On 24/09/18 04:21, Jeff Rizzo wrote:
Trying to get my Jetson TK1 into a more-usable state; it was running a 7.99.23 kernel with 7.0 userland relatively well, but not being one to leave well enough alone, I decided to upgrade to 8.0, figuring that would be, you know, better. :)
There have been reports of netbsd-8 being unstable on tk1 http://mail-index.netbsd.org/port-arm/2018/07/27/msg004967.html I did suggest bisecting in private.
I managed to get an 8.0 kernel (the one from ftp.n.o) running; however, the USB3 hub i had been using stopped working. (It had worked, albeit in USB2 mode, under the 7.99.23 kernel) At a suggestion from jmcneill@, I installed the nvidia driver binaries in /libdata/firmware - once I did that, I still wasn't able to boot with the USB hub connected, but once it booted, when I plug the hub in, I see the uhub4 through uhub6 attach, but none of the devices:uhub4 at uhub2 port 1: VIA Labs, Inc. (0x2109) USB3.0 Hub (0x813), class 9/0, rev 3.00/90.01, addr 1 uhub5 at uhub4 port 1: VIA Labs, Inc. (0x2109) USB3.0 Hub (0x813), class 9/0, rev 3.00/90.01, addr 2 uhub6 at uhub4 port 4: VIA Labs, Inc. (0x2109) USB3.0 Hub (0x813), class 9/0, rev 3.00/90.01, addr 3My keyboard works OK if I plug directly into the one on-board USB port, but I kinda like using a hub. :)
xhcidebug/uhubdebug is probably required here :(
In order to see if this is a fixed issue, I then tried booting a -current image from SD card; this was a little tricky to get booted due to u-boot issues, but I did eventually get it to boot a kernel (I had to set ${soc} and ${fdtfile} in u-boot manually). Booting this kernel shows a few problems with the tegrai2c* driver (times out eventually), but then crashes in drm:[ 1.0000630] tegradrm0 at simplebus0[ 1.0000630] kern info: [drm] Supports vblank timestamp caching Rev 2 (21.10.2013). [ 1.0000630] kern info: [drm] No driver support for vblank timestamp query.[ 1.0000630] uvm_fault(0x819e6ba8, 0, 1) -> e [ 1.0000630] Fatal kernel mode data abort: 'Translation Fault (S)' [ 1.0000630] trapframe: 0x81a33a20 [ 1.0000630] FSR=00000005, FAR=00000110, spsr=20000153 [ 1.0000630] r0 =00000110, r1 =00000000, r2 =00000000, r3 =817e3b58 [ 1.0000630] r4 =9188bf88, r5 =9188bf88, r6 =00000110, r7 =00000000 [ 1.0000630] r8 =8165d144, r9 =9153e448, r10=81a33ba0, r11=81a33adc [ 1.0000630] r12=81a33ae0, ssp=81a33a70, slr=8126f31c, pc =81389270
I've fixed this, I think. Nick