9:10 p.m.
Hi,
Issue observed in call forwarding feature information list parsing.
Due to this, isimodem reports Erase failure to the core even when the
Erase request is successfully handled by the network.
Considering the current logic, this is the fix that can be provided without
changes to other parts of the code. One of the other option is to initialize the status in
the calling function
and check its validity before the usage.
Regards,
Jeevaka
9:10 p.m.
New subject: [PATCH] isi: Fix issue in call forwarding erase response
---
drivers/isimodem/call-forwarding.c | 20 ++++++++++++++------
1 files changed, 14 insertions(+), 6 deletions(-)
diff --git a/drivers/isimodem/call-forwarding.c b/drivers/isimodem/call-forwarding.c
index 1448451..c364bdf 100644
--- a/drivers/isimodem/call-forwarding.c
+++ b/drivers/isimodem/call-forwarding.c
@@ -98,12 +98,20 @@ static gboolean decode_gsm_forwarding_info(const void *restrict data,
uint8_t _ton;
char *_number = NULL;
- if (!g_isi_sb_iter_get_byte(&iter, &_status, 3)
- || !g_isi_sb_iter_get_byte(&iter, &_ton, 4)
- || !g_isi_sb_iter_get_byte(&iter, &_norply, 5)
- || !g_isi_sb_iter_get_byte(&iter, &_numlen, 7)
- || !g_isi_sb_iter_get_alpha_tag(&iter, &_number,
- _numlen * 2, 10))
+ if (status && !g_isi_sb_iter_get_byte(&iter, &_status, 3))
+ return FALSE;
+
+ if (ton && !g_isi_sb_iter_get_byte(&iter, &_ton, 4))
+ return FALSE;
+
+ if (norply && !g_isi_sb_iter_get_byte(&iter, &_norply, 5))
+ return FALSE;
+
+ if (number && !g_isi_sb_iter_get_byte(&iter, &_numlen, 7))
+ return FALSE;
+
+ if (number && !g_isi_sb_iter_get_alpha_tag(&iter,
+ &_number, _numlen * 2, 10))
return FALSE;
if (status)
--
1.7.0.4
12:12 a.m.
New subject: [PATCH] isi: Fix issue in call forwarding erase response
Hi Jeevaka,
2010/11/24 Jeevaka Badrappan <jeevaka.badrappan(a)elektrobit.com>:
---
drivers/isimodem/call-forwarding.c | 20 ++++++++++++++------
1 files changed, 14 insertions(+), 6 deletions(-)
diff --git a/drivers/isimodem/call-forwarding.c b/drivers/isimodem/call-forwarding.c
index 1448451..c364bdf 100644
--- a/drivers/isimodem/call-forwarding.c
+++ b/drivers/isimodem/call-forwarding.c
@@ -98,12 +98,20 @@ static gboolean decode_gsm_forwarding_info(const void *restrict
data,
uint8_t _ton;
char *_number = NULL;
- if (!g_isi_sb_iter_get_byte(&iter, &_status, 3)
- || !g_isi_sb_iter_get_byte(&iter, &_ton, 4)
- || !g_isi_sb_iter_get_byte(&iter, &_norply, 5)
- || !g_isi_sb_iter_get_byte(&iter, &_numlen, 7)
- || !g_isi_sb_iter_get_alpha_tag(&iter, &_number,
- _numlen * 2, 10))
+ if (status && !g_isi_sb_iter_get_byte(&iter,
&_status, 3))
+ return FALSE;
+
+ if (ton && !g_isi_sb_iter_get_byte(&iter, &_ton,
4))
+ return FALSE;
+
+ if (norply && !g_isi_sb_iter_get_byte(&iter,
&_norply, 5))
+ return FALSE;
+
+ if (number && !g_isi_sb_iter_get_byte(&iter,
&_numlen, 7))
+ return FALSE;
+
+ if (number && !g_isi_sb_iter_get_alpha_tag(&iter,
+ &_number, _numlen * 2, 10))
return FALSE;
I think the problem is that we try to decode the alpha tag even when
it is not present, i.e., when _numlen is zero.
In fact, this has already been fixed in my current working tree in a
bit different way. Mostly due to the fact that I added a new
g_isi_sb_iter_get_struct() method, which allows a little easier access
to sub-blocks like the one above.
If you're curious, here's the tree with the fix:
http://meego.gitorious.org/~akiniemi/meego-cellular/ofono-aki/commits/gisi
Thanks for the report in any case! ;)
Cheers,
Aki
1:55 a.m.
New subject: [PATCH] isi: Fix issue in call forwarding erase response
Hi Aki,
I think the problem is that we try to decode the alpha tag
even when it is not present, i.e., when _numlen is zero.
In fact, this has already been fixed in my current working
tree in a bit different way. Mostly due to the fact that I added a new
g_isi_sb_iter_get_struct() method, which allows a little
easier access to sub-blocks like the one above.
If you're curious, here's the tree with the fix:
http://meego.gitorious.org/~akiniemi/meego-cellular/ofono-aki/
commits/gisi
Went through your fix and it looks good.
Regards,
Jeevaka
4316
days inactive
4316
days old
3 comments
3 participants
participants (3)
-
Aki Niemi -
Jeevaka Badrappan -
Jeevaka.Badrappan＠elektrobit.com