| 00:00.32 | kergoth | [g2]: packages/linux/linux-nslu2/nslu2/defconfig (or similar) |
| 00:00.40 | [g2] | rwhitby-around, I read you the latest oe evangelist :) |
| 00:00.55 | [g2] | you are |
| 00:01.11 | rwhitby-around | yep - even contributed a changeset overnight to fix the redbook typo |
| 00:01.38 | [g2] | rwhitby-around, we'll make a sw guy out of you yet! :) |
| 00:02.00 | [g2] | talk about coming over to the dark-side :) |
| 00:02.04 | rwhitby-around | I thought I was the s/w guy? |
| 00:02.31 | [g2] | oh you love that soldering iron, I just know it! |
| 00:02.44 | [g2] | you and kas11 |
| 00:02.45 | tking-sleep | he he |
| 00:03.48 | tking | g2: any chance of any pix of the bare board? |
| 00:04.08 | [g2] | allan was supposed to post those some time |
| 00:04.40 | Tiersten | wooo |
| 00:04.49 | Tiersten | ]evsroi=n.2.352t |
| 00:04.49 | Tiersten | leen_tnebaeln= |
| 00:04.49 | Tiersten | owhi_=df4000c0000002-00 |
| 00:04.55 | Tiersten | It's byteswapped but it works :) |
| 00:05.05 | Tiersten | Just told it to try and read the flash from the jtag port |
| 00:05.43 | tking | dyoung: that info you gave me last night worked on my bdi |
| 00:06.00 | Tiersten | tking: cool |
| 00:06.18 | tking | looks like the jtag is working correctly. |
| 00:06.21 | Tiersten | ya |
| 00:06.23 | [g2] | Tiersten, that may only be in the output |
| 00:06.45 | Tiersten | Pinout is correct it seems. I'm not using TRSTN |
| 00:06.48 | tking | tiersten: did you set it for 16bit mode? |
| 00:07.11 | Tiersten | JTAG Tool only works in 16 bit mode |
| 00:07.29 | tking | I ran out of time today but was looking for a checksum'd image of the 2.3v25. |
| 00:07.57 | Tiersten | It looks okay to me |
| 00:08.01 | tking | I know it should be on the linksys site. |
| 00:08.11 | [g2] | Tiersten, can you read the flash part type ? |
| 00:09.23 | Tiersten | Erm |
| 00:09.43 | dyoung | Very Cool. |
| 00:09.54 | [g2] | Extremely cool! |
| 00:09.57 | dyoung | I might have to buy the [G2] $20 wiggler device. |
| 00:10.04 | Tiersten | I can't read the DID it seem |
| 00:10.07 | rwhitby-around | URL? |
| 00:10.20 | rwhitby-around | [for $20 wiggler] |
| 00:10.27 | Tiersten | You can make it yourself ;) |
| 00:10.32 | tking | url? |
| 00:10.36 | Tiersten | Or get a Altera ByteBlaster cable |
| 00:10.42 | [g2] | I haven't tested it yet but it should work. |
| 00:10.46 | dyoung | https://digilent.us/Sales/Product.cfm?Prod=JTAG3 |
| 00:10.50 | tking | I have a Xilinx Cable...will that work? |
| 00:10.51 | [g2] | exactly |
| 00:10.52 | Tiersten | Any place that stocks FPGA gear will stock JTAG cables |
| 00:11.00 | Tiersten | tking: Yeah. should do |
| 00:11.08 | tking | got several of those |
| 00:11.09 | Tiersten | You'll have to check the pinout but that's it |
| 00:11.19 | tking | this has flying leads. |
| 00:11.26 | tking | for inline or 2X style |
| 00:11.27 | dyoung | I could make it myself. But I'm too lazy. |
| 00:11.48 | dyoung | Did you hve to do something ugly for pull-ups? |
| 00:12.32 | rwhitby-around | I've got access to Xilinx cables at work. $20 seems a fair price for the time I'd waste making one myself. |
| 00:12.45 | [g2] | nod. |
| 00:13.17 | Tiersten | Seems to work okay just directly attaching to it. Didn't bother with any pullups |
| 00:13.17 | tking | me neithre |
| 00:13.17 | [g2] | probably cost me a grand to make with my soldering skillz |
| 00:13.17 | tking | *neither |
| 00:13.37 | [g2] | rwhitby-around, Guess you owe my that soda. |
| 00:13.46 | Tiersten | ? |
| 00:13.53 | tking | tiersten: can you write up something for the wiki? |
| 00:13.56 | Tiersten | Sure |
| 00:13.57 | rwhitby-around | If there's someone going to buy one, can you get an extra one for me? I'll paypal the cost and shipping. |
| 00:14.00 | tking | I'll add what I can afterwards |
| 00:14.02 | rwhitby-around | [g2]? |
| 00:14.23 | [g2] | oh -- now rwhitby-around plays forgetfull |
| 00:14.34 | tking | ha |
| 00:14.56 | rwhitby-around | you tell them [g2] - I want to see how much you embellish the bet :-) |
| 00:15.10 | [g2] | I bet you a soda JTAG would be available due to the manufacturing |
| 00:15.14 | Tiersten | ohh |
| 00:15.37 | *** join/#nslu2-linux bapper (~bapper@id-cralid-cuda1b-47-113.losaca.adelphia.net) |
| 00:15.49 | Tiersten | It doesn't look like they used it to program the flash though. You'd use a proper test point for that instead of just unpopulated resistor positions |
| 00:15.50 | tking | how else are you going to program and test automaticallY/ |
| 00:15.53 | rwhitby-around | yep, a long, long time ago (in nslu2-linux timescale( |
| 00:15.53 | Tiersten | dunno |
| 00:15.58 | tking | true |
| 00:16.06 | [g2] | FreeNode-#nslu2-linux.log:Sep 11 12:12:34 <aCidbOy> I'm addicted to diet soda |
| 00:16.07 | [g2] | FreeNode-#nslu2-linux.log:Sep 18 20:13:34 <[g2]> rwhitby-around, Guess you owe my that soda. |
| 00:16.07 | [g2] | FreeNode-#nslu2-linux.log:Sep 18 20:15:07 <[g2]> I bet you a soda JTAG would be available due to the manufacturing |
| 00:16.07 | tking | and there were NO pin marks on mine |
| 00:16.20 | [g2] | do! |
| 00:16.22 | Tiersten | yeah. none on mine as well |
| 00:16.22 | [g2] | doh! |
| 00:17.09 | tking | so they are using a multi-gang flash fixture and then soldering them in place. |
| 00:17.15 | Tiersten | Probably |
| 00:17.22 | rwhitby-around | [g2]: buy me a wiggler, and I'll put some extra on the Paypal for the soda :-) |
| 00:17.27 | Tiersten | heh |
| 00:18.08 | tking | rwhitby: coke or pepsi? |
| 00:18.27 | rwhitby-around | tking: that's up to [g2], I'll spring for either. |
| 00:18.32 | Tiersten | Now do I try and replace redboot... |
| 00:18.39 | [g2] | Aug 15 09:47:12 <rwhitby> If only we had a JTAG connector to plug it into :-) |
| 00:18.45 | Tiersten | heh |
| 00:19.01 | dyoung | I'm gonna buy one next week. |
| 00:19.11 | Tiersten | What? A pepsi? ;) |
| 00:20.05 | tking-dinner | Tiersten : I'll compare notes when I get back. |
| 00:20.29 | Tiersten | tking-dinner: yup. enjoy your dinner |
| 00:20.49 | tking-dinner | ty\ |
| 00:21.02 | dyoung | Thats why I was asking about ugly pullups. I didnt measure it yet, but it looked like a standard 2mm SMT block would fit. |
| 00:22.10 | rwhitby-around | bbiab |
| 00:33.20 | Tiersten | dyoung: Yeah. I didn't bother with any pullups and it seems to be okay |
| 00:43.05 | dyoung | Yeah, I DO sorta need to buy me some mt dew. I'm totally out. |
| 00:43.28 | dyoung | but now that Teirsten has blazed the trail, I might as well get that JTAG cable while I'm at it. |
| 00:44.01 | [g2] | well, I don't the rwhitby-around owe's me a soda :( |
| 00:44.51 | [g2] | He didn't think JTAG would be on a consumer device, but I don't think I made the bet (other than talking to my screen) |
| 00:45.38 | [g2] | I was almost certain it would be and bet an NSLU2 to find out the traces |
| 00:45.58 | dyoung | That was a pretty pricy bet. |
| 00:46.04 | dyoung | What happened with that unit anyways? |
| 00:46.07 | [g2] | that seems like *forever* ago |
| 00:46.23 | [g2] | that's the unit I sent to allan |
| 00:46.31 | dyoung | Right. But what was broke about it? |
| 00:46.43 | Tiersten | The Linksys APs have JTAG ports as well |
| 00:46.49 | [g2] | nothing originally |
| 00:47.08 | Tiersten | It's even easier than the NSLU2 as they've got a position for a pin header but just not populated |
| 00:47.20 | [g2] | then I figured I'd try my hand at soldering if he was going to torch it anyway |
| 00:47.55 | rwhitby-around | [g2]: what's the price of a good soda your way? |
| 00:48.06 | *** join/#nslu2-linux GermanBirdman (~chatzilla@pcp0010078833pcs.eatntn01.nj.comcast.net) |
| 00:48.17 | [g2] | Tiersten, they also have a bastardized JTAG and no BSDL |
| 00:48.35 | Tiersten | Yeah that's what you get for using Broadcom chips :) |
| 00:49.05 | [g2] | rwhitby-around, $0.60US |
| 00:49.23 | [g2] | Can you even send me a postcard for less ? |
| 00:49.29 | Tiersten | How do you put preformatted text in the wiki? |
| 00:49.36 | Tiersten | It's trying to reformat it and it goes all weird |
| 00:49.41 | jacques_gone | Tiersten, good question |
| 00:49.53 | jacques_gone | even what the docs say didnt work for me |
| 00:50.03 | jacques_gone | I ended up putting leading spaces and that worked |
| 00:50.13 | Tiersten | ah okay. cool. I'll give that a try |
| 00:50.30 | jacques_gone | the [=...=] thing absolutely had no effect |
| 00:50.41 | jacques_gone | nor did @@...@@ |
| 00:51.04 | rwhitby-around | what was Lonewolf's donation status page again? |
| 00:51.08 | [g2] | "/PRE .. |
| 00:51.49 | dyoung | So you sent a perfectly good slug to Allan, all in the name of.. science?? Youre THE MAN. |
| 00:51.56 | dyoung | jbot, g2? |
| 00:51.57 | jbot | Easy to use, portable and powerful 2D graphics library. URL: http://www.ap.univie.ac.at/users/ljubo/g2/g2.shtml |
| 00:51.58 | [g2] | http://www.greyskydesigns.com/~lonewolf/nslu2.php |
| 00:52.10 | dyoung | jbot, [g2]? |
| 00:52.11 | jbot | i guess [g2] is really cool, or an all around nice guy!, or a funny guy, or the master of NSLU2 Buildroot. |
| 00:52.57 | rwhitby-around | [g2]: what's your paypal address? |
| 00:52.58 | dyoung | jbot, [g2] is also a Slug Sacrificer. |
| 00:52.59 | jbot | dyoung: okay |
| 00:53.03 | rwhitby-around | [/msg me] |
| 00:53.07 | [g2] | LOL |
| 00:53.18 | dyoung | jbot, [g2]? |
| 00:53.20 | jbot | extra, extra, read all about it, [g2] is really cool, or an all around nice guy!, or a funny guy, or the master of NSLU2 Buildroot. a Slug Sacrificer. |
| 00:53.56 | dyoung | ;-) |
| 00:54.04 | [g2] | :) |
| 00:54.20 | GermanBirdman | I have a question. I don't own one of these beauties yet, but it is ordered. I saw on the yahoo group discussion that basically only ext2, ext3 and vfat is compiled in, no iso9660 needed to understand cdroms. Would it be possible to compile use a kernel module, or would it have to be compiled in? |
| 00:54.32 | dyoung | I changed your .config for busybox in unslung yesterday. I hope it doesnt break anything. I added CONFIG_LFS. |
| 00:56.35 | [g2] | dyoung are you on the bk oe train yet ? |
| 00:56.50 | dyoung | No. |
| 00:56.58 | dyoung | I'm having some serious PEBCAK issues. |
| 00:57.06 | dyoung | PEBKAC. |
| 00:57.28 | rwhitby-around | Unslung 2.0 will be built from OE, and will use the latest ipkg format. |
| 00:57.33 | dyoung | jbot, pebkac? |
| 00:57.34 | jbot | Sounds like the Problem Exists Between Keyboard And Chair |
| 00:57.52 | dyoung | I'll catch up to you guys later. |
| 00:57.59 | dyoung | I need to make this build machine work first. |
| 00:58.26 | dyoung | I dont have enough space for anything right now. |
| 00:58.45 | dyoung | (besides, then I can bug you guys for my dumb questions) |
| 00:58.47 | [g2] | rwhitby-around, So I just need to do a bk edit packages/linux/linux-nslu2-2.4.22/nslu2/defconfig to change the kernel config ? |
| 00:59.14 | [g2] | and no I don't run emacs |
| 00:59.16 | rwhitby-around | well, I offered to PayPal [g2] for a six-pack, but he declined. I kept up my end of the bet :-) |
| 00:59.35 | [g2] | whoa a six-pack! |
| 00:59.44 | rwhitby-around | six pack of soda :-) |
| 00:59.52 | [g2] | Yes you did! Thx |
| 01:00.17 | Tiersten | I've updated the Wiki now |
| 01:00.35 | dyoung | GermanBirdman, You currently cannot compile a module that will be compatible with NSLU2 factory firmware. |
| 01:01.34 | GermanBirdman | thanks... |
| 01:01.52 | rwhitby-around | [g2]: I reckon that would do it. Then oemake -c clean linux-nslu2 should get it ready for rebuilding |
| 01:02.07 | [g2] | rwhitby-around, thx |
| 01:02.21 | [g2] | I was gonna try usbnet and usbserial |
| 01:02.25 | rwhitby-around | I think the linux-nslu2 is the right target. |
| 01:02.50 | dyoung | usbserial worked for me for 2.6.7. pegasus (net) blew up though. |
| 01:03.00 | [g2] | I think so too, that'd be my guess but I think an oemake will probably just make it |
| 01:03.39 | rwhitby-around | I'm not sure yet of how oemake handles dependencies on patch files and stuff like that. |
| 01:03.53 | rwhitby-around | suck it and see |
| 01:04.25 | Tiersten | Hmm... Who had the bare PCB again? |
| 01:04.38 | [g2] | allan on the yahoo message group |
| 01:04.41 | Tiersten | Wonder if there are extra address lines wired to the RAM pads? |
| 01:04.48 | [g2] | yes |
| 01:04.48 | Tiersten | Memory upgrade :) |
| 01:04.51 | Tiersten | cool |
| 01:04.51 | rwhitby-around | [g2]: I wonder if you can push and pull changesets between clones of the same bk repostory. I expect you could. |
| 01:05.06 | [g2] | iirc he already checked and it's in the yahoo messages |
| 01:05.16 | rwhitby-around | Tiersten: someone on the list posted the part number for the chips to do taht |
| 01:05.48 | [g2] | rwhitby-around, I newer to bk than you, but my reading of the doc is that you can between peers based of the same partent |
| 01:05.59 | [g2] | s/partent/parent |
| 01:06.23 | rwhitby-around | yeah - if we can do it by emailing changesets, then even better (don't have to expose a bk repository to the internet) |
| 01:06.42 | rwhitby-around | [bk send] |
| 01:06.50 | Tiersten | rwhitby-around: Must have missed that one. I'll go look in the archives |
| 01:07.45 | rwhitby-around | Aug 20: Adding RAM (can't get yahoo article number from my gmail archive) |
| 01:09.17 | rwhitby-around | hmm - thought the part number was there. Will search further. |
| 01:10.09 | rwhitby-around | Aug 13: Adding more RAM |
| 01:10.12 | rwhitby-around | that's the one |
| 01:10.29 | Tiersten | ah. wasn't subscribed at that point |
| 01:10.41 | dyoung | As soon as I get this stupid machine going I'll jump on the bandwagon. |
| 01:10.51 | rwhitby-around | I've got a gmail archive going right back to the first message |
| 01:10.52 | dyoung | I've totally given up on buildroot linux.mk hacking. |
| 01:14.17 | rwhitby-around | Tiersten: are the JTAG pads on the board in the right order to just solder a header on there for this pinout: https://digilent.us/Sales/Product.cfm?Prod=JTAG3 |
| 01:15.10 | Tiersten | erm let me check |
| 01:15.39 | rwhitby-around | [e.g. can you update the wiki with the physical dimensions and pinout if we want to solder a header directly onto the board] |
| 01:16.04 | Tiersten | I soldered wires as I wanted to be able to shut the case |
| 01:16.22 | Tiersten | There isn't that much between the PCB and casing on that side |
| 01:16.32 | Tiersten | much space between |
| 01:16.35 | rwhitby-around | True - maybe I'll have to have another flat cable coming out of the Disk2 hole |
| 01:17.13 | rwhitby-around | dyoung: So you're going to buy digilent cable this week? |
| 01:17.39 | Tiersten | rwhitby: It's not in the correct order |
| 01:18.08 | Tiersten | The NSLU2 has it in TDI, TMS, TCK, TDO order |
| 01:19.26 | Tiersten | It's the left pads if you've got the PCB orientated with the USB ports at the top BTW |
| 01:19.58 | Tiersten | Not hard to work out anyway :) |
| 01:20.33 | [biovore] | probably should make some more pics about the serial port mod.. its not hard at all.. |
| 01:22.45 | rwhitby-around | yeah, I've been meaning to get around to it. But this kernel hacking stuff is distracting ... |
| 01:24.48 | Tiersten | bbiab |
| 01:28.37 | [g2] | Tiersten, Looks like it see that flash device |
| 01:29.20 | dyoung | Yes, maybe wedesnday or so. |
| 01:30.31 | Tiersten | back |
| 01:30.33 | Tiersten | [g2]: huh? |
| 01:30.54 | rwhitby-around | dyoung: would you be able to buy one for me too? |
| 01:31.03 | [g2] | Device geometry definition: |
| 01:31.03 | [g2] | <PROTECTED> |
| 01:31.03 | [g2] | <PROTECTED> |
| 01:31.03 | [g2] | <PROTECTED> |
| 01:31.05 | rwhitby-around | [I'll PayPal you for it] |
| 01:31.30 | Tiersten | ah yeah. I can't query the part number though. I think there is a config page in there somewhere with the IDs |
| 01:31.31 | Tiersten | *shrug* |
| 01:32.32 | jacques_gone | will wigglers work? |
| 01:32.35 | Tiersten | yeah |
| 01:32.37 | [g2] | Tiersten, what are you planning for the JTAG ? |
| 01:32.52 | Tiersten | Not planning anything. Trying erasing at the moment |
| 01:32.54 | jacques_gone | cool |
| 01:34.17 | [g2] | Tiersten, a read first would have been nice to actually verify the data in the flash |
| 01:34.30 | Tiersten | Err. I've already done that... |
| 01:34.43 | Tiersten | I said it was byteswapped but it's reading fine |
| 01:35.02 | [g2] | I think that's just the display of info. |
| 01:35.03 | jacques_gone | man WTF firefox pr1 doesnt work at all for me |
| 01:35.16 | [biovore] | ? |
| 01:35.25 | [biovore] | no problems here with it.. |
| 01:35.32 | [g2] | Tiersten, did you read out the flash chip ? |
| 01:35.34 | Tiersten | [g2]: It's probably because I'm viewing it on a x86 |
| 01:35.38 | *** join/#nslu2-linux rwhitby-treo (~upirc@203.202.55.131) |
| 01:35.42 | Tiersten | Endianness diff |
| 01:35.43 | [g2] | Tiersten, nod. |
| 01:35.51 | jacques_gone | it's totally screwed |
| 01:36.02 | jacques_gone | it wont even quit |
| 01:36.03 | Tiersten | I've read out redboot and the sysconf bit and they seem to be fine |
| 01:36.22 | [g2] | AWESOME |
| 01:36.23 | Tiersten | It matches up with what I can get by dumping it out via the NSLU2 in Linux |
| 01:36.45 | [g2] | with the dd or the .bin image |
| 01:36.49 | Tiersten | dd |
| 01:36.58 | [g2] | there both the same :) |
| 01:37.03 | [g2] | well nearly |
| 01:37.12 | Tiersten | ya but I'm reading the flash and not whatever Linksys prepared |
| 01:37.48 | [g2] | well after you flash the fw upgrade the pretty identical :) |
| 01:38.00 | Tiersten | Yeah but I wanted to be certain that it was working out :) |
| 01:38.11 | [g2] | excellent move |
| 01:38.26 | Tiersten | heh |
| 01:38.35 | Tiersten | tking has done it as well |
| 01:38.44 | Tiersten | He tried it using a BDI2000 at work |
| 01:41.00 | Tiersten | ah. it does detect what chip it is when it's flashing |
| 01:41.07 | Tiersten | 28F640J3A |
| 01:42.19 | jacques_gone | [biovore], apparently it was really unhappy with something in my old config - it wouldnt do anything - couldn't go to web pages, nothing. I deleted .mozilla/firefox and now it seems OK |
| 01:43.16 | dyoung | I'll pick one up for you too. |
| 01:44.37 | rwhitby-treo | thx - that's the url that you posted above, right? |
| 01:44.52 | [biovore] | oh.. in linux.. yeah.. its kinda bitchy about that.. |
| 01:45.07 | dyoung | Yes, the digilent jtag3 xilinux unit |
| 01:45.13 | dyoung | xilinx even |
| 01:45.14 | jacques_gone | [biovore], it could have warned me |
| 01:45.29 | rwhitby-treo | dyoung: thx - please do |
| 01:45.38 | jacques_gone | dyoung what are you guys getting? I might as well get the same thing |
| 01:45.51 | dyoung | My raid1 on the home pc doesnt autorun. What the hell. |
| 01:45.53 | dyoung | https://digilent.us/Sales/Product.cfm?Prod=JTAG3 |
| 01:47.21 | jacques_gone | dyoung thanks |
| 01:47.59 | jacques_gone | I assume we dont need the -LV, just the regular voltage model? |
| 01:48.10 | Tiersten | It has to support 3.3V |
| 01:48.22 | Tiersten | The plain one should do |
| 01:48.31 | jacques_gone | thanks |
| 01:49.10 | Tiersten | JTAG is an absolute last resort thing anyway. It's slooooooooow to reflash any significant amounts of memory |
| 01:49.26 | jacques_gone | but it's the only way to go for bootloader development |
| 01:50.05 | Tiersten | No it's not. You just have to be extremely lucky :) |
| 01:50.49 | rwhitby-treo | Friends don't friends replace bootloaders without JTAG access. |
| 01:50.57 | rwhitby-treo | don't let ... |
| 01:50.58 | Tiersten | true |
| 01:51.30 | [biovore] | lol rwhitby-treo.. thats a good one.. |
| 01:51.45 | jacques_gone | well experimental bootloaders at least |
| 01:52.00 | Tiersten | Always useful to have a JTAG adapter. Nearly everything has a port these days |
| 01:52.01 | dyoung | Wasnt there a trailer to that line like ... Usually. or something? |
| 01:52.25 | dyoung | Because I dont recall anyone saying "dont do that" the first time I made a brick.... |
| 01:52.38 | jacques_gone | I've flashed lots of bootloaders to my ipaq but only after they were declared safe, and I have seen a lot of ppl brick them |
| 01:53.41 | jacques_gone | same with the tuxscreens |
| 01:55.51 | Tiersten | Yeah. You'd have to ship it back to Compaq to get them fixed |
| 01:56.04 | jacques_gone | yep |
| 01:56.39 | jacques_gone | and since this was for linux development, only the CRL would do it for free, and that took weeks because they were so busy with other things |
| 01:56.57 | Tiersten | Busy with the shear numbers of bricked iPaqs as well i'm sure |
| 01:58.37 | jacques_gone | I don't think they ever had more than about 10 at one time |
| 01:58.51 | jacques_gone | but there was only one guy there who does it |
| 01:59.06 | jacques_gone | and he has to open each one up, attach the JTAG leads, etc ,etc |
| 01:59.21 | jacques_gone | and this isn't his real job :-) |
| 01:59.37 | Tiersten | Must have been fun... |
| 02:00.09 | Tiersten | There don't seem to be any binaries for RedBoot 1.92 on the Intel page. The older release was just source |
| 02:00.45 | jacques_gone | are you looking to compare with that's in flash? the nslu2 one is modified by sercomm |
| 02:00.58 | Tiersten | Is it? |
| 02:01.07 | Tiersten | What have they added? The upgrade and assign commands? |
| 02:02.54 | Tiersten | ah. assign, boot, move and upgrade |
| 02:03.04 | jacques_gone | others here know much more about that - I just got my serial installed last night - but even I saw sercomm mentioned in the help |
| 02:03.30 | jacques_gone | and the bootloader also looks for the trailer or goes into upgrade mode I think |
| 02:03.42 | jacques_gone | and it seems they disabled some stuff |
| 02:03.56 | jacques_gone | all reasons we want to be able to change it |
| 02:04.01 | Tiersten | The redboot actually in the flash has a header of some kind |
| 02:04.22 | Tiersten | There is a big chunk of nulls near the beginning which doesn't exist in the stock intel binaries |
| 02:04.30 | jacques_gone | rwhitby-away, mentioned some commands they added but I don't recall which |
| 02:04.43 | Tiersten | whether this is some sort of header or a side effect of how Linksys/Sercomm compiled it i don't know |
| 02:04.59 | Tiersten | It just lists assign, boot, move and upgrade in the Wiki |
| 02:05.29 | jacques_gone | ah I didn't think to look there |
| 02:06.53 | Tiersten | Writing seems to work |
| 02:07.27 | *** join/#nslu2-linux CodeWhacker (~CodeWhack@nr24-216-68-196-228.fuse.net) |
| 02:08.21 | rwhitby-treo | jacques: we have the source to the modified redboot |
| 02:09.38 | jacques_gone | hmm where is the redboot stuff in the wiki I cant find it |
| 02:10.10 | Tiersten | In information |
| 02:11.10 | rwhitby-away | the modified redboot source is in the latest Linksys source release from a couple of week ago (matches V23R25). |
| 02:11.11 | jacques_gone | all I see is RedBootSequence |
| 02:11.16 | Tiersten | that's it |
| 02:12.09 | jacques_gone | ok I see it now |
| 02:14.21 | jacques_gone | rwhitby-away, that'll be useful - I didn't realize we had that |
| 02:22.56 | rwhitby-away | any good free jtag tools for windows (yes, I know, don't say it) |
| 02:23.19 | Tiersten | JTAG Tools works |
| 02:23.26 | Tiersten | You need to install Cygwin to compile it though |
| 02:23.56 | rwhitby-away | that'll work then - I have Cygwin already |
| 02:25.55 | rwhitby-away | [still waiting for CoLinux to support the I/O ports] |
| 02:26.47 | rwhitby-away | dyoung: so we've confirmed that the cable you are buying is compatible with Xilinx Parallel Cable III, and hence will work with JTAG Tools? |
| 02:29.07 | dyoung | According to the webpage it is. |
| 02:29.08 | Tiersten | If it is definitely compat with a Xilinx Parallel Cable III then yeah, it'll work with JTAG Tools |
| 02:31.20 | dyoung | The intel document talks about a windows program named raven |
| 02:31.34 | dyoung | BTW teirsten, did you find the pre-made binary redboot files on the intel site? |
| 02:31.50 | Tiersten | dyoung: Yeah |
| 02:32.11 | Tiersten | dyoung: Raven is a Macraigor thing |
| 02:32.21 | dyoung | Were you gonna give 'er a go? |
| 02:32.22 | Tiersten | Like the Wiggler also from them |
| 02:32.45 | Tiersten | dyoung: Not sure. Considering it. It takes so long to program just 128K |
| 02:36.12 | Tiersten | It's a lot faster on a BDI2000 but obviously doesn't cost $20 |
| 02:44.22 | Tiersten | Whats the equiv commands to the boot script? If I try the stock redboot from Intel then it won't have the boot script or the boot command |
| 02:45.50 | dyoung | Theres load right? |
| 02:46.21 | Tiersten | Yeah. What's the addresses? |
| 02:46.40 | dyoung | For factory firmware: |
| 02:46.42 | dyoung | load -r -v -b 0x1d00000 /nslu2/vmlinuz.sda |
| 02:46.42 | dyoung | load -r -v -b 0x1000000 /nslu2/ramdisk.g2.rc.gz |
| 02:47.18 | dyoung | then I would exec 0x1d00000 |
| 02:47.28 | dyoung | But Not sure if theres exec. |
| 02:47.45 | Tiersten | exec is there I think... |
| 02:48.13 | Tiersten | It should be there anyway |
| 02:50.53 | Tiersten | ah no. I'm not trying to load an image remotely :) |
| 02:51.39 | Tiersten | Anyway, extremely late so I'm calling it a night |
| 02:51.40 | Tiersten | Night all |
| 02:52.25 | jacques_gone | night Tiersten |
| 02:53.13 | rwhitby-treo | night thx |
| 03:02.23 | *** part/#nslu2-linux rwhitby-treo (~upirc@203.202.55.131) |
| 03:07.50 | *** join/#nslu2-linux rpedde (~ron@adsl-70-240-56-201.dsl.snantx.swbell.net) |
| 03:19.01 | jacques_gone | bah the redboot docs are not too useful yet since everything is disabled |
| 03:19.10 | jacques_gone | no fis list, no fconfig |
| 03:34.22 | rwhitby-away | yeah - the source is the best to look at |
| 03:34.46 | jacques_gone | ah it looks like our sick idea for loading kernels over http might not work - according to redboot docs it cannot route - so the implication is even if you use http the server has to be on the same subnet |
| 03:35.10 | jacques_gone | unless you use an arp proxy... |
| 03:35.16 | jacques_gone | not sure |
| 03:41.31 | *** join/#nslu2-linux jacques_gone (~fontenot@c-67-171-198-193.client.comcast.net) |
| 03:41.35 | jacques_gone | crap |
| 03:41.45 | jacques_gone | why did they have to choose 102.168.0.1 |
| 03:41.48 | jacques_gone | 192.168.0.1 |
| 03:42.43 | jacques_gone | i was playing with ping in redboot and the nslu2 was apparently using that address instead of the one I assigned it |
| 03:42.49 | jacques_gone | and it collided with my router |
| 03:52.06 | *** join/#nslu2-linux CIA-5 (~CIA@to.je.spocco.com) |
| 03:54.03 | [g2] | jacques_gone, oe question ? |
| 03:54.19 | jacques_gone | [g2], ask away I'll try my best :-) |
| 03:54.46 | [g2] | I'd like to run menuconfig for the linux-nslu2 kernel |
| 03:55.10 | [g2] | actually oldconfig first then menuconfig |
| 03:58.57 | jacques_gone | you mean you would like to use a different kernel config than the oe default for nslu2 ? |
| 03:59.22 | [g2] | yes. I tweaked the defconfig but I think I missed something |
| 03:59.38 | dyoung | ip_address -l 192.168.1.77 |
| 04:00.26 | dyoung | I hate my new build machine. |
| 04:00.26 | jacques_gone | [g2], you changed packages/linux/linux-nslu2-2.4.22/nslu2/defconfig and it didn't take effect? |
| 04:00.53 | jacques_gone | dyoung thanks |
| 04:00.58 | [g2] | yes it took effect, but it's wrong :) and I like to fix it :) |
| 04:01.10 | [g2] | is deprecated |
| 04:01.10 | [g2] | usbnet.c: In function `usbnet_probe': |
| 04:01.10 | [g2] | usbnet.c:2518: error: structure has no member named `ethtool_ops' |
| 04:01.10 | [g2] | usbnet.c: At top level: |
| 04:01.10 | [g2] | usbnet.c:2742: error: variable `usbnet_ethtool_ops' has initializer but incomplete type |
| 04:01.10 | [g2] | u |
| 04:01.41 | [g2] | log.do_compile.13063 |
| 04:02.04 | jacques_gone | you want to run menuconfig and then have oe continue building the kernel? |
| 04:02.42 | [g2] | oldconfig then maybe menuconfig |
| 04:03.00 | [g2] | I don't think I can do that in the middle |
| 04:03.10 | jacques_gone | there might be a way but I don't know it :-\ |
| 04:03.19 | [g2] | ok thx |
| 04:03.39 | jacques_gone | but that might be a good question for #oe |
| 04:03.44 | jacques_gone | I would like to know the answer myself |
| 04:04.31 | [g2] | ok one other question. I did a bk edit, is there a way to undo that ? |
| 04:05.05 | [g2] | * it just escapes him atm |
| 04:06.59 | tking-dinner | rwhitby-away : I've got the Xilinx III cables too but just got the BDI2000 to talk to it today. |
| 04:07.05 | tking-dinner | it *should* work. |
| 04:07.33 | *** join/#nslu2-linux CIA-5 (~CIA@to.je.spocco.com) |
| 04:08.48 | [g2] | Aborting Changes |
| 04:08.48 | [g2] | To undo the effect of any editing, i.e. throw away changes in a work directory...do: |
| 04:08.48 | [g2] | <PROTECTED> |
| 04:08.48 | [g2] | <PROTECTED> |
| 04:13.17 | rwhitby-away | g2: you should just be able to go into the build directory and run make menuconfig there like in a normal kernel directory |
| 04:13.38 | rwhitby-away | then copy the .config file back over the defconfig when you are finished, then rebuild |
| 04:13.58 | [g2] | rwhitby-away, the default will use the x86 tools |
| 04:16.27 | [g2] | rwhitby-away, I think we want to run the make like oe. I've got that in the log iirc |
| 04:16.52 | [g2] | that'll probablly work |
| 04:17.54 | [g2] | hmmm... The changes are still there |
| 04:35.37 | [g2] | jacques_gone, still gone ? |
| 04:36.06 | tking | the fun of distributed development. |
| 04:36.23 | [g2] | tking thx for all your work on the wike |
| 04:36.25 | [g2] | i |
| 04:37.41 | tking | thanks. I'm not a Kernel Hacker (although I work with Kernel Hackers!) |
| 04:39.04 | *** join/#nslu2-linux CodeWhacker (~CodeWhack@nr24-216-68-196-228.fuse.net) |
| 04:39.25 | [g2] | I'm just starting |
| 04:39.28 | tking | at work we are under such pressure on the project we are working on that I don't get a chance to understand the process. Here I've started to get an appreciation for the process. |
| 04:39.53 | tking | I write the VHDL for the glue logic. |
| 04:40.01 | [g2] | there's nothing like real proficiency |
| 04:40.15 | [g2] | cool |
| 04:40.56 | tking | for fun I wrote a cpu model (4 bit) in vhdl |
| 04:41.09 | tking | don't do verilog |
| 04:42.00 | [g2] | another project I've been watching is the CPU on a softcore |
| 04:42.27 | [g2] | opencores and microblaze are 2 of the 3 candidates I've been following |
| 04:42.40 | tking | next project at work is uBlaze |
| 04:43.08 | tking | our shop uses embedded ppc and we are probalby going to start using the ppc405 in our products. |
| 04:43.10 | [g2] | the microblaze guys are running uCLinux and have been for months |
| 04:43.40 | tking | our current product is 860T based. |
| 04:44.12 | CodeWhacker | anyone knoe where the xfs patches came from in the oe anyu2 build? |
| 04:44.18 | [g2] | cool, I did a couple years of work with moto stuff 68360, 860 |
| 04:44.23 | CodeWhacker | s/knoe/know/ |
| 04:44.31 | [g2] | a little 603e |
| 04:44.57 | [g2] | CodeWhacker, kergoth put them in oe somehow |
| 04:45.14 | [g2] | I'm running the kernel in oe with his root fs |
| 04:45.44 | [g2] | I was trying to build the usbnet and usbserial drivers unsuccessfully -- due to my ignorance I'm sure |
| 04:45.45 | CodeWhacker | [g2], I think some 2.6 stuff snuck into the 2.4 tree. |
| 04:46.07 | [g2] | in oe ? |
| 04:46.33 | [biovore] | hows the progress on a 2.6 kernel for the nslu2.. any clues? |
| 04:46.38 | CodeWhacker | usbnet doesn't compile. it blows up on a symbol that I can't find anywhere in 2.4, but see references to in 2.6 |
| 04:47.02 | [g2] | ahh.. that's what I was getting |
| 04:47.27 | [g2] | | usbnet.c: In function `usbnet_probe': |
| 04:47.27 | [g2] | | usbnet.c:2518: error: structure has no member named `ethtool_ops' |
| 04:47.27 | [g2] | | usbnet.c: At top level: |
| 04:47.28 | [g2] | | usbnet.c:2742: error: variable `usbnet_ethtool_ops' has initializer but incomplete type |
| 04:47.28 | [g2] | | |
| 04:47.55 | CodeWhacker | that's the one. http://www.ussg.iu.edu/hypermail/linux/kernel/0308.2/0282.html suggests this is a 2.6ism |
| 04:48.34 | CodeWhacker | after a few hours' digging, I found the source of the changes in 2.4.22-xfs-nslu2.patch.bz2 |
| 04:48.36 | [g2] | well, kergoth is working on getting 2.6 in there too, maybe it's just a non-default kernel issue |
| 04:50.08 | CodeWhacker | I guess the next step is to whack those parts of the patch out and see if it compiles |
| 04:50.15 | [g2] | I'm not really planning on using the 2.4.x kernel much if at all really |
| 04:50.39 | [g2] | did you want to try something ? |
| 04:51.02 | CodeWhacker | well, of course! I'm not patient! :) |
| 04:51.18 | [g2] | what did you want to try ? |
| 04:51.34 | CodeWhacker | I wanted to try out the oe build |
| 04:51.44 | CodeWhacker | even without usbnet |
| 04:51.45 | [g2] | it works and runs fine |
| 04:53.02 | [g2] | no networking, do you want the boot messages from startup ? |
| 04:53.07 | CodeWhacker | btw, is there a way to rebuild without rebuilding all the cross tools? the wiki says "remove .../build/tmp", which works but makes for some long rebuild times |
| 04:54.07 | [g2] | I've bk edit'ed a file I don't know how to unedit :) |
| 04:54.17 | [g2] | the nslu2 kernel defconfig :) |
| 04:54.29 | CodeWhacker | bk seems pretty straightforward, 'specially since they have a page showing what CVS commands equate to what bk commands |
| 04:54.51 | CodeWhacker | uh-oh... |
| 04:55.30 | CodeWhacker | delete it and do another pull |
| 05:00.22 | [g2] | is that a regular delete or a bk delete ? |
| 05:01.02 | [g2] | I was also looking to add lsof to the root_fs |
| 05:02.50 | dyoung | lsof is your friend. |
| 05:02.58 | GermanBirdman | Sorry to disturb and I don't even own a box yet (it is ordered) - I'm just wondering for your custom kernel, which linux loader do you use? Lilo, Grub, or is it implemented in redboot? (eventhough from the redboot website it doesn't appear to be a lilo) |
| 05:03.11 | GermanBirdman | just wondering... |
| 05:03.26 | tking | redboot |
| 05:03.40 | tking | that is the bootloader. |
| 05:03.52 | GermanBirdman | oh ok... (stupid me) |
| 05:03.57 | [g2] | GermanBirdman, are you familiar with the wiki |
| 05:04.08 | tking | www.nslu2-linux.org |
| 05:04.18 | GermanBirdman | I read everything I could |
| 05:04.27 | [g2] | http://www.nslu2-linux.org/wiki/Info/RedBootSequence |
| 05:04.58 | GermanBirdman | That entry I haven't seen yet, thanks! |
| 05:05.07 | [g2] | np |
| 05:06.18 | CodeWhacker | regular delete. then bk pull should refresh the copy |
| 05:08.10 | [g2] | thx |
| 05:08.38 | CodeWhacker | only just? you're new to programming? |
| 05:08.45 | CodeWhacker | :) |
| 05:08.56 | [g2] | ash-2.05b$ oemake bootstrap-image 2>&1 | tee 0919_0105_rm_get.log |
| 05:08.56 | [g2] | NOTE: Psyco JIT Compiler (http://psyco.sf.net) not available. Install it to increase performance. |
| 05:08.56 | [g2] | NO |
| 05:09.46 | CodeWhacker | emerge psyco |
| 05:10.22 | [g2] | thx |
| 05:10.52 | [g2] | gentoo does roxor |
| 05:11.00 | dyoung | This makes 4th install attempt. |
| 05:11.09 | tking | as long as your build machine is up to it. |
| 05:11.40 | tking | [g2] |
| 05:11.47 | [g2] | tking |
| 05:12.00 | CodeWhacker | only took a week! ;) |
| 05:12.12 | tking | [g2] as long as your build machine is up to it. |
| 05:13.02 | [g2] | well there's the GRP with does a bunch of binary installs |
| 05:13.12 | [g2] | that's no super intensive |
| 05:13.15 | [g2] | not |
| 05:13.57 | tking | ah |
| 05:14.42 | [g2] | not that any of my machines are using the grp stuff :) |
| 05:14.45 | CodeWhacker | true, and I used a bunch of the binaries. but after 'emerge sync' many of them were out of date. fortunately, I wasn't on a schedule. |
| 05:15.28 | CodeWhacker | with all the deps, 'emerge desklets-psidisplays' took about 36 hours |
| 05:15.34 | [g2] | CodeWhacker, the delete and pull doesn't fix it |
| 05:16.01 | CodeWhacker | crap! lemme go see something... |
| 05:16.18 | [g2] | more just fyi than I need a solution now |
| 05:17.34 | [g2] | and I need got get up to speed on bk |
| 05:17.36 | [g2] | and oe |
| 05:17.49 | [g2] | repeat the mantra |
| 05:19.15 | CodeWhacker | aha! 'bk co defconfig' |
| 05:19.29 | CodeWhacker | delete the frobbed one first |
| 05:26.12 | [g2] | looks like it's time for a memory upgrade |
| 05:27.39 | [g2] | Cpu(s): 24.1% us, 27.1% sy, 0.0% ni, 47.8% id, 0.7% wa, 0.3% hi, 0.0% si |
| 05:27.40 | [g2] | Mem: 514404k total, 501488k used, 12916k free, 22680k buffers |
| 05:27.40 | [g2] | Swap: 248968k total, 97928k used, 151040k free, 66652k cached |
| 05:29.58 | [g2] | yep it's just running 1 process |
| 05:30.23 | [g2] | cpu just half loaded on only 1 cc |
| 05:30.28 | dyoung | wow. |
| 05:30.39 | dyoung | Your swap. Seems so... Small. |
| 05:31.01 | [g2] | there's a funny story behind that |
| 05:31.42 | [g2] | two funny stories.... 1) well if you think *thats* small you should see |
| 05:32.32 | [g2] | 2) sometime in the 6-12 months ago timeframe I was playing with buildroot and build an x86 buildroot |
| 05:32.47 | [g2] | I had 4 primary partitions |
| 05:32.57 | [g2] | I needed a 5th |
| 05:33.09 | dyoung | This sounds bad already. |
| 05:33.11 | [g2] | A tiny one 4MB |
| 05:33.29 | [g2] | soooo..... |
| 05:34.10 | [g2] | I deleted my swap and made it a logical partition and split it 256MB swap and 256MB for my busybox buildroot |
| 05:34.57 | [g2] | then I shrank the buildroot part down to 4MB (of which 50% was free) and booted and ran on it |
| 05:35.27 | [g2] | my *massive* kernel which has tons of stuff built into needed 12MB to boot |
| 05:36.36 | [g2] | So I was running my P4 in busybox networking etc.. usb 2.0, in 12MB (with 500MB memory idle) and a 4MB disk |
| 05:38.03 | [g2] | but then [g2] is a sick puppy like that |
| 05:41.25 | [g2] | first -j4 then distcc with oe |
| 05:41.41 | [g2] | yet another thing to learn |
| 05:47.09 | [g2] | well good nite all! |
| 05:47.34 | CodeWhacker | nite, [g2] |
| 05:47.36 | [g2] | CodeWhacker, that delete and bk co appears to be working! thx |
| 05:47.41 | CodeWhacker | np |
| 05:48.05 | [g2] | it's one he** of a community we got going here |
| 05:48.18 | [g2] | that's in the good sense obviously |
| 05:50.00 | *** join/#nslu2-linux [biovore] (~biovore@h-66-167-232-106.mclnva23.covad.net) [NETSPLIT VICTIM] |
| 05:53.36 | dyoung | Install #5. |
| 06:07.18 | tking | dyoung: of what? |
| 06:10.06 | dyoung | I'm sitll fighting my build system. |
| 06:10.24 | tking | ouch |
| 06:10.28 | dyoung | Its winning. |
| 06:10.36 | dyoung | I'm trying to installed it raid1.. |
| 06:10.49 | dyoung | bbl |
| 06:10.55 | tking | k |
| 06:10.58 | tking | nite |
| 06:14.39 | tking-sleep | rwhitby-away : do we want to call romhack a 3rd development stream? (albeit short lived?) |
| 06:46.34 | *** join/#nslu2-linux _biovore_ (~biovore@h-66-167-232-106.mclnva23.covad.net) |
| 07:42.32 | jacques | well as I expected, tftp is not working for me |
| 07:42.59 | jacques | as it dod not on the wrt, tho I built a different tftpd this time |
| 07:44.15 | jacques | las time someone sent me a debian tftpd binary and that worked right away |
| 07:44.29 | jacques | well last time was tftp client, not server |
| 07:45.00 | jacques | I can see the tftp request in ethereal, but no response |
| 07:45.07 | jacques | and the operation times out |
| 07:47.16 | jacques | oh hmm, do I have to set the address of the server? :-\ |
| 07:49.01 | jacques | hmm in addition, does client need to be in hosts file? |
| 07:50.52 | jacques | now I'm getting "file not found" |
| 07:52.08 | jacques | jbot, test |
| 07:52.09 | jbot | Passed |
| 07:59.05 | jacques | ok I got it working - path problems |
| 08:51.22 | rwhitby-read-log | tking-sleep: RamHackU2 is the third development stream now - the fourth would be what we used to call NSLU2PLUS (we need a better name( |
| 08:52.31 | jacques | starting first build of new oe stuff |
| 08:53.13 | rwhitby-read-log | jacques: are you running ip_address -h <host> -l <nslu2> ? |
| 08:53.45 | jacques | rwhitby-read-log, I am now heh |
| 08:54.05 | jacques | that's what finally made it work along with findout out my tftpd was looking for things in / |
| 08:55.44 | jacques | busybox 60.2 ? WTF? |
| 08:55.56 | jacques | why is nslu2 oe using that version? |
| 08:59.58 | jacques | 499.06K/s lol |
| 09:00.26 | jacques | i like my new broadband |
| 09:00.53 | jacques | weird |
| 09:01.17 | jacques | so we're using busybox 60.2, libc headers 2.6.7, with a 2.4.22 kernel :-\ |
| 09:02.01 | jacques | ok now it's getting busybox 1.00-rc3 - I'm confused |
| 09:02.13 | jacques | (I'm doing the -cfetch) |
| 09:05.55 | jacques | wow what's in this file: 2.4.22-xfs-nslu2.patch.bz2 - one rarely sees a compressed patch > 10MB |
| 09:06.22 | rwhitby | that's the patch from 2.4.22 (maybe with xfs) to the Linksys source. |
| 09:06.50 | rwhitby | jacques: the nslu2 kernel in oe at the moment is designed to replicate the stock kernel. |
| 09:07.15 | rwhitby | one step at a time, with stable baselines .... |
| 09:08.42 | jacques | but 10MB patch to 30MB source ? |
| 09:09.08 | jacques | (compressed sizes) |
| 10:51.16 | *** join/#nslu2-linux ice (~grunff@dsl-082-083-014-082.arcor-ip.net) |
| 11:41.20 | Tiersten | I left it flashing the 1.94 RedBoot BE ROM binary from Intel last night and it doesn't work :| You just get +$T050f:0ffff000;0d:0000b288;#e9 printed out and then it stops. Any input on the serial port and it just prints out +$T050f:0ffff000;0d:0000b288;#e9 again. Don't know if it's because I used the wrong one or not as there were 4 different copies of RedBoot. RAM, ROM and little endian versions of both. *shrug* |
| 11:45.12 | rwhitby | Tiersten: The stock redboot is a RAMROM image. Different from both RAM and ROM images |
| 11:45.29 | Tiersten | ahh okay :) |
| 11:45.35 | Tiersten | That might explain why it borks |
| 11:46.00 | rwhitby | But I'm not sure that should matter, unless you're trying to flash redboot from inside redboot (which is why they have the RAMROM option). |
| 11:46.28 | Tiersten | Is the RedBoot source on the CD? |
| 11:46.57 | rwhitby | it's on the Linksys site |
| 11:47.34 | rwhitby | http://www.linksys.com/support/opensourcecode/nslu2/2.3r25/nslu2-2.3R25.zip |
| 11:47.46 | Tiersten | Might try rebuilding it from source and see exactly what they're doing and why the stock 1.94 doesn't work |
| 11:48.48 | rwhitby | There's a redboot-intel-xscale.tar.gz and redboot-intel-xscale.diff in there. |
| 11:49.30 | rwhitby | diff.bz2 |
| 11:49.37 | Tiersten | TBH upgrading redboot isn't going to really give us much |
| 11:50.03 | Tiersten | The changes in 1.92 -> 1.94 didn't really affect us at all |
| 11:50.09 | rwhitby | ~tbh |
| 11:50.33 | Tiersten | To Be Honest ? |
| 11:50.43 | rwhitby | ah |
| 11:51.35 | rwhitby | Yeah, various people are dead keen on replacing it, but as long as it loads a kernel, it's good enough for me ... |
| 11:52.17 | Tiersten | Pretty much. Some people have said that sercomm have disabled some commands in redboot though |
| 11:52.35 | rwhitby | all the fis stuff apart from fis write and fis erase |
| 11:53.04 | Tiersten | It loads the kernel and thats the main thing. It's not a good idea for people to fiddle with upgrading redboot anyway unless they've got a JTAG cable |
| 11:53.20 | rwhitby | Fully agree |
| 11:55.21 | Tiersten | Now if only Linksys had left some easily accessible pads for the expansion bus... |
| 12:16.28 | *** join/#nslu2-linux Neon12345 (~Neon12345@pD9E395B9.dip.t-dialin.net) |
| 12:27.32 | *** join/#nslu2-linux [g2] (~g2@66.57.6.117) |
| 12:28.15 | rwhitby | Hi [g2]! |
| 12:28.15 | [g2] | Morning! |
| 12:28.15 | Tiersten | hey |
| 12:28.15 | [g2] | Morning! |
| 12:28.22 | rwhitby | [g2]: just sent you a memo - did you get it? |
| 12:28.45 | [g2] | yes I see I've got it |
| 12:29.33 | [g2] | that was in between me readind log and getting in |
| 12:29.51 | rwhitby | Anyway, what I said was after looking at oe some more, I think we need to keep an nslu2 kernel same as stock (so it can be used as a base for small mods), and create separate .oe files for improved kernels. |
| 12:30.09 | rwhitby | So you shouldn't be changing that defconfig file. |
| 12:31.20 | rwhitby | I think you need to create a separate defconfig for what you want to do, and then create a new .oe file that uses it. |
| 12:31.38 | rwhitby | then you need to set virtual/kernel to that new .oe file |