**** BEGIN LOGGING AT Thu Jul 08 02:59:57 2010 Jul 08 07:30:34 lag: Hi! I just look at the syslink patch applied by Tim, and I don't see the modalias, nor the module_init(). Did you test this patch? I do not expect it to be functional... Jul 08 07:31:14 What do you mean you don't see them? Jul 08 07:34:45 Sebjan? Jul 08 07:34:45 lag: in the patch applied on the top of the tree, there is not the MODULE_ALIAS code for example Jul 08 07:35:01 * lag looks Jul 08 07:35:52 sebjan: that's true, please follow the email thread started from lag Jul 08 07:36:42 lag: in you v1 patch, there is MODULE_ALIAS code Jul 08 07:37:01 lag: but in the v2 patch which is in Maverick now, there is no such code Jul 08 07:37:09 lag: i guess sebjan is asking this Jul 08 07:37:23 correct :) Jul 08 07:37:26 lag: btw, thanks a lot for email me the blob file Jul 08 07:37:51 cooloney, sebjan: You're correct, I'll add it as a separate patch right now Jul 08 07:38:08 lag: great, man Jul 08 07:38:12 lag: we also need the module_init(), right? Jul 08 07:38:23 Why do you need that? Jul 08 07:39:40 lag: for the omap4_syslink_init function to be called on startup? Or the __init does it? Jul 08 07:40:37 morning Jul 08 07:41:09 sebjan: The device_initcall(omap4_syslink_init); does it Jul 08 07:47:48 sebjan: http://kernel.ubuntu.com/git?p=lag/ubuntu-maverick.git;a=shortlog;h=refs/heads/ti-omap4-syslink Jul 08 07:48:31 Sorry about that. I did a git reset --mixed Jul 08 07:48:57 Added omap4-common.c and must have missed off ipc_drv.c Jul 08 07:49:16 It was still in my tree though, hence when I tested it, it worked Jul 08 08:02:40 lag: no problems. By the way, we shall stick with syslink drivers static right now, as we are seeing issues when using them as modules. We are investigating that, and will send an update when this will be fixed (then we can activate syslink drivers as modules) Jul 08 08:03:10 No problem - keep me posted Jul 08 08:05:16 lag: yes, sure! Jul 08 08:27:30 sebjan: Can you have a look at the patch on the mailing list please Jul 08 08:27:41 sebjan: If it's okay can you ack it as well? Jul 08 08:30:36 lag: done Jul 08 08:32:54 sebjan: Thanking you! Jul 08 09:34:27 sebjan, i still see hangs with the recent panda kernel Jul 08 09:41:37 a lot less frequent, but they happen Jul 08 09:45:38 ogra: Must be userspace ;) Jul 08 09:45:49 My kernel works flawlessly Jul 08 09:45:51 :) Jul 08 09:46:22 the archive kernel ? Jul 08 09:46:29 installed from the package Jul 08 09:49:54 Nope Jul 08 09:50:19 But it will be the latest Maverick kernel won't it? Jul 08 09:51:10 I'm in the process of flahsing my SD card with the latest archive Jul 08 09:51:50 ok Jul 08 09:52:17 ogra: If one thing (within the kernel) is crashing continuously, raise a bug report and I'll take a look Jul 08 09:53:19 lag, here is another one to play with bug 603062 Jul 08 09:53:21 Launchpad bug 603062 in linux-ti-omap4 (Ubuntu Maverick) (and 1 other project) "oops in parport_pc_probe_port function of parport_pc module (affects: 1) (heat: 8)" [Medium,New] https://launchpad.net/bugs/603062 Jul 08 09:55:13 The new image still doesn't work with my monitor :( Jul 08 09:55:30 And there's no console output either Jul 08 09:56:11 the image should start in console mode by default Jul 08 09:56:23 we dont use splash during resizing yet Jul 08 09:57:12 I have nothing Jul 08 09:57:17 Uncompressing Linux... done, booting the kernel. Jul 08 09:57:26 It's been like this for 5mins Jul 08 09:57:36 thats serial Jul 08 09:57:39 Do you think it's resizing? Jul 08 09:57:40 Yeah Jul 08 09:57:43 we dont use serial on ubuntu images Jul 08 09:57:45 And nothing on the monitor Jul 08 09:57:51 Zero Jul 08 09:57:54 Ziltch Jul 08 09:57:56 Nada Jul 08 09:58:04 do you know a cmdline that works for you ? Jul 08 09:58:08 lag: try enabling DEBUG_LL and DEBUG_LL_CONSOLE Jul 08 09:58:28 ukleinek: From within the kernel? Jul 08 09:58:42 lag: make menuconfig Jul 08 09:58:53 ukleinek, note that ubuntu images dont have console= settings by default Jul 08 09:58:54 fdr editconfigs ;) Jul 08 09:59:10 breaks the splash Jul 08 09:59:25 Still nothing ... Jul 08 09:59:29 * lag whistles Jul 08 09:59:35 lag, lets try to find a cmdline that works for everyone Jul 08 10:00:02 there wont happen anything, if your monitor doesnt init it wont do that later either :) Jul 08 10:00:02 I have a cmdline that works for me Jul 08 10:00:25 lag: add a printch(c) to emit_log_char Jul 08 10:00:27 What will happen if I restart the board during the resizing process? Jul 08 10:00:42 you will have to re-dd Jul 08 10:00:52 the filesystem will be corrupted Jul 08 10:00:52 Grrrrrrrrrrrrr Jul 08 10:00:59 What if I wait Jul 08 10:01:04 It will take the same amount of time Jul 08 10:01:17 it will move on and start X to configure the system eventually Jul 08 10:01:31 which you likely wont see if the framebuffer doesnt work Jul 08 10:02:05 dd the image newly, mount the first partition and add your cmdline changes to boot.scr Jul 08 10:02:16 (before booting it for the first time) Jul 08 10:02:58 dont forget mkimage Jul 08 10:03:35 it would be good if we can find the smallest common denominator that makes all monitors work Jul 08 10:04:03 or if TI can fix the EDID code (preferred indeed) Jul 08 10:06:10 1024x768 should work Jul 08 10:06:41 well, depends on your monitor Jul 08 10:06:58 mine cant do non widescreen modes on the HDMI port afaik Jul 08 10:07:14 but i'll try Jul 08 10:07:52 XGA was default beagleboard resolution Jul 08 10:11:26 lag, what are the options you use Jul 08 10:11:57 setenv bootargs omapdss.hdmimode=0 omapdss.hdmicode=4 vram=15M, omapfb.vram=0:5M noinitrd mem=512M root=/dev/mmcblk0p2 rootwait ro ip=none; mmcinit 0; fatload mmc 0 0x80200000 uImage; bootm 0x80200000 Jul 08 10:12:16 omapfb.vram=0:5M ?!? Jul 08 10:12:23 thats definately to less Jul 08 10:12:24 TI told me to use it Jul 08 10:12:38 Well that works Jul 08 10:12:53 But 32 and 8 work for me also Jul 08 10:13:05 vram=15M is also to less for our kernel Jul 08 10:13:10 setenv bootargs omapdss.hdmimode=0 omapdss.hdmicode=4 vram=32M, omapfb.vram=0:8M noinitrd mem=512M root=/dev/mmcblk0p2 rootwait ro ip=none; mmcinit 0; fatload mmc 0 0x80200000 uImage; bootm 0x80200000 Jul 08 10:13:14 we enable all possible output devices Jul 08 10:13:28 can you try without omapfb.vram=0:8M ? Jul 08 10:13:33 Well your cmdline doesn't work for me so ... Jul 08 10:13:44 I can Jul 08 10:13:45 * ogra will try omapdss.hdmimode=0 omapdss.hdmicode=4 on his monitor Jul 08 10:14:54 * ogra just tried omapdss.hdmimode=0 omapdss.hdmicode=35, that makes the board crash once X is up Jul 08 10:15:10 and cuts of 60% of the bottom of the screen on my monitor Jul 08 10:17:03 lag, also i dont get why mem=512M would work for you Jul 08 10:17:20 i was clearly told that will cause issues and hard locks Jul 08 10:18:01 Probably because I'm not running a GUI Jul 08 10:18:18 What was it again 468? Jul 08 10:18:19 well, robclark said thats reserved ram for the coprocessor Jul 08 10:18:31 i dont think that should harm X Jul 08 10:18:39 or have any influence Jul 08 10:18:49 * lag shurgs Jul 08 10:18:51 It just does Jul 08 10:18:56 What is the correct value? Jul 08 10:19:04 463> Jul 08 10:19:05 ? Jul 08 10:19:07 463 Jul 08 10:19:09 right Jul 08 10:19:57 I'm using a kernel with TI's hack in it at the moment Jul 08 10:20:00 sigh, the last crash trashed my SD Jul 08 10:20:08 Let me build my own and I'll try it out Jul 08 10:20:23 lag, that doesnt help at all Jul 08 10:20:33 please use the same kernel we all use Jul 08 10:20:45 ogra: I'm working with both Jul 08 10:21:03 I have other things going on too :) Jul 08 10:21:46 me too Jul 08 10:22:00 So, I need to be running both kernels Jul 08 10:22:05 but it doesnt help the images if you dotn use and improve the image kernels Jul 08 10:22:22 Doesn't the daily builds use the latest Maverick kernel anyway? Jul 08 10:22:31 the one in the archive Jul 08 10:22:44 which is behind the git tree Jul 08 10:23:19 Where is it built from? Jul 08 10:23:24 the git tree Jul 08 10:23:37 but its not built every day Jul 08 10:24:06 tim builds it from kernel.ubuntu.com from cooloneys tree afaik Jul 08 10:24:14 every few weeks Jul 08 10:25:23 with mode=4 i get a small square in the top left corner of my monitor Jul 08 10:25:43 i dont think any of these options will help, TI needs to fix the EDID detection code properly Jul 08 10:26:08 its bad enough that we have to use *any* cmdline options Jul 08 10:27:11 ogra: http://paste.ubuntu.com/460578/ Jul 08 10:27:14 Try some more Jul 08 10:27:20 See what works for you Jul 08 10:27:37 well, what else works for you ? Jul 08 10:27:43 I don't know yet Jul 08 10:27:47 doesnt help if i try all of them Jul 08 10:27:51 What do I have to mkimage on? Jul 08 10:28:02 do* Jul 08 10:28:05 the edited boot.scr Jul 08 10:28:11 How do you do that? Jul 08 10:28:26 I've only done uInitrd and uImage Jul 08 10:28:34 sudo mkimage -A arm -T script -C none -n "Ubuntu boot script" -d boot.scr /boot.scr Jul 08 10:28:55 the first boot.scr needs to be plain text (no uboot header) Jul 08 10:29:06 thats the one you edited Jul 08 10:29:41 I only have the one with the header on Jul 08 10:29:46 Should I just take that off? Jul 08 10:29:52 yeah Jul 08 10:29:57 just delete it in vi Jul 08 10:30:08 mkimage adds it again Jul 08 10:30:09 vi? pfft Jul 08 10:30:32 well, or use: strings /boot.scr >./boot.scr Jul 08 10:30:40 but that might leave cruft at the top line Jul 08 10:31:57 lag, can you try omapdss.hdmicode=6 ? thats one i could imagine all HDMI monitors should be able to do Jul 08 10:32:15 err. line 6 Jul 08 10:32:22 thats code 28 indeed Jul 08 10:33:24 4 sort of works Jul 08 10:33:27 I can see it Jul 08 10:33:32 4 doesnt work here Jul 08 10:33:34 But it's out of range of my monitor Jul 08 10:33:40 And is shakey Jul 08 10:33:45 well, it draws on the screen but is unusable Jul 08 10:34:26 Running /scripts/local-bottom ... Jul 08 10:34:30 640x480n is nothing we can run the desktop in Jul 08 10:34:45 you wont be able to reach any of the buttons in the apps Jul 08 10:34:59 I'll try 28 when it's finished doing 'its thing' Jul 08 10:35:28 It doesn't matter anyway - my keyboard doesn't work! Jul 08 10:35:29 Grr Jul 08 10:36:49 no space left on device? Jul 08 10:37:02 It's a 16GB card? Jul 08 10:37:12 it didnt resize Jul 08 10:37:21 Why not? Jul 08 10:37:27 MMC issues i guess Jul 08 10:37:38 But it resizes yours? Jul 08 10:37:38 GrueMaster reported similar issues with some SD cards Jul 08 10:37:55 I did get the parport segfault though Jul 08 10:37:55 it resizes the five i have here for testing Jul 08 10:38:20 Do you have a 16GB SDHC card there? Jul 08 10:38:27 GrueMaster seems to have a similar card to yours, please check with him (once he is up) if you got the same Jul 08 10:38:30 no Jul 08 10:38:35 my biggest one is 8G Jul 08 10:38:40 SDHC? Jul 08 10:38:43 yes Jul 08 10:38:48 Odd Jul 08 10:38:59 Can I resize it manually? Jul 08 10:39:05 error -110? Jul 08 10:39:07 no Jul 08 10:39:10 Nope Jul 08 10:39:13 Different bug Jul 08 10:39:46 can you plug the card into your pc and check for /var/log/jasper.log please Jul 08 10:40:07 it should have error messages Jul 08 10:40:32 where exactly do you get the "no space left on device" errors btw Jul 08 10:40:47 any particular process thats causes them ? Jul 08 10:41:31 code 28 doesnt work here either btw Jul 08 10:41:43 chops off the bottom half of the screen Jul 08 10:44:23 lag, where do you get the no spaxce messages ? Jul 08 10:44:49 On the monitor Jul 08 10:44:52 if its udevd thats a tmpfs issue (udev writes to a tmpfs) Jul 08 10:44:56 haha Jul 08 10:45:01 ;) Jul 08 10:45:11 I can't remember what it was doing now Jul 08 10:45:19 hmm Jul 08 10:45:20 It could have been udev Jul 08 10:45:25 well, then only jasper.log can tell Jul 08 10:45:27 The screen has gone now Jul 08 10:45:35 does X start ? Jul 08 10:45:38 Nope Jul 08 10:45:49 the timeframe was also way to short Jul 08 10:45:58 resizing 16G will likely take about 15min Jul 08 10:46:12 you guys are insane with that resizing Jul 08 10:46:18 so please get me the log so i can see what went wrong Jul 08 10:46:27 hrw, why ? Jul 08 10:46:43 hrw, compared to a 2-3h install its quite reasonable Jul 08 10:47:02 ogra: cant I fetch tarball which will be unpacked to card? Jul 08 10:47:07 plus you need an additional target device Jul 08 10:47:23 hrw, that will be as messy as a rootstock image Jul 08 10:47:36 you will never get a proper ubuntu install that way Jul 08 10:48:15 it will be something "ubuntu alike" Jul 08 10:48:34 so now I get 512MB image to dd on card which will start with initrd which will resize that 512MB to whole card? Jul 08 10:49:09 you get a 1.4G image and yes, initrd is essential for a proper ubuntu image Jul 08 10:49:20 ogra: 28 from the cmdline is still out of range Jul 08 10:49:22 However Jul 08 10:49:27 When I do: echo 28 | sudo tee -a /sys/devices/platform/omapdss/display0/custom_edid_timing Jul 08 10:49:30 It looks great Jul 08 10:49:36 oh Jul 08 10:49:38 hmm Jul 08 10:49:39 why I can install x86(-64) with 1CD but arm needs >2CD? Jul 08 10:50:03 hrw, because x86 uses a squashfs and a full installer Jul 08 10:50:20 our image is bz2 and needs to be uncompressed before dd Jul 08 10:50:38 and is fully preinstalled Jul 08 10:50:48 hrw: Our boards don't have CD drives :) Jul 08 10:50:57 well, they have SD slots Jul 08 10:51:10 Yes, hence why we use entire images Jul 08 10:51:15 which we used before in a similar way to CD drives Jul 08 10:51:34 so you had to install a live image to USB from SD card Jul 08 10:51:37 Our full install takes 5 mins. How long does X86 take? Jul 08 10:51:51 a) the installation can take 2-3h depending on the card Jul 08 10:52:11 b) you need to use the SD as a "boot floppy" on boards like panda Jul 08 10:52:26 so you need to trash the installation media at the end of the install Jul 08 10:52:29 very very ugly Jul 08 10:52:38 Yup Jul 08 10:52:50 x86 takes between 20 and 40min Jul 08 10:52:58 with a speedy CD drive Jul 08 10:53:13 it copies the 2G from the squashfs to the HDD Jul 08 10:53:20 and then removes bits Jul 08 10:54:04 in our images the resizing takes the longest time, after that its relatively speedy since only oem-config runs Jul 08 10:54:25 sadly resizing takes at least 10min Jul 08 10:55:06 lag, anything intresting in the jasper.log file ? Jul 08 10:55:36 It was trashed Jul 08 10:55:41 gah Jul 08 10:55:42 I trashed it Jul 08 10:55:48 I am re-running Jul 08 10:56:17 what you can do is to add break=bottom on the cmdline, that will give you a busybox shell Jul 08 10:56:38 jasper.log lives in /dev/.initramfs/jasper.log there Jul 08 10:57:03 you can ctrl-d out of that shell and booting moves on Jul 08 10:57:37 (works also with break=top or premount, depending where you want to inspect the initramfs) Jul 08 10:57:44 Just tested 28 with the archive kernel Jul 08 10:57:46 Nada Jul 08 10:57:50 Got to reflash now :( Jul 08 10:57:52 sad Jul 08 10:58:03 we really need TI to fix that properly then Jul 08 10:58:18 fiddling with cmdline options in the default install is ugly anyway Jul 08 10:58:51 * ogra prefers to not have more than "quiet splash root=UUID=..." Jul 08 10:59:20 and ro Jul 08 11:02:19 Well I have direct contact with the video people at TI ... Jul 08 11:02:49 robclark is resident here too :) Jul 08 11:03:15 Is robclark video people? Jul 08 11:03:44 yep Jul 08 11:04:05 Cool - let's bother him :) Jul 08 11:04:19 Where is he based? Jul 08 11:04:53 tx Jul 08 11:10:24 * ogra played with /sys/devices/platform/omapdss/display0/custom_edid_timing Jul 08 11:10:41 sadly i have the same issue as with setting modes on cmdline Jul 08 11:10:49 screen isnt fullscreen Jul 08 11:11:22 i either have a small square at the top left of the monitor or the screen extends beyond monitor size Jul 08 11:12:00 * ogra wishes DVI would just work Jul 08 11:13:34 :( Jul 08 11:13:50 hmm Jul 08 11:13:57 DVI does work Jul 08 11:14:15 well, the HDMI->DVI cable from my beagle does at least Jul 08 11:14:24 weird Jul 08 11:14:56 ogra: .initramfs doesn't exist Jul 08 11:14:58 but same effect Jul 08 11:15:07 /dev/.initramfs/ Jul 08 11:15:19 init mounts it under /dev Jul 08 11:15:23 its a tmpfs Jul 08 11:16:23 Oh okay Jul 08 11:16:33 I'll try dropping into a Busybox shell then Jul 08 11:16:41 * lag is reflashing Jul 08 11:16:44 This sucks Jul 08 11:17:54 aha, DVI doesnt work on boot at all Jul 08 11:18:06 well, it does, but puts the screen to weird ranges Jul 08 11:20:24 * ogra takes a break Jul 08 11:43:10 * lag pounces on robclark! Jul 08 11:43:27 hi lag Jul 08 11:43:33 Hey robclark Jul 08 11:43:36 hi robclark Jul 08 11:43:44 How are you this afternoon/morning? Jul 08 11:43:47 hi hrw Jul 08 11:44:04 * robclark is just finishing my coffee Jul 08 11:44:17 Can you type and drink? :) Jul 08 11:44:28 Or would you like me to wait for you to finish? Jul 08 11:44:52 no, I can type Jul 08 11:46:28 We're having video issues Jul 08 11:46:45 uh-oh Jul 08 11:46:54 I have an LG FLATRON W2261VP Jul 08 11:47:21 I have tried omapdss.hdmicode=* Jul 08 11:47:53 what resolution? If it could do 1920x1080 then, in theory, it shouldn't even need any bootargs.. Jul 08 11:48:13 robclark: thats 22" so rather not fullhd Jul 08 11:48:16 And depending on what '*' is, I either receive shakey text and the monitor say "out of range" Jul 08 11:48:20 Or I get nothing Jul 08 11:48:38 Yes, it does 1920x1080 Jul 08 11:48:41 Or should do Jul 08 11:49:14 hmm.. in the printk's from the HDMI driver you should see what resolution the driver *thinks* the monitor can do Jul 08 11:49:14 But it doesn't Jul 08 11:49:26 With no args, the monitor just goes to sleep Jul 08 11:49:50 but I think the driver still has troubles with some monitors.. Jul 08 11:49:57 We don't have access to the boot-up messages with this kernel Jul 08 11:50:04 * lag flicks to another kernel Jul 08 11:50:08 hmm... ok.. that is a bit inconvenient Jul 08 11:50:24 It does the same thing on my kernel Jul 08 11:50:32 although to be honest, I'm not a display driver guy.. Jul 08 11:51:09 if you could get a log of the boot msgs, we send that to me (rob@ti.com) and I'll loop in display driver team Jul 08 11:51:30 * robclark thinks what is really needed is to get the display driver team access to more monitors to test with Jul 08 11:51:50 I am in direct contact with Mythri Jul 08 11:51:58 ok, perfect Jul 08 11:52:18 But things are slooooooooow over email Jul 08 11:52:26 she knows the HDMI driver more than I.. Jul 08 11:52:49 ok.. I'll try to pester them to get on IRC.. Jul 08 11:53:08 That could be a good thing Jul 08 11:53:24 I think she's in the far east though isn't she? Jul 08 11:53:33 yes Jul 08 11:53:36 And would have gone home by now? Jul 08 11:53:50 lag, afaik, the trouble with higher resolutions is that the driver can't keep up the refresh rate Jul 08 11:53:55 well, most of them work rather late.. so probably still in the office now Jul 08 11:54:09 it's pretty much up to the monitor whether it will still work at that point Jul 08 11:54:33 cwillu_at_work: So where do we go with this? Jul 08 11:55:22 lag, working from memory here, having spoken to folks before on this. Jul 08 11:55:33 hang on a sec, let me check my logs Jul 08 11:55:41 The odd thing is, a mode that doesn't work on the cmdline using omapdss.hdmicode will work using: echo * | sudo tee -a /sys/devices/platform/omapdss/display0/custom_edid_timing Jul 08 11:56:06 And it will work perfectly then? Jul 08 11:56:32 oh, really? Jul 08 11:56:43 Yeah! What gives? Jul 08 11:56:57 But that's no good for be, as I'm working on boot-up code Jul 08 11:57:03 what does .hdmicode= do that a normal omapdss command line doesn't? Jul 08 11:57:08 me* Jul 08 11:57:23 That's what I'm looking for now Jul 08 11:57:24 ;) Jul 08 11:57:27 * lag is on it Jul 08 11:57:34 omapfb.mode=dvi:1024x768MR-16 Jul 08 11:57:39 is what I'm usually doing Jul 08 11:57:51 although usually 1280x1024 or 1600x900 Jul 08 11:57:53 I'm using hdmi Jul 08 11:58:02 cwillu_at_work: on beagleboard? Jul 08 11:58:02 same thing Jul 08 11:58:05 yes Jul 08 11:58:09 oh, are we not talking beagles? Jul 08 11:58:12 Nope Jul 08 11:58:13 cwillu_at_work: they talk about pandaboard Jul 08 11:58:16 I have Beagle working Jul 08 11:58:24 I'd still imagine it's the same thing though Jul 08 11:58:25 cwillu_at_work: panda can do fullhd-60 Jul 08 11:58:41 just :1280x1024MR-16 instead of DVI: Jul 08 11:59:06 especially if it works echoing to the sysfs Jul 08 12:00:19 I might be out on the road with my boss today, I'll find out in an hour. If not, I'm interested in spelunking into the relevant dss code today :)\ Jul 08 12:04:14 cwillu_at_work: I just tried your args - the monitor just stuck its fingers up at me and went to sleep Jul 08 12:04:57 does it report the refresh rate it's getting before it goes to sleep? Jul 08 12:05:05 and did you try any other options re: the dvi: bit? Jul 08 12:05:58 Yes Jul 08 12:06:01 dvi and hdmi Jul 08 12:06:06 Just did this: echo 40 | sudo tee -a /sys/devices/platform/omapdss/display0/custom_edid_timing Jul 08 12:06:10 And it appeared Jul 08 12:06:15 Looking perfect Jul 08 12:06:17 Grrrrrrrrrrr Jul 08 12:06:38 lag, hmm, so setting it via bootargs seems broken? Jul 08 12:07:38 Correct Jul 08 12:08:25 'TI: please fix your drivers' Jul 08 12:08:32 :) Jul 08 12:14:20 btw, for those who haven't already noticed: #pandaboard Jul 08 12:14:29 lag: http://marcin.juszkiewicz.com.pl/2009/06/12/ti-please-fix-your-usb/ Jul 08 12:14:46 Does anyone have a keyboard and mouse attached to their Panda? Jul 08 12:15:05 lag: yup Jul 08 12:15:11 lag, come to think of it, there should be another file in that folder with the names for each display Jul 08 12:15:27 try doing a "cat *", and see if anything shows up Jul 08 12:15:38 hrw: How funny! Jul 08 12:15:44 * robclark has to run.. be back later Jul 08 12:16:58 EDID-Informationcat: driver: Is a directory Jul 08 12:16:58 1 Jul 08 12:16:58 cat: hpd_enabled: Input/output error Jul 08 12:16:58 cat: mirror: No such file or directory Jul 08 12:17:00 hdmi Jul 08 12:17:03 cat: rotate: No such file or directory Jul 08 12:17:06 cat: subsystem: Is a directory Jul 08 12:17:09 0 Jul 08 12:17:14 148500,1920/88/148/44,1080/4/36/5 Jul 08 12:17:17 DRIVER=hdmi_panel Jul 08 12:17:20 1 Jul 08 12:17:23 cat: wss: No such file or directory Jul 08 12:17:25 Whoops Jul 08 12:18:20 so, my suspicion is that hdmi or hdmi_panel is the right invocation then Jul 08 12:18:56 also, I really really hate it when it takes a million years to page an app back into memory when I have 2gb of ram being wasted on cache Jul 08 12:19:16 (and don't tell me to set swappiness, that's useless as a long term proposition) Jul 08 12:19:51 cwillu_at_work: Tried both of them ... nothing Jul 08 12:20:07 boo. Jul 08 12:20:30 k, I'll poke you in a couple hours if I'm not driving out to the middle of nowhere Jul 08 12:20:59 I'm going to have a poke around in the driver in the meantime Jul 08 12:46:40 lag, I heard from Rob that you had some more issues in HDMI.. Jul 08 12:49:29 lag == lee Jul 08 12:49:37 :) Jul 08 12:49:52 We've been chatting Jul 08 12:50:16 did you try out the timing values i suggested over mail ? Jul 08 12:50:29 Yes Jul 08 12:50:36 They worked when I use sysfs Jul 08 12:50:47 ie the custom_edid_timing right? Jul 08 12:50:47 They don't when I use them on the kernel cmdline Jul 08 12:50:52 Correct Jul 08 12:51:15 I don't see why my monitor should be different to any other Jul 08 12:51:16 ok ie when you use omapdss.hdmicode= and omapdss.hdmimode= Jul 08 12:51:24 Beagleboard works out of the box Jul 08 12:51:34 Yes, they don't work Jul 08 12:51:47 it would not be , can you please share the log ? Jul 08 12:52:01 The most I can get from them is a fuzzy screen and my monitor saying "out of range" Jul 08 12:52:32 When I use which settings? Jul 08 12:52:59 any , where you are getting fuzzy screen and monitor giving "out of range " Jul 08 12:53:09 also what is the timing value that you set with beagle ? Jul 08 12:53:36 I have no idea Jul 08 12:53:39 I'll try to find out for you Jul 08 12:54:06 Also, my monitor doesn't turn itself on Jul 08 12:54:08 no that's fine , log would help , but the same command line i sent for 640 and 480 ie vga setting worked ? Jul 08 12:54:19 I have to turn it off, then on again to get my fuzzy screen Jul 08 12:54:59 with the vga setting too ? Jul 08 12:55:25 Here is fuzzy screen settings: http://paste.ubuntu.com/460620/ Jul 08 12:55:49 mythripk, the problem is that we need something that works generally on the ubuntu images Jul 08 12:55:53 Nothing has 'worked' - some have been better than others Jul 08 12:56:04 there is no way to adjust boot options for every monitor Jul 08 12:57:05 preferably by autodetection and without any boot parameters, though if thats not possible, a bootarg with the smallest common denominator that makes it work on all monitors would also work Jul 08 12:58:19 currently the default setting is for 1080p setting but you could pass the boot args , the setting that will definitely work on All Tv's would be VGA resolution according to HDMI spec , I could make it as default Jul 08 12:59:38 Lag didnt VGA resolution work for you ? From your mail i was under the impression that it did. Jul 08 12:59:43 VGA is to small for our desktop Jul 08 13:00:02 Which number was that? Jul 08 13:00:24 omapdss.hdmicode=4 and omapdss.hdmimode=0 Jul 08 13:00:34 and causes weird effects on some monitors ... i.e. mine doesnt expand to the full monitor size, i get a square in the top left corner of the screen Jul 08 13:00:36 That's the fuzzy screen setting that I use Jul 08 13:01:25 ogra , ya VGA is too small but that is the only setting that by default is supported by all TV's and then Jul 08 13:01:57 mythripk: Jul 08 13:01:59 ubuntu@beagleboard:/sys/devices/platform/omapdss$ cat display0/timings Jul 08 13:01:59 61714,1280/80/48/32,720/3/13/5 Jul 08 13:01:59 ubuntu@beagleboard:/sys/devices/platform/omapdss$ cat display1/timings Jul 08 13:01:59 13500,720/12/68/64,574/5/41/5 . Jul 08 13:02:12 From the Beagle Jul 08 13:02:27 mythripk, 1024x768 would work for us ... Jul 08 13:03:08 ok then can you set omapdss.hdmicode=16 and omapdss.hdmimode=0 ? Jul 08 13:03:41 that gets me the above described effect Jul 08 13:03:48 doesnt expand to the full screen size Jul 08 13:04:07 not sure what it does for lag's monitor Jul 08 13:04:21 or for GrueMaster's who also has display issues Jul 08 13:05:16 Checking Jul 08 13:05:30 http://omappedia.org/wiki/Bootargs_for_enabling_display this has explanation as to what you code you could use for all the DVI settings , How to read and set EDID timings for HDMI section , you could modify the omapdss.hdmimode and code correspondingly for bootargs Jul 08 13:05:44 * lag 's monitor stuck its fingers up at him and went to sleep Jul 08 13:06:14 with 16 and 0 ? Jul 08 13:07:07 mythripk, is there any chance that works automatically at some point ? Jul 08 13:07:32 the EDID data should have all possible modes the monitor can handle or not ? Jul 08 13:07:53 so if you can read it you shoudl be able to switch to a valid mode Jul 08 13:07:56 Lag From the pastebin theEDID is read and being set correctly : EDID Content 256 EDID TIMING DATA FOUND Hdmi_code = 4 mode = 1 Jul 08 13:08:48 ogra, yes the EDID has all the values supported by the monitor , but from the log tha lag sent me earlier its NULL , i was actually surprised Jul 08 13:08:49 I think "Hdmi_code = 4 mode = 1" is coming from my boot args Jul 08 13:08:58 Although mode should be 0? Jul 08 13:09:16 mythripk, aha :) Jul 08 13:09:50 yes are you setting omapdss.hdmimode=0? in the bootargs? Jul 08 13:09:57 I am Jul 08 13:10:30 * ogra sees "How to enable HDMI in hot-plug detect mode" ... Jul 08 13:10:41 i wonder if we could enable that by default in our kernel Jul 08 13:11:11 are there any drawbacks when defaulting that value to 1 Jul 08 13:11:13 ? Jul 08 13:11:17 I just tried setting omapdss.hdmimode=0 and leaving the code - nothing Jul 08 13:12:47 ogra , I sent a patch to lag to default put it in hot plug detect mode as he was sync lost digit earlier , through sysfs you could use echo 1 > /sys/devices/platform/omapdss/display2/hpd_enabled again from http://omappedia.org/wiki/Bootargs_for_enabling_display Jul 08 13:13:19 lag , could you please try with another monitor ? or is it the same behavior with all ? Jul 08 13:13:22 mythripk, right, i saw that Jul 08 13:13:46 The only other HDMI device I have is my TV downstairs 42" Jul 08 13:13:52 mythripk, but it seems like something really userfriendly to enable by default if it doesnt disturb other functionallity Jul 08 13:21:40 ogra, yes we could actually read the first block of EDID and set the timings which we currently are not doing today , for some samsung TV's not working Jul 08 13:22:03 hmm Jul 08 13:22:12 i have no issues with my samsung monitor Jul 08 13:22:45 and apparently i'm the only panda user in the ubuntu team who doesnt have display issues ... which is funny :) Jul 08 13:22:54 we have identified that we were setting AVI checksum wrong that resulted in some samsung TV's not working There is a patch available for that Jul 08 13:22:59 good to hear :) Jul 08 13:23:13 so we should include your patch :) Jul 08 13:24:33 that would solve the issues seen in some samsung TV's that enforce strict AVI checksum Jul 08 13:24:48 great, lag can we add that ? Jul 08 13:25:03 Lag already has the patch Jul 08 13:25:24 right, i just want to make sure it ends up in our tree :) Jul 08 13:54:14 Jul 08 13:54:14 Jul 08 13:54:14 Jul 08 13:54:21 Oh my word! Jul 08 13:54:30 Sorry about that - computer went crazy! Jul 08 13:54:41 I blame the OS Jul 08 13:54:50 Right, I've tried it on a different monitor Jul 08 13:55:53 Lag already has the patch: I do? Jul 08 14:05:09 ogra: Where's mythripk gone? Jul 08 14:25:17 lag, I guess mythripk has gone home now Jul 08 14:25:30 np Jul 08 14:25:47 I'm going to try and debug it myself anyway Jul 08 14:26:03 Who needs expert knowledge? Pfft ... Jul 08 14:26:06 :) Jul 08 14:26:14 heheh Jul 08 14:49:56 hrm, why do these packages have {a} and {u} after them? Jul 08 14:50:12 The following NEW packages will be installed: Jul 08 14:50:12 libdbus-1-dev{a} libeet-dev libeet-doc{a} libeet1{a} libhal-dev libhal-storage-dev libjpeg62-dev{a} Jul 08 14:50:12 The following packages will be REMOVED: Jul 08 14:50:12 freepats{u} gstreamer0.10-ffmpeg{u} gstreamer0.10-plugins-ugly{u} Jul 08 14:51:44 {a} == automatically installed Jul 08 14:52:04 as dependency for other packages Jul 08 14:53:29 and {u}? Jul 08 14:56:39 do not remember Jul 08 14:57:42 ogra: I heard my name, but am too fuzzy (ENOCOFFEE) to see what you wanted. What can I do to help this morning? Jul 08 15:01:46 jussi: {a} automatically installed {u} will purge Jul 08 15:01:50 IIRC Jul 08 15:02:04 Or rather automatically uninstalled Jul 08 15:02:13 {up} is when they also get purged I think Jul 08 15:48:15 lool: thanks. also, weird question of the day,will maemo packages work on an ubuntu install? Jul 08 15:50:06 jussi: no Jul 08 15:50:27 jussi: you need to recompile them Jul 08 15:50:45 some of maemo5 libraries are not available in ubuntu Jul 08 15:50:52 ahh Jul 08 15:51:21 some are closed even Jul 08 15:51:36 Im just trying to find a working "media centre" for arm jaunty - xbmc, enna, moovida etc and it seems impossible to find... Jul 08 15:51:54 what hardware? Jul 08 15:52:00 enna has maemo pkgs, hence the question Jul 08 15:52:07 imx51 based Jul 08 15:52:26 imx51... run maverick or lucid then Jul 08 15:52:41 no, cant unfortunately. Jul 08 15:53:31 jussi: You can create a maemo chroot and run it from an Ubuntu install Jul 08 15:53:40 jussi: Which maemo packages do you mean specifically? Jul 08 15:54:25 lool: stuff in here: http://packages.geexbox.org/pool/main/ Jul 08 15:54:59 jussi: It might be that some binaries require opengles though Jul 08 15:55:04 jussi: Dunno Jul 08 15:58:44 lag: I just tested the last syslink auto-load patches, and they fail in loading the modules. I don't see the syslink_ipc event in the logs, so I suspect the omap_syslink_init function to not be called... Could you make a check please? Jul 08 15:59:12 Of course Jul 08 15:59:23 Which logs are you looking in? Jul 08 15:59:48 grep syslink /var/log/udev Jul 08 16:04:12 sebjan? Jul 08 16:04:43 lag: yes, into this log Jul 08 16:05:03 Okay, re-compiling now Jul 08 16:27:08 GrueMaster, we had a TI display person around and were talking about display probs, i only mentioned your name there Jul 08 16:28:56 sebjan: Jul 08 16:28:58 KERNEL[946685538.408904] add /devices/platform/syslink_ipc (platform) Jul 08 16:28:58 UDEV_LOG=3 Jul 08 16:28:58 ACTION=add Jul 08 16:28:58 DEVPATH=/devices/platform/syslink_ipc Jul 08 16:28:58 SUBSYSTEM=platform Jul 08 16:28:59 MODALIAS=platform:syslink_ipc Jul 08 16:29:03 SEQNUM=726 Jul 08 16:29:36 ok Jul 08 16:29:44 lag: shall I understand that you get the uevent? and the modules loaded? Jul 08 16:30:23 No, because I have a read-only file system at the moment Jul 08 16:30:27 Hang on Jul 08 16:31:56 lag: at least 'dmesg | grep syslink' ouputs something (from the omap4_syslink_init function)? Jul 08 16:33:28 Hmmm... no? Jul 08 16:34:01 Perhaps I am working on stagnant logs Jul 08 16:34:04 I'll look into it Jul 08 16:37:42 lag: do we have any mainline kernels for arm? Or do I need to pull source and build my own? Jul 08 16:38:22 GrueMaster: We have images in the archive Jul 08 16:38:48 GrueMaster: http://cdimage.ubuntu.com/ubuntu-netbook/ports/daily-preinstalled/current/ Jul 08 16:39:54 I'm aware of those. I test those. I'm looking for a mainstream kernel package to verify bug 601226 as affecting upstream. Jul 08 16:39:56 Launchpad bug 601226 in linux-ti-omap4 (Ubuntu Maverick) (and 3 other projects) "Unable to handle kernel NULL pointer dereference in ppdev module (affects: 2) (dups: 1) (heat: 16)" [Medium,Triaged] https://launchpad.net/bugs/601226 Jul 08 16:40:18 GrueMaster, there is only the TI branch for panda Jul 08 16:40:21 no mainline Jul 08 16:41:07 ogra_cmpc: This bug also affects Beagleboard. I made your bug a duplicate of mine, as it is the same issue. Jul 08 16:41:11 GrueMaster, you could indeed build from the ti git tree on gitorious.org Jul 08 16:41:24 * ogra_cmpc isnt sure its the same issue Jul 08 16:41:56 GrueMaster, oh, thats parport_pc ? Jul 08 16:42:05 why doesnt your report say that Jul 08 16:42:27 sebjan: Something strange is going on Jul 08 16:42:57 It did. plars just changed it. Jul 08 16:43:19 And I filed my bug last week before I had an omap4 image or system. Jul 08 16:43:28 lag: sorry I need to go now, can we talk tomorrow? Jul 08 16:44:04 GrueMaster, mumble ... mine even points to the function having the issue, i dont get why he changed it Jul 08 16:44:19 sebjan: Of course Jul 08 16:45:24 lag: ok, talk to you tomorrow then **** ENDING LOGGING AT Thu Jul 08 22:47:23 2010 **** BEGIN LOGGING AT Fri Jul 09 00:18:14 2010 **** ENDING LOGGING AT Fri Jul 09 02:59:57 2010