USB2 backport from 2.4.27 to 2.4.4/2.4.18/2.4.20 SUPPORT

[Jamie]

This is the support thread for the USB2 backport from 2.4.27 to 2.4.4/2.4.18/2.4.20. The files can be found in the development thread. (http://www.dealdatabase.com/forum/showthread.php?t=38167).

The support discussions seem to have become fragmented over multiple threads, so I thought a new thread with the appropriate title could pull all the conversations into one place.

Other relevant threads: USB networking on 4.x @ kernel 2.4.18 (split from 3.x USB thread) (http://www.dealdatabase.com/forum/showthread.php?t=36876) -- 4.x centric
USB device support for Series2 / 3.1.x @ kernel 2.4.4 (http://www.dealdatabase.com/forum/showthread.php?t=24626&highlight=usb) -- 3.x centric
MFS Soapbox Derby: aka what is your download speed? (http://www.dealdatabase.com/forum/showthread.php?t=39328&highlight=Soapbox+Derby) -- Performance discussions
USB2 backport from 2.4.27 to 2.4.4/2.4.18/2.4.20 (http://www.dealdatabase.com/forum/showthread.php?t=38167) -- Development discussions

[Desperado]

New poster here.. I'm trying to get the most recent version of the backport to work with the Airlink ASOHOUSB on an HDRV2 w/ 6.2.

It starts up ok, and I assign a static ip in the rc.sysinit.author file. It actually works ok for a few minutes, and I am able to ping, telnet and ftp to it.

bash-2.02# ifconfig
lo Link encap:Local Loopback
inet addr:127.0.0.1 Bcast:0.0.0.0 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:27 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:2970 dropped:27 overruns:0 carrier:0 coll:0

eth0 Link encap:Ethernet HWaddr 00:50:5B:04:73:66
inet addr:192.168.0.111 Bcast:192.168.0.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:64 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:4473 dropped:41 overruns:0 carrier:0 coll:0


After running for a few minutes, it dies. The IP address reverts to this and I can no longer reach it on the network.

eth0 Link encap:Ethernet HWaddr 00:50:5B:04:73:66
inet addr:169.254.51.198 Bcast:169.254.255.255 Mask:255.255.0.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:240 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:17628 dropped:200 overruns:0 carrier:0 coll:0


If I reissue the ifconfig eth0 192.168.0.111... from serial bash, it works again but only for a few minutes.

Any ideas?

[Jamie]

Any ideas?Did you rename dhclient to keep it from running?

See section 3b) of this (http://www.dealdatabase.com/forum/showthread.php?t=42325) thread.

[wbgolden]

Copied and pasted from my posts in another (wrong) thread.
I'm sure I'm missing something simple here...

Trying to set up an Airlink ASOHOUSB usb2 adapter to work with my philips DSR708 running 6.2. :

I backed up the original USB 'o' files, then transferred Jamie's newest ones (usbcore.o, usbnet.o and usb-ohci.o...I didn't transfer the dummy version of ehci because I didn't have the original in the modules directory) to my modules directory using filezilla (binary mode). Symlinked usbnet.o to ax8817x.o. Added vendor/product ID 0b95 7720 to /etc/hotplug/usb.map.

Created a test.conf file with:

/sbin/insmod /lib/modules/usbcore.o log2_irq_thresh=4
/sbin/insmod /lib/modules/usbnet.o
/sbin/insmod /lib/modules/usb-ohci.o

Rebooted and the Tivo sees the adapter:

eth0: register usbnet usb-00:0d.2-1, ASIX AX88772 USB 2.0 Ethernet
But I cannot connect to the tivo via telnet. Anyone with any idea what could be going on? I can certainly provide more info if needed. It seems like everything is OK, just missing one (maybe small) detail.

All was well before I changed over to the Airlink adapters. I was able to ping both ways and download using tytools, tivowebplus was working (somewhat), HMO and MRV both running fine (thank you, NutKase). I am using a Belkin F5d5050 adapter. It continues to work well if I restore the backups to the modules directory.

With the airlink adapter in there and everything set up as stated above, I am unable to ping to or from the Tivo. (But the link light is active, as some people before me stated with the GIG adapters). My network is set up to assign ip addresses using DHCP, which has worked fine for both Tivos in the past--and present with the 1.1 adapters installed.

PING 192.168.0.1 (192.168.0.1): 56 data bytes
ping: sendto: Network is unreachable
There’s not much hurry for me. With the USB 1.1 adapters in there everything is fine. The adapters were only $4 this weekend at Frye's so it's not like I'm out major money. I'm also not sure the usb2 adapters will help my throughput. I'm running wireless-G access points from both my tivos and seem to limited to about .50 mbps. I may just lurk here like I have been since November and wait for the answers to be … revealed.

Thanks!

[Jamie]

I'm sure I'm missing something simple here...You are at a point here where it is difficult to diagnose further without a serial console cable.

There are several possibilities that might be causing your problems and those of gpontis, described elsewhere:

6.2 network configuration issues
One way to eliminate this as a possibility is to boot up to a bare bash, with no tivo software running, and install and configure the device from the serial console. If that doesn't work either, then it probably isn't a 6.2 only issue.
Link negotiation
I really only tested with 1000 and 100 mbps switches. I didn't test half duplex on hubs. It's possible there is a problem negotiating the link type with the link partner. What are you connecting to?
More than one hardware version.
It could be that there is more than one version of this hardware, and the driver works with some (mine) but not others (yours). I think this is unlikely, but possible. The appears to be revision info in /proc/bus/usb/devices if you have usbdevfs mounted (mount -t usbdevfs usbdevfs /proc/bus/usb). My AGIGAUSB shows: Rev= 0.01.

[wbgolden]

You are at a point here where it is difficult to diagnose further without a serial console cable.
I'll get the serial cable hooked up tonight and see what I can figure out. In the meantime, I'll use company time to figure out how to get to the bare bash you described. Also need to figure out how to stop TeraTerm from truncating all the info the Tivo sends to it during the boot process. All the important stuff gets cut off before I can copy and paste it.

Thanks Jamie

[Jamie]

In the meantime, I'll use company time to figure out how to get to the bare bash you described. That's easy, assuming you already have the 3.1.5 killhdinitrd kernel installed. Just move rc.sysinit out of the way and replace it with one that just runs bash on the serial. For example (untested):#!/bin/bash
export MFS_DEVICE=/dev/hda10
export TIVO_ROOT=
/bin/bash --login -i </dev/ttyS2 &>/dev/ttyS2

[eastwind]

I'll get the serial cable hooked up tonight and see what I can figure out. In the meantime, I'll use company time to figure out how to get to the bare bash you described. Also need to figure out how to stop TeraTerm from truncating all the info the Tivo sends to it during the boot process. All the important stuff gets cut off before I can copy and paste it.

Thanks Jamie
Turn on logging to save the whole session. (File..Log)

ew

[Jamie]

{Moved from here (http://www.dealdatabase.com/forum/showthread.php?p=227609#post227609) to this thread as it seems a better fit.}

Jamie is the log2_irq_thresh=4 setting the default on the latest and future releases too, or just the one in that single post in the dev thread?

I ask because I went through all the soapbox steps except for the insmod switch and I'm not seeing any speed increases (still 1.75MBps ((14mbps))) with using the new mfs_ files. I'm using an unpatched mfs_ftp as I ran into the patching problem.


BTW, does this mean anything? Things appear to be OK despite this, but it looks suspicious. I suspect the device is hooked up later on and not here in the startup:
Code:

PCI: 00:0d.2 PCI cache line size set incorrectly (0 bytes) by BIOS/FW, correcting to 32 usb.c: USB device 2 (vend/prod 0x2001/0x1a00) is not claimed by any active drive r. PokeFirewallPort: Poking hole through firewall at port 2190, protocol udp iptables: No chain/target/match by that name PokeFirewallPort: Poking hole through firewall at port 2190, protocol tcp iptables: No chain/target/match by that name using driver frame iter
I've set the default log2_irq_thresh to 4. I expect it will stay there, but it might change again in the future if it looks advantageous.

The messages look normal. The only one unusual is the PCI cache line size message, but I've seen that before too though.

[Jeff D]

(DOH! Got mixed up on what thread I was posting to...)

Well, against what I thought would be true rebooting helped, the speeds are up to 18+Mbps. I'm above 2MBps so this is a huge improvement, but if I could get above 3MBps I'll try. I'm still looking for a different version of patch to use, I'm currently running AW's AIO package version. Off to that thread to follow up...

[Jamie]

(DOH! Got mixed up on what thread I was posting to...)

Well, against what I thought would be true rebooting helped, the speeds are up to 18+Mbps. I'm above 2MBps so this is a huge improvement, but if I could get above 3MBps I'll try. I'm still looking for a different version of patch to use, I'm currently running AW's AIO package version. Off to that thread to follow up...My mfs_ftp patch only helps insertion speed, and it's still considered experimental, as Riley and others have indicated they've had problems with it. You won't see any extraction speedup using the patched mfs_ftp.

I've forgotten what hardware you have, but I know some folks with hdtivos have said that the 3MB/sec speeds aren't achievable on that hardware. I also know that the series 2.5's are slower.

[Jeff D]

More stupid questions...
1) dlink adapter looking good, the transfer rates quoted above were for this adapter. The airlink adapter isn't so good, still around 1MBps. I've seen this:
USB device 3 (vend/prod 0xb95/0x7720) is not claimed
by any active driver.
And comments about editing the iptables files for these vend/prod problems. How does one edit that file?

2)These adapters should be hotswappable, right? The only way I can get the airlink adapter to work is by rmmod the driver files and re insmod the drivers. Other than that the adapater stay inactive, the tivo does show the line above about unclaimed device.

What can I do to get this working correctly?

[Jamie]

More stupid questions...
1) dlink adapter looking good, the transfer rates quoted above were for this adapter. The airlink adapter isn't so good, still around 1MBps. I've seen this:

And comments about editing the iptables files for these vend/prod problems. How does one edit that file?iptables is the low level linux firewall mechanism. I don't think that's what you mean. Perhaps /etc/hotplug/usb.map? I haven't messed with that file, but I believe it maps vendor/product id's to driver module names.2)These adapters should be hotswappable, right? The only way I can get the airlink adapter to work is by rmmod the driver files and re insmod the drivers. Other than that the adapater stay inactive, the tivo does show the line above about unclaimed device.It sounds to me like you have the wrong usbnet module loaded initially if it doesn't recognize that vendor/product id. Older usbnet modules won't recognize that device (and you'll get that message). Only the latest usbnet.o knows about it.What can I do to get this working correctly?I do all my usb module loading myself, before running the tivo rc.sysinit. That seems to work well for me with 7.1b. At some point I'll probably explore the hotplug stuff to see if I can get hot swapping to work with the new devices, but it isn't that high on my priority list. Maybe someone else will get to it first.

[Desperado]

Did you rename dhclient to keep it from running?


That was it. Thanks!

[wbgolden]

That's easy, assuming you already have the 3.1.5 killhdinitrd kernel installed. Just move rc.sysinit out of the way and replace it with one that just runs bash on the serial. For example (untested):#!/bin/bash
export MFS_DEVICE=/dev/hda10
export TIVO_ROOT=
/bin/bash --login -i </dev/ttyS2 &>/dev/ttyS2

and...
Turn on logging to save the whole session. (File..Log)

ew
Thanks to both of you. I will get those things ready to go and see what happens. Will post any solutions and/or other stuff if I find anything.

wbg

[Jeff D]

Perhaps /etc/hotplug/usb.map? I haven't messed with that file, but I believe it maps vendor/product id's to driver module names.It sounds to me like you have the wrong usbnet module loaded initially if it doesn't recognize that vendor/product id. Older usbnet modules won't recognize that device (and you'll get that message). Only the latest usbnet.o knows about it.

Yes, the map file was what I was looking for. Thanks. I've added it in, I'll report back how it works.

As to the usbnet.o file, the one in lib/modules is the latest one (or so I think)
-rw-r--r-- 1 0 0 41332 Jun 27 19:38 usbnet.o
Is there any way another one could be loading from somewhere else? The only old copy I have is in /lib/modules_bu (backup) and I don't think that would be used, but...
here's the output of find:

tivo-bedroom:/$ find / -name usbnet.o
/lib/modules/usbnet.o
/lib/modules_bu/usbnet.o
/hack/usb-2.4.27/obj/drivers/2.4.20/usbnet.o


I can't remember how to recursive grep with this grep witout the -R option, so I'm having a tough time finding what loads the usbnet.o file (I know it has to be in one of these support threads...)

[santa8claws]

Well I successfully installed the new drivers for my usb200m adapter, but I don't see any difference in FTP/TyTool transfer speed (pegged at 500kByte/sec) compared to the stock drivers... So I'm looking to see what I can fix/tweak...

Here's my boot:

usb.c: registered new driver usbdevfs
usb.c: registered new driver hub
usb.c: starting integrated EHCI driver
ehci_hcd EHCI-Direct: BRCM-EHCI
ehci_hcd EHCI-Direct: irq 8, pci mem fffe8170
usb.c: new USB bus registered, assigned bus number 1
ehci_hcd EHCI-Direct: illegal capability!
ehci_hcd EHCI-Direct: USB 0.0 enabled, EHCI 10.10, driver 2003-Dec-29/2.4
hub.c: USB hub found
hub.c: 2 ports detected
bcm-usb.h: TiVo 2-chip USB Host Controller
usb.c: EHCI initialization succeeded
hub.c: new USB device EHCI-Direct-2, assigned address 2
usb.c: USB device 2 (vend/prod 0x77b/0x2226) is not claimed by any active driver.
usb-ohci.c: USB OHCI at membase 0xfffe8100, IRQ 8
usb-ohci.c: usb-OHCI-Direct, BRCM-OHCI
usb.c: new USB bus registered, assigned bus number 2
hub.c: USB hub found
hub.c: 2 ports detected
bcm-usb.h: TiVo 2-chip USB Host Controller
usb.c: registered new driver usbnet
eth0: register usbnet usb-EHCI-Direct-2, ASIX AX8817x USB 2.0 Ethernet


My ifconfig:

eth0 Link encap:Ethernet HWaddr 00:10:60:84:BE:F4
inet addr:192.168.2.1 Bcast:192.168.2.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:43 errors:0 dropped:0 overruns:0 frame:0
TX packets:61 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
RX bytes:6353 (6.2 Kb) TX bytes:9031 (8.8 Kb)

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)


Any ideas or suggestions?
Thanks!
-- S8C

[Jamie]

As to the usbnet.o file, the one in lib/modules is the latest one (or so I think)
-rw-r--r-- 1 0 0 41332 Jun 27 19:38 usbnet.ook. time for a new theory: The hotplug stuff is unloading the usbnet driver when you unplug the dlink device, but not reloading it when you plug in the airlink.Is there any way another one could be loading from somewhere else? The only old copy I have is in /lib/modules_bu (backup) and I don't think that would be used, but...
here's the output of find:

tivo-bedroom:/$ find / -name usbnet.o
/lib/modules/usbnet.o
/lib/modules_bu/usbnet.o
/hack/usb-2.4.27/obj/drivers/2.4.20/usbnet.o


I can't remember how to recursive grep with this grep witout the -R option, so I'm having a tough time finding what loads the usbnet.o file (I know it has to be in one of these support threads...)Busybox is annoying...

You can pipe find into xargs and run grep there: find / -type f | xargs grep usbnet./

[Jamie]

Well I successfully installed the new drivers for my usb200m adapter, but I don't see any difference in FTP/TyTool transfer speed (pegged at 500kByte/sec) compared to the stock drivers... So I'm looking to see what I can fix/tweak...Transfer performance on the series 2.5 is bad. There doesn't seem to be a lot we can do about it. Still, I get ~1.3 MB/sec, so you should be able to do better than 500KB/sec t if the box is idle. There are three places where performance might be affected: the tivo, the PC, and in-between. You can elliminate the in-between as an issue by testing a direct point-to-point connection with a cross over cable.

[Jeff D]

Jamie, it appears your theory may be on the right track....(see below).

I patched the map file to include the adapter under the ax8817x section.

If I reboot with the airlink adapter connected it starts up working, this despite the report of an unclaimed device. All lights on and working.

If I swap out the airlink for the dlink the dlink powers up and works.
If I then swap out the dlink for the airlink the airlink doesn't come online.

-----

If I start from scratch with the airlink connected I can disconnect and reconnect the adapter and it powers up fine. If I then try to swap for the dlink the dlink doesn't come up.

here's the progress of mod load status as the operations happen:

Airlink boot:
lsmod
Module Size Used by Not tainted
ax8817x 26640 1
usb-ohci 23888 0 (unused)
usbcore 101936 0 [ax8817x usb-ohci]
router 66896 36
therm 1488 0
fan 2240 0
ideturbo 5216 1
irblast 3792 0 (unused)
brcmdrv-rb 1228256 138
cobra 91264 4
oslink 26112 12
ircatch 26832 2
i2c_Series2 14320 0 [therm fan cobra]
modemtty_Series2 4896 0
fpga 1920 2
tivoconfig 960 0 (unused)

Airlink disconnected:
ax8817x 26640 0
usb-ohci 23888 0 (unused)
usbcore 101936 0 [ax8817x usb-ohci]
router 66896 36
therm 1488 0
fan 2240 0
ideturbo 5216 1
irblast 3792 0 (unused)
brcmdrv-rb 1228256 138
cobra 91264 4
oslink 26112 12
ircatch 26832 2
i2c_Series2 14320 0 [therm fan cobra]
modemtty_Series2 4896 0
fpga 1920 2
tivoconfig 960 0 (unused)

Airlink reconnected:
ax8817x 26640 1
usb-ohci 23888 0 (unused)
usbcore 101936 0 [ax8817x usb-ohci]
router 66896 36
therm 1488 0
fan 2240 0
ideturbo 5216 1
irblast 3792 0 (unused)
brcmdrv-rb 1228256 138
cobra 91264 4
oslink 26112 12
ircatch 26832 2
i2c_Series2 14320 0 [therm fan cobra]
modemtty_Series2 4896 0
fpga 1920 2
tivoconfig 960 0 (unused)

Works!
Swap out airlink for dlink

Dlink connected:
Doesn't work!
Module Size Used by Not tainted
ax8817x 26640 0 (unused)
usb-ohci 23888 0 (unused)
usbcore 101936 0 [ax8817x usb-ohci]
router 66896 36
therm 1488 0
fan 2240 0
ideturbo 5216 1
irblast 3792 0 (unused)
brcmdrv-rb 1228256 138
cobra 91264 4
oslink 26112 12
ircatch 26832 2
i2c_Series2 14320 0 [therm fan cobra]
modemtty_Series2 4896 0
fpga 1920 2
tivoconfig 960 0 (unused)

Swap Dlink for Dlink:
Works!
Module Size Used by Not tainted
ax8817x 26640 1
usb-ohci 23888 0 (unused)
usbcore 101936 0 [ax8817x usb-ohci]
router 66896 36
therm 1488 0
fan 2240 0
ideturbo 5216 1
irblast 3792 0 (unused)
brcmdrv-rb 1228256 138
cobra 91264 4
oslink 26112 12
ircatch 26832 2
i2c_Series2 14320 0 [therm fan cobra]
modemtty_Series2 4896 0
fpga 1920 2
tivoconfig 960 0 (unused)

Dlink disconnected:
Notice what happened to ax8817x! usb-ochi also doesn't show ax8817x.

usb-ohci 23888 0 (unused)
usbcore 101936 0 [usb-ohci]
router 66896 36
therm 1488 0
fan 2240 0
ideturbo 5216 1
irblast 3792 0 (unused)
brcmdrv-rb 1228256 138
cobra 91264 4
oslink 26112 12
ircatch 26832 2
i2c_Series2 14320 0 [therm fan cobra]
modemtty_Series2 4896 0
fpga 1920 2
tivoconfig 960 0 (unused)


It seems there is something specific to the dlink unloading that removes ax8817x rather than just decrementing the use counter. This didn't happen when a) airlink adapters are swapped out, or dlink was swapped in and not claimed (didn't start up).

BTW: Since you asked and I never responded... HDVR2 (series 2) running 6.2 (2.4.20 kernel)

[jmrwiseguy]

Trying to update a Hughes HDVR2 6.2 system to USB2 drivers
Using the usbobj2.4.27-20050104.tar.bz2 (http://www.dealdatabase.com/forum/attachment.php?attachmentid=4258) files. The transfer speed I get from either ftp or from tytools is ~600k which is what I was getting before switching drivers. I'm going through a 100mb belkin router which is connected to a 100mb switch and that is connected to my DirecTivo via a linksys USB200M.

Here's what I did: I backed up (moved) from /lib/modules ax8817x.o, usbcore.o, and usb-ohci.o. The ehci-hcd.o file was not in /lib/modules.
Then, I copied in the following to /lib/modules:
./usb-2.4.27/obj/hosts/2.4.20_Series2/usbcore.o
./usb-2.4.27/obj/hosts/2.4.20_Series2/usb-ohci.o
./usb-2.4.27/obj/drivers/2.4.20/usbnet.o
set symlink usbnet.o to ax8817x.o

Finally I restarted Tivo and afterwards I don't see any performance improvement.

Here's the ifconfig output if it helps:
bash-2.02# ifconfig
lo Link encap:Local Loopback
inet addr:127.0.0.1 Bcast:0.0.0.0 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 coll:0

eth0 Link encap:Ethernet HWaddr 00:10:60:85:A9:21
inet addr:192.168.2.101 Bcast:192.168.2.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:2794 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:4765861 dropped:4248 overruns:0 carrier:0 coll:0

Attaching the console boot log.

Any ideas?

[Jeff D]

jmrwiseguy, not that I know what I'm talking about, but you didn't mention anything about the new mfs_ftp files. Did you update those too?

links are in the soapbox derby thread linked in post #1 of this topic.

[jmrwiseguy]

jmrwiseguy, not that I know what I'm talking about, but you didn't mention anything about the new mfs_ftp files. Did you update those too?

links are in the soapbox derby thread linked in post #1 of this topic.

No, I only updated the files that Jamie listed in his original post. I did look at the soapbox derby thread but it looked like it was only for the mfs tools stuff. For now, I'm just using tivoftpd for doing ftp so I didn't think I would need to update the mfs stuff. Perhaps somebody can tell me otherwise.

However, in that soapbox derby post there was a mention of a usbcore.o loading tweak -- I'm not using that -- should I?

Also, I didn't do the vwait patch either because it was for 4.01b.

Thanks,

[Jamie]

No, I only updated the files that Jamie listed in his original post. I did look at the soapbox derby thread but it looked like it was only for the mfs tools stuff. For now, I'm just using tivoftpd for doing ftp so I didn't think I would need to update the mfs stuff. Perhaps somebody can tell me otherwise.

However, in that Soapbox derby post there was a mention of a usbcore.o loading tweak -- I'm not using that -- should I?

Also, I didn't do the vwait patch either because it was for 4.01b.
The soapbox derby thread has a lot of good performance tuning tips and is probably a good place to continue to discuss "how do I get this to perform better?".

I always recommend getting some baseline bandwidth measurements with netperf (http://www.dealdatabase.com/forum/showthread.php?t=39214) before worring about application level performance. If you can't get good netperf numbers, you're unlikely to get good application performance.

As I mentioned earlier:There are three places where performance might be affected: the tivo, the PC, and in-between. You can elliminate the in-between as an issue by testing a direct point-to-point connection with a cross over cable.You might try a crossover cable to eliminate your switch and router devices as culprits in the slowdown. Yoiu might also want to look at the later versions of the drivers, which have some additional support for "pause frame" flow control. It might make a difference for you, depending on your switch and router.

mfs-utils (mfs_* tools) has an improved tserver (for tytool) and improved import/export utilities (for mfs_ftp). This can help your overall transfer performance since they reduce the cpu load on the tivo during a transfer. Once you have good base network performance, it makes sense to start looking for performance improvements on the application side.

I know that 7.1x doesn't have the vwait problem. I'm not sure, but I don't think 6.2 has it either. In general though, it makes sense to run top on the tivo and look for processes that are hogging cpu. Tivo network transfers are almost always cpu limited.

[wbgolden]

You are at a point here where it is difficult to diagnose further without a serial console cable.

There are several possibilities that might be causing your problems and those of gpontis, described elsewhere:

6.2 network configuration issues
One way to eliminate this as a possibility is to boot up to a bare bash, with no tivo software running, and install and configure the device from the serial console. If that doesn't work either, then it probably isn't a 6.2 only issue.
Serial console is all set up and working like a champ (man, I'm really glad I finally took this step...). I got to the bare bash and insmod'ed the drivers. Then I was unsure of what to do, so I set the rc.sysinit file back the way it was before. Thought I probably needed to set up network configs but wasn't really sure (or confident) in what to do.

Link negotiation
I really only tested with 1000 and 100 mbps switches. I didn't test half duplex on hubs. It's possible there is a problem negotiating the link type with the link partner. What are you connecting to?
Belkin Wireless-G router. Symptoms are the same whether I connect wirelessly or wired.
More than one hardware version.
It could be that there is more than one version of this hardware, and the driver works with some (mine) but not others (yours). I think this is unlikely, but possible. The appears to be revision info in /proc/bus/usb/devices if you have usbdevfs mounted (mount -t usbdevfs usbdevfs /proc/bus/usb). My AGIGAUSB shows: Rev= 0.01.
I cannot, for the life of me, get anything to come up when I try to open the devices file, so I can't tell what the version is. Again...I'm sure it's just something I am doing wrong.

I'm thinking that at this point, I need to bone up more on Linux commands. To tell the (brutal) truth, right now I feel like a total noob who's well-read on DDB (many months) but is still missing some really important aspects. Maybe I'm just better at lurking and waiting for answers... :D

Nonetheless, I have attached my boot log from the last reboot. Hey, at least I'm trying, right? :cool: There's a bunch of "Inode" alerts, but I figure I'll cross that bridge at another time.

[EDIT] I should note that reverting back to the original drivers works fine with a Hawking HUF11 or Belkin F5D5050 USB 1.1 adapter.

[Jamie]

Serial console is all set up and working like a champ (man, I'm really glad I finally took this step...). I got to the bare bash and insmod'ed the drivers. Then I was unsure of what to do, so I set the rc.sysinit file back the way it was before. Thought I probably needed to set up network configs but wasn't really sure (or confident) in what to do. The basic idea is to insmod the modules, ifconfig the interface, and start testing. It can be useful to mount /proc and /tmp first. Without /tmp, you have no place to put files.mount -o /var
mount -t proc proc /proc
insmod /lib/modules/usbcore.o
insmod /lib/modules/usbnet.o
ifconfig eth0 192.168.1.200 netmask 255.255.255.0 # substitute your tivo IP here
ping 192.168.1.100 # subtitute your pc IP here
netperf -H 192.168.1.100 -C -c -- -m 131072 # netperf test to your PC running netserver

You can startup up tivoftpd and transfer files if you want/need to.

I'm thinking that at this point, I need to bone up more on Linux commands. To tell the (brutal) truth, right now I feel like a total noob who's well-read on DDB (many months) but is still missing some really important aspects. Maybe I'm just better at lurking and waiting for answers... :We're on the bleeding edge here. I don't mind pointing you in the right direction once in a while, but if you're getting frustrated, let it go for now and the device support will no doubt get smoother downstream.
Nonetheless, I have attached my boot log from the last reboot. Hey, at least I'm trying, right? :cool: There's a bunch of "Inode" alerts, but I figure I'll cross that bridge at another time.Looks like you didn't attach the log. Sounds like the inode alerts are normal e2fsck warnings when the /var file system is repaired after an unclean shutdown. These are par for the course with an unjournalled file system.

[wbgolden]

We're on the bleeding edge here. I don't mind pointing you in the right direction once in a while, but if you're getting frustrated, let it go for now and the device support will no doubt get smoother downstream.
I agree. Thanks for your time, Jamie. I will keep a watch on what's happening with the Airlink device(s), but not spend a bunch more of your (et. al) and my time trying to be "the guy that got the ASOHOUSB working with--his--6.2 Directivo"...lol

[corny]From all I've seen you've done great things on this forum and have been more than patient with people like me. Again, thank you[\corny]

wbg

[Jeff D]

Jamie, any idea where I should go to look into the hotswap problem?

And an update to the other stuff. Today my AV receiver for the bedroom came back so when installing it I made some new network cables for that and the tivo. The airlink adapter was disconnected and reconnected, then I started up some transfers. Was I shocked to see transfer rate of 23Mbps!!! And this on a transfer that was saving to a mapped network drive, even with the extra traffic it was fast. Guess I may have a "bad" cable.

[imaloserbaby]

Not wishing to pull this off thread, but I found that the only way I could get any joy at all with networking on 6.2 was by mimmicking the way 4.x configures the adapters, using the netconfig.tcl script in this thread:

http://www.dealdatabase.com/forum/showthread.php?t=43375&page=6

That said, I have read people who did have success, and I think I also read that the insmod and ifconfig need to happen very early on in the boot sequence (hence Alphawolf's 6.2 init configs that run an /etc/rc.runfirst script). That said, this could be merely relating to MRV which is definitely to be discussed in another thread. Sorry if it's a red herring, but it can't hurt to try this perhaps.

[Jamie]

Jamie, any idea where I should go to look into the hotswap problem?Step 1 would be to learn about how linux hotplug works in general. There's a paper here (http://www.kroah.com/linux/talks/ols_2001_hotplug_paper/hotplug.ps) and a website here (http://linux-hotplug.sourceforge.net/).

Step 2 would be to learn what's different the way TiVo sets things up. The first observation is that /sbin/hotplug is a program rather than a shell script. Is there source available? Not that I know of, but I haven't looked very hard. Without source, you can try to reverse engineer it by treating it as a black box (feed it inputs/env settings and see what it seems to do) and/or disassemble and study the code. You'll find that the /etc/hotplug/usb.map file is referenced from tivoapp, not hotplug. Perhaps routerplus/rpsniff could help understand the IPC.

I haven't completed these two first steps myself, and as I said, it's not high on my priority list at present, as I can live without hotplugging for now.

[Jeff D]

wbgolden, I too have got the ASOHOUSB and AGIGAUSB adapters both running on my 6.2 directivos. So it can be done, it was very simple for me... it's good you have serial bash setup otherwise it's a pain to do (unless you script it)

For those that need it...
In the /etc/hotplug/usb.map file you need to add these two lines to the
device ax8817x section:

product 0b95 7720 Airlink ASOHOUSB
product 0b95 1780 Airlink AGIGAUSB

[Jamie]

wbgolden, I too have got the ASOHOUSB and AGIGAUSB adapters both running on my 6.2 directivos. So it can be done, it was very simple for me... it's good you have serial bash setup otherwise it's a pain to do (unless you script it)Glad you got it working. Did you sort out the mysteries you reported in post #20?

[Jeff D]

Glad you got it working. Did you sort out the mysteries you reported in post #20?

Nope, hotswapping the dlink adapter out kills something. I can swap the airlink adapters all day. If you see in my code section of post #20 that unplugging the airlink adapters correctly "frees" and dec's the use counter and and then plugging in inc's the use count. The dlink just wipes out the module competely when removed. There was the one odd time where the dlink adapter wasn't identified when plugged in, but I'm going to say that was user error.

As you suggested... how many times are you really going to be hotswapping this thing?!?! Other than testing I don't know why either. So, I've just let it go. I have been looking at the links you provided, but I suspect I won't find time to get into it. The ramp up on linux development would be enough to turn me off to the whole thing...

BTW, I haven't tried the GIGA at giga speeds, still just using it at the 100 speed.

[Jeff D]

wbgolden, if it helps, I've documented my steps, I can send them to you if you want to see how someone else has done it...

[wbgolden]

wbgolden, if it helps, I've documented my steps, I can send them to you if you want to see how someone else has done it...
That would be cool. I could compare what you've done to get them working with that I am doing to not get them working. I've PM'ed you my email address. Thanks :D

[hxmiller]

Any idea what driver might work with this? Is it proprietary?

http://www.prolific.com.tw/eng/Products.asp?ID=18

[ochs38578]

Hi,

It looks like I am getting a significant number of transmit errors / dropped xmit packets using the usbnet driver on a hr10-250. I have usbcore.o, ehci-dummy.o and usbnet.o loaded, and nothing else other than the default modules. The network card is a netgear FA-120. Any ideas what is causing this or if there is a fix for it? As you can see below there are well over 2000 of them in 3 seconds while the box is pretty much idle from a network standpoint

bash-2.02# ifconfig -a
lo Link encap:Local Loopback
inet addr:127.0.0.1 Bcast:0.0.0.0 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:1 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:98 dropped:1 overruns:0 carrier:0 coll:0

eth0 Link encap:Ethernet HWaddr 00:0F:B5:84:CC:DD
inet addr:192.168.1.100 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:31727 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:80886700 dropped:62678 overruns:0 carrier:0 coll:0

bash-2.02# date
Sun Jul 10 17:31:16 UTC 2005
bash-2.02# ifconfig -a
lo Link encap:Local Loopback
inet addr:127.0.0.1 Bcast:0.0.0.0 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:1 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:98 dropped:1 overruns:0 carrier:0 coll:0

eth0 Link encap:Ethernet HWaddr 00:0F:B5:84:CC:DD
inet addr:192.168.1.100 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:31754 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:80888433 dropped:62697 overruns:0 carrier:0 coll:0

bash-2.02#

Thanks,
James

[eastwind]

Hi,

It looks like I am getting a significant number of transmit errors / dropped xmit packets using the usbnet driver on a hr10-250. I have usbcore.o, ehci-dummy.o and usbnet.o loaded, and nothing else other than the default modules. The network card is a netgear FA-120. Any ideas what is causing this or if there is a fix for it? As you can see below there are well over 2000 of them in 3 seconds while the box is pretty much idle from a network standpoint

..snip..

Thanks,
James
Might be a bad copy of ifconfig. I think I remember reading about one that falsely reports a lot of errors. Mine is version [bash DTiVo(7): ~] $ifconfig --version
net-tools 1.432
ifconfig 1.29 (1997-09-23) and it reports a lot of errors also.

ew

EDIT:
Just installed the newer verions and it reports no errors:[bash DTiVo(7): ~] $ifconfig --version
net-tools 1.60
ifconfig 1.42 (2001-04-13) This is the one in usbobj2.4.27-20050619.tar.bz2.

[JustDan]

I just finished moving my systems over to pure gigabit, with the
SMC switches.

I suspect a problem with my Windows box as all ftp/http sessions
to JF enabled systems hangs. If I lower the MTU on the servers
to 8000 it works.

I also noticed the high errors on the AirLink, but realized something
else-
eth0 Link encap:Ethernet HWaddr 00:12:0E:02:C7:66
inet addr:199.254.167.50 Bcast:199.254.167.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:43394 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:147761691 dropped:97897 overruns:0 carrier:0 coll:
0

Note the TX Packet count. It looks like the counters are 'shifted'

Either way I still get my best performance using MTU=1500.
Specs-
HR10-250
AirLink GB USB 2.0
SMC 8508T switches (2)
Tytool (HD Beta3)
Broadcom 5700 series nic in Windows PC

Dan

[eastwind]

I also noticed the high errors on the AirLink, but realized something
else-Dan
Looks like the same problem as ochs38578. Use the ifconfig from the usb drivers package (in nonusb/user). It is the counters in a different format than expected.

ew

[Jamie]

Any idea what driver might work with this? Is it proprietary?

http://www.prolific.com.tw/eng/Products.asp?ID=18I see that BSD supports it through the cdce module, and all CDC (Communction Device Class) support in linux is in the usbnet driver. So, if anything, I'd expect in the usbnet module in linux to suport it. If it does not, it might just be a matter of adding the vendor/product id (that was the case for the ALI host-to-host cables). If you have such a cable, send me the vendor/product id's and we can try. The vendor has a data sheet, and bsd supports it, so it might not be hard to add linux support if it isn't already there.

[Jamie]

I just finished moving my systems over to pure gigabit, with the
SMC switches. ...Are you sure your windows systems are using jumbo frames? I suspect not if you get better performance with an mtu of 1500. The device does support jumbo frames. I think with my intel GigE devices I had to muck with some settings in the device manager to get XP to use jumbo frames. Examine all the options for that device carefully in the windows device manager.

As EW noted, the ifconfig issue is well known and has been covered before: use the ifconfig included in the backport package.

[JustDan]

Are you sure your windows systems are using jumbo frames? I suspect not if you get better performance with an mtu of 1500. The device does support jumbo frames. I think with my intel GigE devices I had to muck with some settings in the device manager to get XP to use jumbo frames. Examine all the options for that device carefully in the windows device manager.


I believe it is using Jumbo Frames. The embedded Gb card didn't
support them, so I dug out an older 3Com card I had which claims to.
I plan to try an Intel card to make sure though.

I should have been clearer, I was getting close to 3Mb/s with the
USB200M, but around 2Mb/s using the Airlink and 8k jumbo frames.
I only get around 1Mb/s using the 1500 MTU.

Dan

[Jamie]

I've attached some tools that may help identify whether you are really getting jumbo frames end-to-end. These are the ping and tracepath compiled for a series 2 from the Fedora Core 3 iputils-20020927-16 sources.

This ping supports a 'no fragment' hint that can be used to see if packets are getting fragmented. If you send too large a packet, instead of fragmenting it, you'll get a message back that the message was too long:bash-2.02# ping -M do -s 8973 192.168.1.100
PING 192.168.1.100 (192.168.1.100) 8973(9001) bytes of data.
ping: local error: Message too long, mtu=9000This message is one byte too long for my mtu.

Note that some versions of ping (for example, the windows versions (http://www.microsoft.com/resources/documentation/windows/xp/all/proddocs/en-us/ping.mspx)) may use -f to indicate don't fragment. In the ping I'm posting here, -f means "flood ping" which you probably only want to use when testing for packet loss.

Here's the -M options from the man page: -M hint
Select Path MTU Discovery strategy. hint may be either do
(prohibit fragmentation, even local one), want (do PMTU discov-
ery, fragment locally when packet size is large), or dont (do
not set DF flag).

I also included tracepath, which does path mtu discovery and displays the lowest mtu along the path:bash-2.02# tracepath 192.168.1.100
1?: [LOCALHOST] pmtu 9000
1: fedora.home.lan (192.168.1.100) 2.709ms reached
Resume: pmtu 9000 hops 1 back 1

[JustDan]

It appeared I was half JF enabled.

Using the tools provided I can confirm that I cannot use anything
larger than 1500 to connect to my desktop. Oddly when using the
windows ping with -f -s I can send JF up to 28 bytes under the
configured MTU.

So the NIC appears to be sending JFs, but cannot receive them.
Since that doesn't make sense, I started turning off the protection
tools required by work (McAfee, Windows Firewall, etc...)

And the culprit is- Windows Firewall. With it off I am seeing over
5Mb/s.

Thanks for the tools and help. Time for some serious extration

[Jamie]

...
And the culprit is- Windows Firewall. With it off I am seeing over
5Mb/s ....
Somehow it doesn't surprise me that windows firewall was the culprit :-)

I'm glad to see that the HDTiVo's are capable of good performance. In past (http://www.dealdatabase.com/forum/showthread.php?p=199851#post199851) reports it was capped at 2.4MB/sec.

[JustDan]

With your first release of the unified tools and updated USB drives
I could get 3Mb/s with an idle system. When the topic of the
Airlink surfaced, I had high hopes.

So far I have seen a peak of 5.4Mb/s, but sustained 5.25~5.3Mb/s
are normal so far. I ended up with a MTU of 8970. It appears the
NIC drivers on my PC include the header and have an upper limit of
9k.

So thanks again for your efforts. It is really cool to be able to
extract, but to be able to do it in less than 1/2 real time makes
it a joy.

[jkozee]

Has anyone had any luck with the backported drivers on a series2.5 (Humax T800) running 7.1b?

The only hacks I have on the unit are a hacked prom, tivoftpd, busybox, and an rc.sysinit.author that starts ftp, tnlited, and bash.

The symptoms I have seem to be strange. The device comes up ok, but when I start using the eth interface it stops responding. For instance I ftp into the Tivo all is fine until the LIST command. I can telnet in but if I do ls, it lists a few files then just stops. If I telnet in and do a ping it is fine until I interrupt it or if I issue a ping for say 5 packets then it stops when tring to return to the prompt. None of the logs nor ifconfig seem to indicate any problems.

I've tried a 200M and GIGAUSB with the latest releases from Jamie and ADH and the results are the same. I don't believe I had this problem until the 7.1b update. Any suggestions would be greatly appreciated. I can post additional info if needed when I am at the unit.

I'd also like to say thanks to all the gurus here that given this mere mortal the enjoyment (and frustration) of hacking my box.

[eastwind]

Not sure about this as I don't have a prom-modded Humax. Do you know about the netfilter-enable and iptables fixes? Did you make either or both of these alterations. (You might not be able to connect at all without these mods...just bouncing ideas.)

ew

[jkozee]

Do you know about the netfilter-enable and iptables fixes? Did you make either or both of these alterations. (You might not be able to connect at all without these mods...just bouncing ideas.)

ew

I think you are correct, without fixing the netfilter-enable I wouldn't be able to connect. I hacked netfilter-enable myself (I couldn't delete or use netfilter-disable) and I tried the iptables hack which bypasses netfilter-enable altogether (I believe). So, I think I'm safe here. Here's the iptables hack for those who find themselves here and need it.


cd /sbin
mv iptables iptables.old
echo -e '#!/bin/bash\nexit 0' > iptables
chmod 755 iptables

I also tried ADH's 2.4.20 kernel without netfilter as well as rolling my own. Keep bouncing ideas, I'm hoping I there is something simple to get these working!

[Jamie]

Has anyone had any luck with the backported drivers on a series2.5 (Humax T800) running 7.1b? I've had problems with the backport drivers my 540 series2.5 7.1b unit too.

What's working well for me is to use the TiVo supplied usb host controller modules (usbcore, ehci-hcd, usb-ohci) and replacing only the usbnet module with the one from the backport package. I'm currently using it in this mode with a AGIGAUSB and it seems to be reliable, though the performance still doesn't match a Series2. For example, I get 3-4MB/sec extraction speed with mfs_ftp instead of 5-6 MB/sec with a series2.

[jkozee]

What's working well for me is to use the TiVo supplied usb host controller modules (usbcore, ehci-hcd, usb-ohci) and replacing only the usbnet module with the one from the backport package.
THANKS! I will try this tonight. Do you think the mismatch is to blame, or is it the harwdware of the s2.5 itself?

[Jamie]

THANKS! I will try this tonight. Do you think the mismatch is to blame, or is it the harwdware of the s2.5 itself?The bad performance seems to just be a 'feature' of the s2.5 hardware. I know I get about 2/3s the memory bandwidth out of 'hdparm -T /dev/hda' on a s2.5 compared to a s2. My unverified conjecture is that the unified architecture (the BCM7317 on a s2.5 does the work of several chips from a s2) ends up restricting memory bandwidth available to the cpu.

[jkozee]

Jamie,

You hit the nail on the head. After adding the agigausb to the usb.map and just using the usbnet module, I now have a stable network connection on my s2.5 with 7.1b. Thanks for the assistance.

[cyklfreak]

ok Jamie, bear with me for a second,

why would I get power to the adapter but not able to communicate?

is there a boot log that I look at or turn on to see what is needed fron the serial cable(again, I cant get the one I bought to work)

thanks

B

[Jamie]

ok Jamie, bear with me for a second,

why would I get power to the adapter but not able to communicate?

is there a boot log that I look at or turn on to see what is needed fron the serial cable(again, I cant get the one I bought to work)

thanks

BYour first order of business is to get the serial console cable working. I'd sugest you research that first. If you are really stuck, post in the newbie forum. I can't be of any help until you have serial console access.

The ax88772 support is relatively new. Works for me and several others, but it's possible you got a different revision of the hardware or something. More likely, you're just doing something wrong.

There are logs in /var/log. It's not nearly as useful as having bash on the serial console, but you could pull the drive to access the logs to see if there was anything informative there.

[jasch]

Are these changes compatible with 6.1?

I had networking working on my R10, and decided to try the new drivers... They've worked on all my other DTivo units flawlessly, but on the R10 I get this while booting (and no network)...

I installed them the same way I've installed the others (being as the R10 has the same 3.1.5 kernel I used the 2.4.20 drivers)

EDIT: Never mind. Stupid mistake (needed to use the Gen04 files for this one). It's working OK now.

Thanks


Found hpk front panel model 1
/lib/modules/usbcore.o: unresolved symbol pci_register_driver
/lib/modules/usbcore.o: unresolved symbol pci_enable_device
/lib/modules/usbcore.o: unresolved symbol pci_read_config_byte
/lib/modules/usbcore.o: unresolved symbol pci_read_config_dword
/lib/modules/usbcore.o: unresolved symbol pci_set_dma_mask
/lib/modules/usbcore.o: unresolved symbol pci_unregister_driver
/lib/modules/usbcore.o: unresolved symbol pci_set_master
/lib/modules/usbcore.o: unresolved symbol pci_write_config_dword
/lib/modules/usbcore.o: unresolved symbol pci_set_mwi
/lib/modules/usb-ohci.o: unresolved symbol usb_claim_bandwidth
/lib/modules/usb-ohci.o: unresolved symbol pci_register_driver
/lib/modules/usb-ohci.o: unresolved symbol usb_release_bandwidth
/lib/modules/usb-ohci.o: unresolved symbol usb_check_bandwidth
/lib/modules/usb-ohci.o: unresolved symbol pci_enable_device
/lib/modules/usb-ohci.o: unresolved symbol usb_alloc_bus
/lib/modules/usb-ohci.o: unresolved symbol pci_disable_device
/lib/modules/usb-ohci.o: unresolved symbol usb_free_dev
/lib/modules/usb-ohci.o: unresolved symbol usb_inc_dev_use
/lib/modules/usb-ohci.o: unresolved symbol usb_deregister_bus
/lib/modules/usb-ohci.o: unresolved symbol usb_disconnect
/lib/modules/usb-ohci.o: unresolved symbol usb_connect
/lib/modules/usb-ohci.o: unresolved symbol usb_new_device
/lib/modules/usb-ohci.o: unresolved symbol pci_find_slot
/lib/modules/usb-ohci.o: unresolved symbol pci_unregister_driver
/lib/modules/usb-ohci.o: unresolved symbol usb_root_hub_string
/lib/modules/usb-ohci.o: unresolved symbol pci_set_master
/lib/modules/usb-ohci.o: unresolved symbol usb_alloc_dev
/lib/modules/usb-ohci.o: unresolved symbol usb_register_bus
/lib/modules/usb-ohci.o: unresolved symbol usb_free_bus
PokeFirewallPort: Poking hole through firewall at port 2190, protocol udp
PokeFirewallPort: Poking hole through firewall at port 2190, protocol tcp

[fredfillis]

I've followed the basic instructions to install the ubs2.0 drivers to my modules folder but I'm not getting much in the way of performance gain, at least nowhere near anyone else.

Basically all I have done is to delete the old, put in the new and symlink usbnet.o to ax8817x.0. I have not implemented the usbcore.o loading tweak: (insmod /lib/modules/usbcore.o log2_irq_thresh=4) since I read on one of these threads that it is not required under 6.2

When I'm transferring from tivo to pc with mfs_ftp I get about 0.7MB/s and about half that going the other way. I have also installed the unified mfs_* utils. The speed I'm getting now is def. better than before under what I assume where usb1.1 drivers

My netperf numbers are pretty ordinary too.

bash-2.02# netperf -H 192.168.100.110
TCP STREAM TEST to 192.168.100.110
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec

8192 16384 16384 9.99 14.03

Netserver running on tivo
C:\Documents and Settings\Ray\Desktop\netperf>netperf -H 192.168.100.103
TCP STREAM TEST to 192.168.100.103
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec

87380 8192 8192 10.00 9.32

The tivo gets it's IP from the router (reserved ip for the tivo's mac address) and there is also a switch in the circuit. I'm using a USB200M on the tivo.

The speed I'm getting is pretty ordinary.

Do I need to insmod all those drivers in a test.conf file as others are doing? Not sure exactly where to start to get a better result.

[wesmoc]

I went ahead and updated the modules based upon the latest release to attempt to enable USB 2.0 on my Series 2 which I just upgraded to 6.2 (hey, I backed up the entire modules directory, so I was ok).

I'm running v6.2 with a Linksys USB200M (wired) adapter. From what I can see, USB 2.0 was happily enabled upon reboot. (BTW, I didn't make any symbolic link to ax8817x.o module.. I've seen mention of it by others, but it was never in the instructions or initial "how-to" type of postings)

usb.c: registered new driver usbdevfs
usb.c: registered new driver hub
usb.c: starting integrated EHCI driver
ehci_hcd 00:0d.2: PCI device 1033:00e0
ehci_hcd 00:0d.2: irq 6, pci mem b4002000
usb.c: new USB bus registered, assigned bus number 1
PCI: 00:0d.2 PCI cache line size set incorrectly (0 bytes) by BIOS/FW, correcting to 32
ehci_hcd 00:0d.2: USB 2.0 enabled, EHCI 0.95, driver 2003-Dec-29/2.4
hub.c: USB hub found
hub.c: 5 ports detected
usb.c: EHCI initialization succeeded
usb-ohci.c: USB OHCI at membase 0xb4000000, IRQ 5
usb-ohci.c: usb-00:0d.0, PCI device 1033:0035
usb.c: new USB bus registered, assigned bus number 2
hub.c: USB hub found
hub.c: 3 ports detected
hub.c: new USB device 00:0d.2-1, assigned address 2
usb.c: USB device 2 (vend/prod 0x77b/0x2226) is not claimed by any active driver.
usb-ohci.c: USB OHCI at membase 0xb4001000, IRQ 3
usb-ohci.c: usb-00:0d.1, PCI device 1033:0035
usb.c: new USB bus registered, assigned bus number 3
hub.c: USB hub found
hub.c: 2 ports detected
usb.c: registered new driver ax8817x
AX8817x USB driver v0.9.4
eth0: Link established, 100 Mbit, full duplex, flow control disabled


The throughput did improve, but not near what others have been reporting. Originally, after my 6.2 upgrade (I cannot remember what it was like before), I was pulling down at like 0.45MB/sec (via tytool), but after the backport driver install, I'm getting 1.46MB/sec.

Is there anything else that can be tweaked (eg: do the "vwait" and "usbcore loading tweak" fixes apply to 6.2 in any way.. if so, how?)?


Just to check to see if I have this right.. in the obj distribution, you're saying, for a 6.2 install:

cd /lib; tar cpf modules_bak.tar modules
cp -r /var/tmp/usb-2.4.27/obj/hosts/2.4.20_Series2/*.o /lib/modules
cp -r /var/tmp/usb-2.4.27/obj/drivers/2.4.20/*.o /lib/modules (Is this overkill? Copying all of them over shouldn't hurt anything...)


And, from there, nothing else is needed to enable USB 2?

[eastwind]

I went ahead and updated the modules based upon the latest release to attempt to enable USB 2.0 on my Series 2 which I just upgraded to 6.2 (hey, I backed up the entire modules directory, so I was ok).

I'm running v6.2 with a Linksys USB200M (wired) adapter. From what I can see, USB 2.0 was happily enabled upon reboot. (BTW, I didn't make any symbolic link to ax8817x.o module.. I've seen mention of it by others, but it was never in the instructions or initial "how-to" type of postings)

<snip>

The throughput did improve, but not near what others have been reporting. Originally, after my 6.2 upgrade (I cannot remember what it was like before), I was pulling down at like 0.45MB/sec (via tytool), but after the backport driver install, I'm getting 1.46MB/sec.

Is there anything else that can be tweaked (eg: do the "vwait"
and "usbcore loading tweak" fixes apply to 6.2 in any way.. if so, how?)?
You're not seeing the throughput increases because you're still using the stock ax8817x.o module. If you want to use the built in support from usbnet.o (and get the enhancements from that module) you'll need to symlink like you've already been instructed.

ew

[Jamie]

... (BTW, I didn't make any symbolic link to ax8817x.o module.. I've seen mention of it by others, but it was never in the instructions or initial "how-to" type of postings)...
...
usb.c: registered new driver ax8817x
AX8817x USB driver v0.9.4
eth0: Link established, 100 Mbit, full duplex, flow control disabled
...Is there anything else that can be tweaked (eg: do the "vwait" and "usbcore loading tweak" fixes apply to 6.2 in any way.. if so, how?)?It looks to me like you are using the tivo supplied ax8817x driver instead of the usbnet driver. It's older and may not perform as well. I'd suggest you remove it (since you already backed it up) and make the symbolic link from usbnet.o to ax8817x.o. :# cd /lib/modules
# rm ax8817x.o
# ln -s usbnet.o ax8817x.oThe vwait issue doesn't apply on 6.2, AFAIK. There is no need to tweak module parameters to get better performance with the latest drivers.

[wesmoc]

Cool.. Fixed, and rebooted. I'll give it an hour to calm down, but, even with it only up for less than 5 minutes, I'm getting 1.75MB/sec, which is a little better. :)

[Gunnyman]

Can someone clear something up for me please?
A Dtivo running 6.2 hacked with killhdinitrd runs the 2.4.20 kernel right?
this kernel has support for wireless G on 7.x tivo software.
What is keeping these drivers from being inserted into 6.2 thus bringing wireless G to that software?
What am I missing?

[Jamie]

Can someone clear something up for me please?
A Dtivo running 6.2 hacked with killhdinitrd runs the 2.4.20 kernel right?
this kernel has support for wireless G on 7.x tivo software.
What is keeping these drivers from being inserted into 6.2 thus bringing wireless G to that software?
What am I missing?It's been suggested before: link (http://www.dealdatabase.com/forum/showthread.php?t=45033). I'm not sure that anyone has gotten it to work.

[Gunnyman]

Yeah,
Thanks Jamie I saw that thread the other day, I hope this becomes a reality because my house is nearly impossible to get cat 5 where I want it.

[richr69]

Trying to get the Airlink USB 2.0 adapter to work properly (ASOHOUSB)

I performed the following for the Series 2 Tivo w/2.4.20 kernel, using USB2 backport 2.4.27

- back up /lib/modules
- delete ax8817x.o, usbnet.o, usbcore.o, usb-ohci.o, usb-storage and (if it exists) ehci-hcd.o
- copied in the new modules (drivers/hosts): usbnet.o, usb-storage.o, usb-ochi.o, usbcore.o and ehci-dummy.o
- ln -s usbnet.o ax8817x.o

Added the following to the /etc/hotplug/usb.map file under the device ax8817x section:

product 0b95 7720 Airlink ASOHOUSB
product 0b95 1780 Airlink AGIGAUSB

After performing the above the Tivo boots up successfully and the link LED is displayed on the adaptor. My router LEDs also light up. However, my SMC7004VBR router does not show an IP address in the DHCP client log for the Tivo. Also the bash output indicates the following:

USB device 3 (vend/prod 0xb95/0x7720) is not claimed by any active driver.

When I enter ifconfig at the bash prompt, it does display eth0 with an IP address, but its not what my router would assign. I tried pinging it but nothing happens. So I decided to rename the tivo dhclient as follows.

/sbin/dhclient to /sbin/dhclient.orig

Again the Tivo boots up successfully. Again the bash output indicates the following:

USB device 3 (vend/prod 0xb95/0x7720) is not claimed by any active driver.

At this point my router DHCP client log still does not display an IP address for the tivo.

I decide to try the following:

ifconfig eth0 192.168.2.150

At this point I was able to telnet.

So my questions are:
1.) Why does my bash log still indicate
USB device 3 (vend/prod 0xb95/0x7720) is not claimed by any active driver.

2.) Why isn’t an ip address automatically assigned to the tivo. Am I missing something in my steps that causing the DHCP to anot assign an IP address automatically.

[Jamie]

1.) Why does my bash log still indicate
USB device 3 (vend/prod 0xb95/0x7720) is not claimed by any active driver.I believe this is normal. The message is printed before hotplug loads the driver.2.) Why isn’t an ip address automatically assigned to the tivo. Am I missing something in my steps that causing the DHCP to anot assign an IP address automatically.What tivo software version and kernel are you using? There seems to be an issue with dhcp and the 3.1.5 kernel on 7.x: to use dhclient, you either need to find and load a 2.4.20 af_packet.o module, or find a kernel compiled with CONFIG_PACKET=y. This is something that changed between 3.1.5 and 7.x: packet socket support, required by the dhclient, is compiled into the newer kernels, while it used to be in the af_packet.o module.

If you run dhclient by hand on the serial console and see something like this:socket: Address family not supported by protocol
Make sure to set CONFIG_PACKET=y and CONFIG_FILTER=y in your kernel configuration!
exiting.then you likely have this problem.

[richr69]

I'm running 6.2 w/2.4.20 kernel that has been killhdinitrd.

I'll try what you suggested. Thanks for your help.

[ebeighe]

I, too, got one of these Airlink ASOHOUSB usb2 adapters. ($5 at Fry's, who could resist?)
I set up my /lib/modules by over-copying (after backing up the originals, of course!) all the backported obj's for 2.4.20 / Series 2 -- I have a Toshiba SD-H400.
After edited the /etc/hotplug/usb.map, and added a sym link from ax8817x.o to usbnet.o -- everything works! Thanks for the objs!

Anyway, I'm not understanding what is the significance of the ehci-dummy.o file? My originaly /libs have a ehci-hcd.o (which is still there).

Should i be deleting the original ehci-hcd.o?

The airlink adapter appears to be working off of the usb-ohci.o driver -- is that right? I thought since my new adapter was USB 2.0 it would be working off of ehci?

[Jamie]

I, too, got one of these Airlink ASOHOUSB usb2 adapters. ($5 at Fry's, who could resist?)
I set up my /lib/modules by over-copying (after backing up the originals, of course!) all the backported obj's for 2.4.20 / Series 2 -- I have a Toshiba SD-H400.
After edited the /etc/hotplug/usb.map, and added a sym link from ax8817x.o to usbnet.o -- everything works! Thanks for the objs!

Anyway, I'm not understanding what is the significance of the ehci-dummy.o file? My originaly /libs have a ehci-hcd.o (which is still there).

Should i be deleting the original ehci-hcd.o?

The airlink adapter appears to be working off of the usb-ohci.o driver -- is that right? I thought since my new adapter was USB 2.0 it would be working off of ehci?The backport usbcore.o module includes the ehci functionality. You don't want to load ehci-hcd.o too. I'm not quite sure what would happen, but I don't recommened it. Since some versions of the tivo software will try to load ehci-hcd.o for you, the intent of ehci-dummy.o was to provide a replacement ehci-hcd.o that would do no harm if loaded. See post #15 from ADH in the devel thread. Also search for ehci-dummy in the README.

I think the safest thing to do is to delete the tivo supplied ehci-hcd.o and link ehci-dummy.o to ehci-hcd.o. OTOH, if it's working as is, maybe all is ok.

[nmrgeek]

Ok, I upgraded my dtivo from 3.1.1 to 6.2 using PlainBills walkthrough. No problems with that. Now I wanted to get my Linksys WUSB 11 v2.8 working. It worked perfectly before the upgrade and the link light just kept flashing but no link. I tried netconfig.tcl to set it and that did not work. The device came up as eth0. I could use ifconfig to change settings but no network. I tried the new usb 2.0 drivers with no help. If I try to used wlancfg utilites I get (after I changed all the references to wlan0 to eth0).

tivo:/etc/wlan$ ./wlan.init
insmod: a module named usbcore already exists
insmod: a module named usb-ohci already exists
init_module: prism2_usb.o: 0.1.14-pre6-tivo Loaded
init_module: dev_info is: prism2_usb
message=lnxreq_ifstate
ifstate=enable
resultcode=no_value
error: Autojoin indicated failure!

What is going on? How can I see what is going on? I have been working on unix type systems for over 13 years and I am getting nowhere. iwconfig does not work either. ifconfig wlan0 plumb does not setup the adaptor (not sure if mips has this built in but I had to try.) I do not understand why the wireless gets set up at eth0.

Thanks,

nmrgeek

[nmrgeek]

Ok, I just loaded af_packet.o and tried dhclient just for fun and got:

tivo:/lib/modules$ dhclient
Internet Software Consortium DHCP Client 2.0
Copyright 1995, 1996, 1997, 1998, 1999 The Internet Software Consortium.
All rights reserved.

Please contribute if you find this software useful.
For info, please visit http://www.isc.org/dhcp-contrib.html

Listening on LPF/lo/<null>
Sending on LPF/lo/<null>
Listening on LPF/eth0/00:0c:41:5a:8f:02
Sending on LPF/eth0/00:0c:41:5a:8f:02
Sending on Socket/fallback/fallback-net
DHCPDISCOVER on lo to 255.255.255.255 port 67 interval 7
DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 7
DHCPDISCOVER on lo to 255.255.255.255 port 67 interval 18
DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 14
DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 21
DHCPDISCOVER on lo to 255.255.255.255 port 67 interval 7
DHCPDISCOVER on lo to 255.255.255.255 port 67 interval 18
DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 16
DHCPDISCOVER on lo to 255.255.255.255 port 67 interval 8
No DHCPOFFERS received.
No working leases in persistent database - sleeping.

Does this shine light on anything?

[Jamie]

[...]This indicates you have a device configured (eth0) and it is broadcasting dhcp packets, but no one is responding.

[artdent002]

After extensive lurking, lots of reading, and quite a bit of experimentation, I've got my dtivo s2 running 3.1.15, all the hacks I could hope for, easy communication over the network (via USB1.1)... all in all I've been patting myself on the back for getting this far without needing to ask for help.

But now I've done a couple of stupid things: (1) I decided to return my airlink AHOSOUSB to Fry's after only spending a couple of hours trying to get it to work... Lot's of great ideas here from Jamie and others, but I threw in the towel before I got it working (I also paid about $15 for it... but I'm not counting that as one of my 2 stupid things).
(2) I ordered a Linksys USB200m online - I had seen several posts from people who were having no trouble with theirs, so I thought it would be a safe bet. I was dumbfounded when it didn't work, and only then saw a handful of posts on the problem with ver 2 of the adapter. Undaunted I started scouring the net for linux drivers that would work with this thing... nothing.

Has anybody had any luck getting one of these to work? Is it going to take somebody writing new drivers?

Any ideas at all would be appreciated... otherwise it goes back to Outpost on Tuesday...

\Art

[Jamie]

Any ideas at all would be appreciated... otherwise it goes back to Outpost on Tuesday...I can have a look to see if it is easy to support in the backport drivers. I'd need the vendor/product id's. The are in /var/log/kernel.log. Look for a message such as: usb.c: USB device 2 (vend/prod 0xb95/0x1780) is not claimed by any active driver.Your vend/prod numbers will be different, and that's what I'm intrested in. Also, looking at the windows drivers can sometimes be helpful to identify the chip used. All I really need is the .inf file. I looked on the linksys site, but the only drivers they have there are for the V1 usb200m.

[artdent002]

Thanks for taking a look...

From the inf file it looks like the VID is 13B1 and the PID is 0018, but I've attached the inf file for you to see as well...

I've got the drive pulled now so that I can fix a few other things I mucked up, and I'm not sure I know how to mount the partition that tivo mounts to /var/log. Once I have it up and running again with my old usb1.1 adapter, I'll update this with the vendor and product numbers from the log.

UPDATE: From the log it looks like vend/prod is 0x13b1/0x18

\Art

[nmrgeek]

Thanks Jamie!!

After reading the forum for a couple of hours I discovered that the correct drivers were not loading. I created a complete kudge. Load the correct drivers from rc.sysinit.author to start up the wireless (insmod drivers, ifconfig, route, iwconfig to add the ssid) then restart dhclient and volia! Network works great. I think the vnetusba.o does not work well with the Linksys WUSB 11 v2.8 but Jamie's USB 2.0 drivers work great. Getting about 0.5 mb/s using wireless and tytools. Not bad for no wires. Wife hates wires.

Thanks,

nmrgeek

[Jamie]

Thanks for taking a look...

From the inf file it looks like the VID is 13B1 and the PID is 0018, but I've attached the inf file for you to see as well...

I've got the drive pulled now so that I can fix a few other things I mucked up, and I'm not sure I know how to mount the partition that tivo mounts to /var/log. Once I have it up and running again with my old usb1.1 adapter, I'll update this with the vendor and product numbers from the log.I can't find any chip information in the .inf file. My best guess is that it is ax88772 based (like the AHOSOUSB). We can try adding that vendor/product id to see if it works. PM me if you feel like experimenting. Otherwise, return it and buy one of these (http://www.newegg.com/Product/Product.asp?Item=N82E16833147011). They work out of the box both with the tivo drivers and with the backport drivers. They're also more durable than the usb200m, which has a delicate rj45 connector.

[eastwind]

Thanks for taking a look...
I've got the drive pulled now so that I can fix a few other things I mucked up, and I'm not sure I know how to mount the partition that tivo mounts to /var/log. Once I have it up and running again with my old usb1.1 adapter, I'll update this with the vendor and product numbers from the log.

\Art
Create a mount point and mount /dev/hd(TiVo drive letter)9 to get to the /var partition on the TiVo drive:

mkdir /mnt/tivovar
mount /dev/hdc9 /mnt/tivovar
cd /mnt/tivovar/log

(Assumes TiVo drive is secondary master)

ew

[malefactor]

Got all this working with the airlink gig-e adapter, but HMO doesn't work (in case I'm using the wrong terms, automatic other-subnet-connected-tivos-appearing-in-now-showing).

If I reenable dhclient and let the box find its own IP address, HMO works fine.

Is this a known problem? Are there simple ways to fix it?

Using S2 dtivo, 6.2 + hacks.

[Jamie]

Got all this working with the airlink gig-e adapter, but HMO doesn't work (in case I'm using the wrong terms, automatic other-subnet-connected-tivos-appearing-in-now-showing).

If I reenable dhclient and let the box find its own IP address, HMO works fine.

Is this a known problem? Are there simple ways to fix it?

Using S2 dtivo, 6.2 + hacks.
http://www.dealdatabase.com/forum/showthread.php?p=222641#post222641

[artdent002]

Create a mount point and mount /dev/hd(TiVo drive letter)9 to get to the /var partition on the TiVo drive:
[/b]
mkdir /mnt/tivovar
mount /dev/hdc9 /mnt/tivovar
cd /mnt/tivovar/log
[/b]
(Assumes TiVo drive is secondary master)

ew
Thanks - I knew how to make a directory and mount a device partition, but I didn't realize the /var partition was #9... I had foolishly been trying to mount MFS partitions.

[eastwind]

Thanks - I knew how to make a directory and mount a device partition, but I didn't realize the /var partition was #9... I had foolishly been trying to mount MFS partitions.
FYI, pdisk -l /dev/hdb will list all the partitions on the TiVo drive (if it's on primary slave).

ew

[Jamie]

The latest drivers left me without Internet connection at all...

(using the LINKSYS USB200M).

I get connection, but when I try to do something, the connection get's lost.

(doing a LS, or transferring a file).

It's seems like the driver "overloads" and shuts off...I gather from your other posts that you are running 7.2.0 on a Series 2.5.

I haven't had much luck with the backport usbcore modules on Series 2.5 hardware since 7.1b. I haven't spent much time looking at it. It's possible it's a regression in the backport code, though I haven't changed any of the host controller code.

The good news is that the stock TiVo usbcore/ehci-hcd/etc works fine with the backport usbnet.o. See the series of posts that starts here (http://www.dealdatabase.com/forum/showthread.php?p=229069#post229069).

[jasch]

I gather from your other posts that you are running 7.2.0 on a Series 2.5.

I haven't had much luck with the backport usbcore modules on Series 2.5 hardware since 7.1b. I haven't spent much time looking at it. It's possible it's a regression in the backport code, though I haven't changed any of the host controller code.

The good news is that the stock TiVo usbcore/ehci-hcd/etc works fine with the backport usbnet.o. See the series of posts that starts here (http://www.dealdatabase.com/forum/showthread.php?p=229069#post229069).

Thanks for your reply, and sorry for posting on the other thread. I could swear I clicked on the appropiate link before posting seems not).

Altough I rolled back to the previous drivers, I am still getting problems with the Gen04 drivers. The strange thing, is that is has been working perfectly for the last couple of weeks... Unless I am missing a file. (I noticed this insmod: /lib/modules/ehci-hcd.o: No such file or directory on my bash).

[Jamie]

Thanks for your reply, and sorry for posting on the other thread. I could swear I clicked on the appropiate link before posting seems not).

Altough I rolled back to the previous drivers, I am still getting problems with the Gen04 drivers. The strange thing, is that is has been working perfectly for the last couple of weeks... Unless I am missing a file. (I noticed this insmod: /lib/modules/ehci-hcd.o: No such file or directory on my bash).Go back to the stock tivo drivers for everything. Get that to work. Then replace usbnet.o with the one from the backport package.

/lib/modules/ehci-hcd.o should be a link to /platform/lib/modules/ehci-hcd.o

[jasch]

Go back to the stock tivo drivers for everything. Get that to work. Then replace usbnet.o with the one from the backport package.

/lib/modules/ehci-hcd.o should be a link to /platform/lib/modules/ehci-hcd.o

I did the symlink, and replaced usbnet.o.

Altough everything seems to be working OK, I got this on serial:

/lib/modules/ehci-hcd.o: /lib/modules/ehci-hcd.o: unresolved symbol usb_hcd_pci_probe
/lib/modules/ehci-hcd.o: /lib/modules/ehci-hcd.o: unresolved symbol usb_hcd_giveback_urb
/lib/modules/ehci-hcd.o: /lib/modules/ehci-hcd.o: unresolved symbol usb_hcd_pci_remove

[Jamie]

I did the symlink, and replaced usbnet.o.

Altough everything seems to be working OK, I got this on serial:

/lib/modules/ehci-hcd.o: /lib/modules/ehci-hcd.o: unresolved symbol usb_hcd_pci_probe
/lib/modules/ehci-hcd.o: /lib/modules/ehci-hcd.o: unresolved symbol usb_hcd_giveback_urb
/lib/modules/ehci-hcd.o: /lib/modules/ehci-hcd.o: unresolved symbol usb_hcd_pci_removeIt looks to me like you did not properly roll back to the tivo modules. Those symbols are defined in the tivo supplied usbcore.o. Here are the correct md5sums from a 7.2.0-oth-01-2-540 system:bash-2.02# md5sum /platform/lib/modules/{usb*,ehci*}
24dc8dba495458d4149ec10bd81f032c /platform/lib/modules/usb-ohci.o
68988a40ea0f511edd69d13e68921bc3 /platform/lib/modules/usbcore.o
24fb545984c275fade7517a089ac7701 /platform/lib/modules/ehci-hcd.oIf yours are different, or if you don't have these files properly linked to /lib/modules, you've botched your rollback to the stock tivo drivers.

[jasch]

If yours are different, or if you don't have these files properly linked to /lib/modules, you've botched your rollback to the stock tivo drivers.

I think my backup is from 7.1 (yeah, yeah I know). Perhaps some kind soul could PM me the original files... <grin>

[drpud]

Having problems getting the newer version of the linksys adapter to work. I am running D* 6.2. I have replaced the usbnet.o file with the recent file (0915) posted by Jamie. I have also added the new entry for the v2 adpater into the usb.map file in etc/hotplug. I have network access using my older 200M adapter but cannot seem to get the new one (v2) to work. The lan lights on the adapter are lit and it seems to be getting power just no access. What am I missing as I know I have missed something simple? I have searched quite a bit but cannot seem to find anyone with a similar issue.

Jeff

[Jamie]

Having problems getting the newer version of the linksys adapter to work. I am running D* 6.2. I have replaced the usbnet.o file with the recent file (0915) posted by Jamie. I have also added the new entry for the v2 adpater into the usb.map file in etc/hotplug. I have network access using my older 200M adapter but cannot seem to get the new one (v2) to work. The lan lights on the adapter are lit and it seems to be getting power just no access. What am I missing as I know I have missed something simple? I have searched quite a bit but cannot seem to find anyone with a similar issue.Post your serial console output and/or kernel/messages logs that show what happens during a boot. I just got one, but it works fine for me under 7.2.0.

[drpud]

Thanks for getting back so quickly. I have attached the serial output of the device. Can't seem to locate anything too screwy when the USB modules load but, hey, I'm an idi0t. Let me know what you think. Thanks!

Jeff

[coolnetman]

Relative noob to Tivo hacking, in fact 1st Tivo hacked ever..DSR704....hacked running 6.2, leaving 6.2 in-place in-tact while dd'd a 3.1.5 2.4.20 compatible kernel and monte'd Jamie's custom kernel with said updated monte/kmonte 2.4.20 compatible chainload pack.

That said I have not yet had ANY USB work and I started out with the backport because of necessity of 2.4.20 compatibility and got Tivo booting no problem but after integrating the 2.4.27 backport 2.4.20Series2 compat. usb (my original 2.4.4-3.0 errors disappeared, the drivers load..) but I receive the following:

/lib/modules/usb-storage.o: unresolved symbol scsi_unregister_module
/lib/modules/usb-storage.o: unresolved symbol scsi_register
/lib/modules/usb-storage.o: unresolved symbol scsi_register_module
usb.c: USB device 2 (vend/prod 0x50d/0x121) is not claimed by any active driver.
hub.c: new USB device 00:0d.1-1, assigned address 2
usb.c: USB device 2 (vend/prod 0x66b/0x2206) is not claimed by any active driver
Invoking startup scripts for:
platform 'trinity'
implementation 'Series2'....etc.."

Note ehci-usb initializes properly. :confused: I have 2 different USB-to-LAN adapters trying either desperately....

Now, my usb.map HAS those products listed in the proper syntax and I duplicated everything that was under the pegasus section to the ax8817x and vice versa to cover all my bases.

I'm using the 09/15/2005 load of Jamie's USB backport.

I'm both stuck and stumped...

PLEASE help the NOOB! I will be forever greatful...

Thanks in advance,
Cman

[eastwind]

Relative noob to Tivo hacking, in fact 1st Tivo hacked ever..DSR704....hacked running 6.2, leaving 6.2 in-place in-tact while dd'd a 3.1.5 2.4.20 compatible kernel and monte'd Jamie's custom kernel with said updated monte/kmonte 2.4.20 compatible chainload pack.

That said I have not yet had ANY USB work and I started out with the backport because of necessity of 2.4.20 compatibility and got Tivo booting no problem but after integrating the 2.4.27 backport 2.4.20Series2 compat. usb (my original 2.4.4-3.0 errors disappeared, the drivers load..) but I receive the following:

/lib/modules/usb-storage.o: unresolved symbol scsi_unregister_module
/lib/modules/usb-storage.o: unresolved symbol scsi_register
/lib/modules/usb-storage.o: unresolved symbol scsi_register_module
usb.c: USB device 2 (vend/prod 0x50d/0x121) is not claimed by any active driver.
hub.c: new USB device 00:0d.1-1, assigned address 2
usb.c: USB device 2 (vend/prod 0x66b/0x2206) is not claimed by any active driver
Invoking startup scripts for:
platform 'trinity'
implementation 'Series2'....etc.."

Note ehci-usb initializes properly. :confused: I have 2 different USB-to-LAN adapters trying either desperately....

Now, my usb.map HAS those products listed in the proper syntax and I duplicated everything that was under the pegasus section to the ax8817x and vice versa to cover all my bases.

I'm using the 09/15/2005 load of Jamie's USB backport.

I'm both stuck and stumped...

PLEASE help the NOOB! I will be forever greatful...

Thanks in advance,
Cman
What is the kernel version of the monted to kernel? Does it match the drivers you're trying to load?

Why are you monte'ing in the first place (I know there are reasons to, but it's not always necessary).

If you list the device everywhere "to cover all bases", will the correct driver know it's the right one? Or will they all try to grab the device?

The messages about not claimed by any active driver don't mean much--just that the device was found before the driver loaded. It will usually grab it when it loads.

The usb_storage modules only need to be loaded if you're going to use them. And if you really need it you'll need to load scsi_mod.o before usb_storage.o (and you might need sd_mod.o also--can't remember).

Post your rc.sysinit.author.

What is your specific problem? Can't telnet? Can't MRV?

Do you have access via serial cable (bash on serial)?



ew

[Jamie]

Now, my usb.map HAS those products listed in the proper syntax and I duplicated everything that was under the pegasus section to the ax8817x and vice versa to cover all my bases.Eastwind has the right questions.

My suggestion: start simple and worry about making it faster later. You don't need to monte. You don't need the backport drivers. Both of those things give you some performance advantages, but unless you happen to have a dongle that the tivo software doesn't support, they aren't necessary to get basic functionality. Make it work as close to "stock" as possible first, then worry about the performance tweaks.

[coolnetman]

My kernel log:

"kernel: klogd 1.3-3, log source = /proc/kmsg started.
one) kernel: Cannot find map file.
one) kernel: Loaded 138 symbols from 11 modules.
one) kernel: CPU revision is: 00005430
one) kernel: FPU revision is: 00005410
one) kernel: Primary instruction cache 32kb, linesize 32 bytes.
one) kernel: Primary data cache 32kb, linesize 32 bytes.
one) kernel: Linux version 2.4.20 (root@fedora.home.lan) (gcc version 3.0) #6 We
one) kernel: Determined physical RAM map:
one) kernel: memory: 04000000 @ 00000000 (usable)
one) kernel: On node 0 totalpages: 16384
one) kernel: zone(0): 16384 pages.
one) kernel: zone(1): 0 pages.
one) kernel: zone(2): 0 pages.
one) kernel: Kernel command line: root=/dev/hda4 console=2,115200 dsscon=true ru
one) kernel: Monotonic time calibrated: 81.00 counts per usec
one) kernel: Calibrating delay loop... 161.38 BogoMIPS
(none) kernel: Contiguous region 1: 8388608 bytes @ address 0x80
Jan 2 00:00:18 (none) kernel: Contiguous region 2: 1048576 bytes @ address 0x81
Jan 2 00:00:18 (none) kernel: Contiguous region 8: 10485760 bytes @ address 0x8
Jan 2 00:00:18 (none) kernel: Contiguous region of 19922944 bytes total reserve
Jan 2 00:00:18 (none) kernel: Memory: 43704k/65536k available (1341k kernel cod
Jan 2 00:00:18 (none) kernel: Dentry cache hash table entries: 8192 (order: 4,
Jan 2 00:00:18 (none) kernel: Inode cache hash table entries: 4096 (order: 3, 3
Jan 2 00:00:18 (none) kernel: Mount-cache hash table entries: 1024 (order: 1, 8
Jan 2 00:00:18 (none) kernel: Buffer-cache hash table entries: 4096 (order: 2,
Jan 2 00:00:18 (none) kernel: Page-cache hash table entries: 16384 (order: 4, 6
Jan 2 00:00:18 (none) kernel: Checking for 'wait' instruction... unavailable.
Jan 2 00:00:18 (none) kernel: POSIX conformance testing by UNIFIX
Jan 2 00:00:18 (none) kernel: PCI: Probing PCI hardware
Jan 2 00:00:18 (none) kernel: ttyS00 at iomem 0xb4100100 (irq = 79) is a 16550A
Jan 2 00:00:18 (none) kernel: ttyS00 at port 0xbc010000 (irq = 133) is a unknow
Jan 2 00:00:18 (none) kernel: ttyS00 at iomem 0xb4100140 (irq = 81) is a 16550A
Jan 2 00:00:18 (none) kernel: ttyS00 at iomem 0xb4100120 (irq = 80) is a 16550A
Jan 2 00:00:18 (none) kernel: Linux NET4.0 for Linux 2.4
Jan 2 00:00:18 (none) kernel: Based upon Swansea University Computer Society NE
Jan 2 00:00:18 (none) kernel: Initializing RT netlink socket
Jan 2 00:00:18 (none) kernel: Starting kswapd
Jan 2 00:00:18 (none) kernel: Journalled Block Device driver loaded
Jan 2 00:00:18 (none) kernel: Serial driver version 5.05c (2001-07-08) with MAN
(none) kernel: ttyS00 at 0xb4100100 (irq = 79) is a 16550A
Jan 2 00:00:18 (none) kernel: ttyS01 at 0xbc010000 (irq = 133) is a unknown
Jan 2 00:00:18 (none) kernel: ttyS02 at 0xb4100140 (irq = 81) is a 16550A
Jan 2 00:00:18 (none) kernel: ttyS03 at 0xb4100120 (irq = 80) is a 16550A
Jan 2 00:00:18 (none) kernel: Uniform Multi-Platform E-IDE driver Revision: 6.3
Jan 2 00:00:18 (none) kernel: ide: Assuming 33MHz system bus speed for PIO mode
Jan 2 00:00:18 (none) kernel: hda: WDC WD1200JB-00EVA0, ATA DISK drive
Jan 2 00:00:18 (none) kernel: ide0 at 0x400-0x407,0x438 on irq 87
Jan 2 00:00:18 (none) kernel: hda: 234441648 sectors (120034 MB) w/8192KiB Cach
Jan 2 00:00:18 (none) kernel: Partition check:
Jan 2 00:00:18 (none) kernel: hda: [mac] hda1 hda2 hda3 hda4 hda5 hda6 hda7 hd
Jan 2 00:00:18 (none) kernel: RAMDISK driver initialized: 16 RAM disks of 4096K
Jan 2 00:00:18 (none) kernel: PPP generic driver version 2.4.2
Jan 2 00:00:18 (none) kernel: PPP Deflate Compression module registered
Jan 2 00:00:18 (none) kernel: NET4: Linux TCP/IP 1.0 for NET4.0
Jan 2 00:00:18 (none) kernel: IP Protocols: ICMP, UDP, TCP
Jan 2 00:00:18 (none) kernel: IP: routing cache hash table of 512 buckets, 4Kby
Jan 2 00:00:18 (none) kernel: TCP: Hash tables configured (established 4096 bin
Jan 2 00:00:18 (none) kernel: NET4: Unix domain sockets 1.0/SMP for Linux NET4.
Jan 2 00:00:18 (none) kernel: VFS: Mounted root (ext2 filesystem) readonly.
Jan 2 00:00:18 (none) kernel: Freeing unused kernel memory: 60k freed
Jan 2 00:00:18 (none) kernel: Starting rc.sysinit
Jan 2 00:00:18 (none) kernel: Running boot Stage A_PreKickstart scripts
(none) kernel: Scanning for configuration files
Jan 2 00:00:18 (none) kernel: Chainloaded environment in place, skipping monte.
Jan 2 00:00:18 (none) kernel: Starting rc.runfirst
Jan 2 00:00:18 (none) kernel: Bringing up bash. Use 'skill -9 rc.sysinit' to ha
Jan 2 00:00:18 (none) kernel: Loading USB Drivers...
Jan 2 00:00:18 (none) kernel: bash: no job control in this shell
Jan 2 00:00:18 (none) kernel: bash-2.02# <6>usb.c: registered new driver usbdev
Jan 2 00:00:18 (none) kernel: usb.c: registered new driver hub
Jan 2 00:00:18 (none) kernel: usb.c: starting integrated EHCI driver
Jan 2 00:00:18 (none) kernel: ehci_hcd 00:0d.2: PCI device 1033:00e0
Jan 2 00:00:18 (none) kernel: ehci_hcd 00:0d.2: irq 6, pci mem b4002000
Jan 2 00:00:18 (none) kernel: usb.c: new USB bus registered, assigned bus numbe
Jan 2 00:00:18 (none) kernel: PCI: 00:0d.2 PCI cache line size set incorrectly
Jan 2 00:00:18 (none) kernel: ehci_hcd 00:0d.2: USB 2.0 enabled, EHCI 0.95, dri
Jan 2 00:00:18 (none) kernel: hub.c: USB hub found
Jan 2 00:00:18 (none) kernel: hub.c: 5 ports detected
Jan 2 00:00:18 (none) kernel: usb.c: EHCI initialization succeeded
Jan 2 00:00:18 (none) kernel: usb.c: registered new driver usbnet
Jan 2 00:00:18 (none) kernel: pegasus.c: v0.4.30 (2003/04/01):Pegasus/Pegasus I
Jan 2 00:00:18 (none) kernel: usb.c: registered new driver pegasus
Jan 2 00:00:18 (none) kernel: usb.c: registered new driver usbnet
Jan 2 00:00:18 (none) kernel: usb-ohci.c: USB OHCI at membase 0xb4000000, IRQ 5
Jan 2 00:00:18 (none) kernel: usb-ohci.c: usb-00:0d.0, PCI device 1033:0035
usb.c: new USB bus registered, assigned bus numbe
Jan 2 00:00:18 (none) kernel: hub.c: USB hub found
Jan 2 00:00:18 (none) kernel: hub.c: 3 ports detected
Jan 2 00:00:18 (none) kernel: usb-ohci.c: USB OHCI at membase 0xb4001000, IRQ 3
Jan 2 00:00:18 (none) kernel: usb-ohci.c: usb-00:0d.1, PCI device 1033:0035
Jan 2 00:00:18 (none) kernel: usb.c: new USB bus registered, assigned bus numbe
Jan 2 00:00:18 (none) kernel: hub.c: USB hub found
Jan 2 00:00:18 (none) kernel: hub.c: 2 ports detected
Jan 2 00:00:18 (none) kernel: hub.c: new USB device 00:0d.0-1, assigned address
Jan 2 00:00:18 (none) kernel: pegasus.c: eth0: Belkin F5D5050 USB Ethernet
Jan 2 00:00:18 (none) kernel: pegasus.c: setup Pegasus II specific registers
Jan 2 00:00:18 (none) kernel: /lib/modules/usb-storage.o: unresolved symbol scs
Jan 2 00:00:18 (none) kernel: /lib/modules/usb-storage.o: unresolved symbol scs
Jan 2 00:00:18 (none) kernel: /lib/modules/usb-storage.o: unresolved symbol scs
Jan 2 00:00:18 (none) kernel: Invoking startup scripts for:
Jan 2 00:00:18 (none) kernel: platform 'trinity'
Jan 2 00:00:18 (none) kernel: implementation 'Series2'
Jan 2 00:00:18 (none) kernel: implementer 'TiVo'
Jan 2 00:00:18 (none) kernel: Releasing /initrd and clearing ramdisk, if they e
Jan 2 00:00:18 (none) kernel: warning: can't open /var/mtab: No such file or di
Jan 2 00:00:18 (none) kernel: umount: /initrd: not mounted
Jan 2 00:00:18 (none) kernel: Activating swap partitions
Jan 2 00:00:18 (none) kernel: Adding Swap: 130044k swap-space (priority -1)
(none) kernel: Loading core system drivers
Jan 2 00:00:18 (none) kernel: Parallel modem driver loaded, 1 ports starting at
Jan 2 00:00:18 (none) kernel: Loading ircatch
Jan 2 00:00:18 (none) kernel: Found rev 'F' Si2433 modem on /dev/cua1
Jan 2 00:00:18 (none) kernel: Checking for Kickstart panic signal
Jan 2 00:00:18 (none) kernel: Running boot Stage B_PostKickstart scripts
Jan 2 00:00:18 (none) kernel: Cleanup /dev/hda9 pass 1
Jan 2 00:00:18 (none) kernel: ext2fs_check_if_mount: No such file or directory
Jan 2 00:00:18 (none) kernel: /dev/hda9 was not cleanly unmounted, check forced
Jan 2 00:00:18 (none) kernel: Inode 2066, i_blocks wrong 768 (counted=766). Se
Jan 2 00:00:18 (none) kernel:
Jan 2 00:00:18 (none) kernel: Inode 6148, i_blocks wrong 158 (counted=148). Se
Jan 2 00:00:18 (none) kernel:
Jan 2 00:00:18 (none) kernel: Inode 6155, i_blocks wrong 46 (counted=34). Set
Jan 2 00:00:18 (none) kernel:
Jan 2 00:00:18 (none) kernel: Inode 6146, i_blocks wrong 408 (counted=406). Se
Jan 2 00:00:18 (none) kernel:
Jan 2 00:00:18 (none) kernel: Inode 6152, i_blocks wrong 688 (counted=678). Se
Jan 2 00:00:18 (none) kernel:
Jan 2 00:00:18 (none) kernel: Fix summary information? yes
Jan 2 00:00:18 (none) kernel:
Jan 2 00:00:18 (none) kernel: /dev/hda9: 97/32768 files (13.4% non-contiguous),
Jan 2 00:00:18 (none) kernel: Cleanup /dev/hda9 pass 2
(none) kernel: ext2fs_check_if_mount: No such file or directory
Jan 2 00:00:18 (none) kernel: /dev/hda9: clean, 97/32768 files, 12480/131072 bl
Jan 2 00:00:18 (none) kernel: /dev/hda9 is clean after pass 2
Jan 2 00:00:18 (none) kernel: Mounting /var
Jan 2 00:00:18 (none) kernel: /dev/hda9 on /var type ext2 (rw)
Jan 2 00:00:18 (none) kernel: Cleaning up files in /var
Jan 2 00:00:18 (none) kernel: Checking space in /var
Jan 2 00:00:18 (none) kernel: Mounting initial environment
Jan 2 00:00:18 (none) kernel: Starting logging daemons
Jan 2 00:00:18 (none) kernel: Found Silicon Labs "Si2433" modem, rev F, skippin
Jan 2 00:00:18 (none) kernel: Scanning for phase1 repair scripts
Jan 2 00:00:18 (none) kernel: Running boot Stage C_MediaInitialization scripts
Jan 2 00:00:19 (none) kernel: Loading input section drivers
Jan 2 00:00:21 (none) kernel: Bringing up eth0...
Jan 2 00:00:21 (none) kernel: Internet Software Consortium DHCP Client 2.0
Jan 2 00:00:21 (none) kernel: Copyright 1995, 1996, 1997, 1998, 1999 The Intern
Jan 2 00:00:21 (none) kernel: All rights reserved.
Jan 2 00:00:21 (none) kernel:
Jan 2 00:00:21 (none) kernel: Please contribute if you find this software usefu
Jan 2 00:00:21 (none) kernel: For info, please visit http://www.isc.org/dhcp-co
Jan 2 00:00:21 (none) kernel:
Jan 2 00:00:21 (none) kernel: cobra module was successfully installed, LNBPA 0x
Jan 2 00:00:21 (none) kernel: Loading output section drivers
Jan 2 00:00:22 (none) kernel: Splash the screen
Jan 2 00:00:23 (none) kernel: Running boot Stage D_PreMfs scripts
Jan 2 00:00:23 (none) kernel: Remote control is TIVO
Jan 2 00:00:23 (none) kernel: MFS partition on /dev/hda10
Jan 2 00:00:24 (none) kernel: socket: Address family not supported by protocol
Jan 2 00:00:24 (none) kernel: Make sure to set CONFIG_PACKET=y and CONFIG_FILTE
Jan 2 00:00:24 (none) kernel: exiting.
Jan 2 00:00:25 (none) kernel: Loading Trinity dssapp
Jan 2 00:00:25 (none) kernel: Look for debug board
Jan 2 00:00:25 (none) kernel: /tvbin/dssappAV: DSS Interface Version 1.24, comp
Jan 2 00:00:25 (none) kernel: Setting up network for debug
Jan 2 00:00:25 (none) kernel: Starting Telnet Listener...
Jan 2 00:00:25 (none) kernel: Starting /proc Listener...
Jan 2 00:00:25 (none) kernel: Loading irblast
Jan 2 00:00:26 (none) kernel: Running without ideturbo mode
Jan 2 00:00:26 (none) kernel: Loading fan
Jan 2 00:00:26 (none) kernel: Loading therm
Jan 2 00:00:26 (none) kernel: Loading TvBus router
Jan 2 00:00:26 (none) kernel: Updating system clock
Sep 20 16:47:38 (none) kernel: Time set to: Tue Sep 20 16:47:38 2005
Sep 20 16:47:38 (none) kernel: Enabling local route
Sep 20 16:47:38 (none) kernel: Setting TCP keepalive parameters
Sep 20 16:47:38 (none) kernel: Checking for additional disk
(none) kernel: Start fan control
Sep 20 16:47:40 (none) kernel: First temperature parameters set:
Sep 20 16:47:40 (none) kernel: Terminal temp: 71
Sep 20 16:47:40 (none) kernel: Critical temp: 62
Sep 20 16:47:40 (none) kernel: Logging temp: 60
Sep 20 16:47:40 (none) kernel: Target temp: 50
Sep 20 16:47:40 (none) kernel: Lowest fan speed: 7
Sep 20 16:47:40 (none) kernel: /tvbin/fancontrol is running in the background.
Sep 20 16:47:40 (none) kernel: Starting TvLauncher
Sep 20 16:47:41 (none) kernel: PokeFirewallPort: Poking hole through firewall at
Sep 20 16:47:41 (none) kernel: iptables v1.2.6a: can't initialize iptables table
Sep 20 16:47:41 (none) kernel: Perhaps iptables or your kernel needs to be upgra
Sep 20 16:47:42 (none) kernel: Waiting for launcher to start.
Sep 20 16:47:42 (none) kernel: Launcher is running.
Sep 20 16:47:43 (none) kernel: Scanning for phase2 repair scripts
Sep 20 16:47:43 (none) kernel: Running boot Stage E_PreApplication scripts
Sep 20 16:47:45 (none) kernel: Checking for database conversions
Sep 20 16:47:45 (none) kernel: No upgrade to load
Sep 20 16:47:45 (none) kernel: Not upgrading software
Sep 20 16:47:46 (none) kernel: Scanning for phase3 repair scripts
Sep 20 16:47:46 (none) kernel: Running boot Stage F_ApplicationLaunch scripts
Sep 20 16:47:46 (none) kernel: Starting Services.
Sep 20 16:47:50 (none) kernel: Bringing up telnet daemon...
Sep 20 16:47:50 (none) kernel: Couldn't bind to socket.
Sep 20 16:47:50 (none) kernel: Blocking Updates via IP
Sep 20 16:47:50 (none) kernel: rc.runfirst is complete
Sep 20 16:47:54 (none) kernel: Microcode version is TiVo!
Sep 20 16:47:55 (none) kernel: Found hpk front panel model 1
Sep 20 16:48:05 (none) kernel: PokeFirewallPort: Poking hole through firewall at
Sep 20 16:48:05 (none) kernel: iptables v1.2.6a: can't initialize iptables table
Sep 20 16:48:05 (none) kernel: Perhaps iptables or your kernel needs to be upgra
Sep 20 16:48:05 (none) kernel: PokeFirewallPort: Poking hole through firewall at
Sep 20 16:48:05 (none) kernel: iptables v1.2.6a: can't initialize iptables table
Sep 20 16:48:05 (none) kernel: Perhaps iptables or your kernel needs to be upgra
Sep 20 16:48:19 (none) kernel: using driver frame iter
Sep 20 16:48:19 (none) kernel: ApgLogFrequency( 0 sec )
Sep 20 16:48:19 (none) kernel: Finished creating ApgDriverFrameIter
Sep 20 16:48:19 (none) kernel: Using a smartSorter!
(none) kernel: Scanning for phase4 repair scripts

[coolnetman]

So here's the deal:

First, Thanks to ALL for such quick responses! YOU ROCK!

I'm bash'd in via console cable/serial cable. Network manufacturer sales engineer by trade, so new to tivo hacking but not new to the scene so you can slam me as technically as you like and I appreciate the help GREATLY!! :)

kernel version = Linux version 2.4.20 (root@fedora.home.lan) (gcc version 3.0) #6 Wed Jan 26 17:5 2:06 MST 2005

according to 'cat /proc/version'.

I wasn't loading the scsi or smod drivers like a fool so usb storage would never work but I would like to experiment later and thought Iwould try to kill as many birds as possible with one stone while the box is open so I loaded alot of the objects now...like maybe adding a dvdwritter l8r via my USB storage box :) (any advice here would kick a$$ as well)...but advice on pairing down and starting basic is ALWAYS best. PALM prc installs work very similar I'm afraid and TREOs are even less forgiving...

rc.sysinit.author is:

"#!/bin/bash
#Set environment variables and fork rc.runlast. Saves a neglegible
#amount of time required for booting when you do this. Setting the
#path here makes it stick.
#

export
PATH=$PATH:/tivo-bin:/hack:/var/hack:/var/hack/bin:/utils:/bin:/sbin:/tvbin:/tiv
/etc/rc.d/rc.runlast &"

As I 'm using tivo legend Jamie (you are in these threads, mad props) chainloader for tivo kernel monte'ing. However, maybe I need not monte here as I'm using 2.4.20 which has alot of the additions compiled-in that most other kernels do not. Very good point, too newbie so don't know...

here is rc.runfirst (Called before sysinit etc...as per Jamie's build):

#!/bin/bash
#Everything here will run before anything else in the init process.
#This is a good place to start a bash shell, since it will run a while
#before the mfs daemon and tivoapp actually start. This gives you a
#chance to fix things before the tivo stops in its boot process if
#anything is broken, without resorting to removing your tivo hard disk.

echo "Starting rc.runfirst"

#This path does not stick; it is only here so that commands in this script work.
#It will be necessary to set it again later.
export PATH=/bin:/sbin:/tvbin:/devbin:/busybox/:/hacks:/hack:/var/hack:/var/hack

#This sets all hacks to run at the lowest priority, ensuring a "smoother"
#tivo operation.
/busybox/setpri fifo 1 $$

echo "Bringing up bash. Use 'skill -9 rc.sysinit' to halt the boot process."
exec /bin/bash --login </dev/ttyS2 >/dev/ttyS2&

echo "Loading USB Drivers..."
insmod /lib/modules/usbcore.o log2_irq_thresh=4
insmod /lib/modules/ax8817x.o
insmod /lib/modules/pegasus.o
insmod /lib/modules/usbnet.o
insmod /lib/modules/usb-ohci.o
insmod /lib/modules/usb-storage.o
insmod /lib/modules/cdrom.o

sleep 15

#insmod /lib/modules/pegasus.o


echo "Bringing up eth0..."
#ifconfig eth0 192.168.1.153 broadcast 192.168.1.255 netmask 255.255.255.0

#Uncomment the following line to enable DHCP (otherwise the IP is static)
dhclient

sleep 15

echo "Bringing up telnet daemon..."
tnlited 23 /busybox/securelogin

echo "Blocking Updates via IP"
route add -host 204.176.49.2 gw 127.0.0.1
route add -net 204.176.49.0 gw 127.0.0.1 netmask 255.255.255.0

echo "rc.runfirst is complete" "

Now loading usbnet & ax8817x probably redundant and re-enterant for the same driver but desperate, newbie, and trying everything....

I also wasn't loading the pegasus.o driver before I posted the last post. So finallly a light turns on....pegasus loads Belkin 5050D lights up and a MAC address is now active, only after a manual 'ifconfig eth0 up' however DHCP even when connected directly to a 4-port switch/router w/a built-in DHCP server is non-existent, and when I manually assign an IP via ifconfig eth0, my kernel complains "no job control" so (I.e. I have not run ANY Superpatch/guidepatch/hack .tcls yet, not until I get basic IP connectivity)when I attempt to ping from the Tivo to router or otherwise I can't "^c" or "^z" to kill it. and have to reboot to get shell back. some help here please it suketh...

So basic problem was NO USB device activity. Now the device comes up, its mac is active as eth0 after a manual 'up' command, plus some added stupidity for not loading the 'pegasus.o' driver, but connectivity is ZERO even with a manual IP load. And the router/switch is 100% operational as tested with laptop using to configure everthing.

Also I renamed 'netfilter-enabled' to 'netfilternotenabled' so IPTables isn't even loaded... wanted to ensure filtering would not hinder me and then add rules and re-enable later....make sense?

My kernel log:

"kernel: klogd 1.3-3, log source = /proc/kmsg started.
one) kernel: Cannot find map file.
one) kernel: Loaded 138 symbols from 11 modules.
one) kernel: CPU revision is: 00005430
one) kernel: FPU revision is: 00005410
one) kernel: Primary instruction cache 32kb, linesize 32 bytes.
one) kernel: Primary data cache 32kb, linesize 32 bytes