[Bug target/123010] RV64: 32-bit multiply by 2 generates slli+srai+slli and doesn't use slliw or addw

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123010

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
           Severity|normal                      |enhancement
   Last reconfirmed|                            |2025-12-04
             Status|UNCONFIRMED                 |NEW

--- Comment #7 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Even more this is fine:
```
#include <stdint.h>

#define N 2

uint32_t mulu(uint32_t *a) {
    return *a * N;
}

int32_t muls(int32_t *a) {
  return *a * N;
}
```
and:
```
#include <stdint.h>

#define N 2
uint32_t b;
int32_t c;
void mulu(uint32_t a) {
    b= a * N;
}

void muls(int32_t a) {
  c= a * N;
}
```
are ok.

So this is just an interaction between argument passing and returns together
with the shift.