Re: BEAGLEBONE 8.0_BETA kernel fails

[Nick Hudson <skrll%netbsd.org@localhost>]

On 08/01/17 00:55, John D. Baker wrote:
> On Mon, 31 Jul 2017, Nick Hudson wrote:
>
> > VERBOSE_INIT_ARM output please
[snip]
> devmap: 48000000 -> 48ffffff @ e0000000
> pmap_map_chunk: pa=0x48000000 va=0xe0000000 size=0x1000000 resid=0x1000000 prot=0x3 cache=0
> sS

This is the mapping used for device in question... looks ok.
> devmap: 4a000000 -> 4affffff @ e1000000
> pmap_map_chunk: pa=0x4a000000 va=0xe1000000 size=0x1000000 resid=0x1000000 prot=0x3 cache=0
> sS
> devmap: 4c000000 -> 4c0fffff @ e2000000
> pmap_map_chunk: pa=0x4c000000 va=0xe2000000 size=0x100000 resid=0x100000 prot=0x3 cache=0
> S

> omapdmtimer0 at obio1 addr 0x48040000-0x48040fff intr 68: DMTIMER2
> omapdmtimer2 at obio1 addr 0x48044000-0x48044fff intr 92: DMTIMER4

These are the physical addresses for the devices.

[snip]

> Fatal kernel mode data abort: 'External Non-Linefetch Abort (S)'
> trapframe: 0x80a4de70
> FSR=00001808, FAR=e0044010, spsr=20000113

It fails trying to access 0xe0044010... no idea why? Maybe some board 
clock initialisation is missing?

> r0 =00000000, r1 =e0044000, r2 =00000010, r3 =00000001
> r4 =000003e8, r5 =9ff9a1a0, r6 =80311914, r7 =00000001
> r8 =806fe59c, r9 =9ef40ed8, r10=87ffc01b, r11=80a4dedc
> r12=80a4dee0, ssp=80a4dec0, slr=803063cc, pc =80311914
>
> Stopped in pid 0.1 (system) at  netbsd:generic_bs_w_4:  str     r3, [r1, r2]
> db>
Maybe you could git bisect it to see what broke it?

Nick