From 81c2c32de9c1058c33fcf77ada31186b4ae1f1fe Mon Sep 17 00:00:00 2001
From: Matthias Kretz <kretz@kde.org>
Date: Wed, 3 Feb 2021 15:49:30 +0000
Subject: [PATCH] libstdc++: Fix mask reduction of simd_mask<double> on POWER7

POWER7 does not support __vector long long reductions, making the
generic _S_popcount implementation ill-formed. Specializing _S_popcount
for PPC allows optimization and avoids the issue.

libstdc++-v3/ChangeLog:

	* include/experimental/bits/simd.h: Add __have_power10vec
	conditional on _ARCH_PWR10.
	* include/experimental/bits/simd_builtin.h: Forward declare
	_MaskImplPpc and use it as _MaskImpl when __ALTIVEC__ is
	defined.
	(_MaskImplBuiltin::_S_some_of): Call _S_popcount from the
	_SuperImpl for optimizations and correctness.
	* include/experimental/bits/simd_ppc.h: Add _MaskImplPpc.
	(_MaskImplPpc::_S_popcount): Implement via vec_cntm for POWER10.
	Otherwise, for >=int use -vec_sums divided by a sizeof factor.
	For <int use -vec_sums(vec_sum4s(...)) to sum all mask entries.
---
 libstdc++-v3/include/experimental/bits/simd.h |  5 +++
 .../include/experimental/bits/simd_builtin.h  |  6 ++--
 .../include/experimental/bits/simd_ppc.h      | 35 ++++++++++++++++++-
 3 files changed, 43 insertions(+), 3 deletions(-)

diff --git a/libstdc++-v3/include/experimental/bits/simd.h b/libstdc++-v3/include/experimental/bits/simd.h
index 149396d6f82..becd1d6a4bb 100644
--- a/libstdc++-v3/include/experimental/bits/simd.h
+++ b/libstdc++-v3/include/experimental/bits/simd.h
@@ -477,6 +477,11 @@ constexpr inline bool __support_neon_float =
   false;
 #endif
 
+#ifdef _ARCH_PWR10
+constexpr inline bool __have_power10vec = true;
+#else
+constexpr inline bool __have_power10vec = false;
+#endif
 #ifdef __POWER9_VECTOR__
 constexpr inline bool __have_power9vec = true;
 #else
diff --git a/libstdc++-v3/include/experimental/bits/simd_builtin.h b/libstdc++-v3/include/experimental/bits/simd_builtin.h
index efca65fa6e3..7f728a10488 100644
--- a/libstdc++-v3/include/experimental/bits/simd_builtin.h
+++ b/libstdc++-v3/include/experimental/bits/simd_builtin.h
@@ -920,6 +920,7 @@ template <typename _Abi> struct _MaskImplX86;
 template <typename _Abi> struct _SimdImplNeon;
 template <typename _Abi> struct _MaskImplNeon;
 template <typename _Abi> struct _SimdImplPpc;
+template <typename _Abi> struct _MaskImplPpc;
 
 // simd_abi::_VecBuiltin {{{
 template <int _UsedBytes>
@@ -959,10 +960,11 @@ template <int _UsedBytes>
     using _CommonImpl = _CommonImplBuiltin;
 #ifdef __ALTIVEC__
     using _SimdImpl = _SimdImplPpc<_VecBuiltin<_UsedBytes>>;
+    using _MaskImpl = _MaskImplPpc<_VecBuiltin<_UsedBytes>>;
 #else
     using _SimdImpl = _SimdImplBuiltin<_VecBuiltin<_UsedBytes>>;
-#endif
     using _MaskImpl = _MaskImplBuiltin<_VecBuiltin<_UsedBytes>>;
+#endif
 #endif
 
     // }}}
@@ -2899,7 +2901,7 @@ template <typename _Abi>
       _GLIBCXX_SIMD_INTRINSIC static bool
       _S_some_of(simd_mask<_Tp, _Abi> __k)
       {
-	const int __n_true = _S_popcount(__k);
+	const int __n_true = _SuperImpl::_S_popcount(__k);
 	return __n_true > 0 && __n_true < int(_S_size<_Tp>);
       }
 
diff --git a/libstdc++-v3/include/experimental/bits/simd_ppc.h b/libstdc++-v3/include/experimental/bits/simd_ppc.h
index b92fc19ccb8..ef52d129a85 100644
--- a/libstdc++-v3/include/experimental/bits/simd_ppc.h
+++ b/libstdc++-v3/include/experimental/bits/simd_ppc.h
@@ -30,6 +30,7 @@
 #ifndef __ALTIVEC__
 #error "simd_ppc.h may only be included when AltiVec/VMX is available"
 #endif
+#include <altivec.h>
 
 _GLIBCXX_SIMD_BEGIN_NAMESPACE
 
@@ -114,10 +115,42 @@ template <typename _Abi>
     // }}}
   };
 
+// }}}
+// _MaskImplPpc {{{
+template <typename _Abi>
+  struct _MaskImplPpc : _MaskImplBuiltin<_Abi>
+  {
+    using _Base = _MaskImplBuiltin<_Abi>;
+
+    // _S_popcount {{{
+    template <typename _Tp>
+      _GLIBCXX_SIMD_INTRINSIC static int _S_popcount(simd_mask<_Tp, _Abi> __k)
+      {
+	const auto __kv = __as_vector(__k);
+	if constexpr (__have_power10vec)
+	  {
+	    return vec_cntm(__to_intrin(__kv), 1);
+	  }
+	else if constexpr (sizeof(_Tp) >= sizeof(int))
+	  {
+	    using _Intrin = __intrinsic_type16_t<int>;
+	    const int __sum = -vec_sums(__intrin_bitcast<_Intrin>(__kv), _Intrin())[3];
+	    return __sum / (sizeof(_Tp) / sizeof(int));
+	  }
+	else
+	  {
+	    const auto __summed_to_int = vec_sum4s(__to_intrin(__kv), __intrinsic_type16_t<int>());
+	    return -vec_sums(__summed_to_int, __intrinsic_type16_t<int>())[3];
+	  }
+      }
+
+    // }}}
+  };
+
 // }}}
 
 _GLIBCXX_SIMD_END_NAMESPACE
 #endif // __cplusplus >= 201703L
 #endif // _GLIBCXX_EXPERIMENTAL_SIMD_PPC_H_
 
-// vim: foldmethod=marker sw=2 noet ts=8 sts=2 tw=80
+// vim: foldmethod=marker foldmarker={{{,}}} sw=2 noet ts=8 sts=2 tw=100
-- 
2.30.2