From 416091ebdb9e901b29d026633e73167d6353a0b0 Mon Sep 17 00:00:00 2001 From: Zhijin Zeng Date: Fri, 12 Apr 2024 14:59:33 +0800 Subject: [PATCH] arm: enable more intrinsic function for armv7 Signed-off-by: Zhijin Zeng --- simde/arm/neon/ld2_lane.h | 12 ++++++------ simde/arm/neon/ld3.h | 4 ++-- simde/arm/neon/ld3_lane.h | 12 ++++++------ simde/arm/neon/ld4.h | 4 ++-- simde/arm/neon/st3.h | 4 ++-- simde/arm/neon/st4.h | 4 ++-- simde/arm/neon/types.h | 2 ++ 7 files changed, 22 insertions(+), 20 deletions(-) diff --git a/simde/arm/neon/ld2_lane.h b/simde/arm/neon/ld2_lane.h index f2c2a1a4d..81b29dd20 100644 --- a/simde/arm/neon/ld2_lane.h +++ b/simde/arm/neon/ld2_lane.h @@ -285,10 +285,10 @@ simde_int16x8x2_t simde_vld2q_lane_s16(int16_t const ptr[HEDLEY_ARRAY_PARAM(2)], } return r; } -#if defined(SIMDE_ARM_NEON_A64V8_NATIVE) +#if defined(SIMDE_ARM_NEON_A32V7_NATIVE) #define simde_vld2q_lane_s16(ptr, src, lane) vld2q_lane_s16(ptr, src, lane) #endif -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vld2q_lane_s16 #define vld2q_lane_s16(ptr, src, lane) simde_vld2q_lane_s16((ptr), (src), (lane)) #endif @@ -365,10 +365,10 @@ simde_uint16x8x2_t simde_vld2q_lane_u16(uint16_t const ptr[HEDLEY_ARRAY_PARAM(2) } return r; } -#if defined(SIMDE_ARM_NEON_A64V8_NATIVE) +#if defined(SIMDE_ARM_NEON_A32V7_NATIVE) #define simde_vld2q_lane_u16(ptr, src, lane) vld2q_lane_u16(ptr, src, lane) #endif -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vld2q_lane_u16 #define vld2q_lane_u16(ptr, src, lane) simde_vld2q_lane_u16((ptr), (src), (lane)) #endif @@ -564,10 +564,10 @@ simde_poly16x8x2_t simde_vld2q_lane_p16(simde_poly16_t const ptr[HEDLEY_ARRAY_PA } return r; } -#if defined(SIMDE_ARM_NEON_A64V8_NATIVE) +#if defined(SIMDE_ARM_NEON_A32V7_NATIVE) #define simde_vld2q_lane_p16(ptr, src, lane) vld2q_lane_p16(ptr, src, lane) #endif -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vld2q_lane_p16 #define vld2q_lane_p16(ptr, src, lane) simde_vld2q_lane_p16((ptr), (src), (lane)) #endif diff --git a/simde/arm/neon/ld3.h b/simde/arm/neon/ld3.h index d53dbf76c..a60c2aa0d 100644 --- a/simde/arm/neon/ld3.h +++ b/simde/arm/neon/ld3.h @@ -268,7 +268,7 @@ simde_vld3_s64(int64_t const *ptr) { return r; #endif } -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vld3_s64 #define vld3_s64(a) simde_vld3_s64((a)) #endif @@ -400,7 +400,7 @@ simde_vld3_u64(uint64_t const *ptr) { return r; #endif } -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vld3_u64 #define vld3_u64(a) simde_vld3_u64((a)) #endif diff --git a/simde/arm/neon/ld3_lane.h b/simde/arm/neon/ld3_lane.h index 7541bb207..4950792a8 100644 --- a/simde/arm/neon/ld3_lane.h +++ b/simde/arm/neon/ld3_lane.h @@ -285,10 +285,10 @@ simde_int16x8x3_t simde_vld3q_lane_s16(int16_t const ptr[HEDLEY_ARRAY_PARAM(3)], } return r; } -#if defined(SIMDE_ARM_NEON_A64V8_NATIVE) +#if defined(SIMDE_ARM_NEON_A32V7_NATIVE) #define simde_vld3q_lane_s16(ptr, src, lane) vld3q_lane_s16(ptr, src, lane) #endif -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vld3q_lane_s16 #define vld3q_lane_s16(ptr, src, lane) simde_vld3q_lane_s16((ptr), (src), (lane)) #endif @@ -365,10 +365,10 @@ simde_uint16x8x3_t simde_vld3q_lane_u16(uint16_t const ptr[HEDLEY_ARRAY_PARAM(3) } return r; } -#if defined(SIMDE_ARM_NEON_A64V8_NATIVE) +#if defined(SIMDE_ARM_NEON_A32V7_NATIVE) #define simde_vld3q_lane_u16(ptr, src, lane) vld3q_lane_u16(ptr, src, lane) #endif -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vld3q_lane_u16 #define vld3q_lane_u16(ptr, src, lane) simde_vld3q_lane_u16((ptr), (src), (lane)) #endif @@ -564,10 +564,10 @@ simde_poly16x8x3_t simde_vld3q_lane_p16(simde_poly16_t const ptr[HEDLEY_ARRAY_PA } return r; } -#if defined(SIMDE_ARM_NEON_A64V8_NATIVE) +#if defined(SIMDE_ARM_NEON_A32V7_NATIVE) #define simde_vld3q_lane_p16(ptr, src, lane) vld3q_lane_p16(ptr, src, lane) #endif -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vld3q_lane_p16 #define vld3q_lane_p16(ptr, src, lane) simde_vld3q_lane_p16((ptr), (src), (lane)) #endif diff --git a/simde/arm/neon/ld4.h b/simde/arm/neon/ld4.h index 7446b9bd5..777c24f73 100644 --- a/simde/arm/neon/ld4.h +++ b/simde/arm/neon/ld4.h @@ -232,7 +232,7 @@ simde_vld4_s64(int64_t const ptr[HEDLEY_ARRAY_PARAM(4)]) { return s_; #endif } -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vld4_s64 #define vld4_s64(a) simde_vld4_s64((a)) #endif @@ -344,7 +344,7 @@ simde_vld4_u64(uint64_t const ptr[HEDLEY_ARRAY_PARAM(4)]) { return s_; #endif } -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vld4_u64 #define vld4_u64(a) simde_vld4_u64((a)) #endif diff --git a/simde/arm/neon/st3.h b/simde/arm/neon/st3.h index 940016b33..29301ae6c 100644 --- a/simde/arm/neon/st3.h +++ b/simde/arm/neon/st3.h @@ -283,7 +283,7 @@ simde_vst3_s64(int64_t ptr[HEDLEY_ARRAY_PARAM(3)], simde_int64x1x3_t val) { #endif #endif } -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vst3_s64 #define vst3_s64(a, b) simde_vst3_s64((a), (b)) #endif @@ -440,7 +440,7 @@ simde_vst3_u64(uint64_t ptr[HEDLEY_ARRAY_PARAM(3)], simde_uint64x1x3_t val) { #endif #endif } -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vst3_u64 #define vst3_u64(a, b) simde_vst3_u64((a), (b)) #endif diff --git a/simde/arm/neon/st4.h b/simde/arm/neon/st4.h index 1cb3eab23..6b0b4706b 100644 --- a/simde/arm/neon/st4.h +++ b/simde/arm/neon/st4.h @@ -236,7 +236,7 @@ simde_vst4_s64(int64_t *ptr, simde_int64x1x4_t val) { #endif #endif } -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vst4_s64 #define vst4_s64(a, b) simde_vst4_s64((a), (b)) #endif @@ -352,7 +352,7 @@ simde_vst4_u64(uint64_t *ptr, simde_uint64x1x4_t val) { #endif #endif } -#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES) +#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) #undef vst4_u64 #define vst4_u64(a, b) simde_vst4_u64((a), (b)) #endif diff --git a/simde/arm/neon/types.h b/simde/arm/neon/types.h index e37f379a0..5a5954ac9 100644 --- a/simde/arm/neon/types.h +++ b/simde/arm/neon/types.h @@ -1291,6 +1291,8 @@ typedef union { #endif #if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES) typedef simde_float32_t float32_t; + typedef simde_poly8_t poly8_t; + typedef simde_poly16_t poly16_t; typedef simde_int8x8_t int8x8_t; typedef simde_int16x4_t int16x4_t;