ARM exception vector question -- never mind

Gene Smith gds at chartertn.net
Tue Aug 5 02:43:33 UTC 2008


Never mind. I see it in the bsp's start/start.S.
...

gds wrote, On 08/04/2008 06:13 PM:
> For my type of arm (ARM946ES) there needs to be exception jump structure 
> at the bottom of memory (address 0) to handle the exceptions (data 
> aborts, irq, etc). Looking through the ARM specific rtems code I can't 
> seem to find anything that defines this type of thing (see below, where 
> workingEntry is located at 0). This is from a linux bsp that uses the 946es.
> 
> Do the current ARM bsp's in rtems not need this type of structure? Or 
> possibly it's assumed to be setup by a bootloader? Or am I just not 
> seeing something like this in rtems code?
> 
> I also looked at the contributed NDS project which uses ARM946ES and 
> can't find this.
> 
> The items at resetJump: and below I do see how they are initialized in 
> function _CPU_ISR_install_vector in score/cpukit/score/cpu/arm/cpu.c.
> 
> Thanks,
> -gene
> 
> .global workingEntry
> workingEntry:
> 	ldr	r15, resetJump
> 	ldr r15, undefinedInstructJump
> 	ldr r15, softwareInterrupt
> 	ldr r15, prefetchAbort
> 	ldr r15, dataAbort
> 	ldr r15, unknownVector
> 	ldr r15, normalIntr
> 	ldr r15, fiqIntr
> resetJump:
> .int contEntry
> undefinedInstructJump:
> .int contUndefInstr
> softwareInterrupt:
> .int contSWI
> prefetchAbort:
> .int contPrefetch
> dataAbort:
> .int contData
> unknownVector:
> .int contUnknownVector
> normalIntr:
> .int contIntr
> fiqIntr:
> .int contFiq
> 
> _______________________________________________
> rtems-users mailing list
> rtems-users at rtems.com
> http://rtems.rtems.org/mailman/listinfo/rtems-users
> 




More information about the users mailing list