[dpdk-dev,09/11] lib: add armv8 rte_vect.h
Commit Message
From: Benjamin Boren <Ben.Boren@intel.com>
Signed-off-by: Benjamin Boren <Ben.Boren@intel.com>
Signed-off-by: David Hunt <david.hunt@intel.com>
---
.../common/include/arch/arm64/rte_vect.h | 102 +++++++++++++++++++++
1 file changed, 102 insertions(+)
create mode 100644 lib/librte_eal/common/include/arch/arm64/rte_vect.h
Comments
On Fri, 23 Oct 2015 15:17:11 +0100
David Hunt <david.hunt@intel.com> wrote:
> +
> +typedef float32x4_t __m128;
> +
> +typedef int32x4_t __m128i;
> +
> +typedef __m128i xmm_t;
> +
> +#define XMM_SIZE (sizeof(xmm_t))
> +#define XMM_MASK (XMM_SIZE - 1)
> +
> +typedef union rte_xmm {
> + xmm_t x;
> + uint8_t u8[XMM_SIZE / sizeof(uint8_t)];
> + uint16_t u16[XMM_SIZE / sizeof(uint16_t)];
> + uint32_t u32[XMM_SIZE / sizeof(uint32_t)];
> + uint64_t u64[XMM_SIZE / sizeof(uint64_t)];
> + double pd[XMM_SIZE / sizeof(double)];
> +} rte_xmm_t __aligned(16);
> +
> +#define _mm_srli_epi32(a, imm) { (__m128i)vshrq_n_u32((uint32x4_t)a, imm) }
> +
> +#define _mm_srli_si128(a, imm) { (__m128i)vextq_s8((int8x16_t)a, \
> + vdupq_n_s8(0), (imm)) }
> +
> +static inline __m128i
> +_mm_set_epi32(int i3, int i2, int i1, int i0);
> +static inline int
> +_mm_cvtsi128_si64(__m128i a);
> +
> +static inline __m128i
> +_mm_set_epi32(int i3, int i2, int i1, int i0)
> +{
> + int32_t __aligned(16) data[4] = { i0, i1, i2, i3 };
> + return vld1q_s32(data);
> +}
> +
> +static inline int
> +_mm_cvtsi128_si64(__m128i a)
> +{
> + return vgetq_lane_s64(a, 0);
> +}
> +
> +static inline __m128i
> +_mm_and_si128(__m128i a, __m128i b)
> +{
> + return (__m128i)vandq_s32(a, b);
> +}
> +
What is the purpose of those wrappers? I mean, does it help to compile
some SSE-dependent libraries in DPDK for NEON?
Jan
new file mode 100644
@@ -0,0 +1,102 @@
+/*
+ * BSD LICENSE
+ *
+ * Copyright(c) 2015 Intel Corporation. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * * Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef _RTE_VECT_ARM64_H_
+#define _RTE_VECT_ARM64_H_
+
+/**
+ * @file
+ *
+ * RTE SSE/AVX related header.
+ */
+
+
+#include <arm_neon.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef float32x4_t __m128;
+
+typedef int32x4_t __m128i;
+
+typedef __m128i xmm_t;
+
+#define XMM_SIZE (sizeof(xmm_t))
+#define XMM_MASK (XMM_SIZE - 1)
+
+typedef union rte_xmm {
+ xmm_t x;
+ uint8_t u8[XMM_SIZE / sizeof(uint8_t)];
+ uint16_t u16[XMM_SIZE / sizeof(uint16_t)];
+ uint32_t u32[XMM_SIZE / sizeof(uint32_t)];
+ uint64_t u64[XMM_SIZE / sizeof(uint64_t)];
+ double pd[XMM_SIZE / sizeof(double)];
+} rte_xmm_t __aligned(16);
+
+#define _mm_srli_epi32(a, imm) { (__m128i)vshrq_n_u32((uint32x4_t)a, imm) }
+
+#define _mm_srli_si128(a, imm) { (__m128i)vextq_s8((int8x16_t)a, \
+ vdupq_n_s8(0), (imm)) }
+
+static inline __m128i
+_mm_set_epi32(int i3, int i2, int i1, int i0);
+static inline int
+_mm_cvtsi128_si64(__m128i a);
+
+static inline __m128i
+_mm_set_epi32(int i3, int i2, int i1, int i0)
+{
+ int32_t __aligned(16) data[4] = { i0, i1, i2, i3 };
+ return vld1q_s32(data);
+}
+
+static inline int
+_mm_cvtsi128_si64(__m128i a)
+{
+ return vgetq_lane_s64(a, 0);
+}
+
+static inline __m128i
+_mm_and_si128(__m128i a, __m128i b)
+{
+ return (__m128i)vandq_s32(a, b);
+}
+
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_VECT_ARM64_H_*/
+