[PATCH 1/3] PPC: permit %dm0 as insn operand

[Surya Kumari Jangala]

Hi Jan,
Thanks a lot for finding, and fixing, this issue!
Apologies for not getting back sooner on this patch.

Regards,
Surya

On 19/01/26 3:13 pm, Jan Beulich wrote:
> On 12.01.2026 13:25, Jan Beulich wrote:
>> When the DM registers were added, sorting of the table was broken. With
>> the present arrangement, this leads to (only) %dm0 not being found by the
>> binary search that both use sites of the table entertain.
> 
> Short of any reply I'm going to commit at least this one now, for it to be
> in 2.46. I'll wait some more with the other two patches here, as there it
> seems a little more likely that I might be moving things in a direction
> not wanted by you, the maintainers.
> 
> Jan
> 
>> ---
>> Should there perhaps be verification of the table being sorted, in debug
>> builds?
>>
>> --- a/gas/config/tc-ppc.c
>> +++ b/gas/config/tc-ppc.c
>> @@ -338,7 +338,6 @@ static const struct pd_reg pre_defined_r
>>    { "ctr", 9, PPC_OPERAND_SPR },
>>    { "dar", 19, PPC_OPERAND_SPR },
>>    { "dec", 22, PPC_OPERAND_SPR },
>> -  { "dsisr", 18, PPC_OPERAND_SPR },
>>  
>>    /* Dense Math Registers.  */
>>    { "dm0", 0, PPC_OPERAND_DMR },
>> @@ -350,6 +349,8 @@ static const struct pd_reg pre_defined_r
>>    { "dm6", 6, PPC_OPERAND_DMR },
>>    { "dm7", 7, PPC_OPERAND_DMR },
>>  
>> +  { "dsisr", 18, PPC_OPERAND_SPR },
>> +
>>    /* Floating point registers */
>>    { "f.0", 0, PPC_OPERAND_FPR },
>>    { "f.1", 1, PPC_OPERAND_FPR },
>>
>