optimized zgemv_n kernel for sandybridge

diff --git a/kernel/x86_64/KERNEL.SANDYBRIDGE b/kernel/x86_64/KERNEL.SANDYBRIDGE
index d4fbca7..b654d35 100644 (file)
--- a/kernel/x86_64/KERNEL.SANDYBRIDGE
+++ b/kernel/x86_64/KERNEL.SANDYBRIDGE
@@ -1,6 +1,9 @@
 SGEMVNKERNEL = sgemv_n.c
 SGEMVTKERNEL = sgemv_t.c
 
+ZGEMVNKERNEL = zgemv_n.c
+
+
 SGEMMKERNEL    =  sgemm_kernel_16x4_sandy.S
 SGEMMINCOPY    =  ../generic/gemm_ncopy_16.c
 SGEMMITCOPY    =  ../generic/gemm_tcopy_16.c

diff --git a/kernel/x86_64/zgemv_n.c b/kernel/x86_64/zgemv_n.c
index 7b89070..9098368 100644 (file)
--- a/kernel/x86_64/zgemv_n.c
+++ b/kernel/x86_64/zgemv_n.c
@@ -31,9 +31,12 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 #if defined(HASWELL)
 #include "zgemv_n_microk_haswell-2.c"
+#elif defined(SANDYBRIDGE)
+#include "zgemv_n_microk_sandy-2.c"
 #endif
 
 
+
 #define NBMAX 1024
 
 #ifndef HAVE_KERNEL_16x4

diff --git a/kernel/x86_64/zgemv_n_microk_sandy-2.c b/kernel/x86_64/zgemv_n_microk_sandy-2.c
index f90e221..352c60f 100644 (file)
--- a/kernel/x86_64/zgemv_n_microk_sandy-2.c
+++ b/kernel/x86_64/zgemv_n_microk_sandy-2.c
@@ -50,39 +50,42 @@ static void zgemv_kernel_16x4( BLASLONG n, FLOAT **ap, FLOAT *x, FLOAT *y)
        ".align 16                                      \n\t"
        ".L01LOOP%=:                                    \n\t"
 
+        "prefetcht0      256(%4,%0,8)                   \n\t"
        "vmovups        (%4,%0,8), %%ymm8               \n\t" // 2 complex values form a0
        "vmovups      32(%4,%0,8), %%ymm9               \n\t" // 2 complex values form a0
 
        "vmulpd           %%ymm8 , %%ymm0 , %%ymm12     \n\t"
        "vmulpd           %%ymm8 , %%ymm1 , %%ymm13     \n\t"
+        "prefetcht0      256(%5,%0,8)                   \n\t"
        "vmulpd           %%ymm9 , %%ymm0 , %%ymm14     \n\t"
-       "vmulpd           %%ymm9 , %%ymm1 , %%ymm15     \n\t"
-
        "vmovups        (%5,%0,8), %%ymm8               \n\t" // 2 complex values form a0
+       "vmulpd           %%ymm9 , %%ymm1 , %%ymm15     \n\t"
        "vmovups      32(%5,%0,8), %%ymm9               \n\t" // 2 complex values form a0
 
        "vmulpd           %%ymm8 , %%ymm2 , %%ymm10     \n\t"
        "vaddpd           %%ymm12, %%ymm10, %%ymm12     \n\t"
        "vmulpd           %%ymm8 , %%ymm3 , %%ymm11     \n\t"
        "vaddpd           %%ymm13, %%ymm11, %%ymm13     \n\t"
+        "prefetcht0      256(%6,%0,8)                   \n\t"
        "vmulpd           %%ymm9 , %%ymm2 , %%ymm10     \n\t"
        "vaddpd           %%ymm14, %%ymm10, %%ymm14     \n\t"
+       "vmovups        (%6,%0,8), %%ymm8               \n\t" // 2 complex values form a0
        "vmulpd           %%ymm9 , %%ymm3 , %%ymm11     \n\t"
        "vaddpd           %%ymm15, %%ymm11, %%ymm15     \n\t"
 
-       "vmovups        (%6,%0,8), %%ymm8               \n\t" // 2 complex values form a0
        "vmovups      32(%6,%0,8), %%ymm9               \n\t" // 2 complex values form a0
 
        "vmulpd           %%ymm8 , %%ymm4 , %%ymm10     \n\t"
        "vaddpd           %%ymm12, %%ymm10, %%ymm12     \n\t"
        "vmulpd           %%ymm8 , %%ymm5 , %%ymm11     \n\t"
        "vaddpd           %%ymm13, %%ymm11, %%ymm13     \n\t"
+        "prefetcht0      256(%7,%0,8)                   \n\t"
        "vmulpd           %%ymm9 , %%ymm4 , %%ymm10     \n\t"
        "vaddpd           %%ymm14, %%ymm10, %%ymm14     \n\t"
+       "vmovups        (%7,%0,8), %%ymm8               \n\t" // 2 complex values form a0
        "vmulpd           %%ymm9 , %%ymm5 , %%ymm11     \n\t"
        "vaddpd           %%ymm15, %%ymm11, %%ymm15     \n\t"
 
-       "vmovups        (%7,%0,8), %%ymm8               \n\t" // 2 complex values form a0
        "vmovups      32(%7,%0,8), %%ymm9               \n\t" // 2 complex values form a0
 
        "vmulpd           %%ymm8 , %%ymm6 , %%ymm10     \n\t"
@@ -94,7 +97,7 @@ static void zgemv_kernel_16x4( BLASLONG n, FLOAT **ap, FLOAT *x, FLOAT *y)
        "vmulpd           %%ymm9 , %%ymm7 , %%ymm11     \n\t"
        "vaddpd           %%ymm15, %%ymm11, %%ymm15     \n\t"
 
-       "prefetcht0      192(%3,%0,8)                   \n\t"
+       "prefetcht0      256(%3,%0,8)                   \n\t"
        "vmovups          (%3,%0,8),  %%ymm10           \n\t"
        "vmovups        32(%3,%0,8),  %%ymm11           \n\t"