**** BEGIN LOGGING AT Sun Dec 28 02:59:57 2008 Dec 28 10:26:46 juhosg * r13755 /trunk/target/linux/ar71xx/image/Makefile: [ar71xx] generate firmware image for the D-Link DIR-615 C1 board Dec 28 10:32:15 juhosg * r13756 /trunk/target/linux/ar71xx/ (config-2.6.27 config-2.6.28): [ar71xx] don't override CONFIG_NF_CT_ACCT Dec 28 20:27:42 nbd: Bist Du auf dem congress? Dec 28 20:32:40 nbd: Wir sollten dringent mal ueber uci sprechen. Alex macht da grade einen Aufstand ueber Dinge, die er nur zur Haelfte versteht. Dec 28 20:44:24 MateIn4: ping Dec 28 20:45:45 bartman007: pong Dec 28 20:47:04 xMff: Bist Du auf dem Congress? Dec 28 21:03:40 MateIn4: had some questions about the jtag reset process for the mi424, but we got it thanks. may have a few additional questions shortly Dec 28 21:04:01 x-alina: xMff is at the congress Dec 28 21:06:23 MateIn4: do you have a redboot image design to run from ram without being flashed? Dec 28 21:08:11 bartman007: have both ram and rom version on the wiki Dec 28 21:09:33 bartman007: didn't really address your question, the ram version assumes DRAM has been initialized. Dec 28 21:09:36 the redram image tries to initialize the sdram controller Dec 28 21:09:44 good Dec 28 21:10:48 one more thing, both version use the wrong serial port if you have that connected. I never use the serial and just telnet into the box Dec 28 21:10:49 which means it must be written to flash, correct? Dec 28 21:11:13 you may be able to load the RAM version from jtag Dec 28 21:11:30 I just always wrote to FLASH Dec 28 21:13:20 if the image is loaded into ram, initializing the sdram controller is problematic... but nbd just noticed noticed that the jtag device he is using does support flash :) Dec 28 21:17:19 good Dec 28 21:17:44 MateIn4: I will have time to recheck everything when i got back home from ccc.. also got a wrv54g, so can fix that up at last, too Dec 28 21:18:16 kaloz: good Dec 28 21:18:49 your script should work on that one as well, right? Dec 28 21:19:10 yes, i've actually tested that option as well Dec 28 21:19:24 oh, so you have one? Dec 28 21:19:41 yes, I do Dec 28 21:20:06 very important - you need to use firmware 2.03 since it's the only one that supports telnet Dec 28 21:20:19 xMff: Sorry, falls Du was gesagt hast, habe ich es leider nicht mitbekommen, da mir der olsrd die IPv6 default route geloescht hat Dec 28 21:21:28 MateIn4: ok, so downgrade/upgrade, then i need your script and probably a binary Dec 28 21:21:41 MateIn4: if you have a pretested one, i would appreciate that Dec 28 21:22:02 exactly, with the script you can get a dump of the existing image Dec 28 21:22:34 no, I haven't actually built images for the wrv54g. I only got one of those and I've lent it out Dec 28 21:24:11 kaloz: let me look into some code I did a few months back. I think I did a port of redboot to it but I'll have to dig it out Dec 28 21:24:35 someone sent me an email claiming that it worked Dec 28 22:02:14 MateIn4: so... your redrom image is crashing on my board Dec 28 22:02:35 in what way? give me some hint Dec 28 22:03:04 it always displays a + on the serial port, and sometimes random text before Dec 28 22:03:39 I never used the serial port, wait a bit and telnet to 192.168.1.1 Dec 28 22:03:47 use the WAN port! Dec 28 22:03:56 it is still running from the flash mapping when it crashes, so it looks like it hasn't finished copying itself into ram yet Dec 28 22:04:18 the jtag device is showing that the device is crashing. Dec 28 22:04:22 it runs from flash, does not copy itself to RAM Dec 28 22:04:47 can you build a romram image? Dec 28 22:04:51 ok, and the image is written properly Dec 28 22:05:07 no ROMRAM support for ixp4xx devices that I can tell Dec 28 22:05:44 IIRC intel has some reference romram images on their site, my memory may be foggy Dec 28 22:05:52 the image is written properly, it does verify. Dec 28 22:06:50 I couldn't find any ROMRAM, I tried. The sources do contain config files for ROMRAM but they don't work. Dec 28 22:07:21 would giving you the faulting address help? Dec 28 22:08:22 perhaps Dec 28 22:09:23 bartman007: earlier you said that the box was bricked, I hope there's no HW problem with it Dec 28 22:09:54 MateIn4: that is possible, but the user that shipped it to me indicated that it did previously boot Dec 28 22:10:16 0x50019388 Dec 28 22:10:30 he thought the problems were firmware related, but was conjecture. Dec 28 22:11:10 always the same address Dec 28 22:11:16 no deviations Dec 28 22:14:27 may want to try and just running it without jtag, it's in the pci_init routine but I'm not sure if it's exactly the same address Dec 28 22:15:23 i did try that Dec 28 22:15:48 I see Dec 28 22:16:01 are you sure you have a rev A unit? Dec 28 22:17:07 how can i tell? Dec 28 22:17:18 Kaloz: I can't verify it myself... but I believe that 'mankash' in #openwrt has finally got his wrv54gs to boot openwrt... he said he had some small issues left... but quite possibly... you may get some results soon... mankash said he was going to fill MateIn4 in on the details... since it was his port that helped him get it working... Dec 28 22:18:43 nbd: good question, I only got rev A units, but rev D look quite different Dec 28 22:19:13 rev D don't have an extra PHY for WAN port Dec 28 22:19:24 this one does Dec 28 22:19:59 maybe it's an issue with the power supply Dec 28 22:20:02 because it's not the original one Dec 28 22:20:14 maybe the minipci card draws too much power and causes it to crash Dec 28 22:20:40 I've had 7 units and only 3 good power supplies, fried 2 Dec 28 22:21:01 was the mini-pci changed? Dec 28 22:23:35 the power supply shows 5V 3A Dec 28 22:24:33 the pci bus has the mini-pci slot plus the two MoCA controllers Dec 28 22:34:18 MateIn4: then revD has to have a seaparate mach id later when you have one Dec 28 22:35:28 kaloz: could have redboot set revision TAG Dec 28 22:37:57 MateIn4: right, forgot that :) Dec 28 22:39:10 Kaloz: I saw on GPL sources that they check for HW revision by looking at the MoCA controllers, revs AB have older type. Dec 28 22:39:54 of course, it might be easier to check if the phy is there or not Dec 28 22:46:34 MateIn4: I'll try another power adapter on this board when I get a chance, I may have to combine a couple that we have here to supply sufficient amperage :) Dec 28 22:47:43 bartman007: it's a strange problem, 5 other users have successfully used that redboot image. interestingly, they all had rev A! Dec 28 22:48:10 I'll vouch for that... Dec 28 22:48:26 a bad psu could defintiely explain the "flakey" experiences the original user indicated Dec 28 22:48:44 I got three running it... and one had a bad psu... Dec 28 22:49:36 if I was at home I'd just hook it up to my test-bed psu, but berlin is a long ways away from home. Dec 28 22:50:39 good luck! check some other routers, I know that some USR models are also 5V 3A with same connector Dec 28 22:52:03 the usr's have 12vdc/3.33amp psu's Dec 28 22:52:13 hahaha, got a user in #openwrt asking about mi424 as we speak. Dec 28 22:53:05 Bartman007: I ordered one of these for 9$ from china... if it works... we'll be getting a bunch more.. http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=260276345014 Dec 28 22:53:21 9 including shipping Dec 28 22:53:28 the USR8052 has the same power supply Dec 28 22:53:39 oops, it's the USR8054 Dec 28 22:54:22 ohhh... okay... different model than the one's we worked on... Dec 29 01:26:02 agb * r13757 /trunk/package/kernel/modules/ (netdevices.mk netsupport.mk network.mk): [package] kernel/modules: split network makefile into netdevices and netsupport **** ENDING LOGGING AT Mon Dec 29 02:59:56 2008