12:21 p.m.
This set of patches have been sent this summer but not upstreamed.
As few bugs have been opened which could be fixed by those patches, I resend them
for review.
- For Speedup dongles, the logic in udevng is to assign the modem port on the
latest interface. This is not true for SpeedUp SU-7300U (BUG #23168) and
SpeedUp 9800 (BUG #23167). So, I added 2 specific rules for both dongles.
- For dongle ZTE MF190, 6 interfaces are present and we shall use the interface
ttyUSB4 for modem channel. The logic in udevng is to assign the modem channel
on the latest interface. To not break the support of ZTE modems with 5
interfaces (which are using the 4th one for PPP, like K3570), I can't extend
the current logic to take into account the ttyUSB4. So, I added a specific rule
for ZTE MF190.
I added also a specific rule for the dongle ZTE MF668 as the modem channel is
not on the latest interface (BUG #23683)
- I modified also udevng.c in order to avoid aux/modem channel assignment done
according OFONO_LABEL to be overridden by the default assignment.
Philippe Nunes (5):
udev: Add rules to support ZTE MF668 dongle
udev: Add rules to support ZTE MF190 dongle
udevng.c: tty assignment according OFONO_LABEL should take precedence
udev: Add rules to support Speedup 7300 dongle
udev: Add rules to support SpeedUp 9800 dongle
plugins/ofono.rules | 20 ++++++++++++++++
plugins/udevng.c | 61 +++++++++++++++++++++++++++++++-------------------
2 files changed, 58 insertions(+), 23 deletions(-)
12:20 p.m.
Hi Philippe,
This set of patches have been sent this summer but not upstreamed.
As few bugs have been opened which could be fixed by those patches, I resend them
for review.
- For Speedup dongles, the logic in udevng is to assign the modem port on the
latest interface. This is not true for SpeedUp SU-7300U (BUG #23168) and
SpeedUp 9800 (BUG #23167). So, I added 2 specific rules for both dongles.
- For dongle ZTE MF190, 6 interfaces are present and we shall use the interface
ttyUSB4 for modem channel. The logic in udevng is to assign the modem channel
on the latest interface. To not break the support of ZTE modems with 5
interfaces (which are using the 4th one for PPP, like K3570), I can't extend
the current logic to take into account the ttyUSB4. So, I added a specific rule
for ZTE MF190.
I added also a specific rule for the dongle ZTE MF668 as the modem channel is
not on the latest interface (BUG #23683)
why is all this reasoning here and not in the commit message of the
patches? I expect that every patch has a proper commit message that
explains its goal. Just a non-meaningful subject line is not enough.
Regards
Marcel
2:51 p.m.
Hi Marcel,
On 11/29/2011 01:20 PM, Marcel Holtmann wrote:
Hi Philippe,
> This set of patches have been sent this summer but not upstreamed.
> As few bugs have been opened which could be fixed by those patches, I resend them
> for review.
>
> - For Speedup dongles, the logic in udevng is to assign the modem port on the
> latest interface. This is not true for SpeedUp SU-7300U (BUG #23168) and
> SpeedUp 9800 (BUG #23167). So, I added 2 specific rules for both dongles.
>
> - For dongle ZTE MF190, 6 interfaces are present and we shall use the interface
> ttyUSB4 for modem channel. The logic in udevng is to assign the modem channel
> on the latest interface. To not break the support of ZTE modems with 5
> interfaces (which are using the 4th one for PPP, like K3570), I can't extend
> the current logic to take into account the ttyUSB4. So, I added a specific rule
> for ZTE MF190.
> I added also a specific rule for the dongle ZTE MF668 as the modem channel is
> not on the latest interface (BUG #23683)
why is all this reasoning here and not in the commit message of the
patches? I expect that every patch has a proper commit message that
explains its goal. Just a non-meaningful subject line is not enough.
Yes, you're right, I should move this explanation in the commit
messages. Before, sending a new set of patches accordingly, I would like
to raise an issue regarding the patches for ofono.rules:
Your commit "build: do not install ofono.rules anymore" regarding
Makefile.am makes none effective my proposed changes in ofono rules.
Could you explain why the file ofono.rules is not installed anymore
whereas we are still supposed to consider some specific properties
exposed by this file?
Thanks,
Regards,
Philippe.
12:21 p.m.
New subject: [PATCH 1/5] udev: Add rules to support ZTE MF668 dongle
---
plugins/ofono.rules | 9 +++++++++
1 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/plugins/ofono.rules b/plugins/ofono.rules
index 30a1b7e..c7fac57 100644
--- a/plugins/ofono.rules
+++ b/plugins/ofono.rules
@@ -15,6 +15,15 @@ KERNEL=="shrm0", ENV{OFONO_DRIVER}="u8500"
LABEL="ofono_isi_end"
+SUBSYSTEM!="tty", GOTO="ofono_tty_end"
+KERNEL!="ttyUSB[0-9]*", GOTO="ofono_tty_end"
+
+# ZTE
+ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="02", ENV{OFONO_LABEL}="modem"
+ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="aux"
+
+LABEL="ofono_tty_end"
+
SUBSYSTEM!="usb", GOTO="ofono_end"
ENV{DEVTYPE}!="usb_device", GOTO="ofono_end"
--
1.7.1
5:49 a.m.
New subject: [PATCH 1/5] udev: Add rules to support ZTE MF668 dongle
Hi Philippe,
plugins/ofono.rules | 9 +++++++++
1 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/plugins/ofono.rules b/plugins/ofono.rules
index 30a1b7e..c7fac57 100644
--- a/plugins/ofono.rules
+++ b/plugins/ofono.rules
@@ -15,6 +15,15 @@ KERNEL=="shrm0", ENV{OFONO_DRIVER}="u8500"
LABEL="ofono_isi_end"
+SUBSYSTEM!="tty", GOTO="ofono_tty_end"
+KERNEL!="ttyUSB[0-9]*", GOTO="ofono_tty_end"
+
+# ZTE
+ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="02", ENV{OFONO_LABEL}="modem"
+ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="aux"
+
+LABEL="ofono_tty_end"
+
what are we trying to fix here. Please include /proc/bus/usb/devices or
usb-devices and some oFono debug lines that show that without this we
are doing a wrong assignment.
Regards
Marcel
5:34 p.m.
New subject: [PATCH 1/5] udev: Add rules to support ZTE MF668 dongle
Hi Marcel,
On 12/19/2011 06:49 AM, Marcel Holtmann wrote:
Hi Philippe,
> plugins/ofono.rules | 9 +++++++++
> 1 files changed, 9 insertions(+), 0 deletions(-)
>
> diff --git a/plugins/ofono.rules b/plugins/ofono.rules
> index 30a1b7e..c7fac57 100644
> --- a/plugins/ofono.rules
> +++ b/plugins/ofono.rules
> @@ -15,6 +15,15 @@ KERNEL=="shrm0", ENV{OFONO_DRIVER}="u8500"
>
> LABEL="ofono_isi_end"
>
> +SUBSYSTEM!="tty", GOTO="ofono_tty_end"
> +KERNEL!="ttyUSB[0-9]*", GOTO="ofono_tty_end"
> +
> +# ZTE
> +ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="02", ENV{OFONO_LABEL}="modem"
> +ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="aux"
> +
> +LABEL="ofono_tty_end"
> +
what are we trying to fix here. Please include /proc/bus/usb/devices or
usb-devices and some oFono debug lines that show that without this we
are doing a wrong assignment.
We are trying to fix the BUG#23683. In the attached oFono AT log, you
can notice that the modem port is not responsive.
No answer to the command:
Modem: > ATZ E0 +CMEE=1\r
usb-devices gives:
T: Bus=02 Lev=02 Prnt=02 Port=00 Cnt=01 Dev#= 11 Spd=480 MxCh= 0
D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1
P: Vendor=19d2 ProdID=0017 Rev=00.00
S: Manufacturer=ZTE,Incorporated
S: Product=ZTE HSPA Technologies MSM
S: SerialNumber=P680A1ZTED010000
C: #Ifs= 5 Cfg#= 1 Atr=e0 MxPwr=500mA
I: If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
I: If#= 1 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
I: If#= 2 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
I: If#= 3 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
I: If#= 4 Alt= 0 #EPs= 2 Cls=08(stor.) Sub=06 Prot=50 Driver=usb-storage
In practice, the modem port should be assigned to /dev/ttyUSB2 and not
to the latest interface as done by udevng.
Regards,
Philippe.
Regards
Marcel
_______________________________________________
ofono mailing list
ofono(a)ofono.org
http://lists.ofono.org/listinfo/ofono
12:21 p.m.
New subject: [PATCH 2/5] udev: Add rules to support ZTE MF190 dongle
---
plugins/ofono.rules | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/plugins/ofono.rules b/plugins/ofono.rules
index c7fac57..079db07 100644
--- a/plugins/ofono.rules
+++ b/plugins/ofono.rules
@@ -22,6 +22,9 @@ KERNEL!="ttyUSB[0-9]*", GOTO="ofono_tty_end"
ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="02", ENV{OFONO_LABEL}="modem"
ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="aux"
+ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="04", ENV{OFONO_LABEL}="modem"
+ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="aux"
+
LABEL="ofono_tty_end"
SUBSYSTEM!="usb", GOTO="ofono_end"
--
1.7.1
5:50 a.m.
New subject: [PATCH 2/5] udev: Add rules to support ZTE MF190 dongle
Hi Philippe,
plugins/ofono.rules | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/plugins/ofono.rules b/plugins/ofono.rules
index c7fac57..079db07 100644
--- a/plugins/ofono.rules
+++ b/plugins/ofono.rules
@@ -22,6 +22,9 @@ KERNEL!="ttyUSB[0-9]*", GOTO="ofono_tty_end"
ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="02", ENV{OFONO_LABEL}="modem"
ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="aux"
+ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="04", ENV{OFONO_LABEL}="modem"
+ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="aux"
+
LABEL="ofono_tty_end"
SUBSYSTEM!="usb", GOTO="ofono_end"
same here. The modem AUX port is always the second TTY and the modem TTY
is always the last one. Where is this not true?
Regards
Marcel
5:45 p.m.
New subject: [PATCH 2/5] udev: Add rules to support ZTE MF190 dongle
On 12/19/2011 06:50 AM, Marcel Holtmann wrote:
Hi Philippe,
> plugins/ofono.rules | 3 +++
> 1 files changed, 3 insertions(+), 0 deletions(-)
>
> diff --git a/plugins/ofono.rules b/plugins/ofono.rules
> index c7fac57..079db07 100644
> --- a/plugins/ofono.rules
> +++ b/plugins/ofono.rules
> @@ -22,6 +22,9 @@ KERNEL!="ttyUSB[0-9]*", GOTO="ofono_tty_end"
> ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="02", ENV{OFONO_LABEL}="modem"
> ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="aux"
>
> +ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="04", ENV{OFONO_LABEL}="modem"
> +ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="aux"
> +
> LABEL="ofono_tty_end"
>
> SUBSYSTEM!="usb", GOTO="ofono_end"
same here. The modem AUX port is always the second TTY and the modem TTY
is always the last one. Where is this not true?
Actually, with this dongle ZTE MF190, we have 6 TTY (usb0 -> usb5) and
usb-devices gives:
T: Bus=02 Lev=02 Prnt=02 Port=00 Cnt=01 Dev#= 32 Spd=480 MxCh= 0
D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1
P: Vendor=19d2 ProdID=0124 Rev=00.00
S: Manufacturer=ZTE,Incorporated
S: Product=ZTE WCDMA Technologies MSM
S: SerialNumber=P671A2TMED010000
C: #Ifs= 7 Cfg#= 1 Atr=e0 MxPwr=500mA
I: If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
I: If#= 1 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
I: If#= 2 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
I: If#= 3 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
I: If#= 4 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
I: If#= 5 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
I: If#= 6 Alt= 0 #EPs= 2 Cls=08(stor.) Sub=06 Prot=50 Driver=usb-storage
Now, ttyUSB5 is not responsive. So, for this particular dongle, the
modem won't be on the latest interface. In practice, we shall use the
interface ttyUSB4 for PPP.
Windows Device Manager gives the following results:
Vendor Id: 19d2 / Product Id: 0124 (same as after usb_modeswitch)
ZTE Diagnostic port - 00
ZTE AT port - 01
ZTE ATExt2 port - 02
ZTE USB SmartCard Reader - 03
ZTE Proprietary USB Modem - 04
ZTE Mobile broadband Network manager - 05
So, it confirms that the modem port com is on ttyUSB4.
To not break the support of ZTE modems with 5 interfaces (which are
using the 4th one for PPP, like K3570), I can't extend the current logic
in udevng to take into account the ttyUSB4. So, I added a specific rule
for ZTE MF190.
Regards,
Philippe.
Regards
Marcel
_______________________________________________
ofono mailing list
ofono(a)ofono.org
http://lists.ofono.org/listinfo/ofono
12:21 p.m.
New subject: [PATCH 3/5] udevng.c: tty assignment according OFONO_LABEL should take precedence
---
plugins/udevng.c | 61 +++++++++++++++++++++++++++++++++--------------------
1 files changed, 38 insertions(+), 23 deletions(-)
diff --git a/plugins/udevng.c b/plugins/udevng.c
index be87320..6a3c970 100644
--- a/plugins/udevng.c
+++ b/plugins/udevng.c
@@ -235,6 +235,7 @@ static gboolean setup_sierra(struct modem_info *modem)
static gboolean setup_huawei(struct modem_info *modem)
{
const char *mdm = NULL, *pcui = NULL, *diag = NULL;
+ const char *default_pcui = NULL;
GSList *list;
DBG("%s", modem->syspath);
@@ -264,19 +265,24 @@ static gboolean setup_huawei(struct modem_info *modem)
if (mdm != NULL && pcui != NULL)
break;
} else if (g_strcmp0(info->interface, "255/255/255") == 0) {
- if (g_strcmp0(info->number, "00") == 0)
+ if (mdm == NULL && g_strcmp0(info->number, "00") == 0)
mdm = info->devnode;
- else if (g_strcmp0(info->number, "01") == 0)
- pcui = info->devnode;
- else if (g_strcmp0(info->number, "02") == 0)
- pcui = info->devnode;
- else if (g_strcmp0(info->number, "03") == 0)
- pcui = info->devnode;
- else if (g_strcmp0(info->number, "04") == 0)
- pcui = info->devnode;
+ else if (pcui == NULL) {
+ if (g_strcmp0(info->number, "01") == 0)
+ default_pcui = info->devnode;
+ else if (g_strcmp0(info->number, "02") == 0)
+ default_pcui = info->devnode;
+ else if (g_strcmp0(info->number, "03") == 0)
+ default_pcui = info->devnode;
+ else if (g_strcmp0(info->number, "04") == 0)
+ default_pcui = info->devnode;
+ }
}
}
+ if (pcui == NULL && default_pcui != NULL)
+ pcui = default_pcui;
+
if (mdm == NULL || pcui == NULL)
return FALSE;
@@ -291,7 +297,7 @@ static gboolean setup_huawei(struct modem_info *modem)
static gboolean setup_speedup(struct modem_info *modem)
{
- const char *aux = NULL, *mdm = NULL;
+ const char *aux = NULL, *mdm = NULL, *default_mdm = NULL;
GSList *list;
DBG("%s", modem->syspath);
@@ -311,15 +317,20 @@ static gboolean setup_speedup(struct modem_info *modem)
if (aux != NULL)
break;
} else if (g_strcmp0(info->interface, "255/255/255") == 0) {
- if (g_strcmp0(info->number, "01") == 0)
+ if (aux == NULL && g_strcmp0(info->number, "01") == 0)
aux = info->devnode;
- else if (g_strcmp0(info->number, "02") == 0)
- mdm = info->devnode;
- else if (g_strcmp0(info->number, "03") == 0)
- mdm = info->devnode;
+ else if (mdm == NULL) {
+ if (g_strcmp0(info->number, "02") == 0)
+ default_mdm = info->devnode;
+ else if (g_strcmp0(info->number, "03") == 0)
+ default_mdm = info->devnode;
+ }
}
}
+ if (mdm == NULL && default_mdm != NULL)
+ mdm = default_mdm;
+
if (aux == NULL || mdm == NULL)
return FALSE;
@@ -385,9 +396,10 @@ static gboolean setup_alcatel(struct modem_info *modem)
if (aux != NULL)
break;
} else if (g_strcmp0(info->interface, "255/255/255") == 0) {
- if (g_strcmp0(info->number, "03") == 0)
+ if (aux == NULL && g_strcmp0(info->number, "03") == 0)
aux = info->devnode;
- else if (g_strcmp0(info->number, "05") == 0)
+ else if (mdm == NULL &&
+ g_strcmp0(info->number, "05") == 0)
mdm = info->devnode;
}
}
@@ -425,9 +437,10 @@ static gboolean setup_novatel(struct modem_info *modem)
if (aux != NULL)
break;
} else if (g_strcmp0(info->interface, "255/255/255") == 0) {
- if (g_strcmp0(info->number, "00") == 0)
+ if (aux == NULL && g_strcmp0(info->number, "00") == 0)
aux = info->devnode;
- else if (g_strcmp0(info->number, "01") == 0)
+ else if (mdm == NULL &&
+ g_strcmp0(info->number, "01") == 0)
mdm = info->devnode;
}
}
@@ -465,9 +478,10 @@ static gboolean setup_nokia(struct modem_info *modem)
if (aux != NULL)
break;
} else if (g_strcmp0(info->interface, "10/0/0") == 0) {
- if (g_strcmp0(info->number, "02") == 0)
+ if (mdm == NULL && g_strcmp0(info->number, "02") == 0)
mdm = info->devnode;
- else if (g_strcmp0(info->number, "04") == 0)
+ else if (aux == NULL &&
+ g_strcmp0(info->number, "04") == 0)
aux = info->devnode;
}
}
@@ -505,13 +519,14 @@ static gboolean setup_telit(struct modem_info *modem)
if (aux != NULL)
break;
} else if (g_strcmp0(info->interface, "255/255/255") == 0) {
- if (g_strcmp0(info->number, "00") == 0)
+ if (mdm == NULL && g_strcmp0(info->number, "00") == 0)
mdm = info->devnode;
else if (g_strcmp0(info->number, "01") == 0)
diag = info->devnode;
else if (g_strcmp0(info->number, "02") == 0)
gps = info->devnode;
- else if (g_strcmp0(info->number, "03") == 0)
+ else if (aux == NULL
+ && g_strcmp0(info->number, "03") == 0)
aux = info->devnode;
}
}
--
1.7.1
5:48 a.m.
New subject: [PATCH 3/5] udevng.c: tty assignment according OFONO_LABEL should take precedence
Hi Philippe,
plugins/udevng.c | 61
+++++++++++++++++++++++++++++++++--------------------
1 files changed, 38 insertions(+), 23 deletions(-)
diff --git a/plugins/udevng.c b/plugins/udevng.c
index be87320..6a3c970 100644
--- a/plugins/udevng.c
+++ b/plugins/udevng.c
@@ -235,6 +235,7 @@ static gboolean setup_sierra(struct modem_info *modem)
static gboolean setup_huawei(struct modem_info *modem)
{
const char *mdm = NULL, *pcui = NULL, *diag = NULL;
+ const char *default_pcui = NULL;
GSList *list;
DBG("%s", modem->syspath);
@@ -264,19 +265,24 @@ static gboolean setup_huawei(struct modem_info *modem)
if (mdm != NULL && pcui != NULL)
break;
} else if (g_strcmp0(info->interface, "255/255/255") == 0) {
- if (g_strcmp0(info->number, "00") == 0)
+ if (mdm == NULL && g_strcmp0(info->number, "00") == 0)
mdm = info->devnode;
- else if (g_strcmp0(info->number, "01") == 0)
- pcui = info->devnode;
- else if (g_strcmp0(info->number, "02") == 0)
- pcui = info->devnode;
- else if (g_strcmp0(info->number, "03") == 0)
- pcui = info->devnode;
- else if (g_strcmp0(info->number, "04") == 0)
- pcui = info->devnode;
+ else if (pcui == NULL) {
+ if (g_strcmp0(info->number, "01") == 0)
+ default_pcui = info->devnode;
+ else if (g_strcmp0(info->number, "02") == 0)
+ default_pcui = info->devnode;
+ else if (g_strcmp0(info->number, "03") == 0)
+ default_pcui = info->devnode;
+ else if (g_strcmp0(info->number, "04") == 0)
+ default_pcui = info->devnode;
+ }
}
}
+ if (pcui == NULL && default_pcui != NULL)
+ pcui = default_pcui;
+
what is this default_pcui doing here. I am failing to understand its
usage.
if (mdm == NULL || pcui == NULL)
return FALSE;
@@ -291,7 +297,7 @@ static gboolean setup_huawei(struct modem_info *modem)
static gboolean setup_speedup(struct modem_info *modem)
{
- const char *aux = NULL, *mdm = NULL;
+ const char *aux = NULL, *mdm = NULL, *default_mdm = NULL;
GSList *list;
DBG("%s", modem->syspath);
@@ -311,15 +317,20 @@ static gboolean setup_speedup(struct modem_info *modem)
if (aux != NULL)
break;
} else if (g_strcmp0(info->interface, "255/255/255") == 0) {
- if (g_strcmp0(info->number, "01") == 0)
+ if (aux == NULL && g_strcmp0(info->number, "01") == 0)
aux = info->devnode;
- else if (g_strcmp0(info->number, "02") == 0)
- mdm = info->devnode;
- else if (g_strcmp0(info->number, "03") == 0)
- mdm = info->devnode;
+ else if (mdm == NULL) {
+ if (g_strcmp0(info->number, "02") == 0)
+ default_mdm = info->devnode;
+ else if (g_strcmp0(info->number, "03") == 0)
+ default_mdm = info->devnode;
+ }
}
}
+ if (mdm == NULL && default_mdm != NULL)
+ mdm = default_mdm;
+
Same here. This makes the code pretty much unreadable.
if (aux == NULL || mdm == NULL)
return FALSE;
@@ -385,9 +396,10 @@ static gboolean setup_alcatel(struct modem_info *modem)
if (aux != NULL)
break;
} else if (g_strcmp0(info->interface, "255/255/255") == 0) {
- if (g_strcmp0(info->number, "03") == 0)
+ if (aux == NULL && g_strcmp0(info->number, "03") == 0)
aux = info->devnode;
- else if (g_strcmp0(info->number, "05") == 0)
+ else if (mdm == NULL &&
+ g_strcmp0(info->number, "05") == 0)
mdm = info->devnode;
}
}
@@ -425,9 +437,10 @@ static gboolean setup_novatel(struct modem_info *modem)
if (aux != NULL)
break;
} else if (g_strcmp0(info->interface, "255/255/255") == 0) {
- if (g_strcmp0(info->number, "00") == 0)
+ if (aux == NULL && g_strcmp0(info->number, "00") == 0)
aux = info->devnode;
- else if (g_strcmp0(info->number, "01") == 0)
+ else if (mdm == NULL &&
+ g_strcmp0(info->number, "01") == 0)
mdm = info->devnode;
}
}
@@ -465,9 +478,10 @@ static gboolean setup_nokia(struct modem_info *modem)
if (aux != NULL)
break;
} else if (g_strcmp0(info->interface, "10/0/0") == 0) {
- if (g_strcmp0(info->number, "02") == 0)
+ if (mdm == NULL && g_strcmp0(info->number, "02") == 0)
mdm = info->devnode;
- else if (g_strcmp0(info->number, "04") == 0)
+ else if (aux == NULL &&
+ g_strcmp0(info->number, "04") == 0)
aux = info->devnode;
}
}
@@ -505,13 +519,14 @@ static gboolean setup_telit(struct modem_info *modem)
if (aux != NULL)
break;
} else if (g_strcmp0(info->interface, "255/255/255") == 0) {
- if (g_strcmp0(info->number, "00") == 0)
+ if (mdm == NULL && g_strcmp0(info->number, "00") == 0)
mdm = info->devnode;
else if (g_strcmp0(info->number, "01") == 0)
diag = info->devnode;
else if (g_strcmp0(info->number, "02") == 0)
gps = info->devnode;
- else if (g_strcmp0(info->number, "03") == 0)
+ else if (aux == NULL
+ && g_strcmp0(info->number, "03") == 0)
aux = info->devnode;
}
}
This becomes unreadable. What are we trying to achieve here.
Regards
Marcel
4:32 p.m.
New subject: [PATCH 3/5] udevng.c: tty assignment according OFONO_LABEL should take precedence
Hi Marcel,
On 12/19/2011 06:48 AM, Marcel Holtmann wrote:
Hi Philippe,
> plugins/udevng.c | 61 +++++++++++++++++++++++++++++++++--------------------
> 1 files changed, 38 insertions(+), 23 deletions(-)
>
> diff --git a/plugins/udevng.c b/plugins/udevng.c
> index be87320..6a3c970 100644
> --- a/plugins/udevng.c
> +++ b/plugins/udevng.c
> @@ -235,6 +235,7 @@ static gboolean setup_sierra(struct modem_info *modem)
> static gboolean setup_huawei(struct modem_info *modem)
> {
> const char *mdm = NULL, *pcui = NULL, *diag = NULL;
> + const char *default_pcui = NULL;
> GSList *list;
>
> DBG("%s", modem->syspath);
> @@ -264,19 +265,24 @@ static gboolean setup_huawei(struct modem_info *modem)
> if (mdm != NULL&& pcui != NULL)
> break;
> } else if (g_strcmp0(info->interface, "255/255/255") == 0) {
> - if (g_strcmp0(info->number, "00") == 0)
> + if (mdm == NULL&& g_strcmp0(info->number, "00") == 0)
> mdm = info->devnode;
> - else if (g_strcmp0(info->number, "01") == 0)
> - pcui = info->devnode;
> - else if (g_strcmp0(info->number, "02") == 0)
> - pcui = info->devnode;
> - else if (g_strcmp0(info->number, "03") == 0)
> - pcui = info->devnode;
> - else if (g_strcmp0(info->number, "04") == 0)
> - pcui = info->devnode;
> + else if (pcui == NULL) {
> + if (g_strcmp0(info->number, "01") == 0)
> + default_pcui = info->devnode;
> + else if (g_strcmp0(info->number, "02") == 0)
> + default_pcui = info->devnode;
> + else if (g_strcmp0(info->number, "03") == 0)
> + default_pcui = info->devnode;
> + else if (g_strcmp0(info->number, "04") == 0)
> + default_pcui = info->devnode;
> + }
> }
> }
>
> + if (pcui == NULL&& default_pcui != NULL)
> + pcui = default_pcui;
> +
what is this default_pcui doing here. I am failing to understand its
usage.
We need to prevent the assignment done through the label, to be
overridden by the 'default' assignment (the one done by the implicit rule).
That's why I introduced new conditions:
if (mdm == NULL && g_strcmp0(info->number, "00") == 0)
and
else if (pcui == NULL) {
In practice, if the pcui is already assigned through the label, the
'default' assignment' algorithm is not considered.
On the contrary, the 'default' assignment is done using a dedicated
variable 'default_pcui'.
If we were still using the variable 'pcui', we couldn't update anymore
this variable through the 'default' assignment due to the new condition
'if (pcui == NULL)'.
> if (mdm == NULL || pcui == NULL)
> return FALSE;
>
> @@ -291,7 +297,7 @@ static gboolean setup_huawei(struct modem_info *modem)
>
> static gboolean setup_speedup(struct modem_info *modem)
> {
> - const char *aux = NULL, *mdm = NULL;
> + const char *aux = NULL, *mdm = NULL, *default_mdm = NULL;
> GSList *list;
>
> DBG("%s", modem->syspath);
> @@ -311,15 +317,20 @@ static gboolean setup_speedup(struct modem_info *modem)
> if (aux != NULL)
> break;
> } else if (g_strcmp0(info->interface, "255/255/255") == 0) {
> - if (g_strcmp0(info->number, "01") == 0)
> + if (aux == NULL&& g_strcmp0(info->number, "01") == 0)
> aux = info->devnode;
> - else if (g_strcmp0(info->number, "02") == 0)
> - mdm = info->devnode;
> - else if (g_strcmp0(info->number, "03") == 0)
> - mdm = info->devnode;
> + else if (mdm == NULL) {
> + if (g_strcmp0(info->number, "02") == 0)
> + default_mdm = info->devnode;
> + else if (g_strcmp0(info->number, "03") == 0)
> + default_mdm = info->devnode;
> + }
> }
> }
>
> + if (mdm == NULL&& default_mdm != NULL)
> + mdm = default_mdm;
> +
Same here. This makes the code pretty much unreadable.
This logic becomes however necessary as I introduced dedicated rules for
Speedup and Huawei based on the label.
Without this patch, the 'label' assignment doesn't work necessarily (in
fact, it depends of the order of the interfaces).
> if (aux == NULL || mdm == NULL)
> return FALSE;
>
> @@ -385,9 +396,10 @@ static gboolean setup_alcatel(struct modem_info *modem)
> if (aux != NULL)
> break;
> } else if (g_strcmp0(info->interface, "255/255/255") == 0) {
> - if (g_strcmp0(info->number, "03") == 0)
> + if (aux == NULL&& g_strcmp0(info->number, "03") == 0)
> aux = info->devnode;
> - else if (g_strcmp0(info->number, "05") == 0)
> + else if (mdm == NULL&&
> + g_strcmp0(info->number, "05") == 0)
> mdm = info->devnode;
> }
> }
> @@ -425,9 +437,10 @@ static gboolean setup_novatel(struct modem_info *modem)
> if (aux != NULL)
> break;
> } else if (g_strcmp0(info->interface, "255/255/255") == 0) {
> - if (g_strcmp0(info->number, "00") == 0)
> + if (aux == NULL&& g_strcmp0(info->number, "00") == 0)
> aux = info->devnode;
> - else if (g_strcmp0(info->number, "01") == 0)
> + else if (mdm == NULL&&
> + g_strcmp0(info->number, "01") == 0)
> mdm = info->devnode;
> }
> }
> @@ -465,9 +478,10 @@ static gboolean setup_nokia(struct modem_info *modem)
> if (aux != NULL)
> break;
> } else if (g_strcmp0(info->interface, "10/0/0") == 0) {
> - if (g_strcmp0(info->number, "02") == 0)
> + if (mdm == NULL&& g_strcmp0(info->number, "02") == 0)
> mdm = info->devnode;
> - else if (g_strcmp0(info->number, "04") == 0)
> + else if (aux == NULL&&
> + g_strcmp0(info->number, "04") == 0)
> aux = info->devnode;
> }
> }
> @@ -505,13 +519,14 @@ static gboolean setup_telit(struct modem_info *modem)
> if (aux != NULL)
> break;
> } else if (g_strcmp0(info->interface, "255/255/255") == 0) {
> - if (g_strcmp0(info->number, "00") == 0)
> + if (mdm == NULL&& g_strcmp0(info->number, "00") == 0)
> mdm = info->devnode;
> else if (g_strcmp0(info->number, "01") == 0)
> diag = info->devnode;
> else if (g_strcmp0(info->number, "02") == 0)
> gps = info->devnode;
> - else if (g_strcmp0(info->number, "03") == 0)
> + else if (aux == NULL
> + && g_strcmp0(info->number, "03") == 0)
> aux = info->devnode;
> }
> }
This becomes unreadable. What are we trying to achieve here.
For Alcatel, Novatel, Nokia, Telit, I don't have a bug stating that we
could have an overriding issue.
I just propagated the logic explained above for more safety.
Regards,
Philippe.
Regards
Marcel
_______________________________________________
ofono mailing list
ofono(a)ofono.org
http://lists.ofono.org/listinfo/ofono
12:21 p.m.
New subject: [PATCH 4/5] udev: Add rules to support Speedup 7300 dongle
---
plugins/ofono.rules | 4 ++++
1 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/plugins/ofono.rules b/plugins/ofono.rules
index 079db07..7f284d4 100644
--- a/plugins/ofono.rules
+++ b/plugins/ofono.rules
@@ -25,6 +25,10 @@ ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="0
ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="04", ENV{OFONO_LABEL}="modem"
ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="aux"
+# SpeedUp 7300
+ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9e00",
ENV{ID_USB_INTERFACE_NUM}=="00", ENV{OFONO_LABEL}="modem"
+ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9e00",
ENV{ID_USB_INTERFACE_NUM}=="03", ENV{OFONO_LABEL}="aux"
+
LABEL="ofono_tty_end"
SUBSYSTEM!="usb", GOTO="ofono_end"
--
1.7.1
5:51 a.m.
New subject: [PATCH 4/5] udev: Add rules to support Speedup 7300 dongle
Hi Philippe,
plugins/ofono.rules | 4 ++++
1 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/plugins/ofono.rules b/plugins/ofono.rules
index 079db07..7f284d4 100644
--- a/plugins/ofono.rules
+++ b/plugins/ofono.rules
@@ -25,6 +25,10 @@ ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0017",
ENV{ID_USB_INTERFACE_NUM}=="0
ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="04", ENV{OFONO_LABEL}="modem"
ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="aux"
+# SpeedUp 7300
+ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9e00",
ENV{ID_USB_INTERFACE_NUM}=="00", ENV{OFONO_LABEL}="modem"
+ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9e00",
ENV{ID_USB_INTERFACE_NUM}=="03", ENV{OFONO_LABEL}="aux"
+
LABEL="ofono_tty_end"
SUBSYSTEM!="usb", GOTO="ofono_end"
since the SpeedUp modem seem to be fully broken, we better just only do
label assignment and create a plugins/ofono-speedup.rules. I do not
wanna intermix this just because of total broken hardware.
Regards
Marcel
12:21 p.m.
New subject: [PATCH 5/5] udev: Add rules to support SpeedUp 9800 dongle
---
plugins/ofono.rules | 4 ++++
1 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/plugins/ofono.rules b/plugins/ofono.rules
index 7f284d4..9e7a651 100644
--- a/plugins/ofono.rules
+++ b/plugins/ofono.rules
@@ -29,6 +29,10 @@ ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="0
ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9e00",
ENV{ID_USB_INTERFACE_NUM}=="00", ENV{OFONO_LABEL}="modem"
ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9e00",
ENV{ID_USB_INTERFACE_NUM}=="03", ENV{OFONO_LABEL}="aux"
+# SpeedUp 9800
+ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9800",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="modem"
+ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9800",
ENV{ID_USB_INTERFACE_NUM}=="02", ENV{OFONO_LABEL}="aux"
+
LABEL="ofono_tty_end"
SUBSYSTEM!="usb", GOTO="ofono_end"
--
1.7.1
5:52 a.m.
New subject: [PATCH 5/5] udev: Add rules to support SpeedUp 9800 dongle
Hi Philippe,
plugins/ofono.rules | 4 ++++
1 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/plugins/ofono.rules b/plugins/ofono.rules
index 7f284d4..9e7a651 100644
--- a/plugins/ofono.rules
+++ b/plugins/ofono.rules
@@ -29,6 +29,10 @@ ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0124",
ENV{ID_USB_INTERFACE_NUM}=="0
ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9e00",
ENV{ID_USB_INTERFACE_NUM}=="00", ENV{OFONO_LABEL}="modem"
ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9e00",
ENV{ID_USB_INTERFACE_NUM}=="03", ENV{OFONO_LABEL}="aux"
+# SpeedUp 9800
+ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9800",
ENV{ID_USB_INTERFACE_NUM}=="01", ENV{OFONO_LABEL}="modem"
+ATTRS{idVendor}=="1c9e", ATTRS{idProduct}=="9800",
ENV{ID_USB_INTERFACE_NUM}=="02", ENV{OFONO_LABEL}="aux"
+
LABEL="ofono_tty_end"
SUBSYSTEM!="usb", GOTO="ofono_end"
same applies here. See my other comment.
And btw. does anybody bothered to look into the inf files from the
Windows driver to figure out what the actual assignments are?
Regards
Marcel
3889
days inactive
3909
days old
15 comments
2 participants
participants (2)
-
Marcel Holtmann -
Philippe Nunes