n Sep 17 20:05:33 daemon.debug: ../git/drivers/ubloxmodem/network-registration.c:ctze_notify() tz +08 dst 1 time 19/09/17,22:02:36 Sep 17 20:05:33 daemon.debug: ../git/src/network.c:ofono_netreg_time_notify() net time 2019-09-17 22:02:36 utcoff 7200 dst 1 Sep 17 20:05:33 daemon.debug: ../git/src/network.c:ofono_netreg_status_notify() /ublox_0 status 5 tech 2 lac 0 ci 16 Sep 17 20:05:33 daemon.debug: ../git/src/gprs.c:netreg_status_changed() 5 (roaming) Sep 17 20:05:33 daemon.debug: ../git/src/gprs.c:gprs_netreg_update()
attach: 1, driver_attached: 1
We now got registered, but we believe the driver is attached, so no action is taken and the driver will not be attached...
and the Attached D-Bus property will remain False.
--Richard
Den tis 17 sep. 2019 kl 23:48 skrev <[email protected]>:
Quoted 53 lines
From: Richard Röjfors <[email protected]>
In the same way as we consider the driver attached when the gprs status indicates we are registered, we should consider it deattached when the status indicates unregistration.
If we don't, then we would not always recover from the case when deattaching the driver fails. We would just revert back the driver attached status to true, and "ignore" if the status indicates the opposite when we check the registration status afterwards.
--- src/gprs.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)src/gprs.c +0
-6@@ -2568,12 +2568,6 @@ void ofono_gprs_status_notify(struct ofono_gprs *gprs, int status) gprs->status = status; -if (status !=NETWORK_REGISTRATION_STATUS_REGISTERED && - status !=NETWORK_REGISTRATION_STATUS_ROAMING ) { -gprs_attached_update (gprs); -return ; - } -/* * If we're already taking action, e.g. attaching or detaching, then
* ignore this notification for now, we will take appropriate
actionsrc/gprs.c +6
-0@@ -2582,6 +2576,12 @@ void ofono_gprs_status_notify(struct ofono_gprs *gprs, int status)if (gprs->flags &GPRS_FLAG_ATTACHING )return ; +if (status !=NETWORK_REGISTRATION_STATUS_REGISTERED && + status !=NETWORK_REGISTRATION_STATUS_ROAMING ) { +ofono_gprs_detached_notify (gprs); +return ; + } +/* We registered without being powered */ if (gprs->powered == FALSE)goto detach;--
2.20.1
Denis Kenzior
AppliedHi Richard,
Quoted 6 lines · richard.rojfors(a)gmail.com
From: Richard Röjfors <[email protected]>
In the same way as we consider the driver attached when the gprs status indicates we are registered, we should consider it deattached when the status indicates unregistration.
Quoted 9 lines
If we don't, then we would not always recover from the case when deattaching the driver fails. We would just revert back the driver attached status to true, and "ignore" if the status indicates the opposite when we check the registration status afterwards.
--- src/gprs.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
Applied, thanks.
Regards,
-Denis