00:07:45 | sammy | Good Morning all .........!! |
00:08:23 | BZFlag | morning sammy. TuxScreen blob is booting thanx to Russ. |
00:08:40 | BZFlag | nite... ;-) |
03:01:24 | sammy | welcome back ...!! |
03:01:38 | sammy | are U sleeping or not ? |
03:02:48 | BZFlag | I should bem but I'm not. |
03:02:58 | BZFlag | er I should be, but I'm not. |
03:03:37 | Russ | working on diff, touchscreen, and frontlight |
03:03:42 | Russ | er diffs |
03:04:26 | sammy | or something like insomnolence ? |
03:04:32 | Russ | BZFlag: how you doing on sa1100fb.cL |
03:04:36 | Russ | er, .c |
03:07:56 | sammy | if I porting blob and kernel on LART and then want show something like "Hello World" on screen ,can I use QT GUI from iPAQ on it ? |
03:08:15 | Russ | what screen? |
03:08:43 | sammy | like LCD or termil screen on PC |
03:08:50 | Russ | which one |
03:09:14 | sammy | LCD ... |
03:09:28 | Russ | why wouldn't you be able to? |
03:10:28 | sammy | but |
03:11:21 | BZFlag | Russ: not working on it yet. |
03:11:31 | BZFlag | rebuilding with jffs2 support. |
03:12:13 | BZFlag | Russ: what happened to the blob jffs2 patches? Should we look at those next? |
03:12:53 | Russ | soon, I use them at work, along with the cramfs ones |
03:13:23 | BZFlag | My son burned his hands on sparklers tonight, so I'm not expecting much sleep or work will get done tonight. |
03:13:59 | BZFlag | patches: nice. I should get back to work on the tftp patches too. |
03:16:08 | Russ | ouch |
03:17:26 | BZFlag | yeah. he was doing so well with just holding the tip, but then after like 20 sparklers he had one go out and grabbed the top to bring it back. |
03:17:51 | BZFlag | It hurt just watching him, we grabbed fast and shouted, but it was too late. |
03:18:17 | sammy | and how about your son ? is he alright ? |
03:18:51 | BZFlag | sammy: he will be, but he'll be sore tomorrow and we probably won't get much sleep tonight. |
03:19:17 | BZFlag | I did manage to rock him to sleep a little while ago, so he is sleeping now. |
03:19:54 | BZFlag | Russ: what what the name of that jffs2 image you had? |
03:20:09 | BZFlag | and is there a tarball so we can fix the terminal speed? |
03:22:24 | sammy | God prays your cute son !! |
03:22:49 | Russ | BZFlag: the nico image? |
03:24:24 | BZFlag | yeah. |
03:24:57 | Russ | sample_img-01.jffs2 |
03:25:11 | Russ | getting ready to make some images of my own |
03:26:47 | BZFlag | sammy: thanx. |
03:26:51 | BZFlag | cool. |
03:28:46 | Russ | building uClibc now |
03:35:01 | BZFlag | ponders changing all foo (foo@bar) entries in AUTHOR to foo <foo@bar> as they should be. ;-) |
03:35:47 | Russ | bah |
03:37:38 | BZFlag | heh, did it. there's a useless patch for ya. ;-) |
03:37:51 | BZFlag | bah on me or uClibc? |
03:38:03 | Russ | on the changing things |
03:39:15 | BZFlag | I'm bored while waiting for my fs to upload |
03:39:56 | Russ | you done with sa1100fb yet? getting impatient |
03:40:52 | BZFlag | heh. did you check out the ts blob version to get the splash code? |
03:41:11 | Russ | no |
03:41:36 | Russ | UCB1200 generic module installed |
03:41:36 | Russ | ucb1200 touch screen driver initialized |
03:41:36 | Russ | ucb1200 adc driver initialized |
03:41:36 | Russ | UCB1200 audio driver version 2.2 initialized |
03:41:54 | Russ | Frontlight Driver Initialized... |
03:42:06 | BZFlag | and then? |
03:42:29 | Russ | the jffs2 scan get stuck... |
03:42:34 | Russ | NetWinder Floating Point Emulator V0.95 (c) 1998-1999 Rebel.com |
03:42:35 | Russ | ... |
03:43:01 | Russ | the Frontlight driver should allow the setting of brightness, contrast, and power management |
03:53:25 | BZFlag | blob needs erase... |
03:53:31 | Russ | ? |
03:53:44 | Russ | builds a busybox |
03:53:57 | BZFlag | when you flash the jffs2 image it should erase the rest of the partition. |
03:54:10 | BZFlag | else the jffs2 scan will have trouble. |
03:54:16 | BZFlag | does it do that now? |
03:54:58 | Russ | I think jffs2 might do that on its own |
03:55:13 | Russ | ...unless there was a jffs2 partition there before.... |
03:56:02 | BZFlag | if it finds any non-jffs2 blocks it prints warnings etc. |
03:56:12 | Russ | thats ok |
03:56:17 | BZFlag | if it finds a completely empty block it uses it. |
03:56:38 | BZFlag | else it ignores the block IIRC. |
03:57:15 | BZFlag | there was a bug posted where jffs2 would erase non-jffs2 filesystems, so the erase unknown block code was removed. |
03:57:26 | Russ | it was? |
03:57:29 | Russ | thats not right |
03:57:42 | BZFlag | it should erase other filesystems? |
03:57:44 | Russ | oh, wait, I see what you are saying |
03:57:56 | Russ | I think it sees if it can find something jffs2 first |
03:58:59 | BZFlag | I think it will try to erase any blocks after the valid image till the end of the partition, but if it finds one random suspected good block, that will mess up. |
03:59:19 | BZFlag | hence the need for the bootloader to erase the remaining blocks. |
04:00:28 | BZFlag | I just uploaded a 3M file of 0xff, that did it, but took a while as you can imagine. |
04:02:27 | Russ | the nico image is nearly 3M |
04:04:02 | Russ | building tinylogin.... |
04:11:20 | BZFlag | yeah, I just remounted / rw and it reports 0 free |
04:15:24 | Russ | I noticed |
04:18:24 | BZFlag | though it did find like 20 blocks that it marked dirty!? |
04:18:54 | Russ | then its probably an image thats aleady been used |
04:19:06 | Russ | er, why is it dirty |
04:19:43 | BZFlag | dont know. |
04:19:56 | BZFlag | I wrote all 0xff first remember. |
04:25:20 | BZFlag | empty blocks should get marked dirty till GC erases them later. |
04:41:16 | BZFlag | I mailed the jffs list, and forgot to mention the kernel version... ;-/ |
04:41:20 | BZFlag | night! |
06:48:19 | sammy | {{{{{{{ sammy }}}}}}} |
08:12:00 | erikm | damn |
08:12:26 | sammy | what's up !? |
08:12:34 | erikm | cvs update on the wrong branch |
08:18:30 | sammy | erikm: where U define the Ser1UTDR1 and UTCR0_1StpBit |
08:21:03 | sammy | and as U say that CVS add the command and support on blob |
08:21:53 | sammy | means can get new blob from CVS again the same way ....? |
09:37:22 | sammy | go .... |
17:43:27 | erikm | relocation went ok |
17:43:35 | erikm | now compiling a kernel |
17:44:40 | Russ|werk | and supposively paramaters will pass? |
17:45:58 | erikm | yes |
17:46:00 | erikm | I hope so |
17:48:31 | erikm | let's try it immediately with the hard disk as root |
17:48:36 | erikm | no risk no fun, right? |
17:53:02 | erikm | ok, half of it works |
17:53:12 | erikm | ide_setup: idebus=33 |
17:53:20 | erikm | so the commandline node works |
17:53:34 | erikm | apparently the memory nodes don't |
17:54:07 | erikm | and the root node also works |
17:57:42 | erikm | puts some debug printks into the kernel |
18:02:23 | Russ|werk | you didn't before? |
18:03:11 | erikm | nope |
18:09:11 | erikm | woohoo! |
18:09:14 | erikm | blob> boot console=ttySA0,9600 idebus=33 root=/dev/hda1 single |
18:09:23 | erikm | Kernel command line: console=ttySA0,9600 idebus=33 root=/dev/hda1 single |
18:09:29 | erikm | Memory: 8MB 8MB 8MB 8MB 64MB = 96MB total |
18:09:46 | erikm | Give root password for maintenance |
18:09:46 | erikm | (or type Control-D for normal startup): |
18:20:37 | Russ|werk | very nice |
18:20:49 | Russ|werk | can you email me the patch to lart.c? |
18:22:03 | erikm | Russ|werk: not yet, it needs some other kernel work as well |
18:22:22 | Russ|werk | like what? |
18:22:52 | erikm | the node ID is not correctly set when using ATAG_MEM nodes |
18:23:22 | Russ|werk | I'm confused |
18:23:50 | Russ|werk | if you use ATAG_MEM nodes, do you have to change you <sa1100_arch>.c file around? |
18:24:29 | erikm | yes, first of all you have to tell where your bootloader puts the arguments |
18:25:07 | erikm | Russ|werk: and because blob is the first bootloader on SA1100 that properly uses the ATAG_* stuff, we're running into a bug right now |
18:32:23 | erikm | re BZFlag |
18:32:40 | erikm | BZFlag: so marc also ordered a tuxscreen |
18:35:37 | erikm | BZFlag: (marc == marc joosen) |
18:37:11 | Russ|werk | so all the arches that want to use ATAG_MEM need to modify their arch file with an ifdef atag bootloader? |
18:37:33 | rmk | no, we're solving it... |
18:37:40 | erikm | no, they can also use the older, less flexible format |
18:38:28 | erikm | but the new format is pretty neat. the old one couldn't describe more than one memory zone, the new one has no problems with that |
18:38:49 | rmk | and its far more flexible and extendable |
18:39:01 | erikm | sure |
18:39:19 | erikm | rmk: I suppose you're going to ditch the old one in 2.5? |
18:39:40 | rmk | nope - I made a promise to keep it around for the next 5 years |
18:40:06 | rmk | there's too many boot loaders out there to modify ;( |
18:40:16 | erikm | put such a comment in the source |
18:42:24 | Russ|werk | btw, what exactly is an assabet frontlight |
18:42:25 | rmk | patch? ;) |
18:43:03 | erikm | rmk: will do |
18:43:09 | rmk | thanks |
19:33:01 | BZFlag | erikm: I got marc's order. will ship tomorrow. |
19:33:20 | erikm | BZFlag: he will be enlightened ;) |
19:33:43 | Russ|werk | enlightened? |
19:33:45 | erikm | BZFlag: (and I will have something to hack on when I'm in the US) |
20:59:25 | erikm | bye |