Send connman mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
https://lists.01.org/mailman/listinfo/connman
or, via email, send a message with subject or body 'help' to
[email protected]
You can reach the person managing the list at
[email protected]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of connman digest..."
Today's Topics:
1. [PATCHv2] gdhcp: Don't try to remove timer again (Saurav Babu)
2. [PATCH] config: Use g_try_malloc0 instead of g_malloc0
(Niraj Kumar Goit)
3. Re: [PATCHv2] gdhcp: Don't try to remove timer again
(Patrik Flykt)
4. [PATCHv2] gdhcp: Don't try to remove timer again (Saurav Babu)
5. Re: [PATCH] config: Use g_try_malloc0 instead of g_malloc0
(Patrik Flykt)
----------------------------------------------------------------------
Message: 1
Date: Mon, 25 Jan 2016 16:28:37 +0530
From: Saurav Babu <[email protected]>
To: [email protected]
Cc: [email protected]
Subject: [PATCHv2] gdhcp: Don't try to remove timer again
Message-ID:
<[email protected]>
GLib-CRITICAL warning message is obtained in below scenario:
1. service is connected and link local address is obtained.
2. Try to disconnect service.
(connmand:8377): GLib-CRITICAL **: Source ID 289 was not found when attempting
to remove it
(connmand:8377): GLib-CRITICAL **: Source ID 303 was not found when attempting
to remove it
When Link Local IP address is obtained then both dhcp_client->timeout
assigned for DISCOVER_TIMEOUT and ANNOUNCE_INTERVAL are already removed
when discover_timeout() and ipv4ll_announce_timeout() function returns
FALSE but the dhcp_client->timeout is not assigned to 0. Now when
dhcp_release() calls remove_timeouts() function then
dhcp_client->timeout is tried to remove again resulting in GLib-CRITICAL
warning. This patch explicitly sets dhcp_client->timeout to 0 in
discover_timeout() and ipv4ll_announce_timeout() function.
---
gdhcp/client.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/gdhcp/client.c b/gdhcp/client.c
index f9cba89..a0adb65 100644
--- a/gdhcp/client.c
+++ b/gdhcp/client.c
@@ -2625,6 +2625,7 @@ static gboolean discover_timeout(gpointer user_data)
* if the server is non-authoritative it will ignore the request if the
* option is present.
*/
+ dhcp_client->timeout = 0;
g_dhcp_client_start(dhcp_client, NULL);
return FALSE;
@@ -2681,6 +2682,7 @@ static gboolean ipv4ll_announce_timeout(gpointer
dhcp_data)
dhcp_client->ipv4ll_available_cb(dhcp_client,
dhcp_client->ipv4ll_available_data);
dhcp_client->conflicts = 0;
+ dhcp_client->timeout = 0;
return FALSE;
}
--
1.9.1
------------------------------
Message: 2
Date: Mon, 25 Jan 2016 17:40:05 +0530
From: Niraj Kumar Goit <[email protected]>
To: [email protected]
Cc: [email protected]
Subject: [PATCH] config: Use g_try_malloc0 instead of g_malloc0
Message-ID: <[email protected]>
---
src/config.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/config.c b/src/config.c
index 344b8ce..3e30847 100644
--- a/src/config.c
+++ b/src/config.c
@@ -1328,7 +1328,10 @@ static int try_provision_service(struct
connman_config_service *config,
if (config->virtual) {
struct connect_virtual *virtual;
- virtual = g_malloc0(sizeof(struct connect_virtual));
+ virtual = g_try_malloc0(sizeof(struct connect_virtual));
+ if(!virtual)
+ return -ENOMEM;
+
virtual->service = service;
virtual->vfile = config->virtual_file;
--
1.7.9.5
------------------------------
Message: 3
Date: Mon, 25 Jan 2016 14:24:18 +0200
From: Patrik Flykt <[email protected]>
To: Saurav Babu <[email protected]>
Cc: [email protected], [email protected]
Subject: Re: [PATCHv2] gdhcp: Don't try to remove timer again
Message-ID: <[email protected]>
Content-Type: text/plain; charset="UTF-8"
On Mon, 2016-01-25 at 16:28 +0530, Saurav Babu wrote:
> GLib-CRITICAL warning message is obtained in below scenario:
> 1. service is connected and link local address is obtained.
> 2. Try to disconnect service.
>
> (connmand:8377): GLib-CRITICAL **: Source ID 289 was not found when
> attempting to remove it
>
> (connmand:8377): GLib-CRITICAL **: Source ID 303 was not found when
> attempting to remove it
>
> When Link Local IP address is obtained then both dhcp_client->timeout
> assigned for DISCOVER_TIMEOUT and ANNOUNCE_INTERVAL are already removed
> when discover_timeout() and ipv4ll_announce_timeout() function returns
> FALSE but the dhcp_client->timeout is not assigned to 0. Now when
> dhcp_release() calls remove_timeouts() function then
> dhcp_client->timeout is tried to remove again resulting in GLib-CRITICAL
> warning. This patch explicitly sets dhcp_client->timeout to 0 in
> discover_timeout() and ipv4ll_announce_timeout() function.
> ---
The commit message may still be correct for both versions of the patch,
but please add an explanation about the differences between these two
versions here after the '--'. Usually a new version requires some part
of the commit message to be amended...
Cheers,
Patrik
> gdhcp/client.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/gdhcp/client.c b/gdhcp/client.c
> index f9cba89..a0adb65 100644
> --- a/gdhcp/client.c
> +++ b/gdhcp/client.c
> @@ -2625,6 +2625,7 @@ static gboolean discover_timeout(gpointer user_data)
> * if the server is non-authoritative it will ignore the request if the
> * option is present.
> */
> + dhcp_client->timeout = 0;
> g_dhcp_client_start(dhcp_client, NULL);
>
> return FALSE;
> @@ -2681,6 +2682,7 @@ static gboolean ipv4ll_announce_timeout(gpointer
> dhcp_data)
> dhcp_client->ipv4ll_available_cb(dhcp_client,
> dhcp_client->ipv4ll_available_data);
> dhcp_client->conflicts = 0;
> + dhcp_client->timeout = 0;
>
> return FALSE;
> }
------------------------------
Message: 4
Date: Mon, 25 Jan 2016 18:04:15 +0530
From: Saurav Babu <[email protected]>
To: [email protected]
Cc: [email protected]
Subject: [PATCHv2] gdhcp: Don't try to remove timer again
Message-ID: <[email protected]>
GLib-CRITICAL warning message is obtained in below scenario:
1. service is connected and link local address is obtained.
2. Try to disconnect service.
(connmand:8377): GLib-CRITICAL **: Source ID 289 was not found when attempting
to remove it
(connmand:8377): GLib-CRITICAL **: Source ID 303 was not found when attempting
to remove it
When Link Local IP address is obtained then both dhcp_client->timeout
assigned for DISCOVER_TIMEOUT and ANNOUNCE_INTERVAL are already removed
when discover_timeout() and ipv4ll_announce_timeout() function returns
FALSE but the dhcp_client->timeout is not assigned to 0. Now when
dhcp_release() calls remove_timeouts() function then
dhcp_client->timeout is tried to remove again resulting in GLib-CRITICAL
warning. This patch explicitly sets dhcp_client->timeout to 0 in
discover_timeout() and ipv4ll_announce_timeout() function.
---
dhcp_client->timeout must be assigned to 0 before g_dhcp_client_start() because
there is a possibility that dhcp_client->timeout is started again in
g_dhcp_client_start() if dhcp_client->retry_times is not equal to
DISCOVER_RETRIES.
In earlier patch dhcp_client->timeout value was again set to 0 after the
timeout was
started properly.
gdhcp/client.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/gdhcp/client.c b/gdhcp/client.c
index f9cba89..a0adb65 100644
--- a/gdhcp/client.c
+++ b/gdhcp/client.c
@@ -2625,6 +2625,7 @@ static gboolean discover_timeout(gpointer user_data)
* if the server is non-authoritative it will ignore the request if the
* option is present.
*/
+ dhcp_client->timeout = 0;
g_dhcp_client_start(dhcp_client, NULL);
return FALSE;
@@ -2681,6 +2682,7 @@ static gboolean ipv4ll_announce_timeout(gpointer
dhcp_data)
dhcp_client->ipv4ll_available_cb(dhcp_client,
dhcp_client->ipv4ll_available_data);
dhcp_client->conflicts = 0;
+ dhcp_client->timeout = 0;
return FALSE;
}
--
1.9.1
------------------------------
Message: 5
Date: Mon, 25 Jan 2016 14:38:04 +0200
From: Patrik Flykt <[email protected]>
To: Niraj Kumar Goit <[email protected]>
Cc: [email protected], [email protected]
Subject: Re: [PATCH] config: Use g_try_malloc0 instead of g_malloc0
Message-ID: <[email protected]>
Content-Type: text/plain; charset="UTF-8"
Hi,
On Mon, 2016-01-25 at 17:40 +0530, Niraj Kumar Goit wrote:
> ---
> src/config.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/src/config.c b/src/config.c
> index 344b8ce..3e30847 100644
> --- a/src/config.c
> +++ b/src/config.c
> @@ -1328,7 +1328,10 @@ static int try_provision_service(struct
> connman_config_service *config,
> if (config->virtual) {
> struct connect_virtual *virtual;
>
> - virtual = g_malloc0(sizeof(struct connect_virtual));
> + virtual = g_try_malloc0(sizeof(struct connect_virtual));
ConnMan is in really big trouble if allocation of 16 bytes fails.
Nothing will work if the memory is exahausted anyway, so aborting from
g_malloc0 is a reasonable option here.
> + if(!virtual)
> + return -ENOMEM;
> +
That said, these checks are then totally unnecessary for small sizes.
Especially here, as the allocation lives only for one iteration over the
main loop using g_timeout_add(0, ...).
When the requested size with g_malloc is always greater than zero, it
must have succeeded if we can check it... So these patches should
instead clean up unnecessary NULL checks when using g_malloc on small
memory allocations.
> virtual->service = service;
> virtual->vfile = config->virtual_file;
>
Patrik
------------------------------
Subject: Digest Footer
_______________________________________________
connman mailing list
[email protected]
https://lists.01.org/mailman/listinfo/connman
------------------------------
End of connman Digest, Vol 3, Issue 19
**************************************
