Hi Andrew, I retried the 3.14 kernel.
I'm getting the following error: WARNING: CPU: 0 PID: 110 at drivers/usb/gadget/f_fs.c:2595 ffs_ep0_write+0x44d/0x474() [ 149.939235] Modules linked in: [ 149.942448] CPU: 0 PID: 110 Comm: adbd Tainted: G W 3.14.35+ #6 [ 149.949616] [<c00119c5>] (unwind_backtrace) from [<c000f63b>] (show_stack+0xb/0xc) [ 149.957570] [<c000f63b>] (show_stack) from [<c05005ab>] (dump_stack+0x4f/0x64) [ 149.965181] [<c05005ab>] (dump_stack) from [<c0030b4f>] (warn_slowpath_common+0x4b/0x68) [ 149.973686] [<c0030b4f>] (warn_slowpath_common) from [<c0030b7d>] (warn_slowpath_null+0x11/0x14) [ 149.982940] [<c0030b7d>] (warn_slowpath_null) from [<c036f70d>] (ffs_ep0_write+0x44d/0x474) [ 149.991729] [<c036f70d>] (ffs_ep0_write) from [<c00cafb1>] (vfs_write+0x6d/0x130) [ 149.999599] [<c00cafb1>] (vfs_write) from [<c00cb2c9>] (SyS_write+0x2d/0x64) [ 150.006994] [<c00cb2c9>] (SyS_write) from [<c000cc81>] (ret_fast_syscall+0x1/0x46) [ 150.014950] ---[ end trace 3555ece349f08477 ]--- [ 151.039009] read descriptors [ 151.042065] read strings I believe it has to do with the FunctionFS. What is it used for? Can I disable it? I believe it is only used so that ADB would work right? To provide you with some extra info, I had to build the kernel with the folowing modifications to the defconfig: I commented: # CONFIG_USB_G_NOKIA=m //This led to an inconsistent state I included this: CONFIG_DMA_OMAP=y //This was a required dependency Cheers, Michael 2015-03-18 2:39 GMT+01:00 Andrew Henderson <[email protected]>: > Are you using an LCD cape? I have seen flickering similar to what you > described with the 3.8 kernel when I forced a video resolution via the > "video=" kernel command line option (in uEnv.txt in the boot partition) for > the LCD3/4/7 CircuitCo capes and the 4DCAPE-43T/70T capes from 4D Systems. > For those display capes, the resolution is pulled from the device tree > overlay for the cape automatically, so you should not specify a video > resolution. If you are using HDMI, you may not be getting the proper EDID > resolution information from your display. In that case, the TILCDC driver > will default to 1024x768, which your display (if it is an HDMI display) may > not properly support. Either way, I have seen that refresh problem when the > logical resolution that the TILCDC driver expects mismatches the physical > resolution of your display device. > > I have not seen that particular USB error before, but I also have not > worked with the 3.10 kernel on the BBB. I've gone straight from 3.8 to > 3.13/14. > > Andrew > > On Tuesday, March 17, 2015 at 4:45:42 PM UTC-4, Michael Gobbers wrote: >> >> Hi Andrew, >> >> Thanks for the reply. Indeed Keith Conger supplied me with the patches. >> I have gotten a bit further now as I'm getting the hand of configuring >> and building the kernel for the beaglebone black. I'm now able to boot all >> the way to the android home screen. >> >> However, I have major graphical glitches. It seems that the boot >> animation is being displayed together with the home screen. Their also >> seemed to be a lot of flickering on the screen. So there is something >> seriously wrong there. >> >> The biggest issue for me that I came across was the fact that my usb host >> port didn't seem to work. So no mouse available there. >> Today I tried the 3.10 kernel and there I do not have the graphical >> glitches. However, again there is no host port available. >> >> from the ftdi output I did find the following while booting the kernel: >> >> [ 2.725494] musb-hdrc musb-hdrc.0.auto: musb_init_controller failed with >> status -517 >> >> [ 2.733755] platform musb-hdrc.0.auto: Driver musb-hdrc requests probe >> deferral >> >> Have you ever come across this error? >> > -- Michael Gobbers <http://be.linkedin.com/in/michaelgobbers> +32477188592 Desguinlei 6 - 2018 Antwerpen -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
