Joined: 07 Jun 2006 Posts: 23 Location: New York, NY
Posted: Wed Nov 15, 2006 7:13 Post subject:
I reported earlier that Build 4538 successfully enables the WAN port on my F5D7230-4 v1111tt. Unfortunately, I also found that it seems to lose it's connection several minutes, (I'm using it in wireless bridge mode) so I went back to dd-wrt.v23_std-3969.bin which doesn't enable the WAN port.
I also noticed that builds 3969 and 4538 have the new time zone settings for the US, but the "Save Settings" button on the Basic Setup page doesn't work when the router is in wireless bridge mode.
max2010 wrote:
Eko,
thanks for your answers.
I just reportet what I found different from Belkin firmware in case somebody else wants to compare it with his results.
I have 1 question will the micro(2mb) build also support the Asus WL-520G, ifso where can i download it, the latest builds of dd-wrt reboots the router constantly.
If you can provide nvram and cfe dump.... or the unit. But I hate 2 mb flash units.
nvram from de config.cfg also good? CFE dont have that (no jtag cable) i have the source from the WL-520 and in the image folder there are a few files 1 linux.trx(reference firmaware what i have running now with Broadcom www) 1 linux.bin 1 CFE.bin 1 cfe_4704.bin 1 cfez.bin and 1 pmon_lg.bin.
Good news
-- correction of dd-wrt and Belkin firmware comparison --
Transferspeed on WAN port is higher then I had messured before. I found that I had an old "static DNS1 = 192.168.3.1" entry in "Network Address Server Settings (DHCP)". I did not notice that this setting was still in from some other test.
Now, I get between 6000 and 9000 kbit/s with test from http://www.speakeasy.net/speedtest, that is more than with Belking firmware.
My highest throughput with another test site is is around 13000 and sometimes over 14000 kbit/s. See here http://www.speedtest.net/result/59209735.png
Test setup
Beta DD-WRT v23 SP3 (11/14/06) std - build 4788
Connection to cablemodem in 100Mbit full duplex mode
PC wire connection to Lan port 1, under Linux (Fedora core 6) _________________ SB5120 motorola cablemodem
F5D7230-4 V1232 Belkin, DD-WRT v23 SP3beta std - build 4788
AMD Sempron 2600+ running Win98SE or FedoraCore6
Wan port works, but something is still not perfect.
I think there is some pins on the AC101L left floating or in the wrong state.
My Wan port is in 100Mbit/s Halfduplex mode after a Reboot or PowerOn.
As a result I see Collisons and Errors in "ifconfig eth1".
Here is Belkin ifconfig: -- look at Collisions and RX Errors, Dropped, Frame --
~ # ifconfig eth1
eth1 Link encap:Ethernet HWaddr 00:11:09:90:02:21
inet addr:67.168.184.152 Bcast:67.168.185.255 Mask:255.255.254.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:30825 errors:142 dropped:71 overruns:0 frame:71
TX packets:4752 errors:1 dropped:0 overruns:0 carrier:0
collisions:95 txqueuelen:1000
RX bytes:10669656 (10.1 MiB) TX bytes:1427942 (1.3 MiB)
Interrupt:4
Here is what Belkin WAN port shows seen from PC: -- look at last line --
[root@silver ~]# mii-tool -vv
eth0: negotiated 100baseTx-HD, link ok
registers for MII PHY 1:
3000 786d 0000 8201 01e1 40a1 0001 0000
0000 0000 0000 0000 0000 0000 0000 0000
000c 0080 0001 0023 0c59 0406 0200 8560
8a7b 01f9 0000 0000 0000 0000 0000 0000
product info: vendor 00:00:20, model 32 rev 1
basic mode: autonegotiation enabled
basic status: autonegotiation complete, link ok
capabilities: 100baseTx-FD 100baseTx-HD 10baseT-FD 10baseT-HD
advertising: 100baseTx-FD 100baseTx-HD 10baseT-FD 10baseT-HD
link partner: 100baseTx-HD 10baseT-HD
Now, Thats how I get it to work correctly
0. Power up Belkin.
1. Telnet to Belkin with no wire connected to Wan port. Do reboot from shell. And wait until it is active and pings.
2. Connect PC to Wan port, make sure it is in 100baseTX-FD mode with "mii-tool -vv". Otherwise repeat step 1.
3. Now connect Cablemodem or Linksys router to Wan port.
Here is Belkin ifconfig: -- NO Collisions and NO Errors after so many RX packets --
~ # ifconfig eth1
eth1 Link encap:Ethernet HWaddr 00:11:09:90:02:21
inet addr:67.168.184.152 Bcast:67.168.185.255 Mask:255.255.254.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:4640374 errors:0 dropped:0 overruns:0 frame:0
TX packets:255969 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:706605827 (673.8 MiB) TX bytes:181236122 (172.8 MiB)
Interrupt:4
Here is what Belkin WAN port shows seen from PC: -- look at last line --
[root@silver ~]# mii-tool -vv
eth0: negotiated 100baseTx-FD, link ok
registers for MII PHY 1:
3100 786d 0000 8201 01e1 41e1 0001 0000
0000 0000 0000 0000 0000 0000 0000 0000
000c 0080 0001 0023 0c59 0406 0200 8560
8a7b 01f9 0000 0000 0000 0000 0000 0000
product info: vendor 00:00:20, model 32 rev 1
basic mode: autonegotiation enabled
basic status: autonegotiation complete, link ok
capabilities: 100baseTx-FD 100baseTx-HD 10baseT-FD 10baseT-HD
advertising: 100baseTx-FD 100baseTx-HD 10baseT-FD 10baseT-HD
link partner: 100baseTx-FD 100baseTx-HD 10baseT-FD 10baseT-HD
Compare with Belkin LAN port seen from PC: -- look at last line --
[root@silver ~]# mii-tool -vv
eth0: negotiated 100baseTx-FD, link ok
registers for MII PHY 1:
3100 786d 0000 8201 01e1 45e1 0001 0000
0000 0000 0000 0000 0000 0000 0000 0000
000c 0080 0000 0023 0c59 0406 0200 8560
8a7b 01f9 0000 0000 0000 0000 0000 0000
product info: vendor 00:00:20, model 32 rev 1
basic mode: autonegotiation enabled
basic status: autonegotiation complete, link ok
capabilities: 100baseTx-FD 100baseTx-HD 10baseT-FD 10baseT-HD
advertising: 100baseTx-FD 100baseTx-HD 10baseT-FD 10baseT-HD
link partner: 100baseTx-FD 100baseTx-HD 10baseT-FD 10baseT-HD flow-control
tested with DD-WRT v23 SP3 (11/14/06) std - build 4788 and PC running Fedora 6
Maybe setting these Bits to 1 can help, just in case hardware is messed up ???
The pins on the chip set default values at reset time. As I read the datasheet, this way they would be set in the correct state after reset, regardless of how the pins are connected or left floating?
(control register)
0.12 Enable Auto-Negotiate process (overrides 0.13 and 0.8 )
(Auto-Negotiation-Advertisement-Register)
4.8 100BASE-TX full-duplex capable
4.7 100BASE-TX half-duplex capable
4.6 10BASE-T full-duplex capable
4.5 10BASE-T half-duplex capable _________________ SB5120 motorola cablemodem
F5D7230-4 V1232 Belkin, DD-WRT v23 SP3beta std - build 4788
AMD Sempron 2600+ running Win98SE or FedoraCore6
Well I ended up building the JTAG cable from HairyDairyMade's design.
Backed up CFE, NVRAM, Kernel, then erased NVRAM. Got the unit to respond to pings again
Next I loaded up DD-WRT micro and rebooted. Everything looked great. I could get to
DD-WRT console. I was about to backup the working CFE, NVRAM, Kernel of DD-WRT when one
of my solder points came loose. I was a little anxious so I rushed to solder job and didn't
realize some solder underneath the wire had inadvertently shorted TDO and TDI.
Unfortunately I didn't realize this until the WRT54G program was complaining it couldn't communicate
with the unit. I then realized there were no lights (power, LAN, etc.) when I plugged in the unit.
Things didn't look good. I immediately removed all the solder points and cleaned up the area.
Still no lights. I also removed the shielding plate thinking some solder may have gone through
to the other side and shorted something. Everything looked fine, but still no lights.
Now I'm realizing that no lights might not actually mean total failure. It could be possible
the unit will still respond to JTAG even if no lights show up at all. It occured to me that
in my rush to try and get power LEDs to show up, I never reconnected the JTAG cable properly
to see if the unit was responding despite lack of lights.
It is a pain to solder those tiny holes but I'll go ahead and do it again if there is some
hope.
What do you think from your experiences. Am I totally screwed if no lights show up when
plugged in, or could it still be responding to JTAG? What would theoretically happen if
one shorted TDO/TDI?
Maybe I'm being naive, but I'm thinking I'm probably not the first one to ever do this in
the history of JTAG debricking so someone probably has some advice to share.
In case anyone else ever runs into such problems...
I went ahead and resoldered the JTAG connection.
I was pleasantly surprised to find out the unit is still responding to JTAG.
I'm in the process of /flash:wholeflash right now.
Hopefully that will fix things up. You know, they say you learn more by making mistakes than having everything go right. I think that is very true.
I'm quite happy the first thing I did was create backups of CFE, NVRAM, Kernel, and Wholeflash. I highly suggest this for anyone considering JTAG. I know it is in the documentation but sometimes people get anxious and skip over it.
I'll post an update on whether the flash was successful and whether the unit comes back to life.
I have a Belkin 7230 1000 series which I have now successfully flashed with dd-wrt.
I am connected with Bigpond in Australia and the problem is with either the original belkin firmware or dd-wrt firmware that my ISP wont allow a dhcp lease any longer than 1 hour.
When the router is removed from the modem (SB5101 Cable modem) and the computer is directly connected to the modem, the DHCP lease will extend for 6 hours no probs.
This is extremely annoying as obviously I get disconnected from any games etc.
so someone probably has some advice to share.