Re: [beagleboard] firmware loads fail, remoteproc error codes

2019-05-29 Thread Hugh Frater
If anyone stumbles upon this thread - the attached script will allow init.d to start the PRU at boot time which the new pru_rproc driver doesn't seem to do. In debian 8.7 with 4.4 kernels, it needed a reboot once the kernel module had loaded because of a bug creating the rpmsg entry in /dev.

Re: [beagleboard] firmware loads fail, remoteproc error codes

2019-05-28 Thread Hugh Frater
Fixed it... sudo sh -c "echo 'start' > state" from within /sys/devices/platform/ocp/4a326004.pruss-soc-bus/4a30.pruss/4a338000.pru/remoteproc/remoteproc2 On Tuesday, 28 May 2019 17:25:44 UTC+1, Hugh Frater wrote: > > My takeaway from Mark's excellent writeup is that I should be able to

Re: [beagleboard] firmware loads fail, remoteproc error codes

2019-05-28 Thread Hugh Frater
My takeaway from Mark's excellent writeup is that I should be able to copy my .out file into /lib/firmware with the correct naming convention. I should then be able to echo 'start' or 'stop' to the correct state file. Which seems to be located at:

Re: [beagleboard] firmware loads fail, remoteproc error codes

2019-05-28 Thread Hugh Frater
Reading it now, I'll be back... Cheers for all your help up till now On Tuesday, 28 May 2019 16:53:47 UTC+1, RobertCNelson wrote: > > On Tue, May 28, 2019 at 10:41 AM Hugh Frater > wrote: > > > > Tried both those fixes Robert, no joy. Is it still acceptable practice > to symlink

Re: [beagleboard] firmware loads fail, remoteproc error codes

2019-05-28 Thread Hugh Frater
Tried both those fixes Robert, no joy. Is it still acceptable practice to symlink am335x-pru1-fw to our .out file? Google led me to this post: http://e2e.ti.com/support/processors/f/791/t/783113?Linux-AM3358-remoteproc-remoteproc0-Direct-firmware-load-for-tmp-PRU-LED1-out-failed-with-error-2

[beagleboard] firmware loads fail, remoteproc error codes

2019-05-28 Thread Hugh Frater
Does anyone know what the remoteproc error codes are? pru_rproc doesn't start the pru's at boot time, I assume this is the new behaviour in 4.14 with the uBoot overlays. I've updated initramfs since symlinking my binary to /lib/firmware/am335x-pru1-fw, that hasn't helped. I've done an