| Message ID | 20210701061648.9447-31-hongtao.liu@intel.com |
|---|---|
| State | New |
| Headers | show |
| Series | Support all AVX512FP16 intrinsics. | expand |
I'm going to check in 10 patches.
[PATCH 30/62] AVX512FP16: Add vcvtsh2si/vcvtsh2usi/vcvtsi2sh/vcvtusi2sh.
[PATCH 31/62] AVX512FP16: Add testcase for
vcvtsh2si/vcvtsh2usi/vcvtsi2sh/vcvtusi2sh.
[PATCH 32/62] AVX512FP16: Add
vcvttph2w/vcvttph2uw/vcvttph2dq/vcvttph2qq/vcvttph2udq/vcvttph2uqq
[PATCH 33/62] AVX512FP16: Add testcase for
vcvttph2w/vcvttph2uw/vcvttph2dq/vcvttph2udq/vcvttph2qq/vcvttph2uqq
[PATCH 34/62] AVX512FP16: Add vcvttsh2si/vcvttsh2usi
[PATCH 35/62] AVX512FP16: Add vcvtph2pd/vcvtph2psx/vcvtpd2ph/vcvtps2phx
[PATCH 36/62] AVX512FP16: Add testcase for
vcvtph2pd/vcvtph2psx/vcvtpd2ph/vcvtps2phx
[PATCH 37/62] AVX512FP16: Add vcvtsh2ss/vcvtsh2sd/vcvtss2sh/vcvtsd2sh.
[PATCH 38/62] AVX512FP16: Add testcase for
vcvtsh2sd/vcvtsh2ss/vcvtsd2sh/vcvtss2sh
[PATCH 39/62] AVX512FP16: Add intrinsics for casting between vector
float16 and vector float32/float64/integer.
Bootstrapped and regtested on x86_64-pc-linux-gnu{-m32,}.
Newly added runtime testcase passed on sde{-m32,}.
On Thu, Jul 1, 2021 at 2:17 PM liuhongt <hongtao.liu@intel.com> wrote:
>
> gcc/ChangeLog:
>
> * config/i386/avx512fp16intrin.h (_mm_cvtsh_i32): New intrinsic.
> (_mm_cvtsh_u32): Likewise.
> (_mm_cvt_roundsh_i32): Likewise.
> (_mm_cvt_roundsh_u32): Likewise.
> (_mm_cvtsh_i64): Likewise.
> (_mm_cvtsh_u64): Likewise.
> (_mm_cvt_roundsh_i64): Likewise.
> (_mm_cvt_roundsh_u64): Likewise.
> (_mm_cvti32_sh): Likewise.
> (_mm_cvtu32_sh): Likewise.
> (_mm_cvt_roundi32_sh): Likewise.
> (_mm_cvt_roundu32_sh): Likewise.
> (_mm_cvti64_sh): Likewise.
> (_mm_cvtu64_sh): Likewise.
> (_mm_cvt_roundi64_sh): Likewise.
> (_mm_cvt_roundu64_sh): Likewise.
> * config/i386/i386-builtin-types.def: Add corresponding builtin types.
> * config/i386/i386-builtin.def: Add corresponding new builtins.
> * config/i386/i386-expand.c (ix86_expand_round_builtin):
> Handle new builtin types.
> * config/i386/sse.md
> (avx512fp16_vcvtsh2<sseintconvertsignprefix>si<rex64namesuffix><round_name>):
> New define_insn.
> (avx512fp16_vcvtsh2<sseintconvertsignprefix>si<rex64namesuffix>_2): Likewise.
> (avx512fp16_vcvt<floatsuffix>si2sh<rex64namesuffix><round_name>): Likewise.
>
> gcc/testsuite/ChangeLog:
>
> * gcc.target/i386/avx-1.c: Add test for new builtins.
> * gcc.target/i386/sse-13.c: Ditto.
> * gcc.target/i386/sse-23.c: Ditto.
> * gcc.target/i386/sse-14.c: Add test for new intrinsics.
> * gcc.target/i386/sse-22.c: Ditto.
> ---
> gcc/config/i386/avx512fp16intrin.h | 158 +++++++++++++++++++++++++
> gcc/config/i386/i386-builtin-types.def | 8 ++
> gcc/config/i386/i386-builtin.def | 8 ++
> gcc/config/i386/i386-expand.c | 8 ++
> gcc/config/i386/sse.md | 46 +++++++
> gcc/testsuite/gcc.target/i386/avx-1.c | 8 ++
> gcc/testsuite/gcc.target/i386/sse-13.c | 8 ++
> gcc/testsuite/gcc.target/i386/sse-14.c | 10 ++
> gcc/testsuite/gcc.target/i386/sse-22.c | 10 ++
> gcc/testsuite/gcc.target/i386/sse-23.c | 8 ++
> 10 files changed, 272 insertions(+)
>
> diff --git a/gcc/config/i386/avx512fp16intrin.h b/gcc/config/i386/avx512fp16intrin.h
> index bd801942365..7524a8d6a5b 100644
> --- a/gcc/config/i386/avx512fp16intrin.h
> +++ b/gcc/config/i386/avx512fp16intrin.h
> @@ -3529,6 +3529,164 @@ _mm512_maskz_cvt_roundepu16_ph (__mmask32 __A, __m512i __B, int __C)
>
> #endif /* __OPTIMIZE__ */
>
> +/* Intrinsics vcvtsh2si, vcvtsh2us. */
> +extern __inline int
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvtsh_i32 (__m128h __A)
> +{
> + return (int) __builtin_ia32_vcvtsh2si32_round (__A, _MM_FROUND_CUR_DIRECTION);
> +}
> +
> +extern __inline unsigned
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvtsh_u32 (__m128h __A)
> +{
> + return (int) __builtin_ia32_vcvtsh2usi32_round (__A,
> + _MM_FROUND_CUR_DIRECTION);
> +}
> +
> +#ifdef __OPTIMIZE__
> +extern __inline int
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvt_roundsh_i32 (__m128h __A, const int __R)
> +{
> + return (int) __builtin_ia32_vcvtsh2si32_round (__A, __R);
> +}
> +
> +extern __inline unsigned
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvt_roundsh_u32 (__m128h __A, const int __R)
> +{
> + return (int) __builtin_ia32_vcvtsh2usi32_round (__A, __R);
> +}
> +
> +#else
> +#define _mm_cvt_roundsh_i32(A, B) \
> + ((int)__builtin_ia32_vcvtsh2si32_round ((A), (B)))
> +#define _mm_cvt_roundsh_u32(A, B) \
> + ((int)__builtin_ia32_vcvtsh2usi32_round ((A), (B)))
> +
> +#endif /* __OPTIMIZE__ */
> +
> +#ifdef __x86_64__
> +extern __inline long long
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvtsh_i64 (__m128h __A)
> +{
> + return (long long)
> + __builtin_ia32_vcvtsh2si64_round (__A, _MM_FROUND_CUR_DIRECTION);
> +}
> +
> +extern __inline unsigned long long
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvtsh_u64 (__m128h __A)
> +{
> + return (long long)
> + __builtin_ia32_vcvtsh2usi64_round (__A, _MM_FROUND_CUR_DIRECTION);
> +}
> +
> +#ifdef __OPTIMIZE__
> +extern __inline long long
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvt_roundsh_i64 (__m128h __A, const int __R)
> +{
> + return (long long) __builtin_ia32_vcvtsh2si64_round (__A, __R);
> +}
> +
> +extern __inline unsigned long long
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvt_roundsh_u64 (__m128h __A, const int __R)
> +{
> + return (long long) __builtin_ia32_vcvtsh2usi64_round (__A, __R);
> +}
> +
> +#else
> +#define _mm_cvt_roundsh_i64(A, B) \
> + ((long long)__builtin_ia32_vcvtsh2si64_round ((A), (B)))
> +#define _mm_cvt_roundsh_u64(A, B) \
> + ((long long)__builtin_ia32_vcvtsh2usi64_round ((A), (B)))
> +
> +#endif /* __OPTIMIZE__ */
> +#endif /* __x86_64__ */
> +
> +/* Intrinsics vcvtsi2sh, vcvtusi2sh. */
> +extern __inline __m128h
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvti32_sh (__m128h __A, int __B)
> +{
> + return __builtin_ia32_vcvtsi2sh32_round (__A, __B, _MM_FROUND_CUR_DIRECTION);
> +}
> +
> +extern __inline __m128h
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvtu32_sh (__m128h __A, unsigned int __B)
> +{
> + return __builtin_ia32_vcvtusi2sh32_round (__A, __B, _MM_FROUND_CUR_DIRECTION);
> +}
> +
> +#ifdef __OPTIMIZE__
> +extern __inline __m128h
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvt_roundi32_sh (__m128h __A, int __B, const int __R)
> +{
> + return __builtin_ia32_vcvtsi2sh32_round (__A, __B, __R);
> +}
> +
> +extern __inline __m128h
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvt_roundu32_sh (__m128h __A, unsigned int __B, const int __R)
> +{
> + return __builtin_ia32_vcvtusi2sh32_round (__A, __B, __R);
> +}
> +
> +#else
> +#define _mm_cvt_roundi32_sh(A, B, C) \
> + (__builtin_ia32_vcvtsi2sh32_round ((A), (B), (C)))
> +#define _mm_cvt_roundu32_sh(A, B, C) \
> + (__builtin_ia32_vcvtusi2sh32_round ((A), (B), (C)))
> +
> +#endif /* __OPTIMIZE__ */
> +
> +#ifdef __x86_64__
> +extern __inline __m128h
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvti64_sh (__m128h __A, long long __B)
> +{
> + return __builtin_ia32_vcvtsi2sh64_round (__A, __B, _MM_FROUND_CUR_DIRECTION);
> +}
> +
> +extern __inline __m128h
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvtu64_sh (__m128h __A, unsigned long long __B)
> +{
> + return __builtin_ia32_vcvtusi2sh64_round (__A, __B, _MM_FROUND_CUR_DIRECTION);
> +}
> +
> +#ifdef __OPTIMIZE__
> +extern __inline __m128h
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvt_roundi64_sh (__m128h __A, long long __B, const int __R)
> +{
> + return __builtin_ia32_vcvtsi2sh64_round (__A, __B, __R);
> +}
> +
> +extern __inline __m128h
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_cvt_roundu64_sh (__m128h __A, unsigned long long __B, const int __R)
> +{
> + return __builtin_ia32_vcvtusi2sh64_round (__A, __B, __R);
> +}
> +
> +#else
> +#define _mm_cvt_roundi64_sh(A, B, C) \
> + (__builtin_ia32_vcvtsi2sh64_round ((A), (B), (C)))
> +#define _mm_cvt_roundu64_sh(A, B, C) \
> + (__builtin_ia32_vcvtusi2sh64_round ((A), (B), (C)))
> +
> +#endif /* __OPTIMIZE__ */
> +#endif /* __x86_64__ */
> +
> +
> #ifdef __DISABLE_AVX512FP16__
> #undef __DISABLE_AVX512FP16__
> #pragma GCC pop_options
> diff --git a/gcc/config/i386/i386-builtin-types.def b/gcc/config/i386/i386-builtin-types.def
> index 57b9ea786e1..74bda59a65e 100644
> --- a/gcc/config/i386/i386-builtin-types.def
> +++ b/gcc/config/i386/i386-builtin-types.def
> @@ -1308,9 +1308,17 @@ DEF_FUNCTION_TYPE (V8HF, V8HI)
> DEF_FUNCTION_TYPE (QI, V8HF, INT, UQI)
> DEF_FUNCTION_TYPE (HI, V16HF, INT, UHI)
> DEF_FUNCTION_TYPE (SI, V32HF, INT, USI)
> +DEF_FUNCTION_TYPE (INT, V8HF, INT)
> +DEF_FUNCTION_TYPE (INT64, V8HF, INT)
> +DEF_FUNCTION_TYPE (UINT, V8HF, INT)
> +DEF_FUNCTION_TYPE (UINT64, V8HF, INT)
> DEF_FUNCTION_TYPE (V8HF, V8HF, V8HF)
> DEF_FUNCTION_TYPE (VOID, PCFLOAT16, V8HF, UQI)
> DEF_FUNCTION_TYPE (V8HF, PCFLOAT16, V8HF, UQI)
> +DEF_FUNCTION_TYPE (V8HF, V8HF, INT, INT)
> +DEF_FUNCTION_TYPE (V8HF, V8HF, INT64, INT)
> +DEF_FUNCTION_TYPE (V8HF, V8HF, UINT, INT)
> +DEF_FUNCTION_TYPE (V8HF, V8HF, UINT64, INT)
> DEF_FUNCTION_TYPE (V2DI, V8HF, V2DI, UQI)
> DEF_FUNCTION_TYPE (V4DI, V8HF, V4DI, UQI)
> DEF_FUNCTION_TYPE (V4SI, V8HF, V4SI, UQI)
> diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def
> index 44c55876e48..3602b40d6d5 100644
> --- a/gcc/config/i386/i386-builtin.def
> +++ b/gcc/config/i386/i386-builtin.def
> @@ -3094,6 +3094,14 @@ BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtdq2ph_v16si_mask_
> BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtudq2ph_v16si_mask_round, "__builtin_ia32_vcvtudq2ph_v16si_mask_round", IX86_BUILTIN_VCVTUDQ2PH_V16SI_MASK_ROUND, UNKNOWN, (int) V16HF_FTYPE_V16SI_V16HF_UHI_INT)
> BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtqq2ph_v8di_mask_round, "__builtin_ia32_vcvtqq2ph_v8di_mask_round", IX86_BUILTIN_VCVTQQ2PH_V8DI_MASK_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8DI_V8HF_UQI_INT)
> BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtuqq2ph_v8di_mask_round, "__builtin_ia32_vcvtuqq2ph_v8di_mask_round", IX86_BUILTIN_VCVTUQQ2PH_V8DI_MASK_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8DI_V8HF_UQI_INT)
> +BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsh2si_round, "__builtin_ia32_vcvtsh2si32_round", IX86_BUILTIN_VCVTSH2SI32_ROUND, UNKNOWN, (int) INT_FTYPE_V8HF_INT)
> +BDESC (OPTION_MASK_ISA_64BIT, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsh2siq_round, "__builtin_ia32_vcvtsh2si64_round", IX86_BUILTIN_VCVTSH2SI64_ROUND, UNKNOWN, (int) INT64_FTYPE_V8HF_INT)
> +BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsh2usi_round, "__builtin_ia32_vcvtsh2usi32_round", IX86_BUILTIN_VCVTSH2USI32_ROUND, UNKNOWN, (int) UINT_FTYPE_V8HF_INT)
> +BDESC (OPTION_MASK_ISA_64BIT, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsh2usiq_round, "__builtin_ia32_vcvtsh2usi64_round", IX86_BUILTIN_VCVTSH2USI64_ROUND, UNKNOWN, (int) UINT64_FTYPE_V8HF_INT)
> +BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsi2sh_round, "__builtin_ia32_vcvtsi2sh32_round", IX86_BUILTIN_VCVTSI2SH32_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8HF_INT_INT)
> +BDESC (OPTION_MASK_ISA_64BIT, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsi2shq_round, "__builtin_ia32_vcvtsi2sh64_round", IX86_BUILTIN_VCVTSI2SH64_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8HF_INT64_INT)
> +BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtusi2sh_round, "__builtin_ia32_vcvtusi2sh32_round", IX86_BUILTIN_VCVTUSI2SH32_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8HF_UINT_INT)
> +BDESC (OPTION_MASK_ISA_64BIT, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtusi2shq_round, "__builtin_ia32_vcvtusi2sh64_round", IX86_BUILTIN_VCVTUSI2SH64_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8HF_UINT64_INT)
>
> BDESC_END (ROUND_ARGS, MULTI_ARG)
>
> diff --git a/gcc/config/i386/i386-expand.c b/gcc/config/i386/i386-expand.c
> index 7d9e1bd6a2d..b83c6d9a92b 100644
> --- a/gcc/config/i386/i386-expand.c
> +++ b/gcc/config/i386/i386-expand.c
> @@ -10489,16 +10489,24 @@ ix86_expand_round_builtin (const struct builtin_description *d,
> {
> case UINT64_FTYPE_V2DF_INT:
> case UINT64_FTYPE_V4SF_INT:
> + case UINT64_FTYPE_V8HF_INT:
> case UINT_FTYPE_V2DF_INT:
> case UINT_FTYPE_V4SF_INT:
> + case UINT_FTYPE_V8HF_INT:
> case INT64_FTYPE_V2DF_INT:
> case INT64_FTYPE_V4SF_INT:
> + case INT64_FTYPE_V8HF_INT:
> case INT_FTYPE_V2DF_INT:
> case INT_FTYPE_V4SF_INT:
> + case INT_FTYPE_V8HF_INT:
> nargs = 2;
> break;
> case V32HF_FTYPE_V32HF_V32HF_INT:
> case V8HF_FTYPE_V8HF_V8HF_INT:
> + case V8HF_FTYPE_V8HF_INT_INT:
> + case V8HF_FTYPE_V8HF_UINT_INT:
> + case V8HF_FTYPE_V8HF_INT64_INT:
> + case V8HF_FTYPE_V8HF_UINT64_INT:
> case V4SF_FTYPE_V4SF_UINT_INT:
> case V4SF_FTYPE_V4SF_UINT64_INT:
> case V2DF_FTYPE_V2DF_UINT64_INT:
> diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
> index 8b23048a232..b312d26b806 100644
> --- a/gcc/config/i386/sse.md
> +++ b/gcc/config/i386/sse.md
> @@ -5589,6 +5589,52 @@ (define_insn "*avx512fp16_vcvt<floatsuffix>qq2ph_v2di_mask_1"
> (set_attr "prefix" "evex")
> (set_attr "mode" "TI")])
>
> +(define_insn "avx512fp16_vcvtsh2<sseintconvertsignprefix>si<rex64namesuffix><round_name>"
> + [(set (match_operand:SWI48 0 "register_operand" "=r,r")
> + (unspec:SWI48
> + [(vec_select:HF
> + (match_operand:V8HF 1 "<round_nimm_scalar_predicate>" "v,<round_constraint2>")
> + (parallel [(const_int 0)]))]
> + UNSPEC_US_FIX_NOTRUNC))]
> + "TARGET_AVX512FP16"
> + "%vcvtsh2<sseintconvertsignprefix>si\t{<round_op2>%1, %0|%0, %k1<round_op2>}"
> + [(set_attr "type" "sseicvt")
> + (set_attr "athlon_decode" "double,vector")
> + (set_attr "bdver1_decode" "double,double")
> + (set_attr "prefix_rep" "1")
> + (set_attr "prefix" "evex")
> + (set_attr "mode" "<MODE>")])
> +
> +(define_insn "avx512fp16_vcvtsh2<sseintconvertsignprefix>si<rex64namesuffix>_2"
> + [(set (match_operand:SWI48 0 "register_operand" "=r,r")
> + (unspec:SWI48 [(match_operand:HF 1 "nonimmediate_operand" "v,m")]
> + UNSPEC_US_FIX_NOTRUNC))]
> + "TARGET_AVX512FP16"
> + "%vcvtsh2<sseintconvertsignprefix>si\t{%1, %0|%0, %k1}"
> + [(set_attr "type" "sseicvt")
> + (set_attr "athlon_decode" "double,vector")
> + (set_attr "bdver1_decode" "double,double")
> + (set_attr "prefix_rep" "1")
> + (set_attr "prefix" "evex")
> + (set_attr "mode" "<MODE>")])
> +
> +(define_insn "avx512fp16_vcvt<floatsuffix>si2sh<rex64namesuffix><round_name>"
> + [(set (match_operand:V8HF 0 "register_operand" "=v")
> + (vec_merge:V8HF
> + (vec_duplicate:V8HF
> + (any_float:HF (match_operand:SWI48 2 "<round_nimm_scalar_predicate>" "<round_constraint3>")))
> + (match_operand:V8HF 1 "register_operand" "v")
> + (const_int 1)))]
> + "TARGET_AVX512FP16"
> + "vcvt<floatsuffix>si2sh\t{%2, <round_op3>%1, %0|%0, %1<round_op3>, %2}"
> + [(set_attr "type" "sseicvt")
> + (set_attr "athlon_decode" "*")
> + (set_attr "amdfam10_decode" "*")
> + (set_attr "bdver1_decode" "*")
> + (set_attr "btver2_decode" "double")
> + (set_attr "znver1_decode" "double")
> + (set_attr "prefix" "evex")
> + (set_attr "mode" "HF")])
>
> ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
> ;;
> diff --git a/gcc/testsuite/gcc.target/i386/avx-1.c b/gcc/testsuite/gcc.target/i386/avx-1.c
> index b569cc0bdd9..0aae949097a 100644
> --- a/gcc/testsuite/gcc.target/i386/avx-1.c
> +++ b/gcc/testsuite/gcc.target/i386/avx-1.c
> @@ -731,6 +731,14 @@
> #define __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, D) __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, 8)
> #define __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, 8)
> #define __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, 8)
> +#define __builtin_ia32_vcvtsh2si32_round(A, B) __builtin_ia32_vcvtsh2si32_round(A, 8)
> +#define __builtin_ia32_vcvtsh2si64_round(A, B) __builtin_ia32_vcvtsh2si64_round(A, 8)
> +#define __builtin_ia32_vcvtsh2usi32_round(A, B) __builtin_ia32_vcvtsh2usi32_round(A, 8)
> +#define __builtin_ia32_vcvtsh2usi64_round(A, B) __builtin_ia32_vcvtsh2usi64_round(A, 8)
> +#define __builtin_ia32_vcvtsi2sh32_round(A, B, C) __builtin_ia32_vcvtsi2sh32_round(A, B, 8)
> +#define __builtin_ia32_vcvtsi2sh64_round(A, B, C) __builtin_ia32_vcvtsi2sh64_round(A, B, 8)
> +#define __builtin_ia32_vcvtusi2sh32_round(A, B, C) __builtin_ia32_vcvtusi2sh32_round(A, B, 8)
> +#define __builtin_ia32_vcvtusi2sh64_round(A, B, C) __builtin_ia32_vcvtusi2sh64_round(A, B, 8)
>
> /* avx512fp16vlintrin.h */
> #define __builtin_ia32_vcmpph_v8hf_mask(A, B, C, D) __builtin_ia32_vcmpph_v8hf_mask(A, B, 1, D)
> diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c b/gcc/testsuite/gcc.target/i386/sse-13.c
> index 07e59118438..997fb733132 100644
> --- a/gcc/testsuite/gcc.target/i386/sse-13.c
> +++ b/gcc/testsuite/gcc.target/i386/sse-13.c
> @@ -748,6 +748,14 @@
> #define __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, D) __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, 8)
> #define __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, 8)
> #define __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, 8)
> +#define __builtin_ia32_vcvtsh2si32_round(A, B) __builtin_ia32_vcvtsh2si32_round(A, 8)
> +#define __builtin_ia32_vcvtsh2si64_round(A, B) __builtin_ia32_vcvtsh2si64_round(A, 8)
> +#define __builtin_ia32_vcvtsh2usi32_round(A, B) __builtin_ia32_vcvtsh2usi32_round(A, 8)
> +#define __builtin_ia32_vcvtsh2usi64_round(A, B) __builtin_ia32_vcvtsh2usi64_round(A, 8)
> +#define __builtin_ia32_vcvtsi2sh32_round(A, B, C) __builtin_ia32_vcvtsi2sh32_round(A, B, 8)
> +#define __builtin_ia32_vcvtsi2sh64_round(A, B, C) __builtin_ia32_vcvtsi2sh64_round(A, B, 8)
> +#define __builtin_ia32_vcvtusi2sh32_round(A, B, C) __builtin_ia32_vcvtusi2sh32_round(A, B, 8)
> +#define __builtin_ia32_vcvtusi2sh64_round(A, B, C) __builtin_ia32_vcvtusi2sh64_round(A, B, 8)
>
> /* avx512fp16vlintrin.h */
> #define __builtin_ia32_vcmpph_v8hf_mask(A, B, C, D) __builtin_ia32_vcmpph_v8hf_mask(A, B, 1, D)
> diff --git a/gcc/testsuite/gcc.target/i386/sse-14.c b/gcc/testsuite/gcc.target/i386/sse-14.c
> index 0530192d97e..89a589e0d80 100644
> --- a/gcc/testsuite/gcc.target/i386/sse-14.c
> +++ b/gcc/testsuite/gcc.target/i386/sse-14.c
> @@ -690,6 +690,14 @@ test_1 (_mm512_cvt_roundepi32_ph, __m256h, __m512i, 8)
> test_1 (_mm512_cvt_roundepu32_ph, __m256h, __m512i, 8)
> test_1 (_mm512_cvt_roundepi64_ph, __m128h, __m512i, 8)
> test_1 (_mm512_cvt_roundepu64_ph, __m128h, __m512i, 8)
> +test_1 (_mm_cvt_roundsh_i32, int, __m128h, 8)
> +test_1 (_mm_cvt_roundsh_u32, unsigned, __m128h, 8)
> +#ifdef __x86_64__
> +test_1 (_mm_cvt_roundsh_i64, long long, __m128h, 8)
> +test_1 (_mm_cvt_roundsh_u64, unsigned long long, __m128h, 8)
> +test_2 (_mm_cvt_roundi64_sh, __m128h, __m128h, long long, 8)
> +test_2 (_mm_cvt_roundu64_sh, __m128h, __m128h, unsigned long long, 8)
> +#endif
> test_1x (_mm512_reduce_round_ph, __m512h, __m512h, 123, 8)
> test_1x (_mm512_roundscale_round_ph, __m512h, __m512h, 123, 8)
> test_1x (_mm512_getmant_ph, __m512h, __m512h, 1, 1)
> @@ -734,6 +742,8 @@ test_2 (_mm512_maskz_cvt_roundepi32_ph, __m256h, __mmask16, __m512i, 8)
> test_2 (_mm512_maskz_cvt_roundepu32_ph, __m256h, __mmask16, __m512i, 8)
> test_2 (_mm512_maskz_cvt_roundepi64_ph, __m128h, __mmask8, __m512i, 8)
> test_2 (_mm512_maskz_cvt_roundepu64_ph, __m128h, __mmask8, __m512i, 8)
> +test_2 (_mm_cvt_roundi32_sh, __m128h, __m128h, int, 8)
> +test_2 (_mm_cvt_roundu32_sh, __m128h, __m128h, unsigned, 8)
> test_2x (_mm512_cmp_round_ph_mask, __mmask32, __m512h, __m512h, 1, 8)
> test_2x (_mm_cmp_round_sh_mask, __mmask8, __m128h, __m128h, 1, 8)
> test_2x (_mm_comi_round_sh, int, __m128h, __m128h, 1, 8)
> diff --git a/gcc/testsuite/gcc.target/i386/sse-22.c b/gcc/testsuite/gcc.target/i386/sse-22.c
> index 04e6340516b..fed12744c6c 100644
> --- a/gcc/testsuite/gcc.target/i386/sse-22.c
> +++ b/gcc/testsuite/gcc.target/i386/sse-22.c
> @@ -795,6 +795,14 @@ test_1 (_mm512_cvt_roundepi32_ph, __m256h, __m512i, 8)
> test_1 (_mm512_cvt_roundepu32_ph, __m256h, __m512i, 8)
> test_1 (_mm512_cvt_roundepi64_ph, __m128h, __m512i, 8)
> test_1 (_mm512_cvt_roundepu64_ph, __m128h, __m512i, 8)
> +test_1 (_mm_cvt_roundsh_i32, int, __m128h, 8)
> +test_1 (_mm_cvt_roundsh_u32, unsigned, __m128h, 8)
> +#ifdef __x86_64__
> +test_1 (_mm_cvt_roundsh_i64, long long, __m128h, 8)
> +test_1 (_mm_cvt_roundsh_u64, unsigned long long, __m128h, 8)
> +test_2 (_mm_cvt_roundi64_sh, __m128h, __m128h, long long, 8)
> +test_2 (_mm_cvt_roundu64_sh, __m128h, __m128h, unsigned long long, 8)
> +#endif
> test_1x (_mm512_reduce_round_ph, __m512h, __m512h, 123, 8)
> test_1x (_mm512_roundscale_round_ph, __m512h, __m512h, 123, 8)
> test_1x (_mm512_getmant_ph, __m512h, __m512h, 1, 1)
> @@ -838,6 +846,8 @@ test_2 (_mm512_maskz_cvt_roundepi32_ph, __m256h, __mmask16, __m512i, 8)
> test_2 (_mm512_maskz_cvt_roundepu32_ph, __m256h, __mmask16, __m512i, 8)
> test_2 (_mm512_maskz_cvt_roundepi64_ph, __m128h, __mmask8, __m512i, 8)
> test_2 (_mm512_maskz_cvt_roundepu64_ph, __m128h, __mmask8, __m512i, 8)
> +test_2 (_mm_cvt_roundi32_sh, __m128h, __m128h, int, 8)
> +test_2 (_mm_cvt_roundu32_sh, __m128h, __m128h, unsigned, 8)
> test_2x (_mm512_cmp_round_ph_mask, __mmask32, __m512h, __m512h, 1, 8)
> test_2x (_mm_cmp_round_sh_mask, __mmask8, __m128h, __m128h, 1, 8)
> test_2x (_mm_comi_round_sh, int, __m128h, __m128h, 1, 8)
> diff --git a/gcc/testsuite/gcc.target/i386/sse-23.c b/gcc/testsuite/gcc.target/i386/sse-23.c
> index 684891cc98b..6e8d8a1833c 100644
> --- a/gcc/testsuite/gcc.target/i386/sse-23.c
> +++ b/gcc/testsuite/gcc.target/i386/sse-23.c
> @@ -749,6 +749,14 @@
> #define __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, D) __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, 8)
> #define __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, 8)
> #define __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, 8)
> +#define __builtin_ia32_vcvtsh2si32_round(A, B) __builtin_ia32_vcvtsh2si32_round(A, 8)
> +#define __builtin_ia32_vcvtsh2si64_round(A, B) __builtin_ia32_vcvtsh2si64_round(A, 8)
> +#define __builtin_ia32_vcvtsh2usi32_round(A, B) __builtin_ia32_vcvtsh2usi32_round(A, 8)
> +#define __builtin_ia32_vcvtsh2usi64_round(A, B) __builtin_ia32_vcvtsh2usi64_round(A, 8)
> +#define __builtin_ia32_vcvtsi2sh32_round(A, B, C) __builtin_ia32_vcvtsi2sh32_round(A, B, 8)
> +#define __builtin_ia32_vcvtsi2sh64_round(A, B, C) __builtin_ia32_vcvtsi2sh64_round(A, B, 8)
> +#define __builtin_ia32_vcvtusi2sh32_round(A, B, C) __builtin_ia32_vcvtusi2sh32_round(A, B, 8)
> +#define __builtin_ia32_vcvtusi2sh64_round(A, B, C) __builtin_ia32_vcvtusi2sh64_round(A, B, 8)
>
> /* avx512fp16vlintrin.h */
> #define __builtin_ia32_vcmpph_v8hf_mask(A, B, C, D) __builtin_ia32_vcmpph_v8hf_mask(A, B, 1, D)
> --
> 2.18.1
>
--
BR,
Hongtao
diff --git a/gcc/config/i386/avx512fp16intrin.h b/gcc/config/i386/avx512fp16intrin.h index bd801942365..7524a8d6a5b 100644 --- a/gcc/config/i386/avx512fp16intrin.h +++ b/gcc/config/i386/avx512fp16intrin.h @@ -3529,6 +3529,164 @@ _mm512_maskz_cvt_roundepu16_ph (__mmask32 __A, __m512i __B, int __C) #endif /* __OPTIMIZE__ */ +/* Intrinsics vcvtsh2si, vcvtsh2us. */ +extern __inline int +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsh_i32 (__m128h __A) +{ + return (int) __builtin_ia32_vcvtsh2si32_round (__A, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline unsigned +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsh_u32 (__m128h __A) +{ + return (int) __builtin_ia32_vcvtsh2usi32_round (__A, + _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline int +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundsh_i32 (__m128h __A, const int __R) +{ + return (int) __builtin_ia32_vcvtsh2si32_round (__A, __R); +} + +extern __inline unsigned +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundsh_u32 (__m128h __A, const int __R) +{ + return (int) __builtin_ia32_vcvtsh2usi32_round (__A, __R); +} + +#else +#define _mm_cvt_roundsh_i32(A, B) \ + ((int)__builtin_ia32_vcvtsh2si32_round ((A), (B))) +#define _mm_cvt_roundsh_u32(A, B) \ + ((int)__builtin_ia32_vcvtsh2usi32_round ((A), (B))) + +#endif /* __OPTIMIZE__ */ + +#ifdef __x86_64__ +extern __inline long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsh_i64 (__m128h __A) +{ + return (long long) + __builtin_ia32_vcvtsh2si64_round (__A, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline unsigned long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsh_u64 (__m128h __A) +{ + return (long long) + __builtin_ia32_vcvtsh2usi64_round (__A, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundsh_i64 (__m128h __A, const int __R) +{ + return (long long) __builtin_ia32_vcvtsh2si64_round (__A, __R); +} + +extern __inline unsigned long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundsh_u64 (__m128h __A, const int __R) +{ + return (long long) __builtin_ia32_vcvtsh2usi64_round (__A, __R); +} + +#else +#define _mm_cvt_roundsh_i64(A, B) \ + ((long long)__builtin_ia32_vcvtsh2si64_round ((A), (B))) +#define _mm_cvt_roundsh_u64(A, B) \ + ((long long)__builtin_ia32_vcvtsh2usi64_round ((A), (B))) + +#endif /* __OPTIMIZE__ */ +#endif /* __x86_64__ */ + +/* Intrinsics vcvtsi2sh, vcvtusi2sh. */ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvti32_sh (__m128h __A, int __B) +{ + return __builtin_ia32_vcvtsi2sh32_round (__A, __B, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtu32_sh (__m128h __A, unsigned int __B) +{ + return __builtin_ia32_vcvtusi2sh32_round (__A, __B, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundi32_sh (__m128h __A, int __B, const int __R) +{ + return __builtin_ia32_vcvtsi2sh32_round (__A, __B, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundu32_sh (__m128h __A, unsigned int __B, const int __R) +{ + return __builtin_ia32_vcvtusi2sh32_round (__A, __B, __R); +} + +#else +#define _mm_cvt_roundi32_sh(A, B, C) \ + (__builtin_ia32_vcvtsi2sh32_round ((A), (B), (C))) +#define _mm_cvt_roundu32_sh(A, B, C) \ + (__builtin_ia32_vcvtusi2sh32_round ((A), (B), (C))) + +#endif /* __OPTIMIZE__ */ + +#ifdef __x86_64__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvti64_sh (__m128h __A, long long __B) +{ + return __builtin_ia32_vcvtsi2sh64_round (__A, __B, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtu64_sh (__m128h __A, unsigned long long __B) +{ + return __builtin_ia32_vcvtusi2sh64_round (__A, __B, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundi64_sh (__m128h __A, long long __B, const int __R) +{ + return __builtin_ia32_vcvtsi2sh64_round (__A, __B, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundu64_sh (__m128h __A, unsigned long long __B, const int __R) +{ + return __builtin_ia32_vcvtusi2sh64_round (__A, __B, __R); +} + +#else +#define _mm_cvt_roundi64_sh(A, B, C) \ + (__builtin_ia32_vcvtsi2sh64_round ((A), (B), (C))) +#define _mm_cvt_roundu64_sh(A, B, C) \ + (__builtin_ia32_vcvtusi2sh64_round ((A), (B), (C))) + +#endif /* __OPTIMIZE__ */ +#endif /* __x86_64__ */ + + #ifdef __DISABLE_AVX512FP16__ #undef __DISABLE_AVX512FP16__ #pragma GCC pop_options diff --git a/gcc/config/i386/i386-builtin-types.def b/gcc/config/i386/i386-builtin-types.def index 57b9ea786e1..74bda59a65e 100644 --- a/gcc/config/i386/i386-builtin-types.def +++ b/gcc/config/i386/i386-builtin-types.def @@ -1308,9 +1308,17 @@ DEF_FUNCTION_TYPE (V8HF, V8HI) DEF_FUNCTION_TYPE (QI, V8HF, INT, UQI) DEF_FUNCTION_TYPE (HI, V16HF, INT, UHI) DEF_FUNCTION_TYPE (SI, V32HF, INT, USI) +DEF_FUNCTION_TYPE (INT, V8HF, INT) +DEF_FUNCTION_TYPE (INT64, V8HF, INT) +DEF_FUNCTION_TYPE (UINT, V8HF, INT) +DEF_FUNCTION_TYPE (UINT64, V8HF, INT) DEF_FUNCTION_TYPE (V8HF, V8HF, V8HF) DEF_FUNCTION_TYPE (VOID, PCFLOAT16, V8HF, UQI) DEF_FUNCTION_TYPE (V8HF, PCFLOAT16, V8HF, UQI) +DEF_FUNCTION_TYPE (V8HF, V8HF, INT, INT) +DEF_FUNCTION_TYPE (V8HF, V8HF, INT64, INT) +DEF_FUNCTION_TYPE (V8HF, V8HF, UINT, INT) +DEF_FUNCTION_TYPE (V8HF, V8HF, UINT64, INT) DEF_FUNCTION_TYPE (V2DI, V8HF, V2DI, UQI) DEF_FUNCTION_TYPE (V4DI, V8HF, V4DI, UQI) DEF_FUNCTION_TYPE (V4SI, V8HF, V4SI, UQI) diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def index 44c55876e48..3602b40d6d5 100644 --- a/gcc/config/i386/i386-builtin.def +++ b/gcc/config/i386/i386-builtin.def @@ -3094,6 +3094,14 @@ BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtdq2ph_v16si_mask_ BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtudq2ph_v16si_mask_round, "__builtin_ia32_vcvtudq2ph_v16si_mask_round", IX86_BUILTIN_VCVTUDQ2PH_V16SI_MASK_ROUND, UNKNOWN, (int) V16HF_FTYPE_V16SI_V16HF_UHI_INT) BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtqq2ph_v8di_mask_round, "__builtin_ia32_vcvtqq2ph_v8di_mask_round", IX86_BUILTIN_VCVTQQ2PH_V8DI_MASK_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8DI_V8HF_UQI_INT) BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtuqq2ph_v8di_mask_round, "__builtin_ia32_vcvtuqq2ph_v8di_mask_round", IX86_BUILTIN_VCVTUQQ2PH_V8DI_MASK_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8DI_V8HF_UQI_INT) +BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsh2si_round, "__builtin_ia32_vcvtsh2si32_round", IX86_BUILTIN_VCVTSH2SI32_ROUND, UNKNOWN, (int) INT_FTYPE_V8HF_INT) +BDESC (OPTION_MASK_ISA_64BIT, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsh2siq_round, "__builtin_ia32_vcvtsh2si64_round", IX86_BUILTIN_VCVTSH2SI64_ROUND, UNKNOWN, (int) INT64_FTYPE_V8HF_INT) +BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsh2usi_round, "__builtin_ia32_vcvtsh2usi32_round", IX86_BUILTIN_VCVTSH2USI32_ROUND, UNKNOWN, (int) UINT_FTYPE_V8HF_INT) +BDESC (OPTION_MASK_ISA_64BIT, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsh2usiq_round, "__builtin_ia32_vcvtsh2usi64_round", IX86_BUILTIN_VCVTSH2USI64_ROUND, UNKNOWN, (int) UINT64_FTYPE_V8HF_INT) +BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsi2sh_round, "__builtin_ia32_vcvtsi2sh32_round", IX86_BUILTIN_VCVTSI2SH32_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8HF_INT_INT) +BDESC (OPTION_MASK_ISA_64BIT, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtsi2shq_round, "__builtin_ia32_vcvtsi2sh64_round", IX86_BUILTIN_VCVTSI2SH64_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8HF_INT64_INT) +BDESC (0, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtusi2sh_round, "__builtin_ia32_vcvtusi2sh32_round", IX86_BUILTIN_VCVTUSI2SH32_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8HF_UINT_INT) +BDESC (OPTION_MASK_ISA_64BIT, OPTION_MASK_ISA2_AVX512FP16, CODE_FOR_avx512fp16_vcvtusi2shq_round, "__builtin_ia32_vcvtusi2sh64_round", IX86_BUILTIN_VCVTUSI2SH64_ROUND, UNKNOWN, (int) V8HF_FTYPE_V8HF_UINT64_INT) BDESC_END (ROUND_ARGS, MULTI_ARG) diff --git a/gcc/config/i386/i386-expand.c b/gcc/config/i386/i386-expand.c index 7d9e1bd6a2d..b83c6d9a92b 100644 --- a/gcc/config/i386/i386-expand.c +++ b/gcc/config/i386/i386-expand.c @@ -10489,16 +10489,24 @@ ix86_expand_round_builtin (const struct builtin_description *d, { case UINT64_FTYPE_V2DF_INT: case UINT64_FTYPE_V4SF_INT: + case UINT64_FTYPE_V8HF_INT: case UINT_FTYPE_V2DF_INT: case UINT_FTYPE_V4SF_INT: + case UINT_FTYPE_V8HF_INT: case INT64_FTYPE_V2DF_INT: case INT64_FTYPE_V4SF_INT: + case INT64_FTYPE_V8HF_INT: case INT_FTYPE_V2DF_INT: case INT_FTYPE_V4SF_INT: + case INT_FTYPE_V8HF_INT: nargs = 2; break; case V32HF_FTYPE_V32HF_V32HF_INT: case V8HF_FTYPE_V8HF_V8HF_INT: + case V8HF_FTYPE_V8HF_INT_INT: + case V8HF_FTYPE_V8HF_UINT_INT: + case V8HF_FTYPE_V8HF_INT64_INT: + case V8HF_FTYPE_V8HF_UINT64_INT: case V4SF_FTYPE_V4SF_UINT_INT: case V4SF_FTYPE_V4SF_UINT64_INT: case V2DF_FTYPE_V2DF_UINT64_INT: diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md index 8b23048a232..b312d26b806 100644 --- a/gcc/config/i386/sse.md +++ b/gcc/config/i386/sse.md @@ -5589,6 +5589,52 @@ (define_insn "*avx512fp16_vcvt<floatsuffix>qq2ph_v2di_mask_1" (set_attr "prefix" "evex") (set_attr "mode" "TI")]) +(define_insn "avx512fp16_vcvtsh2<sseintconvertsignprefix>si<rex64namesuffix><round_name>" + [(set (match_operand:SWI48 0 "register_operand" "=r,r") + (unspec:SWI48 + [(vec_select:HF + (match_operand:V8HF 1 "<round_nimm_scalar_predicate>" "v,<round_constraint2>") + (parallel [(const_int 0)]))] + UNSPEC_US_FIX_NOTRUNC))] + "TARGET_AVX512FP16" + "%vcvtsh2<sseintconvertsignprefix>si\t{<round_op2>%1, %0|%0, %k1<round_op2>}" + [(set_attr "type" "sseicvt") + (set_attr "athlon_decode" "double,vector") + (set_attr "bdver1_decode" "double,double") + (set_attr "prefix_rep" "1") + (set_attr "prefix" "evex") + (set_attr "mode" "<MODE>")]) + +(define_insn "avx512fp16_vcvtsh2<sseintconvertsignprefix>si<rex64namesuffix>_2" + [(set (match_operand:SWI48 0 "register_operand" "=r,r") + (unspec:SWI48 [(match_operand:HF 1 "nonimmediate_operand" "v,m")] + UNSPEC_US_FIX_NOTRUNC))] + "TARGET_AVX512FP16" + "%vcvtsh2<sseintconvertsignprefix>si\t{%1, %0|%0, %k1}" + [(set_attr "type" "sseicvt") + (set_attr "athlon_decode" "double,vector") + (set_attr "bdver1_decode" "double,double") + (set_attr "prefix_rep" "1") + (set_attr "prefix" "evex") + (set_attr "mode" "<MODE>")]) + +(define_insn "avx512fp16_vcvt<floatsuffix>si2sh<rex64namesuffix><round_name>" + [(set (match_operand:V8HF 0 "register_operand" "=v") + (vec_merge:V8HF + (vec_duplicate:V8HF + (any_float:HF (match_operand:SWI48 2 "<round_nimm_scalar_predicate>" "<round_constraint3>"))) + (match_operand:V8HF 1 "register_operand" "v") + (const_int 1)))] + "TARGET_AVX512FP16" + "vcvt<floatsuffix>si2sh\t{%2, <round_op3>%1, %0|%0, %1<round_op3>, %2}" + [(set_attr "type" "sseicvt") + (set_attr "athlon_decode" "*") + (set_attr "amdfam10_decode" "*") + (set_attr "bdver1_decode" "*") + (set_attr "btver2_decode" "double") + (set_attr "znver1_decode" "double") + (set_attr "prefix" "evex") + (set_attr "mode" "HF")]) ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; ;; diff --git a/gcc/testsuite/gcc.target/i386/avx-1.c b/gcc/testsuite/gcc.target/i386/avx-1.c index b569cc0bdd9..0aae949097a 100644 --- a/gcc/testsuite/gcc.target/i386/avx-1.c +++ b/gcc/testsuite/gcc.target/i386/avx-1.c @@ -731,6 +731,14 @@ #define __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, D) __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, 8) #define __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, 8) #define __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, 8) +#define __builtin_ia32_vcvtsh2si32_round(A, B) __builtin_ia32_vcvtsh2si32_round(A, 8) +#define __builtin_ia32_vcvtsh2si64_round(A, B) __builtin_ia32_vcvtsh2si64_round(A, 8) +#define __builtin_ia32_vcvtsh2usi32_round(A, B) __builtin_ia32_vcvtsh2usi32_round(A, 8) +#define __builtin_ia32_vcvtsh2usi64_round(A, B) __builtin_ia32_vcvtsh2usi64_round(A, 8) +#define __builtin_ia32_vcvtsi2sh32_round(A, B, C) __builtin_ia32_vcvtsi2sh32_round(A, B, 8) +#define __builtin_ia32_vcvtsi2sh64_round(A, B, C) __builtin_ia32_vcvtsi2sh64_round(A, B, 8) +#define __builtin_ia32_vcvtusi2sh32_round(A, B, C) __builtin_ia32_vcvtusi2sh32_round(A, B, 8) +#define __builtin_ia32_vcvtusi2sh64_round(A, B, C) __builtin_ia32_vcvtusi2sh64_round(A, B, 8) /* avx512fp16vlintrin.h */ #define __builtin_ia32_vcmpph_v8hf_mask(A, B, C, D) __builtin_ia32_vcmpph_v8hf_mask(A, B, 1, D) diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c b/gcc/testsuite/gcc.target/i386/sse-13.c index 07e59118438..997fb733132 100644 --- a/gcc/testsuite/gcc.target/i386/sse-13.c +++ b/gcc/testsuite/gcc.target/i386/sse-13.c @@ -748,6 +748,14 @@ #define __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, D) __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, 8) #define __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, 8) #define __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, 8) +#define __builtin_ia32_vcvtsh2si32_round(A, B) __builtin_ia32_vcvtsh2si32_round(A, 8) +#define __builtin_ia32_vcvtsh2si64_round(A, B) __builtin_ia32_vcvtsh2si64_round(A, 8) +#define __builtin_ia32_vcvtsh2usi32_round(A, B) __builtin_ia32_vcvtsh2usi32_round(A, 8) +#define __builtin_ia32_vcvtsh2usi64_round(A, B) __builtin_ia32_vcvtsh2usi64_round(A, 8) +#define __builtin_ia32_vcvtsi2sh32_round(A, B, C) __builtin_ia32_vcvtsi2sh32_round(A, B, 8) +#define __builtin_ia32_vcvtsi2sh64_round(A, B, C) __builtin_ia32_vcvtsi2sh64_round(A, B, 8) +#define __builtin_ia32_vcvtusi2sh32_round(A, B, C) __builtin_ia32_vcvtusi2sh32_round(A, B, 8) +#define __builtin_ia32_vcvtusi2sh64_round(A, B, C) __builtin_ia32_vcvtusi2sh64_round(A, B, 8) /* avx512fp16vlintrin.h */ #define __builtin_ia32_vcmpph_v8hf_mask(A, B, C, D) __builtin_ia32_vcmpph_v8hf_mask(A, B, 1, D) diff --git a/gcc/testsuite/gcc.target/i386/sse-14.c b/gcc/testsuite/gcc.target/i386/sse-14.c index 0530192d97e..89a589e0d80 100644 --- a/gcc/testsuite/gcc.target/i386/sse-14.c +++ b/gcc/testsuite/gcc.target/i386/sse-14.c @@ -690,6 +690,14 @@ test_1 (_mm512_cvt_roundepi32_ph, __m256h, __m512i, 8) test_1 (_mm512_cvt_roundepu32_ph, __m256h, __m512i, 8) test_1 (_mm512_cvt_roundepi64_ph, __m128h, __m512i, 8) test_1 (_mm512_cvt_roundepu64_ph, __m128h, __m512i, 8) +test_1 (_mm_cvt_roundsh_i32, int, __m128h, 8) +test_1 (_mm_cvt_roundsh_u32, unsigned, __m128h, 8) +#ifdef __x86_64__ +test_1 (_mm_cvt_roundsh_i64, long long, __m128h, 8) +test_1 (_mm_cvt_roundsh_u64, unsigned long long, __m128h, 8) +test_2 (_mm_cvt_roundi64_sh, __m128h, __m128h, long long, 8) +test_2 (_mm_cvt_roundu64_sh, __m128h, __m128h, unsigned long long, 8) +#endif test_1x (_mm512_reduce_round_ph, __m512h, __m512h, 123, 8) test_1x (_mm512_roundscale_round_ph, __m512h, __m512h, 123, 8) test_1x (_mm512_getmant_ph, __m512h, __m512h, 1, 1) @@ -734,6 +742,8 @@ test_2 (_mm512_maskz_cvt_roundepi32_ph, __m256h, __mmask16, __m512i, 8) test_2 (_mm512_maskz_cvt_roundepu32_ph, __m256h, __mmask16, __m512i, 8) test_2 (_mm512_maskz_cvt_roundepi64_ph, __m128h, __mmask8, __m512i, 8) test_2 (_mm512_maskz_cvt_roundepu64_ph, __m128h, __mmask8, __m512i, 8) +test_2 (_mm_cvt_roundi32_sh, __m128h, __m128h, int, 8) +test_2 (_mm_cvt_roundu32_sh, __m128h, __m128h, unsigned, 8) test_2x (_mm512_cmp_round_ph_mask, __mmask32, __m512h, __m512h, 1, 8) test_2x (_mm_cmp_round_sh_mask, __mmask8, __m128h, __m128h, 1, 8) test_2x (_mm_comi_round_sh, int, __m128h, __m128h, 1, 8) diff --git a/gcc/testsuite/gcc.target/i386/sse-22.c b/gcc/testsuite/gcc.target/i386/sse-22.c index 04e6340516b..fed12744c6c 100644 --- a/gcc/testsuite/gcc.target/i386/sse-22.c +++ b/gcc/testsuite/gcc.target/i386/sse-22.c @@ -795,6 +795,14 @@ test_1 (_mm512_cvt_roundepi32_ph, __m256h, __m512i, 8) test_1 (_mm512_cvt_roundepu32_ph, __m256h, __m512i, 8) test_1 (_mm512_cvt_roundepi64_ph, __m128h, __m512i, 8) test_1 (_mm512_cvt_roundepu64_ph, __m128h, __m512i, 8) +test_1 (_mm_cvt_roundsh_i32, int, __m128h, 8) +test_1 (_mm_cvt_roundsh_u32, unsigned, __m128h, 8) +#ifdef __x86_64__ +test_1 (_mm_cvt_roundsh_i64, long long, __m128h, 8) +test_1 (_mm_cvt_roundsh_u64, unsigned long long, __m128h, 8) +test_2 (_mm_cvt_roundi64_sh, __m128h, __m128h, long long, 8) +test_2 (_mm_cvt_roundu64_sh, __m128h, __m128h, unsigned long long, 8) +#endif test_1x (_mm512_reduce_round_ph, __m512h, __m512h, 123, 8) test_1x (_mm512_roundscale_round_ph, __m512h, __m512h, 123, 8) test_1x (_mm512_getmant_ph, __m512h, __m512h, 1, 1) @@ -838,6 +846,8 @@ test_2 (_mm512_maskz_cvt_roundepi32_ph, __m256h, __mmask16, __m512i, 8) test_2 (_mm512_maskz_cvt_roundepu32_ph, __m256h, __mmask16, __m512i, 8) test_2 (_mm512_maskz_cvt_roundepi64_ph, __m128h, __mmask8, __m512i, 8) test_2 (_mm512_maskz_cvt_roundepu64_ph, __m128h, __mmask8, __m512i, 8) +test_2 (_mm_cvt_roundi32_sh, __m128h, __m128h, int, 8) +test_2 (_mm_cvt_roundu32_sh, __m128h, __m128h, unsigned, 8) test_2x (_mm512_cmp_round_ph_mask, __mmask32, __m512h, __m512h, 1, 8) test_2x (_mm_cmp_round_sh_mask, __mmask8, __m128h, __m128h, 1, 8) test_2x (_mm_comi_round_sh, int, __m128h, __m128h, 1, 8) diff --git a/gcc/testsuite/gcc.target/i386/sse-23.c b/gcc/testsuite/gcc.target/i386/sse-23.c index 684891cc98b..6e8d8a1833c 100644 --- a/gcc/testsuite/gcc.target/i386/sse-23.c +++ b/gcc/testsuite/gcc.target/i386/sse-23.c @@ -749,6 +749,14 @@ #define __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, D) __builtin_ia32_vcvtudq2ph_v16si_mask_round(A, B, C, 8) #define __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtqq2ph_v8di_mask_round(A, B, C, 8) #define __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, D) __builtin_ia32_vcvtuqq2ph_v8di_mask_round(A, B, C, 8) +#define __builtin_ia32_vcvtsh2si32_round(A, B) __builtin_ia32_vcvtsh2si32_round(A, 8) +#define __builtin_ia32_vcvtsh2si64_round(A, B) __builtin_ia32_vcvtsh2si64_round(A, 8) +#define __builtin_ia32_vcvtsh2usi32_round(A, B) __builtin_ia32_vcvtsh2usi32_round(A, 8) +#define __builtin_ia32_vcvtsh2usi64_round(A, B) __builtin_ia32_vcvtsh2usi64_round(A, 8) +#define __builtin_ia32_vcvtsi2sh32_round(A, B, C) __builtin_ia32_vcvtsi2sh32_round(A, B, 8) +#define __builtin_ia32_vcvtsi2sh64_round(A, B, C) __builtin_ia32_vcvtsi2sh64_round(A, B, 8) +#define __builtin_ia32_vcvtusi2sh32_round(A, B, C) __builtin_ia32_vcvtusi2sh32_round(A, B, 8) +#define __builtin_ia32_vcvtusi2sh64_round(A, B, C) __builtin_ia32_vcvtusi2sh64_round(A, B, 8) /* avx512fp16vlintrin.h */ #define __builtin_ia32_vcmpph_v8hf_mask(A, B, C, D) __builtin_ia32_vcmpph_v8hf_mask(A, B, 1, D)