Re: kernel 2.6.3x?
On Tue, 10 Nov 2009 09:09:00 +0100 Jose Luis Perez Diez jl...@escomposlinux.org wrote: El Monday, 9 de November de 2009 19:05:04 arne anka va escriure: ... so, simple question: how's the state of the new kernel? anybody working on it? GTA02 boots mainline 2.6.32 and Lars is working on pending patches see http://article.gmane.org/gmane.comp.handhelds.openmoko.kernel/10406 Well my neo freerunner made a nice kernel panic when booting the 2.6.32 with a config that should have everything needed, (I even tried with the config file given in this reply http://article.gmane.org/gmane.comp.handhelds.openmoko.kernel/10407 ) so I really don't know... as for now I am on a custom andy-tracking kernel because I can't find another version source that boot on my neo... (I don't use openembedded I'm a I do it my way kind of guy ^^) Oh I found out that the 2.6.32_rc6 has an undefined reference to GTA02_GPIO_AMP_HP_IN something really bad because it never apprear anywhere else in the code (grep -R proved that) offending file is sound/soc/s3c24xx/neo1973_gta02_wm8753.c compiled with SND_S3C24XX_SOC_NEO1973_GTA02_WM8753=y I alsa found out that the compilation fails at link time if debug is not activated because of an undefined reference to printascii() (defined in assembly only when debug activated) function is called anyway in s3c_pm_dbg() which is called even if debug is deactivated) I don't know if I should post it on the official kernel mailinglist or not so I post it here. signature.asc Description: PGP signature ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: 2.6.31 is out, where is my ultimate kernel?
On Thu, 10 Sep 2009 14:06:55 +0200 Raphaƫl Jacquot sxp...@sxpert.org wrote: On Thu, 2009-09-10 at 15:55 +0400, Paul Fertser wrote: * imporved wifi I don't know anybody who's going to invest real time into rewriting/fixing driver, especially if there're userspace workarounds present. I, for once, am not really motivated unless Atheros gives us a way to use more sane firmware I'd rather use no firmware at all ;) Well... it's a full MAC device so having a firmware is a good thing :p but having an opensource firmware would be the best thing, especially if there are problems with the current one. signature.asc Description: PGP signature ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
RFC Using MPX for accelerometers
I've had an idea that might not be so bad. MPX is due for the next release of Xorg, so why not use it for the accelerometers, it could create a somewhat standard interface for the applications (think about compatibility for other devices than the freerunner) to see the state of the accelerometers. An application that is fullscreen/selected could just grab the input and release it if it is no more selected. The nice thing is that application not compatible with Xi2 would not see the device if it is configured as additionnal pointer and not default one. If an application needs only pattern recognitions and not the position of the pointer, we could create a lib, so that it could be portable for other devices. So comments about that idea? ^^ signature.asc Description: PGP signature ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: [Any distro?] Funny, very old SIM card not detected
On Wed, 17 Dec 2008 20:40:17 +0100 Filip Onkelinx fi...@linux4.be wrote: Hi, could it be that this is an old 5V SIM card? If I'm not mistaken, those do not work with FR; only 3V and 1.8V (like in most recent phones). As I said in my first mail, I use the freerunner as a daily phone with Om2008.9-testing20081216, so this is not an hardware problem given it works. signature.asc Description: PGP signature ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: [Any distro?] Funny, very old SIM card not detected
ah... forgot the list... On Wed, 17 Dec 2008 12:35:26 +0100 Abelenda diego.abele...@gmail.com wrote: On Wed, 17 Dec 2008 03:03:35 -0800 Dima Kogan dko...@cds.caltech.edu wrote: What's the model of the SIM card? I've had an issue where an old SIM card would only register after a few minutes, and some distros had a bug where they'd get confused if the registration didn't happen early enough. Qtopia would register fine, but Android and anything using FSO would not. See http://trac.freesmartphone.org/ticket/255 Any info on how I do get the model of SIM? The only thing written on it is the serial number (this is an orange (CH)). Anyway I don't understand what you mean by register time, is it the time the SIM takes to associate to an antenna ? the time the SIM take to verify the PIN ? This is not in question here, my PIN isn't even asked by most distros... And it takes less than a minute between the moment the pin is asked and the moment the signal strength is displayed (by Om2008.9) (no mesurements here ^^). Ah I forgot to say I didn't update the firmware of the calypso, because #1024 has no reason to affect me and I don't think that a command that is more recent than the SIM will change anything here... signature.asc Description: PGP signature ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
[Any distro?] Funny, very old SIM card not detected
I came across a really funny thing when trying to get rid of the WSOD. First a little explaining about what I know about my problem: android rootfs doesn't detect my sim card (a card that is more than 3 years old), the thing is that the official releases of Om (including the testing ones) detect the sim... but any other distro fails... well in the case of Android, this is more than simply not detecting it, there are 3 cases with equal probability. 1. doesn't tell anything about a sim card 2. doesn't ask me for my pin and says there is no service 3. asks my pin but doesn't try to unlock the sim and asks me my pin immediately after saying failed (I can do it any number of times and it will not lock my sim) others distro doesn't do anything. Well Om200.9-testing 20081216 works for me, I can use my freerunner as a daily phone with it, so I let this matter for later, but just now I flashed andy-tracking kernel (the binary on his storage, to get rid of the WSOD) and Om2008.9-testing doesn't asks me for my pin anymore so where is my question(s) is it normal due to the /sys changes? is it a kernel bug ? (shouldn't as the calypso is connected via serial port there shouldn't be any kernel related problem, except a serial related bug, that should not have been overlooked for so long...). signature.asc Description: PGP signature ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Why kdrive ?
Hi I just recieved an interesting mail in the Xorg MailingList this was the contents : Why not just go for the full X11? Memory wise I found no difference in the memory usage between X11 and kdrive on the OMAP. I am told that the full Xorg isn't very happy when built against uClibc. So here's my question, if the memory footprint of kdrive is the same as the full xorg, why choose kdrive over the full xorg ? There might be interesing features in the full server that aren't there in kdrive. Plus I've read here that the X driver has got duplicate code in the X driver from the framebuffer, so there's no really gain in term of size of the driver either... So here's my question if I am totally wrong or asked a totally stupid question I am sorry, as I don't know all the implications. signature.asc Description: PGP signature ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: Why kdrive ?
On Wed, 15 Oct 2008 11:29:38 +0100 Graeme Gregory [EMAIL PROTECTED] wrote: Ok, I may have mangled the history a little here, but this is the basics I hope. When we first started on this kdrive/tinyX was the playground for new features within the xorg/xf86 family. xorg didn't support touchscreens, or a configless running. The lack of these features was a killer because making a user construct an xorg.conf on device was a pain. kdrive also had the RENDER extension so people could play with shinies. Xglamo was then created in the kdrive family. Now OM is going to invest some time in investigating if Xorg or Kdrive is the best path forward in the future. Xorg has acheived some level of configless running. Has touchscreen drivers. And the shinies extensions. It is being used on other embedded boards built from OE as well. Graeme Makes sense at that time. Nowadays kdrive doesn't seems to be a solution if I refer the the thread in the Xorg MailingList. ^^ It appears that kdrive rendering method KAA has been removed because EXA is much better. Well, in the case of openmoko, this would mean changing the driver for it to work with xorg. (nearly coding a new driver). On the other hand this would mean a modular X server, with graphic drivers loaded as needed, which means broader support without having to compile a X server per device (as kdrive is compile time configuration). That would correct some bugs I saw about keyboards not recognized by Xglamo and so on. Anyway seriously thinking about migrating to Xorg is not a bad idea itself but would mean a lot of work from openmoko because the community cannot help without the specs... too bad... signature.asc Description: PGP signature ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
softfloat or hardfloat ?
Hi, out of curiosity I checked what was the default floating point implementation for the gcc you provide for cross-compiling. [1] It was a shock when I saw hard-float in the spec. Is it really a good default for a processor which doesn't have FPU like the ARM920T ? I read that in running hardfloat code on a cpu that doesn't support floating point arithmetic is 10 times slower than using the floating point libraries, due to the exception catch by the kernel. So if you use a toolchain defaulting to hard implementation you need to the option -msoft-float for every compilation. Here are my questions : What are the options for the compilation of packages ? Does it contain -msoft-float ? If not, is there a specific reason ? [1] http://wiki.openmoko.org/wiki/Toolchain ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community