Skip to content

[Headers][X86] Allow MMX/SSE integer min/max intrinsics to be used in constexpr #156678

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Sep 3, 2025
Merged

[Headers][X86] Allow MMX/SSE integer min/max intrinsics to be used in constexpr #156678

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5bdb914
[Headers][X86] Allow MMX/SSE integer min/max intrinsics to be used in…
markbhasawut Sep 3, 2025
2834b09
Merge branch 'main' into x86-mmx-sse-constexpr-integer-max-min-intrin…
RKSimon Sep 3, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 8 additions & 8 deletions clang/lib/Headers/emmintrin.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2316,8 +2316,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_madd_epi16(__m128i __a,
/// A 128-bit signed [8 x i16] vector.
/// \returns A 128-bit signed [8 x i16] vector containing the greater value of
/// each comparison.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epi16(__m128i __a,
__m128i __b) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_max_epi16(__m128i __a, __m128i __b) {
return (__m128i)__builtin_elementwise_max((__v8hi)__a, (__v8hi)__b);
}

Expand All @@ -2335,8 +2335,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epi16(__m128i __a,
/// A 128-bit unsigned [16 x i8] vector.
/// \returns A 128-bit unsigned [16 x i8] vector containing the greater value of
/// each comparison.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epu8(__m128i __a,
__m128i __b) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_max_epu8(__m128i __a, __m128i __b) {
return (__m128i)__builtin_elementwise_max((__v16qu)__a, (__v16qu)__b);
}

Expand All @@ -2354,8 +2354,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epu8(__m128i __a,
/// A 128-bit signed [8 x i16] vector.
/// \returns A 128-bit signed [8 x i16] vector containing the smaller value of
/// each comparison.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epi16(__m128i __a,
__m128i __b) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_min_epi16(__m128i __a, __m128i __b) {
return (__m128i)__builtin_elementwise_min((__v8hi)__a, (__v8hi)__b);
}

Expand All @@ -2373,8 +2373,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epi16(__m128i __a,
/// A 128-bit unsigned [16 x i8] vector.
/// \returns A 128-bit unsigned [16 x i8] vector containing the smaller value of
/// each comparison.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epu8(__m128i __a,
__m128i __b) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_min_epu8(__m128i __a, __m128i __b) {
return (__m128i)__builtin_elementwise_min((__v16qu)__a, (__v16qu)__b);
}

Expand Down
32 changes: 16 additions & 16 deletions clang/lib/Headers/smmintrin.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -675,8 +675,8 @@ _mm_stream_load_si128(const void *__V) {
/// \param __V2
/// A 128-bit vector of [16 x i8]
/// \returns A 128-bit vector of [16 x i8] containing the lesser values.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epi8(__m128i __V1,
__m128i __V2) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_min_epi8(__m128i __V1, __m128i __V2) {
return (__m128i)__builtin_elementwise_min((__v16qs)__V1, (__v16qs)__V2);
}

Expand All @@ -693,8 +693,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epi8(__m128i __V1,
/// \param __V2
/// A 128-bit vector of [16 x i8].
/// \returns A 128-bit vector of [16 x i8] containing the greater values.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epi8(__m128i __V1,
__m128i __V2) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_max_epi8(__m128i __V1, __m128i __V2) {
return (__m128i)__builtin_elementwise_max((__v16qs)__V1, (__v16qs)__V2);
}

Expand All @@ -711,8 +711,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epi8(__m128i __V1,
/// \param __V2
/// A 128-bit vector of [8 x u16].
/// \returns A 128-bit vector of [8 x u16] containing the lesser values.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epu16(__m128i __V1,
__m128i __V2) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_min_epu16(__m128i __V1, __m128i __V2) {
return (__m128i)__builtin_elementwise_min((__v8hu)__V1, (__v8hu)__V2);
}

Expand All @@ -729,8 +729,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epu16(__m128i __V1,
/// \param __V2
/// A 128-bit vector of [8 x u16].
/// \returns A 128-bit vector of [8 x u16] containing the greater values.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epu16(__m128i __V1,
__m128i __V2) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_max_epu16(__m128i __V1, __m128i __V2) {
return (__m128i)__builtin_elementwise_max((__v8hu)__V1, (__v8hu)__V2);
}

Expand All @@ -747,8 +747,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epu16(__m128i __V1,
/// \param __V2
/// A 128-bit vector of [4 x i32].
/// \returns A 128-bit vector of [4 x i32] containing the lesser values.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epi32(__m128i __V1,
__m128i __V2) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_min_epi32(__m128i __V1, __m128i __V2) {
return (__m128i)__builtin_elementwise_min((__v4si)__V1, (__v4si)__V2);
}

Expand All @@ -765,8 +765,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epi32(__m128i __V1,
/// \param __V2
/// A 128-bit vector of [4 x i32].
/// \returns A 128-bit vector of [4 x i32] containing the greater values.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epi32(__m128i __V1,
__m128i __V2) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_max_epi32(__m128i __V1, __m128i __V2) {
return (__m128i)__builtin_elementwise_max((__v4si)__V1, (__v4si)__V2);
}

Expand All @@ -783,8 +783,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epi32(__m128i __V1,
/// \param __V2
/// A 128-bit vector of [4 x u32].
/// \returns A 128-bit vector of [4 x u32] containing the lesser values.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epu32(__m128i __V1,
__m128i __V2) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_min_epu32(__m128i __V1, __m128i __V2) {
return (__m128i)__builtin_elementwise_min((__v4su)__V1, (__v4su)__V2);
}

Expand All @@ -801,8 +801,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epu32(__m128i __V1,
/// \param __V2
/// A 128-bit vector of [4 x u32].
/// \returns A 128-bit vector of [4 x u32] containing the greater values.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epu32(__m128i __V1,
__m128i __V2) {
static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
_mm_max_epu32(__m128i __V1, __m128i __V2) {
return (__m128i)__builtin_elementwise_max((__v4su)__V1, (__v4su)__V2);
}

Expand Down
10 changes: 4 additions & 6 deletions clang/lib/Headers/xmmintrin.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2353,9 +2353,8 @@ void _mm_sfence(void);
/// \param __b
/// A 64-bit integer vector containing one of the source operands.
/// \returns A 64-bit integer vector containing the comparison results.
static __inline__ __m64 __DEFAULT_FN_ATTRS_SSE2
_mm_max_pi16(__m64 __a, __m64 __b)
{
static __inline__ __m64 __DEFAULT_FN_ATTRS_SSE2_CONSTEXPR
_mm_max_pi16(__m64 __a, __m64 __b) {
return (__m64)__builtin_elementwise_max((__v4hi)__a, (__v4hi)__b);
}

Expand Down Expand Up @@ -2391,9 +2390,8 @@ _mm_max_pu8(__m64 __a, __m64 __b)
/// \param __b
/// A 64-bit integer vector containing one of the source operands.
/// \returns A 64-bit integer vector containing the comparison results.
static __inline__ __m64 __DEFAULT_FN_ATTRS_SSE2
_mm_min_pi16(__m64 __a, __m64 __b)
{
static __inline__ __m64 __DEFAULT_FN_ATTRS_SSE2_CONSTEXPR
_mm_min_pi16(__m64 __a, __m64 __b) {
return (__m64)__builtin_elementwise_min((__v4hi)__a, (__v4hi)__b);
}

Expand Down
5 changes: 5 additions & 0 deletions clang/test/CodeGen/X86/builtin_test_helpers.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -230,6 +230,11 @@ constexpr bool match_v16si(__m512i _v, int a, int b, int c, int d, int e, int f,
return v[0] == a && v[1] == b && v[2] == c && v[3] == d && v[4] == e && v[5] == f && v[6] == g && v[7] == h && v[8] == i && v[9] == j && v[10] == k && v[11] == l && v[12] == m && v[13] == n && v[14] == o && v[15] == p;
}

constexpr bool match_v16su(__m512i _v, unsigned int a, unsigned int b, unsigned int c, unsigned int d, unsigned int e, unsigned int f, unsigned int g, unsigned int h, unsigned int i, unsigned int j, unsigned int k, unsigned int l, unsigned int m, unsigned int n, unsigned int o, unsigned int p) {
__v16su v = (__v16su)_v;
return v[0] == a && v[1] == b && v[2] == c && v[3] == d && v[4] == e && v[5] == f && v[6] == g && v[7] == h && v[8] == i && v[9] == j && v[10] == k && v[11] == l && v[12] == m && v[13] == n && v[14] == o && v[15] == p;
}

constexpr bool match_v32hi(__m512i _v, short __e00, short __e01, short __e02, short __e03, short __e04, short __e05, short __e06, short __e07,
short __e08, short __e09, short __e10, short __e11, short __e12, short __e13, short __e14, short __e15,
short __e16, short __e17, short __e18, short __e19, short __e20, short __e21, short __e22, short __e23,
Expand Down
4 changes: 4 additions & 0 deletions clang/test/CodeGen/X86/mmx-builtins.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -370,6 +370,8 @@ __m64 test_mm_max_pi16(__m64 a, __m64 b) {
return _mm_max_pi16(a, b);
}

TEST_CONSTEXPR(match_v4hi(_mm_max_pi16((__m64)(__v4hi){+1, -2, +3, -4}, (__m64)(__v4hi){-1, 2, -3, 4}), 1, 2, 3, 4));

__m64 test_mm_max_pu8(__m64 a, __m64 b) {
// CHECK-LABEL: test_mm_max_pu8
// CHECK: call <8 x i8> @llvm.umax.v8i8(
Expand All @@ -382,6 +384,8 @@ __m64 test_mm_min_pi16(__m64 a, __m64 b) {
return _mm_min_pi16(a, b);
}

TEST_CONSTEXPR(match_v4hi(_mm_min_pi16((__m64)(__v4hi){+1, -2, +3, -4}, (__m64)(__v4hi){-1, 2, -3, 4}), -1, -2, -3, -4));

__m64 test_mm_min_pu8(__m64 a, __m64 b) {
// CHECK-LABEL: test_mm_min_pu8
// CHECK: call <8 x i8> @llvm.umin.v8i8(
Expand Down
8 changes: 8 additions & 0 deletions clang/test/CodeGen/X86/sse2-builtins.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -855,12 +855,16 @@ __m128i test_mm_max_epi16(__m128i A, __m128i B) {
return _mm_max_epi16(A, B);
}

TEST_CONSTEXPR(match_v8hi(_mm_max_epi16((__m128i)(__v8hi){1, 2, 3, 4, 5, 6, 7, 8}, (__m128i)(__v8hi){1, 2, 3, 5, 8, 12, 20, 32}), 1, 2, 3, 5, 8, 12, 20, 32));

__m128i test_mm_max_epu8(__m128i A, __m128i B) {
// CHECK-LABEL: test_mm_max_epu8
// CHECK: call <16 x i8> @llvm.umax.v16i8(<16 x i8> %{{.*}}, <16 x i8> %{{.*}})
return _mm_max_epu8(A, B);
}

TEST_CONSTEXPR(match_v16qu(_mm_max_epu8((__m128i)(__v16qu){9, 10, 11, 12, 13, 14, 15, 16, 1, 2, 3, 4, 5, 6, 7, 8}, (__m128i)(__v16qu){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}), 9, 10, 11, 12, 13, 14, 15, 16, 9, 10, 11, 12, 13, 14, 15, 16));

__m128d test_mm_max_pd(__m128d A, __m128d B) {
// CHECK-LABEL: test_mm_max_pd
// CHECK: call {{.*}}<2 x double> @llvm.x86.sse2.max.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}})
Expand All @@ -885,12 +889,16 @@ __m128i test_mm_min_epi16(__m128i A, __m128i B) {
return _mm_min_epi16(A, B);
}

TEST_CONSTEXPR(match_v8hi(_mm_min_epi16((__m128i)(__v8hi){1, 2, 3, 4, 5, 6, 7, 8}, (__m128i)(__v8hi){1, 2, 3, 5, 8, 12, 20, 32}), 1, 2, 3, 4, 5, 6, 7, 8));

__m128i test_mm_min_epu8(__m128i A, __m128i B) {
// CHECK-LABEL: test_mm_min_epu8
// CHECK: call <16 x i8> @llvm.umin.v16i8(<16 x i8> %{{.*}}, <16 x i8> %{{.*}})
return _mm_min_epu8(A, B);
}

TEST_CONSTEXPR(match_v16qu(_mm_min_epu8((__m128i)(__v16qu){9, 10, 11, 12, 13, 14, 15, 16, 1, 2, 3, 4, 5, 6, 7, 8}, (__m128i)(__v16qu){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}), 1, 2, 3, 4, 5, 6, 7, 8, 1, 2, 3, 4, 5, 6, 7, 8));

__m128d test_mm_min_pd(__m128d A, __m128d B) {
// CHECK-LABEL: test_mm_min_pd
// CHECK: call {{.*}}<2 x double> @llvm.x86.sse2.min.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}})
Expand Down
16 changes: 16 additions & 0 deletions clang/test/CodeGen/X86/sse41-builtins.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -291,48 +291,64 @@ __m128i test_mm_max_epi8(__m128i x, __m128i y) {
return _mm_max_epi8(x, y);
}

TEST_CONSTEXPR(match_v16qi(_mm_max_epi8((__m128i)(__v16qs){-1, +2, -3, +4, -5, +6, -7, +8, -9, +10, -11, +12, -13, +14, -15, +16}, (__m128i)(__v16qs){+1, -2, +3, -4, +5, -6, +7, -8, +9, -10, +11, -12, +13, -14, +15, -16}), +1, +2, +3, +4, +5, +6, +7, +8, +9, +10, +11, +12, +13, +14, +15, +16));

__m128i test_mm_max_epi32(__m128i x, __m128i y) {
// CHECK-LABEL: test_mm_max_epi32
// CHECK: call <4 x i32> @llvm.smax.v4i32(<4 x i32> %{{.*}}, <4 x i32> %{{.*}})
return _mm_max_epi32(x, y);
}

TEST_CONSTEXPR(match_v4si(_mm_max_epi32((__m128i)(__v4si){-1, +2, -3, +4}, (__m128i)(__v4si){+1, -2, +3, -4}), +1, +2, +3, +4 ));

__m128i test_mm_max_epu16(__m128i x, __m128i y) {
// CHECK-LABEL: test_mm_max_epu16
// CHECK: call <8 x i16> @llvm.umax.v8i16(<8 x i16> %{{.*}}, <8 x i16> %{{.*}})
return _mm_max_epu16(x, y);
}

TEST_CONSTEXPR(match_v8hu(_mm_max_epu16((__m128i)(__v8hu){1, 3, 5, 7, 9, 11, 13, 15}, (__m128i)(__v8hu){3, 4, 5, 6, 7, 8, 9, 10}), 3, 4, 5, 7, 9, 11, 13, 15));

__m128i test_mm_max_epu32(__m128i x, __m128i y) {
// CHECK-LABEL: test_mm_max_epu32
// CHECK: call <4 x i32> @llvm.umax.v4i32(<4 x i32> %{{.*}}, <4 x i32> %{{.*}})
return _mm_max_epu32(x, y);
}

TEST_CONSTEXPR(match_v4su(_mm_max_epu32((__m128i)(__v4su){1, 3, 5, 7}, (__m128i)(__v4su){3, 4, 5, 6}), 3, 4, 5, 7));

__m128i test_mm_min_epi8(__m128i x, __m128i y) {
// CHECK-LABEL: test_mm_min_epi8
// CHECK: call <16 x i8> @llvm.smin.v16i8(<16 x i8> %{{.*}}, <16 x i8> %{{.*}})
return _mm_min_epi8(x, y);
}

TEST_CONSTEXPR(match_v16qi(_mm_min_epi8((__m128i)(__v16qs){-1, +2, -3, +4, -5, +6, -7, +8, -9, +10, -11, +12, -13, +14, -15, +16}, (__m128i)(__v16qs){+1, -2, +3, -4, +5, -6, +7, -8, +9, -10, +11, -12, +13, -14, +15, -16}), -1, -2, -3, -4, -5, -6, -7, -8, -9, -10, -11, -12, -13, -14, -15, -16));

__m128i test_mm_min_epi32(__m128i x, __m128i y) {
// CHECK-LABEL: test_mm_min_epi32
// CHECK: call <4 x i32> @llvm.smin.v4i32(<4 x i32> %{{.*}}, <4 x i32> %{{.*}})
return _mm_min_epi32(x, y);
}

TEST_CONSTEXPR(match_v4si(_mm_min_epi32((__m128i)(__v4si){-1, +2, -3, +4}, (__m128i)(__v4si){+1, -2, +3, -4}), -1, -2, -3, -4 ));

__m128i test_mm_min_epu16(__m128i x, __m128i y) {
// CHECK-LABEL: test_mm_min_epu16
// CHECK: call <8 x i16> @llvm.umin.v8i16(<8 x i16> %{{.*}}, <8 x i16> %{{.*}})
return _mm_min_epu16(x, y);
}

TEST_CONSTEXPR(match_v8hu(_mm_min_epu16((__m128i)(__v8hu){1, 3, 5, 7, 9, 11, 13, 15}, (__m128i)(__v8hu){3, 4, 5, 6, 7, 8, 9, 10}), 1, 3, 5, 6, 7, 8, 9, 10));

__m128i test_mm_min_epu32(__m128i x, __m128i y) {
// CHECK-LABEL: test_mm_min_epu32
// CHECK: call <4 x i32> @llvm.umin.v4i32(<4 x i32> %{{.*}}, <4 x i32> %{{.*}})
return _mm_min_epu32(x, y);
}

TEST_CONSTEXPR(match_v4su(_mm_min_epu32((__m128i)(__v4su){1, 3, 5, 7}, (__m128i)(__v4su){3, 4, 5, 6}), 1, 3, 5, 6));

__m128i test_mm_minpos_epu16(__m128i x) {
// CHECK-LABEL: test_mm_minpos_epu16
// CHECK: call <8 x i16> @llvm.x86.sse41.phminposuw(<8 x i16> %{{.*}})
Expand Down