Verizon V620 wireless card
John Nielsen
Content-Type: text/plain;
charset="us-ascii"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
I just want to report that I got this card working on my laptop running
6-STABLE. (This should be a reply to this message but I couldn't find it
locally):
http://lists.freebsd.org/pipermail/freebsd-mobile/2006-December/009404.html
Unlike the OP, the usb hub was recognized on my machine and gave me a ugen0
device associated with the CDMA modem. So all I had to do was get the modem
to attach to ucom or similar.
After a couple false starts I got a cue from this message:
http://lists.freebsd.org/pipermail/freebsd-hackers/2005-June/012512.html
The attached patch lets the card attach as a ubsa device. I'm also attaching
the ppp.conf file I hacked up (it's been a while since I've made any ppp
chat scripts so I'm sure it could be done more elegantly).
Here are the relevant bits of dmesg with the card inserted:
cbb0: <TI1520 PCI-CardBus Bridge> at device 11.0 on pci0
cardbus0: <CardBus bus> on cbb0
pccard0: <16-bit PCCard bus> on cbb0
cbb1: <TI1520 PCI-CardBus Bridge> at device 11.1 on pci0
cardbus1: <CardBus bus> on cbb1
pccard1: <16-bit PCCard bus> on cbb1
...
cardbus0: CIS pointer is 0!
cardbus0: Resource not specified in CIS: id=10, size=1000
ohci3: <NEC uPD 9210 USB controller> mem 0x88001000-0x88001fff irq 9 at
device 0.1 on cardbus0
ohci3: [GIANT-LOCKED]
usb4: OHCI version 1.0
usb4: <NEC uPD 9210 USB controller> on ohci3
usb4: USB revision 1.0
uhub4: NEC OHCI root hub, class 9/0, rev 1.00/1.00, addr 1
uhub4: 2 ports with 2 removable, self powered
ucom0: Novatel Wireless Inc. Novatel Wireless Merlin CDMA, rev 1.10/0.00,
addr 2
ucom0: ubsa_request: STALLED
ucom0: ubsa_request: STALLED
...
(lots more STALLED messages)
Despite the error messages it connects and works just fine, with performance
similar to what I see in Windows (which isn't great in this area).
evs: Did you take ohci out of your kernel by any chance?
I'm open to feedback on what the STALLED messages might mean and how to get
rid of them.
JN
--Boundary-00=_uUrxFywZMjriaBO
Content-Type: text/x-diff;
charset="us-ascii";
name="v620.patch"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
filename="v620.patch"
--- usbdevs.orig Mon Jan 22 12:57:30 2007
+++ usbdevs Sun Feb 4 00:11:33 2007
@@ -510,6 +510,7 @@
vendor MOBILITY 0x1342 Mobility
vendor LINKSYS4 0x13b1 Linksys
vendor SHARK 0x13d2 Shark
+vendor NOVATEL 0x1410 Novatel Wireless
vendor RADIOSHACK 0x1453 Radio Shack
vendor SILICOM 0x1485 Silicom
vendor RALINK 0x148f Ralink Technology
@@ -1340,6 +1341,9 @@
/* NovaTech Products */
product NOVATECH NV902 0x9020 NovaTech NV-902W
+/* Novatel Products */
+product NOVATEL CDMA_MODEM 0x1110 Novatel Wireless Merlin CDMA
+
/* Olympus products */
product OLYMPUS C1 0x0102 C-1 Digital Camera
product OLYMPUS C700 0x0105 C-700 Ultra Zoom
--- ubsa.c.orig Sun Feb 4 23:19:49 2007
+++ ubsa.c Sun Feb 4 15:20:36 2007
@@ -224,6 +224,8 @@
{ USB_VENDOR_GOHUBS, USB_PRODUCT_GOHUBS_GOCOM232 },
/* Peracom */
{ USB_VENDOR_PERACOM, USB_PRODUCT_PERACOM_SERIAL1 },
+ /* Novatel */
+ { USB_VENDOR_NOVATEL, USB_PRODUCT_NOVATEL_CDMA_MODEM },
/* Vodafone */
{ USB_VENDOR_VODAFONE, USB_PRODUCT_VODAFONE_MC3G },
{ 0, 0 }
--- usb_quirks.c.orig Mon Jan 22 12:57:30 2007
+++ usb_quirks.c Sun Feb 4 16:23:33 2007
@@ -104,6 +104,10 @@
ANY, { UQ_ASSUME_CM_OVER_DATA }},
{ USB_VENDOR_SUNTAC, USB_PRODUCT_SUNTAC_AS64LX,
0x100, { UQ_ASSUME_CM_OVER_DATA }},
+ { USB_VENDOR_NOVATEL, USB_PRODUCT_NOVATEL_CDMA_MODEM,
+ ANY, { UQ_ASSUME_CM_OVER_DATA }},
{ USB_VENDOR_MOTOROLA2, USB_PRODUCT_MOTOROLA2_A41XV32X,
ANY, { UQ_ASSUME_CM_OVER_DATA }},
/* Devices which should be ignored by uhid */
--Boundary-00=_uUrxFywZMjriaBO
Content-Type: text/plain;
charset="us-ascii";
name="ppp.conf"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
filename="ppp.conf"
#################################################################
# PPP Sample Configuration File
# Originally written by Toshiharu OHNO
# Simplified 5/14/1999 by ws...@cdrom.com
#
# See /usr/share/examples/ppp/ for some examples
#
# $FreeBSD: src/etc/ppp/ppp.conf,v 1.10 2004/11/19 17:12:56 obrien Exp $
#################################################################
default:
set log Phase Chat LCP IPCP CCP tun command
ident user-ppp VERSION (built COMPILATIONDATE)
# Ensure that "device" references the correct serial port
# for your modem. (cuad0 = COM1, cuad1 = COM2)
#
set device /dev/cuaU0
# set speed 115200
set speed 230400
set dial "ABORT BUSY ABORT NO\\sCARRIER TIMEOUT 5 \
\"\" AT OK-AT-OK ATE1Q0 OK \\dATDT\\T TIMEOUT 40 CONNECT"
set timeout 180 # 3 minute idle timer (the default)
enable dns # request DNS info (for resolv.conf)
papchap:
#
# edit the next three lines and replace the items in caps with
# the values which have been assigned by your ISP.
#
set dial "ABORT BUSY ABORT NO\\sCARRIER TIMEOUT 5 \
\"\" AT OK-AT-OK ATDT#777\\T TIMEOUT 40 CONNECT"
set phone #777
set login
set authname
set authkey
accept PAP
# set authname USERNAME
# set authkey PASSWORD
set ifaddr 10.0.0.1/0 10.0.0.2/0 255.255.255.0 0.0.0.0
add default HISADDR # Add a (sticky) default route
--Boundary-00=_uUrxFywZMjriaBO
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
_______________________________________________
freebsd...@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-mobile
To unsubscribe, send any mail to "freebsd-mobil...@freebsd.org"
--Boundary-00=_uUrxFywZMjriaBO--
Greg Troxel
similar to (and I think the ancestor of) FreeBSD's ubsa.)
Can you explain the UQ_ASSUME_CM_OVER_DATA quirk? I couldn't find it
showing up in ubsa or ucom, so I don't understand what it does (and if
anything odd should be done by NetBSD).
About the stall: see the following log entry in usbdevs:
Revision 1.254 / (download) - annotate - [select for diffs], Sun Feb 19 14:48:02 2006 UTC (11 months, 2 weeks ago) by iedowse
Branch: MAIN
Changes since 1.253: +1 -0 lines
Diff to previous 1.253 (colored)
Add a UQ_OPEN_CLEARSTALL quirk for devices that need a clear-stall
operation when a pipe is opened, and add an entry for the ST Micro
biometric CPU.
Submitted by: Fredrik Lindberg
MFC after: 1 week
--
Greg Troxel <g...@ir.bbn.com>
John Nielsen
On Monday 05 February 2007 09:44, Greg Troxel wrote:
> I added the vendor ids to NetBSD's usbdevs and ugensa(4). (ugensa is
> similar to (and I think the ancestor of) FreeBSD's ubsa.)
>
> Can you explain the UQ_ASSUME_CM_OVER_DATA quirk? I couldn't find it
> showing up in ubsa or ucom, so I don't understand what it does (and if
> anything odd should be done by NetBSD).
I don't know what it does exactly. The comment in usb_quirks.h just
says "modem device breaks on cm over data". I do know that it's needed for a
lot of cellphone and modem-type usb devices (including my Samsung VI660
(anyone feel like committing usb/74880?)). I tried my V620 with the ubsa
patch but without the quirk and it didn't work at all.
> About the stall: see the following log entry in usbdevs:
>
> Revision 1.254 / (download) - annotate - [select for diffs], Sun Feb 19
> 14:48:02 2006 UTC (11 months, 2 weeks ago) by iedowse Branch: MAIN
> Changes since 1.253: +1 -0 lines
> Diff to previous 1.253 (colored)
>
> Add a UQ_OPEN_CLEARSTALL quirk for devices that need a clear-stall
> operation when a pipe is opened, and add an entry for the ST Micro
> biometric CPU.
It doesn't look like this was ever MFC'ed. Who should I bug to get that taken
care of? If I can test it and it takes care of the error messages then I'll
add it to my patch and submit a PR.
Funny coincidence. It looks like kern/106033 (the PR that led me down the ubsa
path to begin with) was committed a few hours before my message yesterday.
Thanks Lukas!
JN
Greg Troxel
Content-Transfer-Encoding: quoted-printable
John Nielsen <li...@jnielsen.net> writes:
> [Adding usb@ to CC list.]
>
> On Monday 05 February 2007 09:44, Greg Troxel wrote:
>> I added the vendor ids to NetBSD's usbdevs and ugensa(4). (ugensa is
>> similar to (and I think the ancestor of) FreeBSD's ubsa.)
>>
>> Can you explain the UQ_ASSUME_CM_OVER_DATA quirk? I couldn't find it
>> showing up in ubsa or ucom, so I don't understand what it does (and if
>> anything odd should be done by NetBSD).
>
> I don't know what it does exactly. The comment in usb_quirks.h just=20
> says "modem device breaks on cm over data". I do know that it's needed fo=
r a=20
> lot of cellphone and modem-type usb devices (including my Samsung VI660=20
> (anyone feel like committing usb/74880?)). I tried my V620 with the ubsa=
=20
> patch but without the quirk and it didn't work at all.
I've been using a KPC650 (vz) and a sierra aircard (sprint) on NetBSD
without any sign of that quirk with no issues. So, it could be that
the NetBSD code doesn't provoke the behavior for some other reason.
Greg Troxel <g...@ir.bbn.com>
--=-=-=
Content-Type: application/pgp-signature
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (NetBSD)
iD8DBQFFx18I+vesoDJhHiURAkf9AJ9AcOnKExSrceo5aTeU7uHObV0RIACgqifw
nyiUNjejIgxI59pWFzyV4S8=
=R3mU
-----END PGP SIGNATURE-----
--=-=-=--
Volker
> addr 2
> ucom0: ubsa_request: STALLED
> ucom0: ubsa_request: STALLED
> ...
> (lots more STALLED messages)
John,
just added myself to mobile@ (thanks for cross-posting this to usb@
so I got notice of this discussion).
If you're using the V620 (or any other (W-)CDMA) card with ppp, you
should make sure you're using:
set cd 30! (note the exclamation mark at the end)
I've seen _a lot_ of problem reports (not just with FreeBSD) where
ppp does not recognize a connection loss when using W-CDMA cards.
The exclamation mark forces ppp into waiting for a carrier and
following that CD signal.
I suspect your problems might come from a bad signal. I don't know
much about the V620 (just that it's only used in nothern america)
but if it's a high-speed card (transfer rate > 56k, 3G mode?) you
should force your card into GPRS or 3G mode.
By default W-CDMA cards are trying to auto detect the best signal
reception and that might cause often signal loss due to band change.
If you know how to force the card into one or the other mode, please
use that in ppp's init string. If you don't know it, you may try the
commands for the U630 (I guess both cards are using the same command
set). You may find a bit info about the U630 at:
http://bsd.vwsoft.com/3g/merlin_u630.html
Please note the U630 is serial-over-PCMCIA and not usb so the
hardware is not the same. I thought the V620 is also
serial-over-PCMCIA but it seems you proved me wrong. ;)
Greetings,
Volker
John Nielsen
> > ucom0: Novatel Wireless Inc. Novatel Wireless Merlin CDMA, rev
> > 1.10/0.00, addr 2
> > ucom0: ubsa_request: STALLED
> > ucom0: ubsa_request: STALLED
> > ...
> > (lots more STALLED messages)
>
> John,
>
> just added myself to mobile@ (thanks for cross-posting this to usb@
> so I got notice of this discussion).
>
> If you're using the V620 (or any other (W-)CDMA) card with ppp, you
> should make sure you're using:
>
> set cd 30! (note the exclamation mark at the end)
>
> I've seen _a lot_ of problem reports (not just with FreeBSD) where
> ppp does not recognize a connection loss when using W-CDMA cards.
> The exclamation mark forces ppp into waiting for a carrier and
> following that CD signal.
I'll play around with that, thanks.
> I suspect your problems might come from a bad signal. I don't know
> much about the V620 (just that it's only used in nothern america)
> but if it's a high-speed card (transfer rate > 56k, 3G mode?) you
> should force your card into GPRS or 3G mode.
I'll experiment with that too. One thing I'm not sure I mentioned in my
first post is that the STALLED messages only show up when the device is
opened and closed. If I use ppp in manual mode I get about six of them when
I first start ppp but none while I issue any AT commands or while the link
is active. Then I get about six more when I close the link.
It is a (relatively) high-speed card. Under Windows all the fancy
third-party software does is create a regular dial-up networking link with
a serial speed of 230400. Doing the same with ppp under FreeBSD gets me a
usable connection. Under either OS the latency is pretty high (~300ms) and
a few packets get dropped here and there, but it beats straight dial-up (or
heaven forbid, no connection at all!)
> By default W-CDMA cards are trying to auto detect the best signal
> reception and that might cause often signal loss due to band change.
>
> If you know how to force the card into one or the other mode, please
> use that in ppp's init string. If you don't know it, you may try the
> commands for the U630 (I guess both cards are using the same command
> set). You may find a bit info about the U630 at:
> http://bsd.vwsoft.com/3g/merlin_u630.html
Thanks for the link. More documentation is almost always better. :)
> Please note the U630 is serial-over-PCMCIA and not usb so the
> hardware is not the same. I thought the V620 is also
> serial-over-PCMCIA but it seems you proved me wrong. ;)
I was a little surprised when I first plugged it in under Windows and a USB
hub popped up. But hey, I didn't design the card or even choose to buy it
(it's part of the lucky ball-and-chain package I get from work when I'm on
call).
Thanks for your response.
JN
Volker
You may even try to open the device with a speed setting of 1200
baud. I've played with userland ppp and found that when using speeds
above 115kbit ppp might behave oddly. For any tty device which is
not associated with a real UART chip the speed should not matter.
Even in the driver the speed setting function is a dummy function
which does nothing but ppp takes care of the speed.
I suggest to use a lower speed setting with ppp and try again. You
should have no throughput impact when using 38400 baud with ppp. At
least this is what I was experiencing using 3G hardware.
Please try that (plus the 'set cd xx!' setting) and see if the
stalled messages are gone.
> Thanks for the link. More documentation is almost always better. :)
If you have any good docs for the V620 with the differences to the
U630 I'll put them up also if you can give me pointers. I'm trying
to assemble as much 3G infos as possible as this is a really
interesting infrastructure (I like the idea to have the internet to go).
> I was a little surprised when I first plugged it in under Windows and a USB
> hub popped up. But hey, I didn't design the card or even choose to buy it
> (it's part of the lucky ball-and-chain package I get from work when I'm on
> call).
Most current 3G cards do have an interface like that. I think
Qualcomm designed that and most manufacturers took that. Just the
newer Option Globetrotter cards (3G Max, 3G +) do have a different
interface.
Greetings,
Volker