[Bug tree-optimization/110362] New: Range information on lower bytes of __uint128_t

[Bug tree-optimization/110362] New: Range information on lower bytes of __uint128_t

[antoshkka at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110362

            Bug ID: 110362
           Summary: Range information on lower bytes of __uint128_t
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Keywords: missed-optimization
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: antoshkka at gmail dot com
  Target Milestone: ---

Consider the following example:

int test (__uint128_t a, __uint128_t b) {
  __uint128_t __a = b | (a << 32);
  return __a & 0xffffffff;
}

At the moment GCC-14 with -O2 generates the following assembly:

test(unsigned __int128, unsigned __int128):
        mov     rsi, rdi
        mov     rax, rdx
        sal     rsi, 32
        or      rax, rsi
        ret


Which could be simplified to just:

test(unsigned __int128, unsigned __int128):
        mov     rax, rdx
        ret

Godbolt playground: https://godbolt.org/z/K9x5vnhxq

[Bug rtl-optimization/110362] Range information on lower bytes of __uint128_t

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110362

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2023-06-22
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed.

Also on the GIMPLE level we have

  <bb 2> [local count: 1073741824]:
  # RANGE [irange] __int128 unsigned [0, 0][4294967296, +INF] NONZERO
0xffffffffffffffffffffffff00000000
  _1 = a_2(D) << 32;
  __a_4 = _1 | b_3(D);
  _5 = (int) __a_4;

the lowpart cast (int) __a_4 can be simplified to (int) b_3(D) based on
the range info of _1.  For example via

(simplify
 (convert (ior:c with_possible_nonzero_bits@0 @1))
 (if ( ... @0 has all zeros in lower TYPE_PRECSION (type) bits ...)
  (convert @1)))

I suppose a variant with b & (a << 32) is already optimized (to zero)?

[Bug tree-optimization/110362] Range information on lower bytes of __uint128_t

[roger at nextmovesoftware dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110362

Roger Sayle <roger at nextmovesoftware dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|rtl-optimization            |tree-optimization
                 CC|                            |roger at nextmovesoftware dot com

--- Comment #2 from Roger Sayle <roger at nextmovesoftware dot com> ---
The good news is that this has been fixed in the RTL optimizers/x86 backend,
and GCC-14 currently produces the optimal "mov rax, rdx".  However, I agree
with Richard Biener that could/should also be fixed (earlier) at the
tree-level.