**** BEGIN LOGGING AT Fri Apr 15 02:59:58 2016 Apr 15 03:34:29 Anybody know offhand the pin changes between a5c and rev c? Apr 15 03:35:07 Tried to hook it up to I2C and SPI and somehow my device ended up in read-only mode, so I think I accidentally ended up on the MMC pins Apr 15 03:35:13 Even though I was on the P9 rail Apr 15 07:57:19 hi all - i have strange artifacts on TFT LCD Display they look like frequent vertical dim lines - what might cause issue ? Apr 15 07:58:30 yes Apr 15 08:13:18 maquefel: one sec Apr 15 08:15:41 maquefel: http://gerbil.xs4all.nl/lcd-util.tgz run as root, it'll dump some info and ask "press enter to change polarity", if you press enter it'll cycle through four options, see if any of those improve things (press control-C or control-D to exit) Apr 15 08:16:33 (I'm assuming here you're using a BBB, although you didn't specify) Apr 15 08:17:42 zmatt: thanks, yes a bbb with newheaven LCD 7 display 24 bit Apr 15 08:30:08 zmatt: i can get only worser =) Apr 15 08:30:37 may be backlight is damaged Apr 15 08:31:19 but i don't have them in corners and sides of screen Apr 15 08:33:15 well then at least the display controller config isn't the problem... I have no idea what might be Apr 15 08:33:29 maybe contact the cape manufacturer Apr 15 08:38:26 * maquefel contacted self Apr 15 08:38:58 can it be raster mode ? Apr 15 08:39:02 problems ? Apr 15 08:41:09 ? Apr 15 08:41:57 diplaying 24 bit in raster mode Apr 15 08:43:04 hmmm i ll try to remove protection film Apr 15 08:53:03 zmatt: thanks for help Apr 15 08:53:13 usefull utility Apr 15 13:03:24 n00b here Apr 15 13:03:58 any suggestions on which rev of black to purchase? Apr 15 13:05:21 * shiplits waves hello Apr 15 13:08:09 is any revision other than rev C actually still sold? (other than second-hand) Apr 15 13:09:26 in any case, rev C is what you want since earlier revisions have only 2GB of eMMC Apr 15 13:12:47 hello Apr 15 13:13:04 has the new name for the usb network devices changed? Apr 15 13:13:15 i have a USB_ETH from gadget api enabled Apr 15 13:13:34 but I dont have a /sys/class/net/usb0 i have /sys/class/net/uab0 instead Apr 15 13:19:45 Thank you for the information Apr 15 14:21:28 hi, I cross compile Qt program from Debian VM to my bbb (Debian Jessie). Is Someone familiar with arm-linux-gnueabihf ? Apr 15 14:21:41 Because the commnd line is very long for compiling .cpp Apr 15 14:22:15 And I would like to include .ui made with Qt design. But I don't know how. Apr 15 14:22:33 I'm familiar with cross-compilation, but not with Qt Apr 15 14:24:36 zmatt : ok thanks, still searching but I find nothing that could help Apr 15 14:31:03 Hello there, i'm the annoying guy from http://beagleboard.org/Community/Forums?place=msg%2Fbeagleboard%2FLglGzXxQ6zk%2F-PJ34P7uBgAJ ... I still can't figure out what I'm doing wrong - is there some documenation about the different debian versions provided on elinux? Apr 15 14:49:21 you fail to describe what problem you're seeing ("can't get the c++lib to work like before" is not very descriptive) Apr 15 14:49:57 the differences between the three versions are afaik mostly the amount of crap installed in userspace, nothing kernely Apr 15 14:50:50 I use /dev/mem a lot too (I still need to convert to using uio instead) without any problems Apr 15 14:51:04 Kernely. I like it. Apr 15 14:51:32 note that adding an overlay for the UART actually *conflicts* with using the uart via /dev/mem Apr 15 14:52:10 unless you make the overlay change the choice of driver to uio (see e.g. http://pastebin.com/GrHwgYiR ) Apr 15 14:52:39 basically the options are: Apr 15 14:53:08 1. activate the kernel driver, use the kernel driver as intended, don't meddle with the device directly from userspace Apr 15 14:54:15 2. tell the kernel to use the uio driver, open the uio device from userspace, mmap it and use the same way as /dev/mem (except without the requirement to be root, and you can also receive irqs in userspace! :D ) Apr 15 14:54:52 3. fuck the kernel, don't tell it about the device at it, use /dev/mem to directly enable it in PRCM and access the peripheral directly Apr 15 14:58:44 if using uio then PRCM is handled by the kernel; the peripheral will be enabled by the kernel when you open the uio device and disable it again when it's closed Apr 15 15:00:11 one exception I'd make for the "don't meddle from userspace" rule when using a kernel driver is gpio, since it can be used fairly safely from multiple places at the same time Apr 15 15:00:30 and if you're sufficiently nuts (like me), same goes for EDMA Apr 15 15:03:08 oh, he left Apr 15 15:03:13 well that's not useful Apr 15 15:03:32 Ingrate! Apr 15 15:04:02 at what point did he leave? Apr 15 15:04:16 Sadly I know not. I was in fact afk as well. Apr 15 15:04:37 oh well Apr 15 15:04:39 I also hide join/part and other such static. Apr 15 15:05:16 yeah I still wish for the ability to hide join/parts except for people who spoke in the last screenful of lines or so Apr 15 15:46:05 ERROR: "of_update_property" [drivers/net/ethernet/ti/davinci_mdio.ko] undefined! Apr 15 15:46:06 /home/rah/linux-4.4.6/scripts/Makefile.modpost:91: recipe for target '__modpost' failed Apr 15 15:46:09 hrm Apr 15 17:58:17 hello Apr 15 17:58:41 im trying to get usb networking going on a beaglebone Apr 15 17:58:56 im using a yocto build kernel/rootfs Apr 15 18:07:06 Hello, I'm trying to re-compile Richard Nelson's kernel 4.1.18-ti-r57 following the recipe from https://github.com/RobertCNelson/ti-linux-kernel-dev/tree/4.1.18-ti-r57. When running the build_kernel.sh script, the build fails when trying to apply one of the patches. Apr 15 18:09:08 here is the output of the script when it fails: http://pastebin.com/szWXDi08 Apr 15 18:09:33 can anybody help me fix this? Apr 15 18:23:15 raffo: He's "Robert Nelson" not "Richard" anyway. Apr 15 19:28:28 which USB peripheral controller driver is used for the beaglebone black? Apr 15 19:28:47 or is there a 4.1 kernel config online? Apr 15 19:40:21 http://github.com/beagleboard/linux Apr 15 19:42:02 many thanks Apr 15 20:06:29 musb Apr 15 20:09:04 raffo: that's odd, have you previously built a kernel in the same dir? (seems unlikely given that you named it with a version number, but still...) try tossing the KERNEL directory and letting the script recreate it Apr 15 20:09:46 thanks, it's a fresh download. Let me try again. Apr 15 20:10:41 latest release is 4.1.20-ti-r58 btw Apr 15 20:12:10 (latest release in the 4.1-ti series that is) Apr 15 20:13:54 ok. thanks, I'm giving that a try right now. Apr 15 20:16:20 me too, it already passed the patch that failed for you Apr 15 20:16:39 all patches successful Apr 15 20:24:58 yep, the particular release you tried to build actually shows the same failure for me also Apr 15 20:25:07 a faulty release, unusual Apr 15 20:25:29 especially apparently he did build a release from it Apr 15 20:25:41 *especially since Apr 15 20:28:43 but 4.1.20-ti-r58 fixed the problem Apr 15 20:30:52 I had a 4.1.18-ti-r55 tree working, but it broke after I run build_kernel.sh. r56 and r57 also are broken. Apr 15 20:31:15 4.1.20-ti-r58 is compiling now. Apr 15 20:32:02 I've compiled many many versions of his trees (both -ti and -bone), this is the first time I've seen a failure **** ENDING LOGGING AT Sat Apr 16 02:59:58 2016