From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 7857) id DCCFE3858D33; Tue, 6 Jun 2023 13:45:22 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org DCCFE3858D33 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1686059122; bh=OcyCr+/Lf9C5cVr+zjETrvbgXkGCZ7/Kmr/qrdE8H4M=; h=From:To:Subject:Date:From; b=A63pbJ+SAIoYK34gqXbWjPb5YNaRw0rExmDVt+wGF6rLTGt4+m8ivf4TAeilRl5eA Rjg5ET7/P5xFEKvt8AITkJTHpV3SB2wwsey5uUzhahxei/s+yQghPoPNGc2tbGsybx EmzIH4kOs6WhXfDDxMiNz4pd+jRxQnfRJgcuzQnY= MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="utf-8" From: Matthias Kretz To: gcc-cvs@gcc.gnu.org, libstdc++-cvs@gcc.gnu.org Subject: [gcc r14-1577] libstdc++: Replace use of incorrect non-temporal store X-Act-Checkin: gcc X-Git-Author: Matthias Kretz X-Git-Refname: refs/heads/master X-Git-Oldrev: a3df359fec49d68d182138f4defc49603b4ada30 X-Git-Newrev: 27e45b7597d6fb1a71927d658a0294797b720c0a Message-Id: <20230606134522.DCCFE3858D33@sourceware.org> Date: Tue, 6 Jun 2023 13:45:22 +0000 (GMT) List-Id: https://gcc.gnu.org/g:27e45b7597d6fb1a71927d658a0294797b720c0a commit r14-1577-g27e45b7597d6fb1a71927d658a0294797b720c0a Author: Matthias Kretz Date: Fri Jun 2 13:44:22 2023 +0200 libstdc++: Replace use of incorrect non-temporal store The call to the base implementation sometimes didn't find a matching signature because the _Abi parameter of _SimdImpl* was "wrong" after conversion. It has to call into ::_SimdImpl instead of the current ABI tag's _SimdImpl. This also reduces the number of possible template instantiations. Signed-off-by: Matthias Kretz libstdc++-v3/ChangeLog: PR libstdc++/110054 * include/experimental/bits/simd_builtin.h (_S_masked_store): Call into deduced ABI's SimdImpl after conversion. * include/experimental/bits/simd_x86.h (_S_masked_store_nocvt): Don't use _mm_maskmoveu_si128. Use the generic fall-back implementation. Also fix masked stores without SSE2, which were not doing anything before. Diff: --- .../include/experimental/bits/simd_builtin.h | 6 ++-- libstdc++-v3/include/experimental/bits/simd_x86.h | 38 +++------------------- 2 files changed, 7 insertions(+), 37 deletions(-) diff --git a/libstdc++-v3/include/experimental/bits/simd_builtin.h b/libstdc++-v3/include/experimental/bits/simd_builtin.h index 8337fa2d9a6..64ef6efaf8c 100644 --- a/libstdc++-v3/include/experimental/bits/simd_builtin.h +++ b/libstdc++-v3/include/experimental/bits/simd_builtin.h @@ -1628,7 +1628,7 @@ template if constexpr (_UW_size == _TV_size) // one convert+store { const _UW __converted = __convert<_UW>(__v); - _SuperImpl::_S_masked_store_nocvt( + _UAbi::_SimdImpl::_S_masked_store_nocvt( __converted, __mem, _UAbi::_MaskImpl::template _S_convert< __int_for_sizeof_t<_Up>>(__k)); @@ -1643,7 +1643,7 @@ template const array<_UV, _NAllStores> __converted = __convert_all<_UV, _NAllStores>(__v); __execute_n_times<_NFullStores>([&](auto __i) _GLIBCXX_SIMD_ALWAYS_INLINE_LAMBDA { - _SuperImpl::_S_masked_store_nocvt( + _UAbi::_SimdImpl::_S_masked_store_nocvt( _UW(__converted[__i]), __mem + __i * _UW_size, _UAbi::_MaskImpl::template _S_convert< __int_for_sizeof_t<_Up>>( @@ -1651,7 +1651,7 @@ template }); if constexpr (_NAllStores > _NFullStores) // one partial at the end - _SuperImpl::_S_masked_store_nocvt( + _UAbi::_SimdImpl::_S_masked_store_nocvt( _UW(__converted[_NFullStores]), __mem + _NFullStores * _UW_size, _UAbi::_MaskImpl::template _S_convert< diff --git a/libstdc++-v3/include/experimental/bits/simd_x86.h b/libstdc++-v3/include/experimental/bits/simd_x86.h index 77d2f84ab71..2e301e45677 100644 --- a/libstdc++-v3/include/experimental/bits/simd_x86.h +++ b/libstdc++-v3/include/experimental/bits/simd_x86.h @@ -1106,31 +1106,6 @@ template else _mm512_mask_storeu_pd(__mem, __k, __vi); } -#if 0 // with KNL either sizeof(_Tp) >= 4 or sizeof(_vi) <= 32 - // with Skylake-AVX512, __have_avx512bw is true - else if constexpr (__have_sse2) - { - using _M = __vector_type_t<_Tp, _Np>; - using _MVT = _VectorTraits<_M>; - _mm_maskmoveu_si128(__auto_bitcast(__extract<0, 4>(__v._M_data)), - __auto_bitcast(_MaskImpl::template _S_convert<_Tp, _Np>(__k._M_data)), - reinterpret_cast(__mem)); - _mm_maskmoveu_si128(__auto_bitcast(__extract<1, 4>(__v._M_data)), - __auto_bitcast(_MaskImpl::template _S_convert<_Tp, _Np>( - __k._M_data >> 1 * _MVT::_S_full_size)), - reinterpret_cast(__mem) + 1 * 16); - _mm_maskmoveu_si128(__auto_bitcast(__extract<2, 4>(__v._M_data)), - __auto_bitcast(_MaskImpl::template _S_convert<_Tp, _Np>( - __k._M_data >> 2 * _MVT::_S_full_size)), - reinterpret_cast(__mem) + 2 * 16); - if constexpr (_Np > 48 / sizeof(_Tp)) - _mm_maskmoveu_si128( - __auto_bitcast(__extract<3, 4>(__v._M_data)), - __auto_bitcast(_MaskImpl::template _S_convert<_Tp, _Np>( - __k._M_data >> 3 * _MVT::_S_full_size)), - reinterpret_cast(__mem) + 3 * 16); - } -#endif else __assert_unreachable<_Tp>(); } @@ -1233,8 +1208,8 @@ template else if constexpr (__have_avx && sizeof(_Tp) == 8) _mm_maskstore_pd(reinterpret_cast(__mem), __ki, __vector_bitcast(__vi)); - else if constexpr (__have_sse2) - _mm_maskmoveu_si128(__vi, __ki, reinterpret_cast(__mem)); + else + _Base::_S_masked_store_nocvt(__v, __mem, __k); } else if constexpr (sizeof(__v) == 32) { @@ -1259,13 +1234,8 @@ template else if constexpr (__have_avx && sizeof(_Tp) == 8) _mm256_maskstore_pd(reinterpret_cast(__mem), __ki, __vector_bitcast(__v)); - else if constexpr (__have_sse2) - { - _mm_maskmoveu_si128(__lo128(__vi), __lo128(__ki), - reinterpret_cast(__mem)); - _mm_maskmoveu_si128(__hi128(__vi), __hi128(__ki), - reinterpret_cast(__mem) + 16); - } + else + _Base::_S_masked_store_nocvt(__v, __mem, __k); } else __assert_unreachable<_Tp>();