Port-arm archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Tegra kernel panics without DIAGNOSTIC



On Tue, Mar 07, 2017 at 06:17:17AM +0000, John Klos wrote:
> An NVIDIA Tegra system panics with a NetBSD-current kernel built
> from the sources as of three hours ago when compiled without options
> DIAGNOSTIC:
> 
> ...
> gpio25 at tegragpio0 (Z): 8 pins
> gpio26 at tegragpio0 (AA): 8 pins
> gpio27 at tegragpio0 (BB): 8 pins
> gpio28 at tegragpio0 (CC): 8 pins
> gpio29 at tegragpio0 (DD): 8 pins
> gpio30 at tegragpio0 (EE): 8 pins
> simplebus0 at fdt0: regulators
> fdt1 at simplebus0
> fregulator0 at fdt1: +VDD_MUX
> 
> uvm_fault(0x81685928, 0, 1) -> e
> Fatal kernel mode data abort: 'Translation Fault (S)'
> trapframe: 0x816cf948
> FSR=00000005, FAR=00000000, spsr=20000153
> r0 =00001fff, r1 =00000001, r2 =00000000, r3 =00000000
> r4 =816cf9dc, r5 =00020000, r6 =0000e4a0, r7 =815007c8
> r8 =00000000, r9 =81683758, r10=816cfa14, r11=816cf9c4
> r12=000000b3, ssp=816cf998, slr=81683758, pc =813150b0
> 
> Stopped in pid 0.1 (system) at  813150b0:       ldr     r0, [r3]
> db{0}>
> 
> When I hit a key, it immediately rebooted. Ideas?

I can't reproduce with sources of a few minutes ago.  Then again, I'm
using a different U-Boot and FDT than most other people.

Did you start your non-DIAGNOSTIC build from a clean kernel object
directory?  I wouldn't be surprised that DIAGNOSTIC vs. non-DIAGNOSTIC
have a different ABI in places...

	Jonathan Kollasch


Home | Main Index | Thread Index | Old Index