From mboxrd@z Thu Jan 1 00:00:00 1970 From: Florian Klink Subject: Re: cdc-wdm: unable to connect after suspend Date: Wed, 11 Jun 2014 12:08:39 +0200 Message-ID: <53982AA7.3090508@flokli.de> References: <874mzt55kz.fsf@nemi.mork.no> <53971171.5070001@flokli.de> <539712BA.5010004@flokli.de> <87lht350y0.fsf@nemi.mork.no> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="h1fKSL3AnBfuB4U3bLwkD4VRDeOMecqO3" Cc: netdev@vger.kernel.org, modemmanager-devel@lists.freedesktop.org To: =?UTF-8?B?QmrDuHJuIE1vcms=?= Return-path: Received: from asterix.flokli.de ([78.46.104.9]:52580 "EHLO asterix.flokli.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750757AbaFKKIm (ORCPT ); Wed, 11 Jun 2014 06:08:42 -0400 In-Reply-To: <87lht350y0.fsf@nemi.mork.no> Sender: netdev-owner@vger.kernel.org List-ID: This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --h1fKSL3AnBfuB4U3bLwkD4VRDeOMecqO3 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Am 11.06.2014 10:01, schrieb Bj=C3=B8rn Mork: > Florian Klink writes: >> Am 10.06.2014 16:08, schrieb Florian Klink: >>> Am 10.06.2014 14:09, schrieb Bj=C3=83=C5=BErn Mork: >>>> Florian Klink writes: >>>> >>>>> Hi, >>>>> >>>>> I recently bought a notebook (Fujitsu Lifebook T904) with integrate= d >>>>> 3G/4G modem (Sierra Wireless EM7305) thats powered by the cdc-wdm d= river. >>>>> >>>>> It works without any problems on a fresh bootup using Networkmanage= r. >>>>> >>>>> However, after putting the notebook into standby and waking up agai= n, >>>>> I'm unable to get a connection (always reproducible, not signal qua= lity >>>>> related). >>>> >>>> Does it work again if you restart NetworkManager and ModemManager at= >>>> this point? >>> >>> Nope. ModemManager gets confused completely and drops the modem out o= f >>> the list of connections: >>> >>> ModemManager[3067]: Couldn't find support for device at >>> '/sys/devices/pci0000:00/0000:00:19.0': not supported by any plugin >>> ModemManager[3067]: Couldn't find support for device at >>> '/sys/devices/pci0000:00/0000:00:1c.3/0000:03:00.0': not supported by= >>> any plugin >>> ModemManager[3067]: [/dev/cdc-wdm1] Queried max control message size:= 4096 >>> ModemManager[3067]: [/dev/cdc-wdm1] No transaction matched in receive= d >>> message >=20 > Ah, OK. Then I obviously guessed wrong. MM and the firmware does not > agree on the current transaction sequence. I'm not sure why. The modem > is sending unexpected messages, and I assume MM doesn't get the answers= > it is looking for. This could be caused by the driver cancelling the > interrupt URB at an "inconvenient time" for the firmware. But I haven'= t > seen issues like that with the Qualcomm based Sierra modems before. >=20 > And you get into this state consistently on every suspend+resume?=20 >=20 > It would be useful to know which unexpeced messages we receive (errors?= > same message all over? older queued up messages being played back with > unexpected delay?) Any chance you can enable libmbim + ModemManager > debugging to get the MBIM message dumps? Sure, I can provide the message dumps. Is it as simple as described in here (http://www.freedesktop.org/wiki/Software/ModemManager/Debugging/), or do I need to recompile libmbim + ModemManager with some additional (de= bug) flags enabled? >=20 >=20 >>>> Does it help to do >>>> >>>> echo 0 >/sys/bus/usb/devices/x-y/power/persist >>>> >>>> prior to suspending the notebook? You'll have to replace "x-y" with= the >>>> correct USB bus and port number. You can find this in e.g. the dmes= g >>>> output. For example, if your log shows: >>>> >>>> qmi_wwan 2-4:1.8: cdc-wdm0: USB WDM device >>>> >>>> then x-y =3D 2-4. >>> >>> dmesg shows "cdc_mbim 1-6:2.12: cdc-wdm1: USB WDM device", so I did e= cho >>> 0 > /sys/bus/usb/devices/1-6/power/persist. >>> >>> However, after a suspend/resume cycle, connecting didnt work either. >>> Errors were the same as without persist =3D 0. >=20 > Yes, my initial guess was wrong, so this is not the problem. >=20 >>> dmesg doesn't really show an error message from the modem. Seems like= it >>> also has an issue resuming "Bus 001 Device 006: ID 0483:91d1 >>> STMicroelectronics", but this shouldn't cause the problems with the m= odem... >> >> I attached the dmesg output, probably there's till something interesti= ng >> inside ;-) >=20 > Thanks. I wasn't expecting any errors there (I still don't think there= > is any real error here from the drivers point of view), but I wanted to= > see whether the modem is disconnected and rediscovered or just plainly > suspended. >=20 > But??? I cannot find anything related to the modem device in that log?= > Not even anything mentioning the "1-6" USB port. I just sent the part of dmesg where the suspend & resume cycle happened. Of course, I see the device appearing on bootup: [ 11.537522] usb 1-6: new high-speed USB device number 11 using xhci_hc= d [ 11.711214] usb 1-6: config 1 has an invalid interface number: 8 but m= ax is 3 [ 11.711217] usb 1-6: config 1 has no interface number 1 [ 11.711432] usb 1-6: config 2 has an invalid interface number: 12 but = max is 1 [ 11.711434] usb 1-6: config 2 has an invalid interface number: 13 but = max is 1 [ 11.711435] usb 1-6: config 2 has an invalid interface number: 13 but = max is 1 [ 11.711437] usb 1-6: config 2 has no interface number 0 [ 11.711439] usb 1-6: config 2 has no interface number 1 [...] [ 14.735330] usbcore: registered new interface driver cdc_ncm [ 14.736046] usbcore: registered new interface driver cdc_wdm [ 14.738273] cdc_mbim 1-6:2.12: cdc-wdm1: USB WDM device [ 14.738459] cdc_mbim 1-6:2.12 wwan0: register 'cdc_mbim' at usb-0000:0= 0:14.0-6, CDC MBIM, 96:ba:xx:xx:xx:xx [ 14.738477] usbcore: registered new interface driver cdc_mbim [ 14.749795] systemd-udevd[248]: renamed network interface wwan0 to wwp= 0s20u6c2i12 =2E.. but you don't get any dmesg output about it during the suspend/resu= me cycle. Florian --h1fKSL3AnBfuB4U3bLwkD4VRDeOMecqO3 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJTmCqnAAoJEInyWbj6DRgfqTEQAK/2juYTwbjhFpbvsiTiqwlc 44HuZIQw6GUIDpsDb8+zjiDofPmlgSayfg34MAa362UEXowkGGLcSDrR5u4uAlIF FCeB8XZrebSv7zgmWsp+SsYEpqIaDRzEIwbksaFZaBOL7YLHtf3azb7fTJZIQZrO 4oeHDUx4mCaoii9NFqkOyfM1yjn9LUI76Ncaf95uR3vbCZeGRfD89L6W3HuvcDFF 92U5V7vf+fjLQAi4h7BCThUO9VqYwBrGXFsigm6DRbEfUY2bbV/RrAiP3WOuPDiC K7Zl0bYXd51kRHuwpqXIZHkGBYokjnAkqRCr1KGJ2cjGIjec+khB45+9W4sytKSk 4hxeGz9cFzM2cUIyKz7PFm99qYmO3SjmKakWbb0ClHwRwyzCT2i6BA5hbt6BzYwg oOp4Rg7emqqfktqLH6zqgrvm4p9ExnSc+UgJd7I4C+JWzhjTC+rSiqy22r0iNY8F 4ZkU6KfuORX0y4/jKzcBuEcXt0a0ukiniDazKtG1V0Kjx8MW1QDhiLUkV1tNbaPF L7VAsvO33exrGM9aw2CIHYJbkeJkeorK7z3cnFUNmBbmQwUUYQEio+SiArJHCHXT fdjwY+p9G/JgnnBVSi7aDJaYxyhlS2SrbKNs9WezdxonmYx+1WRqS3/qjZiWDguX vmcqKviLzVR/IxZJ0X35 =bL0C -----END PGP SIGNATURE----- --h1fKSL3AnBfuB4U3bLwkD4VRDeOMecqO3--