Merge pull request #4996 from iha-taisei/sdgemv_sve_unroll

Loop-unrolled transposed [SD]GEMV kernels for A64FX and Neoverse V1

kernel/arm64/KERNEL.A64FX

@@ -2,5 +2,5 @@ include $(KERNELDIR)/KERNEL.ARMV8SVE
 
 SGEMVNKERNEL = gemv_n_sve.c
 DGEMVNKERNEL = gemv_n_sve.c
-SGEMVTKERNEL = gemv_t_sve.c
-DGEMVTKERNEL = gemv_t_sve.c
+SGEMVTKERNEL = gemv_t_sve_v4x3.c
+DGEMVTKERNEL = gemv_t_sve_v4x3.c

kernel/arm64/KERNEL.NEOVERSEV1

@@ -1,4 +1,4 @@
 include $(KERNELDIR)/KERNEL.ARMV8SVE
 
-SGEMVTKERNEL = gemv_t_sve.c
-DGEMVTKERNEL = gemv_t_sve.c
+SGEMVTKERNEL = gemv_t_sve_v1x3.c
+DGEMVTKERNEL = gemv_t_sve_v1x3.c

kernel/arm64/gemv_t_sve_v1x3.c

@@ -0,0 +1,152 @@
+/***************************************************************************
+Copyright (c) 2024, The OpenBLAS Project
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+   1. Redistributions of source code must retain the above copyright
+      notice, this list of conditions and the following disclaimer.
+
+   2. 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.
+   3. Neither the name of the OpenBLAS project 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.
+*****************************************************************************/
+
+#include <arm_sve.h>
+
+#include "common.h"
+
+#ifdef DOUBLE
+#define SV_COUNT svcntd
+#define SV_TYPE svfloat64_t
+#define SV_TRUE svptrue_b64
+#define SV_WHILE svwhilelt_b64_s64
+#define SV_DUP svdup_f64
+#else
+#define SV_COUNT svcntw
+#define SV_TYPE svfloat32_t
+#define SV_TRUE svptrue_b32
+#define SV_WHILE svwhilelt_b32_s64
+#define SV_DUP svdup_f32
+#endif
+
+int CNAME(BLASLONG m, BLASLONG n, BLASLONG dummy1, FLOAT alpha, FLOAT *a,
+          BLASLONG lda, FLOAT *x, BLASLONG inc_x, FLOAT *y, BLASLONG inc_y,
+          FLOAT *buffer)
+{
+  BLASLONG i;
+  BLASLONG ix,iy;
+  BLASLONG j;
+  FLOAT *a_ptr;
+  FLOAT temp;
+
+  iy = 0;
+
+  if (inc_x == 1) {
+    BLASLONG width = (n + 3 - 1) / 3;
+
+    FLOAT *a0_ptr = a + lda * width * 0;
+    FLOAT *a1_ptr = a + lda * width * 1;
+    FLOAT *a2_ptr = a + lda * width * 2;
+
+    FLOAT *y0_ptr = y + inc_y * width * 0;
+    FLOAT *y1_ptr = y + inc_y * width * 1;
+    FLOAT *y2_ptr = y + inc_y * width * 2;
+
+    for (j = 0; j < width; j++) {
+      svbool_t pg00 = ((j + width * 0) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg01 = ((j + width * 1) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg02 = ((j + width * 2) < n) ? SV_TRUE() : svpfalse();
+
+      SV_TYPE temp00_vec = SV_DUP(0.0);
+      SV_TYPE temp01_vec = SV_DUP(0.0);
+      SV_TYPE temp02_vec = SV_DUP(0.0);
+
+      i = 0;
+      BLASLONG sve_size = SV_COUNT();
+      while ((i + sve_size * 1 - 1) < m) {
+        SV_TYPE x0_vec = svld1_vnum(SV_TRUE(), x + i, 0);
+
+        SV_TYPE a00_vec = svld1_vnum(pg00, a0_ptr + i, 0);
+        SV_TYPE a01_vec = svld1_vnum(pg01, a1_ptr + i, 0);
+        SV_TYPE a02_vec = svld1_vnum(pg02, a2_ptr + i, 0);
+
+        temp00_vec = svmla_m(pg00, temp00_vec, a00_vec, x0_vec);
+        temp01_vec = svmla_m(pg01, temp01_vec, a01_vec, x0_vec);
+        temp02_vec = svmla_m(pg02, temp02_vec, a02_vec, x0_vec);
+
+        i += sve_size * 1;
+      }
+
+      if (i < m) {
+        svbool_t pg0 = SV_WHILE(i + sve_size * 0, m);
+
+        pg00 = svand_z(SV_TRUE(), pg0, pg00);
+        pg01 = svand_z(SV_TRUE(), pg0, pg01);
+        pg02 = svand_z(SV_TRUE(), pg0, pg02);
+
+        SV_TYPE x0_vec = svld1_vnum(pg0, x + i, 0);
+
+        SV_TYPE a00_vec = svld1_vnum(pg00, a0_ptr + i, 0);
+        SV_TYPE a01_vec = svld1_vnum(pg01, a1_ptr + i, 0);
+        SV_TYPE a02_vec = svld1_vnum(pg02, a2_ptr + i, 0);
+
+        temp00_vec = svmla_m(pg00, temp00_vec, a00_vec, x0_vec);
+        temp01_vec = svmla_m(pg01, temp01_vec, a01_vec, x0_vec);
+        temp02_vec = svmla_m(pg02, temp02_vec, a02_vec, x0_vec);
+      }
+
+      if ((j + width * 0) < n) {
+        temp = svaddv(SV_TRUE(), temp00_vec);
+        y0_ptr[iy] += alpha * temp;
+      }
+      if ((j + width * 1) < n) {
+        temp = svaddv(SV_TRUE(), temp01_vec);
+        y1_ptr[iy] += alpha * temp;
+      }
+      if ((j + width * 2) < n) {
+        temp = svaddv(SV_TRUE(), temp02_vec);
+        y2_ptr[iy] += alpha * temp;
+      }
+      iy += inc_y;
+
+      a0_ptr += lda;
+      a1_ptr += lda;
+      a2_ptr += lda;
+    }
+
+    return(0);
+  }
+
+  a_ptr = a;
+  for (j = 0; j < n; j++) {
+    temp = 0.0;
+    ix = 0;
+    for (i = 0; i < m; i++) {
+      temp += a_ptr[i] * x[ix];
+      ix += inc_x;
+    }
+    y[iy] += alpha * temp;
+    iy += inc_y;
+    a_ptr += lda;
+  }
+  return(0);
+}

kernel/arm64/gemv_t_sve_v4x3.c

@@ -0,0 +1,234 @@
+/***************************************************************************
+Copyright (c) 2024, The OpenBLAS Project
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+   1. Redistributions of source code must retain the above copyright
+      notice, this list of conditions and the following disclaimer.
+
+   2. 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.
+   3. Neither the name of the OpenBLAS project 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.
+*****************************************************************************/
+
+#include <arm_sve.h>
+
+#include "common.h"
+
+#ifdef DOUBLE
+#define SV_COUNT svcntd
+#define SV_TYPE svfloat64_t
+#define SV_TRUE svptrue_b64
+#define SV_WHILE svwhilelt_b64_s64
+#define SV_DUP svdup_f64
+#else
+#define SV_COUNT svcntw
+#define SV_TYPE svfloat32_t
+#define SV_TRUE svptrue_b32
+#define SV_WHILE svwhilelt_b32_s64
+#define SV_DUP svdup_f32
+#endif
+
+int CNAME(BLASLONG m, BLASLONG n, BLASLONG dummy1, FLOAT alpha, FLOAT *a,
+          BLASLONG lda, FLOAT *x, BLASLONG inc_x, FLOAT *y, BLASLONG inc_y,
+          FLOAT *buffer)
+{
+  BLASLONG i;
+  BLASLONG ix,iy;
+  BLASLONG j;
+  FLOAT *a_ptr;
+  FLOAT temp;
+
+  iy = 0;
+
+  if (inc_x == 1) {
+    BLASLONG width = (n + 3 - 1) / 3;
+
+    FLOAT *a0_ptr = a + lda * width * 0;
+    FLOAT *a1_ptr = a + lda * width * 1;
+    FLOAT *a2_ptr = a + lda * width * 2;
+
+    FLOAT *y0_ptr = y + inc_y * width * 0;
+    FLOAT *y1_ptr = y + inc_y * width * 1;
+    FLOAT *y2_ptr = y + inc_y * width * 2;
+
+    for (j = 0; j < width; j++) {
+      svbool_t pg00 = ((j + width * 0) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg10 = ((j + width * 0) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg20 = ((j + width * 0) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg30 = ((j + width * 0) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg01 = ((j + width * 1) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg11 = ((j + width * 1) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg21 = ((j + width * 1) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg31 = ((j + width * 1) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg02 = ((j + width * 2) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg12 = ((j + width * 2) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg22 = ((j + width * 2) < n) ? SV_TRUE() : svpfalse();
+      svbool_t pg32 = ((j + width * 2) < n) ? SV_TRUE() : svpfalse();
+
+      SV_TYPE temp00_vec = SV_DUP(0.0);
+      SV_TYPE temp10_vec = SV_DUP(0.0);
+      SV_TYPE temp20_vec = SV_DUP(0.0);
+      SV_TYPE temp30_vec = SV_DUP(0.0);
+      SV_TYPE temp01_vec = SV_DUP(0.0);
+      SV_TYPE temp11_vec = SV_DUP(0.0);
+      SV_TYPE temp21_vec = SV_DUP(0.0);
+      SV_TYPE temp31_vec = SV_DUP(0.0);
+      SV_TYPE temp02_vec = SV_DUP(0.0);
+      SV_TYPE temp12_vec = SV_DUP(0.0);
+      SV_TYPE temp22_vec = SV_DUP(0.0);
+      SV_TYPE temp32_vec = SV_DUP(0.0);
+
+      i = 0;
+      BLASLONG sve_size = SV_COUNT();
+      while ((i + sve_size * 4 - 1) < m) {
+        SV_TYPE x0_vec = svld1_vnum(SV_TRUE(), x + i, 0);
+        SV_TYPE x1_vec = svld1_vnum(SV_TRUE(), x + i, 1);
+        SV_TYPE x2_vec = svld1_vnum(SV_TRUE(), x + i, 2);
+        SV_TYPE x3_vec = svld1_vnum(SV_TRUE(), x + i, 3);
+
+        SV_TYPE a00_vec = svld1_vnum(pg00, a0_ptr + i, 0);
+        SV_TYPE a10_vec = svld1_vnum(pg10, a0_ptr + i, 1);
+        SV_TYPE a20_vec = svld1_vnum(pg20, a0_ptr + i, 2);
+        SV_TYPE a30_vec = svld1_vnum(pg30, a0_ptr + i, 3);
+        SV_TYPE a01_vec = svld1_vnum(pg01, a1_ptr + i, 0);
+        SV_TYPE a11_vec = svld1_vnum(pg11, a1_ptr + i, 1);
+        SV_TYPE a21_vec = svld1_vnum(pg21, a1_ptr + i, 2);
+        SV_TYPE a31_vec = svld1_vnum(pg31, a1_ptr + i, 3);
+        SV_TYPE a02_vec = svld1_vnum(pg02, a2_ptr + i, 0);
+        SV_TYPE a12_vec = svld1_vnum(pg12, a2_ptr + i, 1);
+        SV_TYPE a22_vec = svld1_vnum(pg22, a2_ptr + i, 2);
+        SV_TYPE a32_vec = svld1_vnum(pg32, a2_ptr + i, 3);
+
+        temp00_vec = svmla_m(pg00, temp00_vec, a00_vec, x0_vec);
+        temp10_vec = svmla_m(pg10, temp10_vec, a10_vec, x1_vec);
+        temp20_vec = svmla_m(pg20, temp20_vec, a20_vec, x2_vec);
+        temp30_vec = svmla_m(pg30, temp30_vec, a30_vec, x3_vec);
+        temp01_vec = svmla_m(pg01, temp01_vec, a01_vec, x0_vec);
+        temp11_vec = svmla_m(pg11, temp11_vec, a11_vec, x1_vec);
+        temp21_vec = svmla_m(pg21, temp21_vec, a21_vec, x2_vec);
+        temp31_vec = svmla_m(pg31, temp31_vec, a31_vec, x3_vec);
+        temp02_vec = svmla_m(pg02, temp02_vec, a02_vec, x0_vec);
+        temp12_vec = svmla_m(pg12, temp12_vec, a12_vec, x1_vec);
+        temp22_vec = svmla_m(pg22, temp22_vec, a22_vec, x2_vec);
+        temp32_vec = svmla_m(pg32, temp32_vec, a32_vec, x3_vec);
+
+        i += sve_size * 4;
+      }
+
+      if (i < m) {
+        svbool_t pg0 = SV_WHILE(i + sve_size * 0, m);
+        svbool_t pg1 = SV_WHILE(i + sve_size * 1, m);
+        svbool_t pg2 = SV_WHILE(i + sve_size * 2, m);
+        svbool_t pg3 = SV_WHILE(i + sve_size * 3, m);
+
+        pg00 = svand_z(SV_TRUE(), pg0, pg00);
+        pg10 = svand_z(SV_TRUE(), pg1, pg10);
+        pg20 = svand_z(SV_TRUE(), pg2, pg20);
+        pg30 = svand_z(SV_TRUE(), pg3, pg30);
+        pg01 = svand_z(SV_TRUE(), pg0, pg01);
+        pg11 = svand_z(SV_TRUE(), pg1, pg11);
+        pg21 = svand_z(SV_TRUE(), pg2, pg21);
+        pg31 = svand_z(SV_TRUE(), pg3, pg31);
+        pg02 = svand_z(SV_TRUE(), pg0, pg02);
+        pg12 = svand_z(SV_TRUE(), pg1, pg12);
+        pg22 = svand_z(SV_TRUE(), pg2, pg22);
+        pg32 = svand_z(SV_TRUE(), pg3, pg32);
+
+        SV_TYPE x0_vec = svld1_vnum(pg0, x + i, 0);
+        SV_TYPE x1_vec = svld1_vnum(pg1, x + i, 1);
+        SV_TYPE x2_vec = svld1_vnum(pg2, x + i, 2);
+        SV_TYPE x3_vec = svld1_vnum(pg3, x + i, 3);
+
+        SV_TYPE a00_vec = svld1_vnum(pg00, a0_ptr + i, 0);
+        SV_TYPE a10_vec = svld1_vnum(pg10, a0_ptr + i, 1);
+        SV_TYPE a20_vec = svld1_vnum(pg20, a0_ptr + i, 2);
+        SV_TYPE a30_vec = svld1_vnum(pg30, a0_ptr + i, 3);
+        SV_TYPE a01_vec = svld1_vnum(pg01, a1_ptr + i, 0);
+        SV_TYPE a11_vec = svld1_vnum(pg11, a1_ptr + i, 1);
+        SV_TYPE a21_vec = svld1_vnum(pg21, a1_ptr + i, 2);
+        SV_TYPE a31_vec = svld1_vnum(pg31, a1_ptr + i, 3);
+        SV_TYPE a02_vec = svld1_vnum(pg02, a2_ptr + i, 0);
+        SV_TYPE a12_vec = svld1_vnum(pg12, a2_ptr + i, 1);
+        SV_TYPE a22_vec = svld1_vnum(pg22, a2_ptr + i, 2);
+        SV_TYPE a32_vec = svld1_vnum(pg32, a2_ptr + i, 3);
+
+        temp00_vec = svmla_m(pg00, temp00_vec, a00_vec, x0_vec);
+        temp10_vec = svmla_m(pg10, temp10_vec, a10_vec, x1_vec);
+        temp20_vec = svmla_m(pg20, temp20_vec, a20_vec, x2_vec);
+        temp30_vec = svmla_m(pg30, temp30_vec, a30_vec, x3_vec);
+        temp01_vec = svmla_m(pg01, temp01_vec, a01_vec, x0_vec);
+        temp11_vec = svmla_m(pg11, temp11_vec, a11_vec, x1_vec);
+        temp21_vec = svmla_m(pg21, temp21_vec, a21_vec, x2_vec);
+        temp31_vec = svmla_m(pg31, temp31_vec, a31_vec, x3_vec);
+        temp02_vec = svmla_m(pg02, temp02_vec, a02_vec, x0_vec);
+        temp12_vec = svmla_m(pg12, temp12_vec, a12_vec, x1_vec);
+        temp22_vec = svmla_m(pg22, temp22_vec, a22_vec, x2_vec);
+        temp32_vec = svmla_m(pg32, temp32_vec, a32_vec, x3_vec);
+      }
+
+      temp00_vec = svadd_x(SV_TRUE(), temp00_vec, temp10_vec);
+      temp01_vec = svadd_x(SV_TRUE(), temp01_vec, temp11_vec);
+      temp02_vec = svadd_x(SV_TRUE(), temp02_vec, temp12_vec);
+      temp20_vec = svadd_x(SV_TRUE(), temp20_vec, temp30_vec);
+      temp21_vec = svadd_x(SV_TRUE(), temp21_vec, temp31_vec);
+      temp22_vec = svadd_x(SV_TRUE(), temp22_vec, temp32_vec);
+      temp00_vec = svadd_x(SV_TRUE(), temp00_vec, temp20_vec);
+      temp01_vec = svadd_x(SV_TRUE(), temp01_vec, temp21_vec);
+      temp02_vec = svadd_x(SV_TRUE(), temp02_vec, temp22_vec);
+
+      if ((j + width * 0) < n) {
+        temp = svaddv(SV_TRUE(), temp00_vec);
+        y0_ptr[iy] += alpha * temp;
+      }
+      if ((j + width * 1) < n) {
+        temp = svaddv(SV_TRUE(), temp01_vec);
+        y1_ptr[iy] += alpha * temp;
+      }
+      if ((j + width * 2) < n) {
+        temp = svaddv(SV_TRUE(), temp02_vec);
+        y2_ptr[iy] += alpha * temp;
+      }
+      iy += inc_y;
+
+      a0_ptr += lda;
+      a1_ptr += lda;
+      a2_ptr += lda;
+    }
+
+    return(0);
+  }
+
+  a_ptr = a;
+  for (j = 0; j < n; j++) {
+    temp = 0.0;
+    ix = 0;
+    for (i = 0; i < m; i++) {
+      temp += a_ptr[i] * x[ix];
+      ix += inc_x;
+    }
+    y[iy] += alpha * temp;
+    iy += inc_y;
+    a_ptr += lda;
+  }
+  return(0);
+}