On 1/17/23 1:42 PM, Richard Henderson wrote:
> Is there a reason why these are separate from m_systemreg?
GDB puts these in a separate file, and J-Link puts them in a separate feature
block.
In general, I think it's nice to separate stuff related to the secure extension
so folks not working with
On 1/17/23 1:40 PM, Richard Henderson wrote:
>> diff --git a/target/arm/cpu.h b/target/arm/cpu.h
>> index bf2bce046d..fdbb0d9107 100644
>> --- a/target/arm/cpu.h
>> +++ b/target/arm/cpu.h
>> @@ -856,6 +856,7 @@ struct ArchCPU {
>> DynamicGDBXMLInfo dyn_sysreg_xml;
>> DynamicGDBXMLI
On 1/17/23 5:37 AM, Peter Maydell wrote:
> In patch 1 you skip the registers that don't exist without
> the main extension, but here you throw them all in regardless.
> Why the difference ?
Ah, yes. This was an oversight. I'm not sure if there are any
chips that support the security extension bu