[Bug target/113180] New: MIPS: bitops on an long long struct uses ins instead dins

[Bug target/113180] New: MIPS: bitops on an long long struct uses ins instead dins

[syq at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113180

            Bug ID: 113180
           Summary: MIPS: bitops on an long long struct uses ins instead
                    dins
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: syq at gcc dot gnu.org
  Target Milestone: ---

```
struct yy {
        int x:32;
        int a:12;
        int b:12;
        int c:4;
        int d:4;
};

struct yy xx (struct yy a, long long b) {
        a.d = b;
        return a;
}
```

One DINS is enough for this ops, while currently the result is:

```
        dsra    $2,$4,32
        daddiu  $sp,$sp,-32
        ins     $2,$5,28,4
        sd      $4,16($sp)
        sw      $2,20($sp)
        ld      $3,16($sp)
        move    $2,$0
        sll     $4,$3,0
        dext    $4,$4,0,32
        dins    $2,$4,0,32
        dsrl    $3,$3,32
        dins    $2,$3,32,32
        jr      $31
        daddiu  $sp,$sp,32
```

[Bug middle-end/113180] MIPS: bitops on an long long struct uses ins instead dins (or with -mstrict-align on aarch64)

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113180

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
          Component|target                      |middle-end
             Target|mips                        |mips aarch64*-*-*
             Status|UNCONFIRMED                 |NEW
           Keywords|                            |missed-optimization
   Last reconfirmed|                            |2023-12-31
            Summary|MIPS: bitops on an long     |MIPS: bitops on an long
                   |long struct uses ins        |long struct uses ins
                   |instead dins                |instead dins (or with
                   |                            |-mstrict-align on aarch64)

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
The same issue happens on aarch64 with -mstrict-align though it is not as bad
as mips but the underlying issue is the same:
```
        sub     sp, sp, #32
        .cfi_def_cfa_offset 32
        str     x0, [sp, 8]
        lsr     x0, x0, 32
        bfi     w0, w1, 28, 4
        str     w0, [sp, 12]
        ldr     x0, [sp, 8]
        add     sp, sp, 32
        ret
```

Confirmed. there might be other bug reports which are very similar too.

[Bug middle-end/113180] MIPS: bitops on an long long struct uses ins instead dins (or with -mstrict-align on aarch64)

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113180

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |enhancement

[Bug middle-end/113180] MIPS: bitops on an long long struct uses ins instead dins (or with -mstrict-align on aarch64)

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113180

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|mips aarch64*-*-*           |mips aarch64*-*-* arm*-*

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
arm is just as bad:
```
xx:
        @ args = 8, pretend = 0, frame = 8
        @ frame_needed = 0, uses_anonymous_args = 0
        @ link register save eliminated.
        sub     sp, sp, #8
        mov     r3, r0
        strd    r1, r2, [sp]
        ldr     r1, [sp, #8]
        ldrb    r2, [sp, #7]    @ zero_extendqisi2
        bfi     r2, r1, #4, #4
        strb    r2, [sp, #7]
        ldrd    r0, r1, [sp]
        stm     r3, {r0, r1}
        mov     r0, r3
        add     sp, sp, #8
        @ sp needed
        bx      lr
```

Basically the variable a is stored to the stack and then modified and then
stored to the stack and then loaded and stored to the return location.

This is all due to STRICT alignment IIRC.

[Bug middle-end/113180] MIPS: bitops on an long long struct uses ins instead dins (or with -mstrict-align on aarch64)

[syq at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113180

--- Comment #3 from YunQiang Su <syq at gcc dot gnu.org> ---
The argument `to` of `expand_assignment` differs between `strict-align` and
`no-strict-align`.

`debug_tree` states that the `strict-align` one has a `MEMREF` RTL, while 
`no-strict-align` has a `reg` one.

Any idea when the RTL is generated from STMT?