**** BEGIN LOGGING AT Wed Sep 23 02:59:58 2009 Sep 23 06:06:11 rektide: thanks for the mainlining post on openrd-org forum Sep 23 06:19:20 np Sep 23 06:19:24 mr not here. ;) Sep 23 06:28:52 is the cryptographic engine that's supposed to be present on the openrd supported at all? Sep 23 06:29:00 i couldn't find anything about it Sep 23 06:32:19 tinker-f595: np Sep 23 06:32:44 arachnist: i believe there are kernel modules for it Sep 23 06:33:44 tinker-f595: Dhaval's reply was fairly pleasant, and quite speedy. Who knows what it'll lead to, but the reply was better than expected. Sep 23 06:34:24 rektide: you don't happen to remember their names, do you? Sep 23 06:34:52 i'll look it up; one moment Sep 23 06:35:15 CONFIG_CRYPTO_DEV_MV_CESA Sep 23 06:35:42 cryptographic api -> hardware crypto devices -> marvell's cryptographic engine Sep 23 06:35:49 hm Sep 23 06:36:13 i'm using marvell's tree. Sep 23 06:36:18 that explains Sep 23 06:36:36 git.marvell.com. fairly painless. Sep 23 06:38:41 rektide: probably it will lead to the work being delayed by the amount of time it took him to read that and reply "we're still working on it" Sep 23 06:39:17 nagging really isn't very helpful if you aren't nagging about your own contributions being merged Sep 23 06:39:18 ron_: if 7 minute loss is the only effect, so be it. Sep 23 06:41:43 nagging me about your perception that i was nagging doesnt alter my perception that i dont feel like i was nagging, as i dont feel like upstream merging was ever really brought to table. Sep 23 06:42:03 :)) Sep 23 06:42:13 but you're welcome to address me and you will assuredly receive my attention and a reply Sep 23 06:44:00 time for sleep; goodnight all Sep 23 06:45:19 *shrug* Sep 23 06:45:38 it's not me that's waiting on other people to do things for them. delay them all you like Sep 23 06:45:56 I already told you yesterday exactly what you made him repeat Sep 23 06:50:19 looking at ym chat log, i had no discussion with you. you did have a chat log with tinker about topic similar to this one. Sep 23 06:51:14 dhaval just spent a lot of effort getting his own git publically facing. better upfront communication about what is most helpful seems logical, so as to avoid further headache Sep 23 06:51:46 given that no ones asked about mainlining, i really dont understand why you're assuming it was his priority Sep 23 06:55:40 what I told tinker wasn't exactly a private message, and you were around when that occurred Sep 23 06:56:13 and since I told him the code had already been submitted for mainline, why do you assume that wasn't happening? Sep 23 06:56:54 (I'll also give you one guess at who nudged dhaval to get a git repo of his work up publicly ;) Sep 23 06:57:31 ron_: never saw it in the amr-linux email list Sep 23 06:57:37 arm-linux Sep 23 06:59:27 I did. I even attached my own initial patch to it to the thread there :) Sep 23 06:59:51 hmm OpenRD Client stuff? Sep 23 07:00:21 and nico's request for rmk to pull it was there too Sep 23 07:00:33 hmmmm Sep 23 07:00:51 just read Dhavel's response to Rektide Sep 23 07:01:02 I'm pretty sure the client patch followed later. I definitely saw it somewhere. but I haven't been paying too much attention to that beyond -base support Sep 23 07:02:08 Problem is Dhavel said he would have the stuff in his git tree by Sept 17 so people could build from there but Sept 17 came and went and not a peep form Dhavel and no update to his tree. Sep 23 07:03:08 I was just looking at the uboot source that is on the cd ...the base tree is from 2005! Sep 23 07:03:34 sure, give him a break. he's clearly pretty new to both git and working with lk mainline. and they are a daunting and demanding bunch Sep 23 07:04:00 it will go up when he's ready and happy with it. pushing him to go faster won't help that :) Sep 23 07:06:32 (and you you look at the code in u-boot on the CD, you'll see why it hasn't been accepted by denx yet either ... it needs a lot of cleanup) Sep 23 07:06:44 erm ,if you, Sep 23 07:12:08 ron_: if you publish a delivery date and can't make it the proper thing to do is to let people know there is going to be a delay. Sep 23 07:13:31 anyway time to go back to sleep Sep 23 07:14:23 if you expect software delivery dates to be set in stone, this won't be your last disappointment ;) Sep 23 07:14:51 ron_ by the way the communtiy members are willing to test the stuff for Dhavel. We have a diverse range of drives etc to try the code with Sep 23 07:15:06 ron_: you do not get it. Sep 23 07:15:23 no. I write and _give_ it. Sep 23 07:15:33 people get it when it's ready ... Sep 23 07:15:43 ron_: delivery dates always slip, he's just saying that communication matters Sep 23 07:15:53 i.e. either don't promise a date at all, or if you do but you slip, let people know Sep 23 07:15:57 ron_: if you were interviewing with me for a job you would of just of talked your way out of the door Sep 23 07:16:18 lennert: knows... Sep 23 07:16:26 yeah. it's only been 6 days though, and the fact that it didn't appear should be a pretty clear sign there is a delay Sep 23 07:17:11 tinker-f595: that's fine. you just talked yourself out of me ever wanting to work for you ;p Sep 23 07:17:12 ron_: you have a very unprofessional attitude. It is safe to say you do not make your living in this business Sep 23 07:18:25 If you are going to miss a deadline you let people know so they can arrange their schedule appropriately Sep 23 07:18:41 *shrug* Sep 23 07:18:55 * ron_ goes back to coding and doing support for paying customers Sep 23 07:24:09 rektide: http://dpaste.com/97097/ <| is that enough, or do i need anything more from the orion tree to make it work? the module seemed to load fine and shows itself in /proc/crypto Sep 23 07:41:19 hi all Sep 23 07:41:45 hi Sep 23 07:42:51 I run the installer V1, mmc mode on a 8G SD card. It works fine but.....I does not boot automatically. I have to type "boot" using the serial console to make it boot Sep 23 07:43:23 can someone help me ? Sep 23 07:43:31 what does 'printenv bootcmd' shows? Sep 23 07:44:50 bootargs is empty, did not try bootcmd. I do not have my plug here. will try tonight. What is the expected bootcmd ? Sep 23 07:46:21 Another question : ext4 is not supported by the kernel packaged withde the kernel ? thanks the V1 installer. So I want to update the kernel. Does the readme file 2.6.31 work or is there any updated procedure to upgra Sep 23 07:46:32 upgrade the kernel. thanks Sep 23 07:47:15 ext4 is not built as a module; you need to rebuild your own kernel for that. Sep 23 07:47:51 in the forums there are lots of links how to build your custom kernel; along with other pages in the wiki. Sep 23 07:48:19 ok, but I read in the forum that from the 2.6.30-3 kernel has ext4 support by default Sep 23 07:50:51 I tried mounting my ext4 drive with ext3 but it fails Sep 23 07:50:53 too bad Sep 23 07:51:14 I guess I'll have to cross compile my kernel Sep 23 07:51:15 :) Sep 23 07:52:36 do you recommend the 2.6.31 release or any previous kernel ? Sep 23 08:04:29 yaayaa: i'm using 2.6.31 on my openrd and it seems to work fine Sep 23 08:04:43 ok thanks Sep 23 08:04:52 i will give it a try Sep 23 08:05:20 but i needed 4 patches Sep 23 08:06:03 (well, actually, 3. the fourth one is what i did copying the hardware crypto module from orion.git) Sep 23 08:13:15 :) Sep 23 08:43:15 can someone cut/paste the bootcmd and bootargs_root content for a plug after the v1.installer procedure with MMC (so 2 partitions on the sd card) ? just to check tonight....I want to boot from sd card only Sep 23 08:54:17 arachnist: ar eyou using 2.6.31 on openrd base or openrd client? Sep 23 08:54:25 eFfeM: client Sep 23 08:54:58 arachnist: with vga working? thought patches for that part were not integrated Sep 23 08:55:27 eFfeM: i only had vga connected once - back with the stock fedora Sep 23 08:55:33 ah ok Sep 23 08:55:44 where did you get your kernel from ? Sep 23 08:55:49 (or kernel src) Sep 23 08:55:59 kernel from kernel.org + 3 patches Sep 23 08:56:41 http://dpaste.com/97115/ Sep 23 08:56:46 http://dpaste.com/97116/ Sep 23 08:56:50 http://dpaste.com/97117/ Sep 23 08:57:23 you can find those patches also here: http://groups.google.com/group/openrd/browse_thread/thread/640eab6a8a0091ba Sep 23 08:58:13 arachnist: i know the thread, what kept me off was that it starts saying sound and vga is not yet supported (in the first post) Sep 23 08:58:55 well, i'm not really interested in sound and vga Sep 23 09:00:29 i need it for my project so for me if it is not there it is a showstopper Sep 23 09:00:53 pity dhaval or jibal are not on this channel Sep 23 09:02:29 there is a whole bunch of peripheral devices that haven't been ported to mainline lk yet Sep 23 09:03:02 if you need them, you'll either need to use the kernel source that came with it, get your hands dirty porting that forward, or be patient ... Sep 23 09:07:00 or you can try the orion.git at git.marvell.com Sep 23 09:07:15 (if it's there) Sep 23 09:08:46 that should have anything that is on its way to mainline already, but not everything the LSP supports yet Sep 23 09:09:34 yeah, i know, it is just that it is not really clear what the status of the various items is and what eta. Sep 23 09:09:43 currently i'm still using the 2.6.22.18 kernel Sep 23 09:19:57 I imagine the code should be available a bit before all the glaring errors in the datasheets are corrected :) Sep 23 09:24:27 rabeeh: on a slightly more constructive note re ^, is there someone I should be punting things like that at? Sep 23 10:54:24 arachnist : do you have the ext4 built in support with your kernel as indicated in the forums or is rabeeh right (need recompilation)? Sep 23 11:09:59 yaayaa: i have ext4 support built-in, but only for /home, /usr and /data0 (a 830GB partition) Sep 23 11:10:10 yaayaa: my / is on ubifs Sep 23 11:10:22 stalker src # df -h / Sep 23 11:10:22 Filesystem Size Used Avail Use% Mounted on Sep 23 11:10:22 ubi0:rootfs 447M 45M 402M 11% / Sep 23 11:11:09 ok, great. I use an SD card for rootfs, ext4 will be only for my external storage hard drive. Sep 23 11:11:28 thank you arachnist Sep 23 11:11:31 n/p Sep 23 17:28:00 damn viric_'s internet sucks Sep 23 18:30:13 Is it the normal behaviour that the plug does not boot automatically ? my bootcmd and bootargs_root are correct, but I have to type "boot" to make it start.... Sep 23 18:30:27 then everything is fine Sep 23 18:30:54 is it expected that it boots automatically when powered ? Sep 23 18:52:07 no one had this problem ? Sep 23 19:06:30 yaayaa: yes Sep 23 19:06:45 yaayaa: don't you have a 3 seconds timeout, which end on 'boot' if you don't type anything? Sep 23 19:07:44 no, 5 second time out, and whn I plug the serial and check with putty, I got the >>Marvell Sep 23 19:07:54 5 second? fine Sep 23 19:07:57 what after 5 seconds? Sep 23 19:07:59 doesn't it boot? Sep 23 19:09:11 well, I do not know what happens. 1) I plug the device 2) I open a serial console with putty 3) I see >>Marvell 4) Nothing more happens, unless of course I type "boot" Sep 23 19:09:19 ah Sep 23 19:09:25 maybe your putty sends a string Sep 23 19:09:32 opening the console Sep 23 19:09:44 What if you leave the plug boot Sep 23 19:09:51 and only after 20 seconds you run puttY? Sep 23 19:10:25 well, if I do not use putty, I can see that nothing happens since I cannot ssh to the plug (even after a looonnnng time Sep 23 19:13:07 yaayaa: why don't you boot first, and then do a reboot? Sep 23 19:13:37 it works fine... until I unplug the sheeva Sep 23 19:13:51 then I plug it back and.. nothing Sep 23 19:14:31 use another software, like teraterm Sep 23 19:14:40 or hyperterminal Sep 23 19:14:45 boot/reboot (when typed in ) works fine. Sep 23 19:14:49 if we're talking about windows, that is Sep 23 19:14:52 I'm under linux Sep 23 19:15:26 ok, then use minicom, picocom, etc Sep 23 19:15:34 anyway, it does not boot when plugged (can check with ssh), no matter the serial console I use Sep 23 19:16:14 thats weird Sep 23 19:16:16 always have to 1) connect a serial console 2) type boot 3) use the plug normally Sep 23 19:16:19 yes indeed Sep 23 19:16:44 especially since i jsut followed the instructions of the V1.0 installer/mmc Sep 23 19:16:56 yaayaa: how do you know its IP? Sep 23 19:17:05 ip address I mean Sep 23 19:17:37 well, I forced my router to use a mac address mapping Sep 23 19:20:37 it may not use dhcp Sep 23 19:20:57 yes it does Sep 23 19:21:03 well, can't you attach the serial terminal after 20 seconds? Sep 23 19:21:26 since powero n Sep 23 19:21:28 power on Sep 23 19:21:48 I tried 30 sec, then 5 minutes, then 30 minutes. same result Sep 23 19:22:29 ok Sep 23 19:22:30 I would be interested in watching the serial log realtime when (not) booting to see what (does not) happen Sep 23 19:22:48 what if you use the reset button? Sep 23 19:22:54 that doesn't reset the serial usb interface Sep 23 19:22:58 but resets the plug Sep 23 19:23:12 that's what I usually do, otherwise I've to close minicom before repowering Sep 23 19:23:18 (because the device name would change) Sep 23 19:23:39 good idea ! then you think I can check the logs realtime while booting ? Sep 23 19:24:07 help Sep 23 19:24:10 I don't know what do you mean by 'logs'... but that should Sep 23 19:24:12 help Sep 23 19:24:30 well, any helpfull verbose text Sep 23 19:24:35 yes Sep 23 19:24:43 displayed at the >>Marvell level Sep 23 19:25:22 just to understand, what is called "before" the boot command ? Sep 23 19:25:43 what the the env variable that configues the auto boot if any ? Sep 23 19:25:47 The plug loads what is stored at the first sector of the Flash Sep 23 19:25:51 usually, there is the u-boot code Sep 23 19:26:10 ok, I configure a sd card only boot Sep 23 19:26:11 the u-boot code initializes some hardware in the plug, shows a bit of info in the serial console, does the timeout, ... Sep 23 19:26:44 maybe the installer screwed something there Sep 23 19:26:50 I think it always autoboots Sep 23 19:27:04 'boot' means 'run bootcmd' Sep 23 19:27:10 argh ... then I really do not understand Sep 23 19:27:10 bootcmd is a variable. Sep 23 19:27:26 If you have that variable with the boot sequence, it boots automatically Sep 23 19:27:27 yes i know, that 's why it is weird Sep 23 19:27:37 It always autobooted for me Sep 23 19:27:50 I have "bootdelay=3" Sep 23 19:27:57 maybe the u-boot code is missing Sep 23 19:28:16 it would be quite strange Sep 23 19:28:34 what happens after your 5 second timeout? Sep 23 19:28:44 if such was the case, would the plug boot when typing "boot" ? Sep 23 19:28:57 I got the marvell>> Sep 23 19:29:31 I'll try the reset trick. just a sec... Sep 23 19:35:23 well, when I reset the plug, .... it just boots fine Sep 23 19:35:48 Is there a thing to do before unplugging the sheeva ? Sep 23 19:36:04 except "halt" of course Sep 23 19:36:36 really weird Sep 23 19:51:00 no Sep 23 19:51:08 yaayaa: nothing special Sep 23 19:51:28 nothing that could "damage the plug" Sep 23 19:51:34 then, I'm stuck Sep 23 19:52:20 I now now that resetting seems to work (though I did not try resetting when on a non-booting state....) Sep 23 19:54:59 good luck Sep 23 20:02:43 thanks :) Sep 23 20:02:57 and thank you for your help Sep 23 20:03:44 np :) **** ENDING LOGGING AT Thu Sep 24 02:59:58 2009