From d850c3b7f47e58556c160f9d03ea20aa52452020 Mon Sep 17 00:00:00 2001
From: rodri <rgl@antares-labs.eu>
Date: Fri, 24 Nov 2023 15:39:06 +0000
Subject: add more avx instructions and a bench9 benchmark file.

---
 sse.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

(limited to 'sse.h')

diff --git a/sse.h b/sse.h
index ffad543..6a03c9e 100644
--- a/sse.h
+++ b/sse.h
@@ -41,6 +41,8 @@
 			BYTE $(((~r)<<7)|((~v)<<3)|((l)<<2)|(p))
 #define VOP(o, m, ro, rm)	BYTE $(o);	\
 			BYTE $(((m)<<6)|((ro)<<3)|(rm))
+#define VOPi(o, m, ro, rm, i)	VOP((o), (m), (ro), (rm));	\
+			BYTE $(i)
 
 /* MOVLPD */
 //opcode = 660F12
@@ -65,6 +67,15 @@
 //imm8   = 0011 0001
 #define DPPD(s, d) OP4i(0x413A, 0x3, (d), (s), 0x31)
 
+/* VMOVAPD */
+#define VMOVUPD_128mr(off, s, d) VEX3(0,0,0,VEX_m_0F,0,0,VEX_L_128,VEX_p_66);			\
+			VOPi(0x10, 0x1, (d), (s), (off))
+#define VMOVAPD_128rr(s, d) VEX3(0,0,0,VEX_m_0F,0,0,VEX_L_128,VEX_p_66);			\
+			VOP(0x28, 0x3, (d), (s))
+/* VDPPD */
+#define VDPPD(s0, s1, d) VEX3(0,0,0,VEX_m_0F3A,0,(s0),VEX_L_128,VEX_p_66);		\
+			VOPi(0x41, 0x3, (d), (s1), 0x31)
+
 /* VFMADD231SD (128 bit) */
 #define VFMADD231SD(s0, s1, d) VEX3(0,0,0,VEX_m_0F38,1,(s0),VEX_L_128,VEX_p_66);	\
 			VOP(0xB9, 0x3, (d), (s1))
-- 
cgit v1.2.3