2010-10-23 05:33:18

by Tommi Keisala

Subject: [PATCH] This patch avoids a crash when org.bluez.Manager GetProperties request is received and there is not yet any adapters ready. Happens often for example when bluetoothd and ofonod is started next ot each other.

---
src/manager.c | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/src/manager.c b/src/manager.c
index aff069c..dd9560f 100644
--- a/src/manager.c
+++ b/src/manager.c
@@ -197,13 +197,14 @@ static DBusMessage *get_properties(DBusConnection *conn,
DBUS_DICT_ENTRY_END_CHAR_AS_STRING, &dict);

array = g_new0(char *, g_slist_length(adapters) + 1);
- for (i = 0, list = adapters; list; list = list->next, i++) {
+ for (i = 0, list = adapters; list; list = list->next) {
struct btd_adapter *adapter = list->data;

if (!adapter_is_ready(adapter))
continue;

array[i] = (char *) adapter_get_path(adapter);
+ i++;
}
dict_append_array(&dict, "Adapters", DBUS_TYPE_OBJECT_PATH, &array, i);
g_free(array);
--
1.7.0.4

2010-10-23 12:31:11

by Johan Hedberg

Subject: Re: [PATCH] This patch avoids a crash when org.bluez.Manager GetProperties request is received and there is not yet any adapters ready. Happens often for example when bluetoothd and ofonod is started next ot each other.

Hi Tommi,

On Sat, Oct 23, 2010, Tommi Keisala wrote:
> ---
> src/manager.c | 3 ++-
> 1 files changed, 2 insertions(+), 1 deletions(-)

You seem to have somehow managed to screw up the commit message this
time: everything in the summary line and nothing in the message body.
Anyway, to avoid further feedback rounds (since the patch itself is fine
now) I went ahead and fixed the commit message and pushed your patch
upstream. Thanks.

Johan