Last month, I had a few problems with 1.3I on my CATS system: activity on
a serial port causing a panic.  Recently, I noticed a change committed to
footbridge/footbridge_irq.S.  Unfortunately, this appears to have moved
rather than removed the problem.  I'm now getting (as pppd starts during
boot):

syscall: swi 0x2 from non USR32 mode
syscall: trapframe = 0xf36bc744
panic: syscall in non USR32 mode
Stopped at _Debugger + 0x10

Show regs gives:

spsr 	0x20000013
r0	0xa
r1	_current_spl_level
r2	0x1
r3 	_db_onpanic
r4	0xf36bc6c4
r5	0x100
r6	0x4
r7	_session0
r8	0x400
r9	0xf36bc720
r10	0xf4000000
r11	0xf36bc6ac
r12	0xf36bc6b0
usr_sp	0xefbfdcb4
usr_lr	0x1546c
svc_sp	0xf36bc6c0
svc_lr	_panic + 0x5c
pc	_Debugger + 0x10
und_sp	0xf37e3ff0
abt_sp	0xf01b0000
irq_sp	0xf01af000

db> tr
_Debugger
_panic 	(+0x14)
_syscall	
_ttstart
_comsoft
_dosoftints
_ppptioctl 	(+0x240)
_pppintr
_dosoftints
_dosoftints
_dosoftints    repeated as far as my patience with the space bar would
		permit...

Attempting to continue doesn't get very far either...

db> c
boot: howto=00000100 curproc=0x0
syncing disks... Fault with intr_depth > 0
Data abort: 'Permission error (page)' status = 00f address=f010e123
					PC=f003d1f8
Stopped at _putc+0xe8: strb r2,[r12,r3]


Anybody got any ideas as to what's going on?

Thanks,

David.


PS - Thanks to whoever (Mark?) fixed the r8=0 case in footbridge_irq.S.
:-)