From 855b0a692b03555b58aea80e95bca9ec3cfca779 Mon Sep 17 00:00:00 2001
From: Mooneer Salem <mooneer@gmail.com>
Date: Fri, 9 Dec 2022 21:48:41 -0800
Subject: [PATCH] Suppress aarch64 specific code for non-aarch64 builds.

---
 src/vec_neon.h | 215 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 215 insertions(+)

diff --git a/src/vec_neon.h b/src/vec_neon.h
index 8a80b29..9c40000 100644
--- a/src/vec_neon.h
+++ b/src/vec_neon.h
@@ -276,6 +276,8 @@ static inline void sparse_sgemv_accum16(float *out, const float *w, int rows, co
 #define MAX_INPUTS 2048
 #define MAX_OUTPUTS 8192
 
+#if __aarch64__
+
 #if __ARM_FEATURE_DOTPROD
 static inline int32x4_t vdotprod(int32x4_t acc, int8x16_t a, int8x16_t b) {
   return vdotq_s32(acc, a, b);
@@ -361,3 +363,216 @@ static inline void sparse_sgemv_accum8x4(float *_out, const qweight *w, int rows
       vst1q_f32(&_out[i+4], vmulq_f32(scale_1, vcvtq_f32_s32(acc1)));
    }
 }
+
+#else
+
+#ifdef DOT_PROD
+
+#define SCALE (128.f*127.f)
+#define SCALE_1 (1.f/128.f/127.f)
+
+
+#ifdef USE_SU_BIAS
+
+static inline void sgemv_accum8x4(float *out, const qweight *w, int rows, int cols, int col_stride, const float *_x)
+{
+   int i, j;
+   unsigned char x[MAX_INPUTS];
+   (void)col_stride;
+   for (i=0;i<rows;i++) out[i] *= SCALE;
+   for (i=0;i<cols;i++) x[i] = 127+(int)floor(.5+127*_x[i]);
+   for (i=0;i<rows;i+=8)
+   {
+      for (j=0;j<cols;j+=4)
+      {
+         float * restrict y;
+         float xj0, xj1, xj2, xj3;
+         xj0 = x[j+0];
+         xj1 = x[j+1];
+         xj2 = x[j+2];
+         xj3 = x[j+3];
+         y = &out[i];
+         y[0] += (w[0]*xj0+w[1]*xj1+w[2]*xj2+w[3]*xj3);
+         y[1] += (w[4]*xj0+w[5]*xj1+w[6]*xj2+w[7]*xj3);
+         y[2] += (w[8]*xj0+w[9]*xj1+w[10]*xj2+w[11]*xj3);
+         y[3] += (w[12]*xj0+w[13]*xj1+w[14]*xj2+w[15]*xj3);
+         y[4] += (w[16]*xj0+w[17]*xj1+w[18]*xj2+w[19]*xj3);
+         y[5] += (w[20]*xj0+w[21]*xj1+w[22]*xj2+w[23]*xj3);
+         y[6] += (w[24]*xj0+w[25]*xj1+w[26]*xj2+w[27]*xj3);
+         y[7] += (w[28]*xj0+w[29]*xj1+w[30]*xj2+w[31]*xj3);
+         w += 32;
+      }
+   }
+   for (i=0;i<rows;i++) out[i] *= SCALE_1;
+}
+
+static inline void sparse_sgemv_accum8x4(float *out, const qweight *w, int rows, int cols, const int *idx, const float *_x)
+{
+   int i, j;
+   unsigned char x[MAX_INPUTS];
+   for (i=0;i<rows;i++) out[i] *= SCALE;
+   for (i=0;i<cols;i++) x[i] = 127+floor(.5+127*_x[i]);
+   for (i=0;i<rows;i+=8)
+   {
+      int colblocks;
+      colblocks = *idx++;
+      for (j=0;j<colblocks;j++)
+      {
+         int pos;
+         float * restrict y;
+         int xj0, xj1, xj2, xj3;
+         pos = (*idx++);
+         xj0 = x[pos+0];
+         xj1 = x[pos+1];
+         xj2 = x[pos+2];
+         xj3 = x[pos+3];
+         y = &out[i];
+         y[0] += (w[0]*xj0+w[1]*xj1+w[2]*xj2+w[3]*xj3);
+         y[1] += (w[4]*xj0+w[5]*xj1+w[6]*xj2+w[7]*xj3);
+         y[2] += (w[8]*xj0+w[9]*xj1+w[10]*xj2+w[11]*xj3);
+         y[3] += (w[12]*xj0+w[13]*xj1+w[14]*xj2+w[15]*xj3);
+         y[4] += (w[16]*xj0+w[17]*xj1+w[18]*xj2+w[19]*xj3);
+         y[5] += (w[20]*xj0+w[21]*xj1+w[22]*xj2+w[23]*xj3);
+         y[6] += (w[24]*xj0+w[25]*xj1+w[26]*xj2+w[27]*xj3);
+         y[7] += (w[28]*xj0+w[29]*xj1+w[30]*xj2+w[31]*xj3);
+         w += 32;
+      }
+   }
+   for (i=0;i<rows;i++) out[i] *= SCALE_1;
+}
+#else /*USE_SU_BIAS*/
+
+static inline void sgemv_accum8x4(float *out, const qweight *w, int rows, int cols, int col_stride, const float *_x)
+{
+   int i, j;
+   signed char x[MAX_INPUTS];
+   (void)col_stride;
+   for (i=0;i<rows;i++) out[i] *= SCALE;
+   for (i=0;i<cols;i++) x[i] = (int)floor(.5+127*_x[i]);
+   for (i=0;i<rows;i+=8)
+   {
+      for (j=0;j<cols;j+=4)
+      {
+         float * restrict y;
+         float xj0, xj1, xj2, xj3;
+         xj0 = x[j+0];
+         xj1 = x[j+1];
+         xj2 = x[j+2];
+         xj3 = x[j+3];
+         y = &out[i];
+         y[0] += (w[0]*xj0+w[1]*xj1+w[2]*xj2+w[3]*xj3);
+         y[1] += (w[4]*xj0+w[5]*xj1+w[6]*xj2+w[7]*xj3);
+         y[2] += (w[8]*xj0+w[9]*xj1+w[10]*xj2+w[11]*xj3);
+         y[3] += (w[12]*xj0+w[13]*xj1+w[14]*xj2+w[15]*xj3);
+         y[4] += (w[16]*xj0+w[17]*xj1+w[18]*xj2+w[19]*xj3);
+         y[5] += (w[20]*xj0+w[21]*xj1+w[22]*xj2+w[23]*xj3);
+         y[6] += (w[24]*xj0+w[25]*xj1+w[26]*xj2+w[27]*xj3);
+         y[7] += (w[28]*xj0+w[29]*xj1+w[30]*xj2+w[31]*xj3);
+         w += 32;
+      }
+   }
+   for (i=0;i<rows;i++) out[i] *= SCALE_1;
+}
+
+static inline void sparse_sgemv_accum8x4(float *out, const qweight *w, int rows, int cols, const int *idx, const float *_x)
+{
+   int i, j;
+   signed char x[MAX_INPUTS];
+   for (i=0;i<rows;i++) out[i] *= SCALE;
+   for (i=0;i<cols;i++) x[i] = floor(.5+127*_x[i]);
+   for (i=0;i<rows;i+=8)
+   {
+      int colblocks;
+      colblocks = *idx++;
+      for (j=0;j<colblocks;j++)
+      {
+         int pos;
+         float * restrict y;
+         int xj0, xj1, xj2, xj3;
+         pos = (*idx++);
+         xj0 = x[pos+0];
+         xj1 = x[pos+1];
+         xj2 = x[pos+2];
+         xj3 = x[pos+3];
+         y = &out[i];
+         y[0] += (w[0]*xj0+w[1]*xj1+w[2]*xj2+w[3]*xj3);
+         y[1] += (w[4]*xj0+w[5]*xj1+w[6]*xj2+w[7]*xj3);
+         y[2] += (w[8]*xj0+w[9]*xj1+w[10]*xj2+w[11]*xj3);
+         y[3] += (w[12]*xj0+w[13]*xj1+w[14]*xj2+w[15]*xj3);
+         y[4] += (w[16]*xj0+w[17]*xj1+w[18]*xj2+w[19]*xj3);
+         y[5] += (w[20]*xj0+w[21]*xj1+w[22]*xj2+w[23]*xj3);
+         y[6] += (w[24]*xj0+w[25]*xj1+w[26]*xj2+w[27]*xj3);
+         y[7] += (w[28]*xj0+w[29]*xj1+w[30]*xj2+w[31]*xj3);
+         w += 32;
+      }
+   }
+   for (i=0;i<rows;i++) out[i] *= SCALE_1;
+}
+#endif /*USE_SU_BIAS*/
+
+#else /*DOT_PROD*/
+
+#define sgemv_accum8x4 sgemv_accum
+
+
+static inline void sparse_sgemv_accum8x4(float *out, const qweight *w, int rows, int ignore, const int *idx, const float *x)
+{
+   int i, j;
+   (void)ignore;
+   for (i=0;i<rows;i+=8)
+   {
+      int cols;
+      cols = *idx++;
+      for (j=0;j<cols;j++)
+      {
+         int pos;
+         float * restrict y;
+         float xj0, xj1, xj2, xj3;
+         pos = (*idx++);
+         xj0 = x[pos+0];
+         xj1 = x[pos+1];
+         xj2 = x[pos+2];
+         xj3 = x[pos+3];
+         y = &out[i];
+         y[0] += w[0]*xj0;
+         y[1] += w[1]*xj0;
+         y[2] += w[2]*xj0;
+         y[3] += w[3]*xj0;
+         y[4] += w[4]*xj0;
+         y[5] += w[5]*xj0;
+         y[6] += w[6]*xj0;
+         y[7] += w[7]*xj0;
+
+         y[0] += w[8]*xj1;
+         y[1] += w[9]*xj1;
+         y[2] += w[10]*xj1;
+         y[3] += w[11]*xj1;
+         y[4] += w[12]*xj1;
+         y[5] += w[13]*xj1;
+         y[6] += w[14]*xj1;
+         y[7] += w[15]*xj1;
+
+         y[0] += w[16]*xj2;
+         y[1] += w[17]*xj2;
+         y[2] += w[18]*xj2;
+         y[3] += w[19]*xj2;
+         y[4] += w[20]*xj2;
+         y[5] += w[21]*xj2;
+         y[6] += w[22]*xj2;
+         y[7] += w[23]*xj2;
+
+         y[0] += w[24]*xj3;
+         y[1] += w[25]*xj3;
+         y[2] += w[26]*xj3;
+         y[3] += w[27]*xj3;
+         y[4] += w[28]*xj3;
+         y[5] += w[29]*xj3;
+         y[6] += w[30]*xj3;
+         y[7] += w[31]*xj3;
+         w += 32;
+      }
+   }
+}
+#endif /*DOT_PROD*/
+
+#endif /* __aarch64__ */