could have a look at this and provide some fix, or at least some
detailed pointers at the work required. I am happy to collaborate and
test the patches as needed.
Best Regards,
Pau
--
- Pau Espin Pedrol http://www.sysmocom.de/
==
] https://osmocom.org/
[2] https://osmocom.org/projects/osmo-gsm-tester
[3] https://git.osmocom.org/osmo-gsm-tester/
[4]
https://git.sysmocom.de/ofono/commit/?h=osmo-gsm-tester&id=e77b82a1bd1986aab42a119e0556c5941795992e
[5] https://git.sysmocom.de/ofono/log/?h=osmo-gsm-tester
[6] https://osmocom.or
syspath is not really needed in there, and catching devpath instead
allows us to easily match against DEVPATH_OLD in follow-up patches.
Furthermore, previous code was misleading retrieving syspath in
remove_device() and then passing it as devpath param of check_remove().
---
plugins/udevng.c | 12
If for whatever reason a network interface belonging to a modem is
renamed after ofono startup, ofono then keeps advertising the old
non-existing interface name instead of the new one.
To catch the changes we use udev's "move" action together with
DEVPATH_OLD variable, which points to the old devpa
inet_ntoa was called twice in a row and as a result both primay and
secondary DNS were pointing to same static buffer containing last
value (secondary DNS).
As a result, ofono always advertised the secondary DNS twice through
DBus ConnectionContext.GetProperties 'DomainNameServers'.
Related: http
---
src/modem.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/modem.c b/src/modem.c
index d5fda7ca..698a7ca8 100644
--- a/src/modem.c
+++ b/src/modem.c
@@ -1876,8 +1876,10 @@ struct ofono_modem *ofono_modem_create(const char *name,
const char *type)
else
ch or mine (or
both). I'll send a v2 of my patch as suggested and let you deliver on
the best approach. If not good solution is provided, I'd need to keep it
in a separate fork, which imho would be really sad as it seems other
people may potentially use this kind of feature.
Regards,
(dbus_validate_path) which is
expected if a bad OFONO_PATHMNAME is supplied. If it's fine for you, I
can add an error message there stating that dbus_validate_path failed
and printing the pathname. Agree?
--
The current udevng.c implementation sets dbus path names for modems
based on type and a number incremented seuqntially for each new modem
found. As a result, the dbus path for a given device is non
deterministic, since it depends on the devices available during ofono
startup.
Furthermore, if a mod
they prevent us to run ofono.
However, if someone with good knowledge on the code is willing to have a
look and fix them, please feel free to do it.
[1] https://osmocom.org/issues/2236
--
- Pau Espin Pedrol http://www.sysmoc
,
Pau
--
- Pau Espin Pedrol http://www.sysmocom.de/
===
* sysmocom - systems for mobile communications GmbH
* Alt-Moabit 93
* 10559 Berlin, Germany
* Sitz / Registered office: Berlin, HRB 134158 B
* Geschaeftsfuehrer
Fixes following issue announced by valgrind:
ofonod[2870]: plugins/gobi.c:shutdown_cb()
ofonod[2870]: src/modem.c:modem_change_state() old state: 0, new state: 0
==2870== Invalid read of size 8
==2870==at 0x453ADE: shutdown_destroy (qmi.c:1254)
==2870==by 0x508A717: ??? (in /lib/x86_64-lin
12 matches
Mail list logo