[PATCH v2] ifcvt: Avoid early multiple-set conversion for extend patterns

[Jeffrey Law]

On 1/15/2026 1:55 PM, Jeffrey Law wrote:
>
>> (subreg:SI (ior:DI (reg/v:DI 135 [ a ])
>>         (reg/v:DI 136 [ b ])) 0)
>
> Unfortunately num_sign_bit_copies for those objects is not returning 
> anything useful in this context.
>
> I want to think about this a bit more.  It really feels like we should 
> have  a better solution than special casing this in ifcvt.
So after a bit of pondering, I think we have a clean way forward. I've 
cc'd Daniel as it's an area he's done some recent work.

In the .optimized dump we have:


>   if (c_2(D) != 0)
>     goto <bb 3>; [50.00%]
>   else
>     goto <bb 4>; [50.00%]
> ;;    succ:       3 [50.0% (guessed)]  count:536870912 (estimated 
> locally, freq 0.5000) (TRUE_VALUE,EXECUTABLE)
> ;;                4 [50.0% (guessed)]  count:536870912 (estimated 
> locally, freq 0.5000) (FALSE_VALUE,EXECUTABLE)
>
> ;;   basic block 3, loop depth 0, count 536870912 (estimated locally, 
> freq 0.5000), maybe hot
> ;;    prev block 2, next block 4, flags: (NEW, REACHABLE, VISITED)
> ;;    pred:       2 [50.0% (guessed)]  count:536870912 (estimated 
> locally, freq 0.5000) (TRUE_VALUE,EXECUTABLE)
>   a_5 = a_3(D) | b_4(D);
> ;;    succ:       4 [always]  count:536870912 (estimated locally, freq 
> 0.5000) (FALLTHRU,EXECUTABLE)
>
> ;;   basic block 4, loop depth 0, count 1073741824 (estimated locally, 
> freq 1.0000), maybe hot
> ;;    prev block 3, next block 1, flags: (NEW, REACHABLE, VISITED)
> ;;    pred:       2 [50.0% (guessed)]  count:536870912 (estimated 
> locally, freq 0.5000) (FALSE_VALUE,EXECUTABLE)
> ;;                3 [always]  count:536870912 (estimated locally, freq 
> 0.5000) (FALLTHRU,EXECUTABLE)
>   # a_1 = PHI <a_3(D)(2), a_5(3)>

That seems like another case we can likely handle with Andrew's new 
infrastructure to drive phi-opt from match.pd and turn that mess into a 
branchless sequence in gimple.

Jeff