I'm not sure about this, but during debugging, I encountered new symbols that are now GPL only (for HDA access) on 2.6.30+ kernels, so we cannot simply strip out the non-gpl symbols- all you'd be left with is a driver for PCI modems . Bjorn. On 02/18/2010 10:04 PM, Marvin Stodolsky wrote: > Nikolay, > > Thanks for the agrsm-test improvments, I'll deb package them. > > RE: > BTW Why agrsm-tool is a separate package? > It is Debian/Ubuntu policy to provide Drivers and Auxilliary files in > separate packages. > > I am working towards getting the agrsm drivers & agrsm-tools into > their Distributions, > as that would greatly diminish the work load of this List. > The major problem with respect to the agrsm-11c1040 is removal of the USB code, > so the compile will be legal. If you and Bjorn could thus help, it > would be very useful. > > MarvS > > On Thu, Feb 18, 2010 at 7:24 PM, Nikolay Zhuravlev <xxor@xxxxxxx> wrote: >> List, >> >> This is an improved [hopefully] version of the agrsm-test tool. >> More info gathering routines are added for the failure mode. >> The script is also modified to print to stdout by default >> instead of a dedicated logfile (this is what most Unix utilities >> do -- they act as filters, so their output can be piped to something >> else). >> >> Run it as follows: >> sudo ./agrsm-test 2>&1 | tee /tmp/my.log >> >> BTW Why agrsm-tool is a separate package? Wouldn't it be easier to >> merge it with the driver *.deb and distribute them as just one *.deb >> package? >> >> -- >> Nick Zhuravlev >> www.auriga.com >> >> On Fri, Feb 19, 2010 at 02:33:32AM +0300, Nikolay Zhuravlev <xxor@xxxxxxx> wrote: >> >>> Kevin, >>> >>> What does the following command say: >>> $ grep CONFIG_SND_HDA_POWER_SAVE /boot/config-`uname -r` >>> >>> Also, did you try to rebuild the driver under the new kernel? >>> I think portions of the driver depend on the >>> CONFIG_SND_HDA_POWER_SAVE option too. So the driver needs to >>> be rebuilt. >>> >>> cd <where the sources are> >>> make clean >>> make >>> sudo make modules_install >>> sudo modprobe agrmodem >>> sudo modprobe agrserial >>> sudo ln -sf /dev/ttyAGS3 /dev/modem >>> sudo ln -sf /dev/ttyAGS3 /dev/ttySAGR >>> sudo wvdialconf /etc/wvdial.conf >>> >>> -- >>> Nick Zhuravlev >>> www.auriga.com >>> >>> On Wed, Feb 17, 2010 at 02:24:31AM -0200, Kevin Fernandez <siegfri3d@xxxxxxxxx> wrote: >>> >>>> Hello Nikolay, >>>> First thanks for helping! >>>> >>>> I tried to recompile the kernel (2.6.31 of the ubuntu sources), i >>>> disabled the SMP support, CONFIG_SND_HDA_POWER_SAVE was already unset >>>> so i tried to set it as 0, using the new kernel i brought done all >>>> interfaces, unloaded many modules, connected successfully, i could >>>> ping and send 10 icmp packets but then the network stopped again and >>>> pppd crashed (just like before), when reconnecting it gave me a >>>> segfault (so again, just like before), the dmesg output is a bit >>>> different: >>>> >>>> [ 539.463626] ------------[ cut here ]------------ >>>> [ 539.463629] kernel BUG at kernel/timer.c:762! >>>> [ 539.463631] invalid opcode: 0000 [#1] >>>> [ 539.463634] last sysfs file: >>>> /sys/devices/LNXSYSTM:00/device:00/PNP0A08:00/device:0b/PNP0C09:00/PNP0C0A:00/power_supply/BAT1/voltage_now >>>> [ 539.463638] Modules linked in: ppp_async crc_ccitt agrserial >>>> agrmodem isofs binfmt_misc snd_hda_codec_realtek snd_hda_intel >>>> snd_hda_codec snd_hwdep snd_pcm snd_timer snd psmouse soundcore >>>> snd_page_alloc fbcon tileblit font bitblit softcursor i915 drm >>>> i2c_algo_bit intel_agp agpgart video output [last unloaded: mii] >>>> [ 539.463658] >>>> [ 539.463662] Pid: 9361, comm: wvdial Tainted: P (2.6.31.9 >>>> #3) MSI Notebook VR601 >>>> [ 539.463665] EIP: 0060:[<c013be43>] EFLAGS: 00210286 CPU: 0 >>>> [ 539.463673] EIP is at add_timer+0x13/0x20 >>>> [ 539.463675] EAX: fb576ae0 EBX: f97bed7c ECX: 00000000 EDX: f97beac0 >>>> [ 539.463678] ESI: f97bed7c EDI: fffffff4 EBP: efb41d88 ESP: efb41d88 >>>> [ 539.463680] DS: 007b ES: 007b FS: 0000 GS: 00e0 SS: 0068 >>>> [ 539.463684] Process wvdial (pid: 9361, ti=efb40000 task=efb38c80 >>>> task.ti=efb40000) >>>> [ 539.463686] Stack: >>>> [ 539.463687] efb41d90 fb44e04e efb41da0 fb508e7c fb508e37 f97bed7c >>>> efb41dc0 fb508f05 >>>> [ 539.463693] <0> 000003e8 00000000 00000000 00000000 12345678 >>>> f97bed7c efb41dc8 fb44e2b8 >>>> [ 539.463698] <0> efb41dd0 fb44e2cd efb41de8 f97bcf60 fffffff4 >>>> efb41de8 f1bd61a4 f97bed7c >>>> [ 539.463704] Call Trace: >>>> [ 539.463739] [<fb44e04e>] ? lt_add_timer+0x2e/0x30 [agrmodem] >>>> [ 539.463782] [<fb508e7c>] ? VMODEM_Start_Timer+0x28/0x33 [agrmodem] >>>> [ 539.463822] [<fb508e37>] ? timertick_function+0x0/0x1d [agrmodem] >>>> [ 539.463863] [<fb508f05>] ? vxdPortOpen+0x3b/0x42 [agrmodem] >>>> [ 539.463893] [<fb44e2b8>] ? wrap_vxdPortOpen+0x8/0x10 [agrmodem] >>>> [ 539.463921] [<fb44e2cd>] ? modemPortOpen+0xd/0x20 [agrmodem] >>>> [ 539.463927] [<f97bcf60>] ? serial8250_startup+0x1b0/0x370 [agrserial] >>>> [ 539.463933] [<c0363792>] ? uart_startup+0x42/0x130 >>>> [ 539.463937] [<c0363d68>] ? uart_open+0xd8/0x340 >>>> [ 539.463941] [<c034bbb6>] ? tty_init_dev+0x76/0x140 >>>> [ 539.463946] [<c01d8a06>] ? mntput_no_expire+0x16/0x80 >>>> [ 539.463949] [<c034be1f>] ? tty_open+0x19f/0x430 >>>> [ 539.463953] [<c01c6f7a>] ? chrdev_open+0x8a/0x120 >>>> [ 539.463958] [<c01c241a>] ? __dentry_open+0xaa/0x220 >>>> [ 539.463962] [<c01c2675>] ? nameidata_to_filp+0x55/0x70 >>>> [ 539.463965] [<c01c6ef0>] ? chrdev_open+0x0/0x120 >>>> [ 539.463969] [<c01cf7e5>] ? do_filp_open+0x535/0x8f0 >>>> [ 539.463973] [<c01cc465>] ? getname+0x25/0xf0 >>>> [ 539.463977] [<c01c21b0>] ? do_sys_open+0x50/0x160 >>>> [ 539.463981] [<c01c5a45>] ? fput+0x15/0x20 >>>> [ 539.463984] [<c01c2329>] ? sys_open+0x29/0x40 >>>> [ 539.463988] [<c010332c>] ? syscall_call+0x7/0xb >>>> [ 539.463990] Code: 40 2b 6e c0 8b 80 08 02 00 00 89 41 2c e9 36 ff >>>> ff ff 8d b6 00 00 00 00 55 8b 10 89 e5 85 d2 75 0a 8b 50 08 e8 ef fe >>>> ff ff 5d c3 <0f> 0b eb fe 89 f6 8d bc 27 00 00 00 00 55 89 e5 e8 18 97 >>>> ff ff >>>> [ 539.464019] EIP: [<c013be43>] add_timer+0x13/0x20 SS:ESP 0068:efb41d88 >>>> [ 539.464060] ---[ end trace c38ee2fc1c238c35 ]--- >>>> >>>> I'll try to see if i can get some logs of pppd when it crashes. >>>> >>>> Any idea would be appreciated! >>>> Thanks, >>>> Kevin >>>> >>>> 2010/2/16, Nikolay Zhuravlev <xxor@xxxxxxx>: >>>>> Things to try that come to my mind: >>>>> >>>>> - try a non-SMP kernel >>>>> - try to build the kernel with the >>>>> CONFIG_SND_HDA_POWER_SAVE unset >>>>> - try to bring down other interfaces (eth0, wlan0) >>>>> before using the modem >>>>> >>>>> -- >>>>> Nick Zhuravlev >>>>> www.auriga.com >>>>> >>>>> On Sun, Feb 14, 2010 at 08:51:02PM -0200, Kevin Fernandez >>>>> <siegfri3d@xxxxxxxxx> wrote: >>>>> >>>>>> Almost there! >>>>>> I don't know if it's your change (probably!); as i also upgraded to >>>>>> the variant 2.6.31-20-generic. >>>>>> I could run wvdialconf without problem, then i tried multiple times to >>>>>> connect but right after connecting to the first site it was always >>>>>> crashing wvdial/pppd (i could see that it resolved the site in firefox >>>>>> and was trying to connect to it). >>>>>> After trying another time i got a segmentation fault: >>>>>> >>>>>> --> WvDial: Internet dialer version 1.60 >>>>>> --> Initializing modem. >>>>>> --> Sending: ATZ >>>>>> ATZ >>>>>> OK >>>>>> --> Sending: ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 >>>>>> ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 >>>>>> OK >>>>>> --> Modem initialized. >>>>>> --> Sending: ATDT17003131 >>>>>> --> Waiting for carrier. >>>>>> ATDT17003131 >>>>>> BUSY >>>>>> --> The line is busy. Trying again. >>>>>> --> Sending: ATDT17003131 >>>>>> --> Waiting for carrier. >>>>>> ATDT17003131 >>>>>> CONNECT 40000 V44 >>>>>> --> Carrier detected. Waiting for prompt. >>>>>> ~[7f]}#@!}!}!} }8}"}&} }*} } }#}$@#}%}&(BAh}'}"}(}"[0c]J~ >>>>>> --> PPP negotiation detected. >>>>>> --> Starting pppd at Sun Feb 14 20:17:40 2010 >>>>>> --> Pid of pppd: 16639 >>>>>> --> Using interface ppp0 >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> local IP address 189.48.97.120 >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> remote IP address 201.4.82.93 >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> primary DNS address 200.202.193.75 >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> secondary DNS address 200.222.0.34 >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> Connect time 1.1 minutes. >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> pppd: px2[08]@[7f]2[08] >>>>>> --> Disconnecting at Sun Feb 14 20:18:47 2010 >>>>>> --> The PPP daemon has died: A modem hung up the phone (exit code = 16) >>>>>> --> man pppd explains pppd error codes in more detail. >>>>>> --> Try again and look into /var/log/messages and the wvdial and pppd >>>>>> man pages for more information. >>>>>> --> Auto Reconnect will be attempted in 5 seconds >>>>>> --> Initializing modem. >>>>>> --> Sending: ATZ >>>>>> ATZ >>>>>> OK >>>>>> --> Sending: ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 >>>>>> ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 >>>>>> OK >>>>>> --> Modem initialized. >>>>>> Erreur de segmentation >>>>>> >>>>>> >>>>>> >>>>>> dmesg: >>>>>> >>>>>> [ 1348.660145] ------------[ cut here ]------------ >>>>>> [ 1348.660150] kernel BUG at >>>>>> /build/buildd/linux-2.6.31/kernel/timer.c:762! >>>>>> [ 1348.660154] invalid opcode: 0000 [#1] SMP >>>>>> [ 1348.660159] last sysfs file: >>>>>> /sys/devices/system/cpu/cpu1/topology/core_id >>>>>> [ 1348.660162] Modules linked in: ppp_deflate zlib_deflate bsd_comp >>>>>> ppp_async crc_ccitt agrserial agrmodem isofs udf crc_itu_t binfmt_misc >>>>>> ppdev vboxnetflt vboxnetadp vboxdrv joydev snd_hda_codec_realtek >>>>>> wlan_scan_sta ath_rate_sample iptable_nat nf_nat nf_conntrack_ipv4 >>>>>> nf_conntrack nf_defrag_ipv4 iptable_mangle snd_hda_intel snd_hda_codec >>>>>> snd_hwdep iptable_filter pcmcia ip_tables x_tables snd_pcm_oss >>>>>> snd_mixer_oss snd_pcm snd_seq_dummy snd_seq_oss sdhci_pci sdhci >>>>>> led_class yenta_socket rsrc_nonstatic pcmcia_core snd_seq_midi >>>>>> snd_rawmidi snd_seq_midi_event snd_seq snd_timer snd_seq_device snd >>>>>> ath_pci wlan ath_hal psmouse serio_raw soundcore snd_page_alloc sbp2 >>>>>> lp parport dm_raid45 xor fbcon tileblit font bitblit softcursor i915 >>>>>> drm i2c_algo_bit ohci1394 ieee1394 r8169 mii intel_agp agpgart video >>>>>> output >>>>>> [ 1348.660246] >>>>>> [ 1348.660251] Pid: 16129, comm: wvdial Tainted: P >>>>>> (2.6.31-20-generic #57-Ubuntu) MSI Notebook VR601 >>>>>> [ 1348.660255] EIP: 0060:[<c0150bb3>] EFLAGS: 00010286 CPU: 1 >>>>>> [ 1348.660262] EIP is at add_timer+0x13/0x20 >>>>>> [ 1348.660265] EAX: fbaaebc0 EBX: f851ae28 ECX: 00000000 EDX: f851ab60 >>>>>> [ 1348.660268] ESI: f851ae28 EDI: fffffff4 EBP: ec82dd88 ESP: ec82dd88 >>>>>> [ 1348.660272] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 >>>>>> [ 1348.660275] Process wvdial (pid: 16129, ti=ec82c000 task=ec838000 >>>>>> task.ti=ec82c000) >>>>>> [ 1348.660278] Stack: >>>>>> [ 1348.660280] ec82dd90 fb98604e ec82dda0 fba40ebc fba40e77 f851ae28 >>>>>> ec82ddc0 fba40f45 >>>>>> [ 1348.660289] <0> 000003e8 00000000 00000000 00000000 12345678 >>>>>> f851ae28 ec82ddc8 fb9862f8 >>>>>> [ 1348.660298] <0> ec82ddd0 fb98630d ec82dde8 f8518fe0 fffffff4 >>>>>> ec82dde8 ef86be04 f851ae28 >>>>>> [ 1348.660309] Call Trace: >>>>>> [ 1348.660346] [<fb98604e>] ? lt_add_timer+0x2e/0x30 [agrmodem] >>>>>> [ 1348.660389] [<fba40ebc>] ? VMODEM_Start_Timer+0x28/0x33 [agrmodem] >>>>>> [ 1348.660432] [<fba40e77>] ? timertick_function+0x0/0x1d [agrmodem] >>>>>> [ 1348.660473] [<fba40f45>] ? vxdPortOpen+0x3b/0x42 [agrmodem] >>>>>> [ 1348.660504] [<fb9862f8>] ? wrap_vxdPortOpen+0x8/0x10 [agrmodem] >>>>>> [ 1348.660533] [<fb98630d>] ? modemPortOpen+0xd/0x20 [agrmodem] >>>>>> [ 1348.660540] [<f8518fe0>] ? serial8250_startup+0x210/0x380 [agrserial] >>>>>> [ 1348.660548] [<c039d0d6>] ? uart_startup+0x46/0x140 >>>>>> [ 1348.660552] [<c039dc12>] ? uart_open+0x102/0x220 >>>>>> [ 1348.660557] [<c0384d13>] ? __tty_open+0x1a3/0x460 >>>>>> [ 1348.660562] [<c0384fef>] ? tty_open+0x1f/0x40 >>>>>> [ 1348.660568] [<c01eae1d>] ? chrdev_open+0xcd/0x190 >>>>>> [ 1348.660572] [<c01e5df9>] ? __dentry_open+0xb9/0x230 >>>>>> [ 1348.660577] [<c01e6055>] ? nameidata_to_filp+0x55/0x70 >>>>>> [ 1348.660581] [<c01ead50>] ? chrdev_open+0x0/0x190 >>>>>> [ 1348.660587] [<c01f3daa>] ? do_filp_open+0x53a/0x890 >>>>>> [ 1348.660592] [<c01e5b90>] ? do_sys_open+0x50/0x150 >>>>>> [ 1348.660597] [<c01e9645>] ? fput+0x15/0x20 >>>>>> [ 1348.660601] [<c01e5a67>] ? filp_close+0x47/0x70 >>>>>> [ 1348.660605] [<c01e5cf9>] ? sys_open+0x29/0x40 >>>>>> [ 1348.660610] [<c01033ac>] ? syscall_call+0x7/0xb >>>>>> [ 1348.660612] Code: 8b 80 08 02 00 00 89 43 2c e9 fa fe ff ff 0f 0b >>>>>> eb fe 8d b6 00 00 00 00 55 8b 10 89 e5 85 d2 75 0a 8b 50 08 e8 af fe >>>>>> ff ff 5d c3 <0f> 0b eb fe 89 f6 8d bc 27 00 00 00 00 55 89 e5 e8 98 8c >>>>>> ff ff >>>>>> [ 1348.660669] EIP: [<c0150bb3>] add_timer+0x13/0x20 SS:ESP 0068:ec82dd88 >>>>>> [ 1348.660677] ---[ end trace e3d2e7c757a613b0 ]--- >>>>>> >>>>>> >>>>>> Any idea? >>>>>> Thanks for the help! >>>>>> >>>>>> 2010/2/14, Nikolay Zhuravlev <xxor@xxxxxxx>: >>>>>>> I am not sure. I think power_save=0 means that >>>>>>> there is no timeout and snd-hda-intel never sleeps. >>>>>>> >>>>>>> My previous kernel (vanilla 2.6.32) did not have >>>>>>> CONFIG_SND_HDA_POWER_SAVE set, so the power_save was >>>>>>> not even an option, and it all worked fine. >>>>>>> >>>>>>> There are also bug reports of some high pitch noise >>>>>>> related to the power_save=10 setting, both in Fedora >>>>>>> and in Ubuntu. >>>>>>> >>>>>>> -- >>>>>>> Nick Zhuravlev >>>>>>> www.auriga.com >>>>>>> >>>>>>> On Sun, Feb 14, 2010 at 02:26:08PM -0400, Bjorn Wielens >>>>>>> <Uniacke1@xxxxxxxx> >>>>>>> wrote: >>>>>>> >>>>>>>> so it looks like if the hda driver is in a sleep state, the agrsm >>>>>>>> driver >>>>>>>> has a fit. >>>>>>>> >>>>>>>> >>>>>>>> I just verified that my SuSE distro does not set the power save in >>>>>>>> modprobe, hence it worked fine out of the box for me. >>>>>>>> >>>>>>>> That said, it must be possible to modify the driver code to "wake" the >>>>>>>> sound card properly- I know for a fact that it has function prototypes >>>>>>>> for powerup/powerdown functions. >>>>>>>> >>>>>>>> Bjorn. >>>>>>>> >>>>>>>> >>>>>>>> On 02/14/2010 01:57 PM, Nikolay Zhuravlev wrote: >>>>>>>>> Valera, >>>>>>>>> >>>>>>>>> Glad it worked for you. It is amazing how much difference a single >>>>>>>>> bit of information can make ;) >>>>>>>>> >>>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>> >> >
Attachment:
signature.asc
Description: OpenPGP digital signature
