On 09/12/2012 08:54 AM, Alexander Graf wrote:
> So the CC pseudo-register is never written to?
They do handle a write to cc_regnum in s390_pseudo_register_write.
They modify psw.mask as one would expect.
r~
On 09/12/2012 05:11 PM, Richard Henderson wrote:
On 09/12/2012 06:25 AM, Alexander Graf wrote:
+case S390_PSWM_REGNUM:
+env->psw.mask = tmpl;
+env->cc_op = (tmpl >> 13) & 3;
Are you sure this is correct? I thought gdbstub would just ignore the cc bits.
Well... no it won't i
On 09/12/2012 06:25 AM, Alexander Graf wrote:
>> +case S390_PSWM_REGNUM:
>> +env->psw.mask = tmpl;
>> +env->cc_op = (tmpl >> 13) & 3;
>
> Are you sure this is correct? I thought gdbstub would just ignore the cc bits.
Well... no it won't ignore the cc bits. But it would appear
On 09/09/2012 11:04 PM, Richard Henderson wrote:
The real gdb protocol doesn't split out pc or cc as real registers.
Those are pseudos that are extracted as needed from the PSW. Don't
modify env->cc_op during read -- that way lies heisenbugs.
Fill in the XXX for the fp registers.
Remove duplic
The real gdb protocol doesn't split out pc or cc as real registers.
Those are pseudos that are extracted as needed from the PSW. Don't
modify env->cc_op during read -- that way lies heisenbugs.
Fill in the XXX for the fp registers.
Remove duplicated defines in cpu.h.
Signed-off-by: Richard Hend
