[Bug middle-end/92436] SIMD integer subtract with constant always becomes add
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92436 Andrew Pinski changed: What|Removed |Added Target|x86, aarch |x86, aarch64 --- Comment #2 from Andrew Pinski --- Testcases: x86: #include __m128i fn(__m128i a) { return _mm_or_si128( _mm_sub_epi8(a, _mm_set1_epi8(99)), _mm_set1_epi8(99) ); } aarch64: #include uint8x16_t fn(uint8x16_t a) { return vorrq_u8( vsubq_u8(a, vdupq_n_u8(99)), vdupq_n_u8(99) ); } Note aarch64 is a regression from GCC 5.
[Bug middle-end/92436] SIMD integer subtract with constant always becomes add
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92436 Richard Biener changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed||2019-11-11 Ever confirmed|0 |1 --- Comment #1 from Richard Biener --- Confirmed, I think we have dups for this (also applies to FP non-vector constants).
[Bug middle-end/92436] SIMD integer subtract with constant always becomes add
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92436 Andrew Pinski changed: What|Removed |Added Severity|normal |enhancement