![]() ![]() unmounts/remounts frequently until UI freezes.works stable when connected on USB2 > USB3 disk gets detected, but is unstable. > as my TV does not seem to support EDID, I hat to ssh in and change the settings to 1080p-noedid > the MAC for eth0 is different from standard HK images (I use my DHCP-server to hand-out "static" IPs based on known MACs) > openelec did not resize the partition automatically used a 16GB SD Card.flashed it with the instructions from #1 post on my linux workstation Whoa! This is some awesome piece of work! If I do, I may proverbially smack someone if they report either of these two problems post-release. I had a video, but it turns out the camera didn't capture in a high enough def, to capture the information anyone : If I can't get any new insight to these problems within the next week or so, I think I'll release with big red letters. I'll try to put up a small video of (2) tonight when I get home. However HW Accel, shouldn't need a cpu core to really work right so it's quite possible there is more at play here. The result is the same using Memeka's or Owersun's MFC implementation and makes me think this maybe a symptom of (1). (Once using a 4.7 GCC I didn't get the error, but I've never been able to re-create that result).Ģ.) MFC Decoding (currently using Owersun's implementation because Memeka's requires patches that breaks the U2/U3 MFC), consistently drops frames. ![]() I've been told this is a gcc bug, yet I've rebuilt with 4.7, 4.8, and 4.9 GCC toolchains and always seem to get this error. There are only really two problem existing yet on the XU3 and one, maybe a symptom of the other:ġ.) Only 1 of the 8 CPU cores is working. I really don't want to release it, but since its a universal image, its holding up the U2/U3 release as well. Time keeps on ticking, and I'm getting to the point I may release 4.2.1.0 whether the XU3 support is perfect or not. I have just figured it out, it is x86 binary and I tried to execute it on odroid (I am actually building it on odroid). I'd look into A.) does this mkbl2 need to be executed (Answer is NO btw), and B.) what gcc compiled it? (you would want host-gcc not armv7a-openelec-linux-gnueabi-gcc).ĭid you change anything in the source that might have had an affect on the output? This one looks like mkbl2 is being executed, but it is not compiled for the host OS. Makefile:349: recipe for target 'u-boot.bin' failed mkbl2: cannot execute binary file: Exec format error home/odroid/Build/openelec-4.1.6/-4.1.6/toolchain/bin/armv7a-openelec-linux-gnueabi-objcopy -gap-fill=0xff -O binary u-boot u-boot.bin Code: Select all /home/odroid/Build/openelec-4.1.6/-4.1.6/toolchain/bin/armv7a-openelec-linux-gnueabi-objcopy -O srec u-boot u-boot.srec ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |