Hi Alok,

On Wed, Aug 24, 2011 at 04:44:07PM +0300, Alok Barsode wrote:
> From: Alok Barsode <alok.bars...@linux.intel.com>
> 
> connman_technology_enable: Enable a technology. Enables all the
> devices in the device_list of the technology.
> connman_technology_enabled: Callback for connman_technology_enable.
> Changes the state of the technology to ENABLED.
> Ditto for connman_technology_disable/connman_technology_disabled.
So I tested those patches, and I have a couple concerns:

1) I had to apply the following patch:

diff --git a/src/device.c b/src/device.c
index cdd9118..52a5029 100644
--- a/src/device.c
+++ b/src/device.c
@@ -663,10 +663,12 @@ int connman_device_set_powered(struct connman_device
*devi
        if (device->powered == powered)
                return -EALREADY;
 
-       /* Reset pending request */
-       g_source_remove(device->pending_timeout);
-       device->pending_timeout = 0;
-       device->powered_pending = PENDING_NONE;
+       if (device->pending_timeout) {
+               /* Reset pending request */
+               g_source_remove(device->pending_timeout);

in order to avoid GLib critical warnings. Try disabling/enabling ethernet and
you will get the same. There is no pending request with synchronous
technologies as ethernet.

2) I found at least one bug, where turning offline mode on while my WiFi
chipset was associating wouldn't turn WiFi off. I got this from test-manager:

AvailableTechnologies
    [ ethernet wifi ]
OfflineMode
    true
Technologies
    /net/connman/technology/bluetooth
        State = offline
        Type = bluetooth
        Name = Bluetooth
        Tethering = false
    /net/connman/technology/wifi
        State = enabled
        Type = wifi
        Name = WiFi
        Tethering = false
    /net/connman/technology/ethernet
        State = enabled
        Type = ethernet
        Name = Wired
        Tethering = false
SessionMode
    0

3) I realized that we never move the technology states to connected. This is
not a regression, but I'd appreciate if you could look at it.


So, I think we should fix 1) and 2) before pushing this patchset as they're
real regressions.

Cheers,
Samuel.

-- 
Intel Open Source Technology Centre
http://oss.intel.com/
_______________________________________________
connman mailing list
connman@connman.net
http://lists.connman.net/listinfo/connman

Reply via email to