src/hotspot/cpu/x86/assembler_x86.hpp

rev 50140 : Vector cast support

@@ -1092,13 +1092,15 @@
   void cvtsi2ssq(XMMRegister dst, Register src);
   void cvtsi2ssq(XMMRegister dst, Address src);
 
   // Convert Packed Signed Doubleword Integers to Packed Double-Precision Floating-Point Value
   void cvtdq2pd(XMMRegister dst, XMMRegister src);
+  void vcvtdq2pd(XMMRegister dst, XMMRegister src, int vector_len);
 
   // Convert Packed Signed Doubleword Integers to Packed Single-Precision Floating-Point Value
   void cvtdq2ps(XMMRegister dst, XMMRegister src);
+  void vcvtdq2ps(XMMRegister dst, XMMRegister src, int vector_len);
 
   // Convert Scalar Single-Precision Floating-Point Value to Scalar Double-Precision Floating-Point Value
   void cvtss2sd(XMMRegister dst, XMMRegister src);
   void cvtss2sd(XMMRegister dst, Address src);
 

@@ -1109,15 +1111,30 @@
 
   // Convert with Truncation Scalar Single-Precision Floating-Point Value to Doubleword Integer
   void cvttss2sil(Register dst, XMMRegister src);
   void cvttss2siq(Register dst, XMMRegister src);
 
+  // Convert vector double to int
   void cvttpd2dq(XMMRegister dst, XMMRegister src);
 
   // Convert vector float and double
   void vcvtps2pd(XMMRegister dst, XMMRegister src, int vector_len);
   void evcvtps2pd(XMMRegister dst, XMMRegister src, int vector_len);
+  void vcvtpd2ps(XMMRegister dst, XMMRegister src, int vector_len);
+  void evcvtpd2ps(XMMRegister dst, XMMRegister src, int vector_len);
+
+  // Convert vector long to vector FP
+  void evcvtqq2ps(XMMRegister dst, XMMRegister src, int vector_len);
+  void evcvtqq2pd(XMMRegister dst, XMMRegister src, int vector_len);
+
+  // Evex casts with truncation
+  void evpmovwb(XMMRegister dst, XMMRegister src, int vector_len);
+  void evpmovdw(XMMRegister dst, XMMRegister src, int vector_len);
+  void evpmovdb(XMMRegister dst, XMMRegister src, int vector_len);
+  void evpmovqd(XMMRegister dst, XMMRegister src, int vector_len);
+  void evpmovqb(XMMRegister dst, XMMRegister src, int vector_len);
+  void evpmovqw(XMMRegister dst, XMMRegister src, int vector_len);
 
   //Abs of packed Integer values
   void pabsb(XMMRegister dst, XMMRegister src);
   void pabsw(XMMRegister dst, XMMRegister src);
   void pabsd(XMMRegister dst, XMMRegister src);

@@ -1619,10 +1636,12 @@
   void vpermq(XMMRegister dst, XMMRegister src, int imm8);
   void vpermd(XMMRegister dst,  XMMRegister nds, XMMRegister src);
   void vpermd(XMMRegister dst,  XMMRegister nds, Address src);
   void vperm2i128(XMMRegister dst,  XMMRegister nds, XMMRegister src, int imm8);
   void vperm2f128(XMMRegister dst, XMMRegister nds, XMMRegister src, int imm8);
+  void vpermilps(XMMRegister dst, XMMRegister src, int imm8, int vector_len);
+  void vpermpd(XMMRegister dst, XMMRegister src, int imm8, int vector_len);
 
   void pause();
 
   // Undefined Instruction
   void ud2();

@@ -1704,10 +1723,13 @@
   void pmovsxbd(XMMRegister dst, XMMRegister src);
   void pmovsxbq(XMMRegister dst, XMMRegister src);
   void vpmovsxbd(XMMRegister dst, XMMRegister src, int vector_len);
   void vpmovsxbq(XMMRegister dst, XMMRegister src, int vector_len);
   void vpmovsxbw(XMMRegister dst, XMMRegister src, int vector_len);
+  void vpmovsxwd(XMMRegister dst, XMMRegister src, int vector_len);
+  void vpmovsxwq(XMMRegister dst, XMMRegister src, int vector_len);
+  void vpmovsxdq(XMMRegister dst, XMMRegister src, int vector_len);
 
   void evpmovwb(Address dst, XMMRegister src, int vector_len);
   void evpmovwb(Address dst, KRegister mask, XMMRegister src, int vector_len);
 
 #ifndef _LP64 // no 32bit push/pop on amd64