| 00:01.46 | drWyrM | do the "failsafe" dropbear thing. |
| 00:03.43 | drWyrM | sn9: are you sure that your _host_ hotplug scripts are not trying to reconfigure the USB port? |
| 00:04.20 | sn9 | they don't do so when i ssh into the moto firmware |
| 00:05.55 | drWyrM | well, this only leaves the "failsafe" dropbear. :) |
| 00:06.09 | sn9 | trying that now |
| 00:07.59 | sn9 | ok, it continued to reset after dropbear started, so i put in a "sleep 5000" and now dropbear won't start |
| 00:08.59 | drWyrM | this leaves your host as the problem. |
| 00:09.06 | sn9 | how? |
| 00:09.30 | sn9 | why isn't dropbear starting? |
| 00:09.39 | sn9 | it started the first time |
| 00:09.50 | drWyrM | sleep 5000 before calling dropbear? |
| 00:09.55 | sn9 | after |
| 00:10.03 | drWyrM | makes no sense. :/ |
| 00:10.56 | sn9 | i'll remove the host key from nfs and try again |
| 00:11.31 | sn9 | it was a zero-byte file anyway |
| 00:12.56 | drWyrM | your phone userspace is not reconfiguring usbnet, not touching the usb port, this only leaves your host as being problematic. Probably something on your host is reconfiguring the usb port and this stops the nfs connection. |
| 00:14.05 | sn9 | it doesn't do that with the stock firmware |
| 00:16.31 | DNAfanboy | drWyrM: sorry, hi! :D |
| 00:17.08 | drWyrM | sn9: it is a different vendor/product ID. |
| 00:17.15 | sn9 | oh? |
| 00:17.34 | drWyrM | DNAfanboy: Hi :) |
| 00:17.36 | sn9 | hmm, 049f:505a |
| 00:18.24 | sn9 | where would i change that? |
| 00:19.04 | DNAfanboy | drWyrM: got one of the first Brazilian-made ROKR Z6 lol |
| 00:19.21 | drWyrM | sn9: On the phone? Or your host? |
| 00:19.47 | sn9 | the vendor/product id of the phone with the openezx kernel |
| 00:20.03 | drWyrM | sn9: changing the IDs on the phone will not work (different driver on the host part) |
| 00:20.40 | sn9 | ah |
| 00:20.55 | drWyrM | sn9: You should try to disable any hotplug stuff on your host, definitely it is the host that is screwing it. |
| 00:21.30 | drWyrM | DNAfanboy: congrats! :) I didnt even knew that they were building E6 here. :) |
| 00:21.50 | DNAfanboy | drWyrM: s/E6/Z6/ |
| 00:22.02 | drWyrM | oh, sorry. |
| 00:22.12 | DNAfanboy | Motorola Brazil hasn't announced it officialy, but FNAC is already selling them |
| 00:22.19 | DNAfanboy | lol for Motorola BR |
| 00:22.57 | *** join/#openezx Sahil_6667 (i=Samba@gateway/tor/x-281e9adfaa85da0e) |
| 00:22.58 | sn9 | drWyrM: what is the correct kernel module on the host? |
| 00:23.34 | DNAfanboy | anyway, now I can test with the A1200 without fear of fscking stuff on it |
| 00:23.53 | sn9 | the stock firmware uses "zaurus" |
| 00:24.33 | drWyrM | 'cdc_ether'/'usbnet'. No need for 'zaurus'. |
| 00:26.01 | sn9 | ? |
| 00:26.23 | drWyrM | zaurus will be loaded in any way on my debian host. |
| 00:26.46 | drWyrM | but this doesnt mean that the driver will behave the same way ;) |
| 00:27.10 | sn9 | but it works for you? |
| 00:27.32 | drWyrM | well, i dont use NFS, but USBNET works fine on a1200 here. |
| 00:28.19 | drWyrM | DNAfanboy: i am working on a bootloader, common to all ezx phones, based on alexzhang version of blob. |
| 00:28.36 | drWyrM | but i am still afraid of flashing the A1200 :) |
| 00:28.55 | DNAfanboy | drWyrM: maybe because the A1200 is cute? :D |
| 00:30.25 | drWyrM | DNAfanboy: nah, it is because there is still much stuff to do on a1200 OpenEZX kernel, and if i brick it, developing will be very hard again :) |
| 00:30.39 | DNAfanboy | drWyrM: lol yes |
| 00:40.42 | sn9 | drWyrM: S55 is what breaks it |
| 00:42.52 | drWyrM | sn9: weird. |
| 00:43.04 | drWyrM | i dont see where it could break usbnet by reading the script. |
| 00:43.35 | drWyrM | sn9: Now, undo everything, and leave just bootmisc.sh out for now :) |
| 00:44.09 | sn9 | trying now, but forgot to remove the dropbear link i put in |
| 00:44.23 | drWyrM | hum... |
| 00:44.27 | drWyrM | sysctl. |
| 00:44.50 | sn9 | there is no /etc/sysctl.conf, so that shouldn't matter |
| 00:45.22 | drWyrM | yes, my userspace also doesnt have it. |
| 00:45.46 | sn9 | now i get the reset after configuring avahi-autoipd |
| 00:49.29 | sn9 | interesting that dropbear failed to generate keys, but when i copied over the keys that dropbear for the stock firmware generated, it went on |
| 00:49.53 | drWyrM | permission issues with your NFS server? |
| 00:50.27 | sn9 | it creates the first file, but it's zero bytes |
| 00:50.39 | sn9 | i think the key generator hangs |
| 00:51.31 | drWyrM | lack of /dev/random bits? |
| 00:51.46 | drWyrM | but, it resets USB with just dropbear? |
| 00:51.46 | sn9 | well, udev fails to start |
| 00:51.56 | sn9 | no |
| 00:52.41 | sn9 | maybe this is an issue: # CONFIG_TMPFS is not set |
| 00:55.03 | drWyrM | yes. |
| 00:55.09 | drWyrM | this should be on. |
| 00:55.48 | sn9 | what about TMPFS_POSIX_ACL? |
| 00:56.11 | sn9 | you had tmpfs off in the a1200 defconfig, btw |
| 00:56.31 | drWyrM | some time ago, we used to leave SHMEM disabled, and RAMFS assumed TMPFS place, but i enabled SHMEM and forgot to enable TMPFS. |
| 00:57.03 | sn9 | should i disable RAMFS ? |
| 00:58.07 | drWyrM | doesnt makes much difference, RAMFS assumes TMPFS place only if SHMEM is set, otherwise it just remain quiet. |
| 00:58.28 | sn9 | what about TMPFS_POSIX_ACL? |
| 01:00.14 | drWyrM | only if userspace requires it. Which i believe it doesnt. |
| 01:03.58 | CIA-38 | OpenEZX: wyrm * r2072 /trunk/src/kernel-2.6/patches/ (5 files): add TMPFS to the defconfigs. |
| 01:11.06 | drWyrM | ao2 already added TMPFS on .23 defconfigs. |
| 01:11.28 | sn9 | i put S55 back, and it does something now |
| 01:11.42 | drWyrM | weird! |
| 01:11.49 | drWyrM | :P |
| 01:12.03 | sn9 | the screen is now completely blank, with a non-blinking cursor in the upper left |
| 01:12.36 | drWyrM | so, in the end it was because of TMPFS? |
| 01:12.41 | sn9 | yes |
| 01:14.21 | drWyrM | DNAfanboy: it is possible that the same TMPFS issue was breaking X on your a1200 ;) |
| 01:14.30 | DNAfanboy | drWyrM: ahn |
| 01:14.40 | sn9 | well, i still don't get X |
| 01:14.58 | drWyrM | sn9: what error? |
| 01:15.07 | sn9 | can't tell |
| 01:15.36 | sn9 | oh, wait, i got dropbear now |
| 01:29.28 | drWyrM | ? |
| 01:48.43 | sn9 | for some odd reason, /var/volatile is not getting populated. only thing there is "cache" |
| 01:52.28 | sn9 | btw: http://homepage.mac.com/danielg4/dmesg-a910-2.6.21 |
| 01:53.10 | sn9 | drWyrM: any ideas? |
| 01:59.31 | drWyrM | the dmesg looks just fine. |
| 01:59.43 | sn9 | /var/volatile |
| 02:03.43 | sn9 | drWyrM: ^^ |
| 02:09.51 | drWyrM | <PROTECTED> |
| 02:10.37 | sn9 | tmpfs is mounted in several points, including /dev |
| 02:11.03 | drWyrM | and what is /var/volatile use then? |
| 02:11.31 | drWyrM | is it getting mounted? |
| 02:11.36 | sn9 | yes |
| 02:11.58 | sn9 | look at the populate-volatile script and see |
| 02:12.22 | drWyrM | hum.. And kernel should populate it? (i didnt understand where the dmesg fits on this) ;) |
| 02:13.05 | sn9 | dmesg doesn't. it was just an aside |
| 02:14.00 | drWyrM | let me see what is on /var/volatile after a successful boot on a780. |
| 02:20.00 | drWyrM | hum, volatile is just the tmpfs mountpoint for storing tmp files. |
| 02:20.31 | drWyrM | so, if it gets mounted, it is working fine. |
| 02:20.36 | sn9 | no |
| 02:21.04 | sn9 | a lot of things are symlinks to things that are supposed to be created in there |
| 02:21.15 | sn9 | look at the populate-volatile script and see |
| 02:21.29 | drWyrM | well, something is wrong with your init.d scripts. |
| 02:21.43 | drWyrM | maybe populate-volatile is not being called at all. |
| 02:22.00 | sn9 | well, dropbear works, and that's later |
| 02:22.23 | drWyrM | :/ |
| 02:22.55 | sn9 | i think the issue is in some way kernel-related |
| 02:23.27 | drWyrM | can you create files on /var/volatile? |
| 02:24.13 | drWyrM | if it shows on 'mount' and you can touch /var/volatile/foo, it is not kernel problem :) |
| 02:35.12 | sn9 | at least not directly |
| 02:42.04 | *** join/#openezx WM (i=wyrm@201.78.24.183) |
| 02:42.59 | WM | last message i got: "i think the issue is in some way kernel-related" |
| 02:43.10 | WM | after that, i said: |
| 02:43.32 | sn9 | [Tue 30 Oct 2007 07:22:31 PM PDT] = <drWyrM> can you create files on /var/volatile? |
| 02:43.32 | sn9 | [Tue 30 Oct 2007 07:23:17 PM PDT] = <drWyrM> if it shows on 'mount' and you can touch /var/volatile/foo, it is not kernel problem :) |
| 02:43.32 | sn9 | [Tue 30 Oct 2007 07:34:15 PM PDT] <sn9> at least not directly |
| 02:43.33 | sn9 | [Tue 30 Oct 2007 07:41:07 PM PDT] * WM (i=wyrm@201.78.24.183) has joined #openezx |
| 02:44.05 | WM | <drWyrM> In fact, even the way that OE userspace bails out with a kernel without TMPFS is userspace problem, assuming that tmpfs is supported by the kernel without checking it first. On embedded systems i believe that tmpfs is not necessary and should be replace by the much simpler ramfs code. |
| 02:44.05 | WM | <drWyrM> I prefer an userspace that can use ramfs instead of tmpfs, but as i am not the one who makes the userspace images.. 20kb more to the kernel image ;) |
| 02:44.05 | WM | <drWyrM> by default the bootloader on EZX loads just 896KB of the kernel partition on the builtin flash. With blob2 i raised it to 1.5MB. |
| 02:44.06 | WM | * drWyrM hopes that it will be enough :D |
| 02:45.06 | WM | so, you cannot create files on /dev/volatile? |
| 02:45.24 | WM | do you have free RAM? |
| 02:46.52 | sn9 | i can create files there. the "cache" file is proof enough |
| 02:49.18 | WM | well, try calling the populate-volatile script manually, i really think that this is not kernel related. |
| 02:56.20 | WM | and only recently i figured that the gpios for one SSP port were interfering with the data sent by another ssp port. |
| 02:56.33 | DNAfanboy | night all! |
| 02:56.44 | WM | night Cesar :) |
| 04:14.47 | sn9 | WM: found the problem |
| 04:15.05 | sn9 | last line of /etc/default/rcS |
| 04:15.14 | sn9 | now i got syslog msgs |
| 04:15.21 | sn9 | upload pending |
| 04:15.59 | WM | sn9: great! Congrats :) |
| 04:18.45 | sn9 | http://homepage.mac.com/danielg4/messages-a910-2.6.21 |
| 04:20.22 | sn9 | WM: whaddaya think? |
| 04:21.16 | sn9 | a1200 is the hostname; forgot to change that |
| 04:26.09 | WM | note that usb 1-3 is the usb HOST port on PXA and altough the kernel messages are together it is not related with usb0. |
| 04:27.03 | WM | the remaining is userspace.. |
| 04:27.44 | WM | loading hcid is not proper befor an hciattach (and adding bcm2045 support to BlueZ), so it is good to disable it for now. |
| 04:28.34 | WM | It looks good, what works and what doesnt?? :) |
| 04:28.55 | sn9 | what about the fb illegal bits msgs? |
| 04:29.54 | WM | These messages are on OpenEZX before than me. ;) |
| 04:30.32 | WM | there is a relatively recent dmesg of OpenEZX on my a780 on http://people.openezx.org/wyrm |
| 04:31.50 | WM | the same errors exist on A780. And a780 FB works fine. It is the kind of error that you know that exist, you know that you have to fix it, but you keep postponing because there are more important stuff. |
| 04:34.01 | sn9 | Xfbdev fails with "no screens found" |
| 04:34.14 | WM | because of what? |
| 04:34.27 | WM | lack of a input device, probably. |
| 04:34.51 | sn9 | an "input device" is a "screen" now? |
| 04:35.34 | sn9 | here's the skinny: |
| 04:35.45 | sn9 | _XSERVTransSocketOpenCOTSServer: Unable to open socket for inet6 |
| 04:35.45 | sn9 | _XSERVTransOpen: transport open failed for inet6/a1200:0 |
| 04:35.45 | sn9 | _XSERVTransMakeAllCOTSServerListeners: failed to open listener for inet6 |
| 04:35.45 | sn9 | error: Invalid argument |
| 04:36.34 | WM | is that all? |
| 04:37.11 | WM | IIRC the "no screens found" show on every X failure. |
| 04:40.14 | sn9 | yes, those are the only X errors |
| 04:40.50 | WM | its 2:40 AM here, I have to work at 9:00. :/ |
| 04:44.13 | WM | sn9: I will do some more testing with X on a1200, and maybe continue the 18bpp support for pxafb asap. |
| 04:44.24 | sn9 | thx |
| 04:44.31 | WM | Thank YOU! :) |
| 04:44.45 | WM | bye, good night. |
| 05:37.01 | *** join/#openezx |dion| (n=dion@inhex.net) |
| 06:20.04 | *** join/#openezx pleemans (n=peter@leemanpe.ugent.be) |
| 06:40.37 | *** join/#openezx upl (n=chatzill@219.143.147.17) |
| 06:43.13 | *** part/#openezx upl (n=chatzill@219.143.147.17) |
| 06:43.21 | *** join/#openezx upl (n=chatzill@219.143.147.17) |
| 06:46.57 | Hopscotch | good morning |
| 07:40.49 | *** part/#openezx upl (n=chatzill@219.143.147.17) |
| 07:51.08 | *** join/#openezx WM|sleeping (i=wyrm@201.19.154.106) |
| 07:55.07 | *** join/#openezx jannu (n=jannu@hoas-fe3cdd00-15.dhcp.inet.fi) |
| 08:06.08 | *** join/#openezx koen (n=koen@dominion.kabel.utwente.nl) |
| 08:20.45 | sn9 | koen: the problem turned out to be that wyrm disabled tmpfs in all the .21 defconfigs |
| 08:21.11 | koen | right |
| 08:21.21 | koen | that's why I don't use the defconfigs from svn |
| 08:21.40 | koen | the ones in svn are great if you only want to see if the kernel boots |
| 08:21.48 | koen | but I want to use userspace |
| 08:22.10 | sn9 | OTOH, it could also be an argument for not making the userspace need tmpfs |
| 08:22.32 | koen | heh |
| 08:22.45 | koen | it it couldn't be that |
| 08:22.59 | koen | if you start that way lets disable lcd, networking, bluetooth, etc |
| 08:24.50 | sn9 | all that needs shm? |
| 08:29.46 | sn9 | anyway, now it boots but X fails to start |
| 08:31.07 | sn9 | _XSERVTransSocketOpenCOTSServer: Unable to open socket for inet6 |
| 08:31.07 | sn9 | _XSERVTransOpen: transport open failed for inet6/a1200:0 |
| 08:31.07 | sn9 | _XSERVTransMakeAllCOTSServerListeners: failed to open listener for inet6 |
| 08:31.07 | sn9 | error: Invalid argument |
| 08:31.07 | sn9 | Fatal server error: |
| 08:31.08 | sn9 | no screens found |
| 08:31.10 | sn9 | XIO: fatal IO error 104 (Connection reset by peer) on X server ":0.0" |
| 08:31.12 | sn9 | <PROTECTED> |
| 09:13.17 | *** join/#openezx sn9 (n=danielg4@gimpelevich.san-francisco.ca.us) |
| 09:17.51 | *** join/#openezx goxboxlive (n=jrs@195.159.97.196) |
| 09:38.28 | *** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1) |
| 09:39.01 | ao2 | hi |
| 09:40.24 | *** join/#openezx florian (n=fuchs@217.146.132.69) |
| 11:14.59 | *** join/#openezx ade|desk (n=adavey@194.200.143.249) |
| 12:35.09 | *** join/#openezx pleemans (n=peter@leemanpe.ugent.be) |
| 12:48.15 | *** join/#openezx goxboxlive (n=jrs@195.159.97.196) |
| 13:10.57 | *** join/#openezx pleemans (n=peter@leemanpe.ugent.be) |
| 13:12.12 | WM | koen: OE kernel is currently (at least yesterday it was) broke for all phones. |
| 13:12.42 | koen | the current kernel works for a780 |
| 13:13.48 | WM | Yesterday it was turning off the phone right after booting. |
| 13:13.54 | WM | what was wrong? |
| 13:14.14 | koen | dyntick |
| 13:52.55 | ao2 | koen, WM, I'm syncing defconfig-a780 in 2.6.23 branch with the one from OE: this is the diff: http://shell.studenti.unina.it/~ospite/tmp/defconfig-a780_sync-with-OE.diff there are some useless settings for us, but they will not hurt and being in sync as much as possible is good. I only leave out the logo stuff |
| 13:53.39 | koen | the wifi bits are their for userspace that needs it enabled to build |
| 13:53.52 | koen | and for the SDIO wifi cards we are going to use ;) |
| 13:54.14 | ao2 | ah koen what about using different localversions between kernels built with OE and our own? You can use -ezx and we will keep using -ezxdev |
| 13:55.00 | koen | WM: the post on the mailinglist showed that he didn't have oss emulation loaded |
| 13:55.26 | koen | even if the e680 had working sound it wouldn't have worked without loading snd-pcm-oss.ko |
| 13:55.33 | ao2 | ok, so if you really do not care please follow my suggestion :) |
| 13:56.28 | koen | I wonder why CONFIG_NO_HZ breaks on ezx |
| 13:56.34 | koen | it works on all my other systems |
| 14:33.35 | *** join/#openezx agd5f (n=deuchera@65.242.105.130) |
| 15:36.52 | *** join/#openezx goxboxlive (n=goxboxli@139.80-202-160.nextgentel.com) |
| 15:56.17 | *** join/#openezx LoneStar (n=xinzhen@218.82.13.84) |
| 16:22.31 | sn9 | koen: sdio wifi? marvell perchance? |
| 16:24.21 | *** join/#openezx drWyrM (i=c88e3a13@gateway/web/cgi-irc/ircatwork.com/x-7f7228a37d48ef4c) |
| 16:24.34 | koen | most likely atheros |
| 16:24.46 | koen | sdio support is going to get merged in 2.6.24 |
| 16:25.52 | sn9 | oh |
| 16:27.01 | sn9 | marvell's sdio driver has gpl source on their website, but the libertas-dev ppl haven't merged it in afaict |
| 16:27.21 | sn9 | and that's the hardware in the A910 |
| 16:29.03 | sn9 | but if sdio in general has to wait until .24, maybe they'll put it in at that point |
| 16:39.52 | drWyrM | koen: why dont you use ramfs instead of shmfs for OE? |
| 17:01.19 | *** join/#openezx z72ka (n=hermanj@r5af183.net.upc.cz) |
| 17:08.55 | koen | drWyrM: certain apps depend on shmfs, so it's not my call to make] |
| 17:14.20 | *** join/#openezx von_fritz (n=fritz@host59-184-dynamic.54-82-r.retail.telecomitalia.it) |
| 17:16.39 | drWyrM | koen: are you sure about that? Have you tried them with ramfs? I believe that they will work. |
| 17:17.40 | *** join/#openezx von__fritz (n=fritz@host59-184-dynamic.54-82-r.retail.telecomitalia.it) |
| 17:26.30 | *** join/#openezx _dion_ (n=dion@inhex.net) |
| 17:31.30 | *** join/#openezx pH5 (n=ph5@e178226052.adsl.alicedsl.de) |
| 17:56.34 | *** part/#openezx ade|desk (n=adavey@194.200.143.249) |
| 18:00.23 | koen | drWyrM: we tried on a gumstix a while back and it didn't work |
| 18:00.40 | koen | but that might also be the fault of those gumstix kernel patches :) |
| 18:07.50 | sn9 | koen: ok, now that that's settled, does my paste of nine and a half hours ago yield any clues? |
| 18:08.15 | koen | the screens thing? |
| 18:08.34 | koen | could you paste the output of fbset? |
| 18:09.44 | sn9 | i hadn't thought of that; will do |
| 18:13.09 | sn9 | this is the 20071001 rootfs, btw |
| 18:14.42 | sn9 | mode "240x320-57" |
| 18:14.42 | sn9 | <PROTECTED> |
| 18:14.42 | sn9 | <PROTECTED> |
| 18:14.42 | sn9 | <PROTECTED> |
| 18:14.42 | sn9 | <PROTECTED> |
| 18:14.43 | sn9 | <PROTECTED> |
| 18:14.45 | sn9 | endmode |
| 18:17.53 | sn9 | koen: ^ |
| 18:18.48 | koen | that's qvga and should work...... |
| 18:19.45 | sn9 | gives the earlier output on /etc/X11/Xserver |
| 18:21.16 | sn9 | koen: ^ |
| 19:08.31 | *** join/#openezx IAW7548 (i=5704bf12@gateway/web/cgi-irc/ircatwork.com/x-e43f10ad28ee227f) |
| 19:45.25 | *** join/#openezx _dion_ (n=dion@inhex.net) |
| 20:00.40 | *** join/#openezx drath_ (i=vmaster@p5B07EC8E.dip.t-dialin.net) |
| 20:05.24 | *** join/#openezx TimRiker (n=timr@psnet.cc) |
| 21:11.07 | *** join/#openezx alphaone (n=daniel@c120.apm.etc.tu-bs.de) |
| 21:41.48 | WM | koen: i will check CONFIG_NO_HZ. probably it is because ezx-bp.c which still depends on jiffies. |
| 21:54.55 | *** part/#openezx agd5f (n=deuchera@65.242.105.130) |
| 22:22.03 | goxboxlive | WM: I am trying to backu pmy partitions on my A780. I am using dd if... of.... isent that the way to do it? |
| 22:22.33 | WM | dd is fine :) |
| 22:23.02 | goxboxlive | but i get illigeal instruction |
| 22:23.11 | goxboxlive | root@a780:/backup$ dd if=/dev/mtd1 of=/backup/mtd1.dd |
| 22:23.12 | goxboxlive | Illegal instruction |
| 22:41.39 | *** join/#openezx agd5f (n=alex@c-69-143-255-50.hsd1.va.comcast.net) |
| 22:44.37 | *** join/#openezx LuitvD (n=luitvd@beigetower/luitvd) |
| 22:44.40 | LuitvD | hi there |
| 22:45.35 | WM | hum.. It shouldnt. |
| 22:45.46 | WM | LuitvD: Hi :) |
| 22:45.55 | LuitvD | hi! |
| 22:46.38 | WM | goxboxlive: maybe .23 broke our mtd driver. |
| 22:46.42 | WM | let me test on .21 |
| 22:48.16 | goxboxlive | WM: it creates a file, and the mtd2 takes a long time. It has ran now for over 20 min. But i can see that the file is getting bigger. And the illegal instruction happends when it finish the backup it seems |
| 22:49.58 | WM | goxboxlive: it is probably working fine. check the size of the files, and see if they match the size on /proc. |
| 22:50.16 | goxboxlive | yes |
| 22:50.45 | WM | check if mtd-utils is installed. You will need it to write the partitions. |
| 22:52.03 | goxboxlive | WM: I found mtd-tools (debina) is that good enough? |
| 22:52.44 | WM | i guess it is :) |
| 22:56.24 | WM | goxboxlive: and you will need cramfs-tools on your host, to rebuild the rootfs |
| 22:56.49 | goxboxlive | ok |
| 22:59.02 | goxboxlive | Does this look good? http://pastebin.no/2742 |
| 23:01.07 | WM | Yes, the sizes match. |
| 23:02.16 | goxboxlive | great |
| 23:02.38 | WM | now rebuild mtd2 on your host, and assure it is not larger than what myflash.tgz readme says. |
| 23:02.54 | goxboxlive | ok |
| 23:02.58 | WM | and rebuild the kernel with the _ALTERNATE mtd mapping :) |
| 23:03.42 | WM | and, note that the mtd2 will no longer be the rootfs, cat /proc/mtd and see the new layout. |
| 23:06.00 | goxboxlive | my mtd3 is smaller that the one in the README |
| 23:08.02 | WM | hum.. |
| 23:08.55 | WM | mtd3 on the original, or the new layout? :) |
| 23:10.45 | LuitvD | meh, I'm gone again I guess |
| 23:11.00 | sn9 | WM: btw, mtd does not work on a910. if it worked on a1200 without the deprecated flash mapper, why did you add it? |
| 23:11.20 | WM | sn9: it is not working on a1200. |
| 23:11.22 | goxboxlive | The one in the list. |
| 23:12.26 | WM | ive added the map, because a user wanted to backup his firmware, but no success. |
| 23:12.44 | sn9 | and that didn't help, either? |
| 23:12.54 | LuitvD | I'll be back next week I guess... I'll have time then to test some stuff on my E6 |
| 23:15.15 | WM | goxboxlive: mtd3 on the original layout will not be touched by the files you will flash. |
| 23:15.36 | goxboxlive | ok |
| 23:16.21 | goxboxlive | WM: so you deleted: libezxwvimpsutil-xscale-r.so.1.0.0 and all the samba content? |
| 23:17.11 | WM | o have not deleted the wvimps lib, i deleted only the samba and wvimps executable. |
| 23:17.24 | WM | it was enough to reduce the size of the rootfs. |
| 23:17.37 | WM | s/o have/i have/ |
| 23:18.27 | goxboxlive | ok |
| 23:22.09 | goxboxlive | WM: both of them ./usr/SYSqtapp/wvimps |
| 23:22.09 | goxboxlive | ./usr/SYSqtapp/wvimps/wvimps and smbd? |
| 23:22.29 | WM | IIRC yes. |
| 23:23.08 | goxboxlive | :-) But what about the samba startup script i /etc/init.d/samba ? Isent it started at boot? |
| 23:24.24 | WM | i havent deleted. probably my phone is showing some errors behind the GUI ;) |
| 23:25.42 | goxboxlive | ok |
| 23:26.27 | *** join/#openezx DNAfanboy (n=cesar@foresight/developer/DNAFanboy) |
| 23:32.19 | WM | goxboxlive: use people.openezx.org/wyrm/00020000.bin instead of the one inside myflash.tgz |
| 23:32.44 | goxboxlive | WM: ok thx |
| 23:32.59 | WM | the only change is that the one inside myflash.tgz shows a blue screen, and the other doesnt touch the FB. |
| 23:34.02 | goxboxlive | ok goo |
| 23:34.04 | goxboxlive | d |
| 23:40.08 | DNAfanboy | hm, hi all |
| 23:41.01 | DNAfanboy | the kernel on the Angstrom unstable snapshot expects ext2 or ext3 root? and in what partition of the microSD? |
| 23:41.11 | DNAfanboy | (I'm on an A1200, BTW) |
| 23:41.53 | WM | ext2 2nd partition |
| 23:42.00 | DNAfanboy | WM: thanks! |
| 23:44.00 | WM | DNAfanboy: but i think that the OE kernels for a1200 are still broken. |
| 23:44.45 | WM | .23 is broken for a1200 even on our svn. |
| 23:46.04 | DNAfanboy | WM: so I should use one of your kernels? |
| 23:47.12 | WM | DNAfanboy: ATM, the best option is the latest .21 kernel from my images dir. |
| 23:47.29 | WM | DNAfanboy: or compile one yourself :) |
| 23:58.05 | WM | goxboxlive: have you bricked it already? ;) |