1651 void pinsrb(XMMRegister dst, Address src, int imm8); 1652 // SSE 2 insert 1653 void pinsrw(XMMRegister dst, Register src, int imm8); 1654 void pinsrw(XMMRegister dst, Address src, int imm8); 1655 1656 // SSE4.1 packed move 1657 void pmovzxbw(XMMRegister dst, XMMRegister src); 1658 void pmovzxbw(XMMRegister dst, Address src); 1659 1660 void vpmovzxbw( XMMRegister dst, Address src, int vector_len); 1661 void vpmovzxbw(XMMRegister dst, XMMRegister src, int vector_len); 1662 void evpmovzxbw(XMMRegister dst, KRegister mask, Address src, int vector_len); 1663 1664 void evpmovwb(Address dst, XMMRegister src, int vector_len); 1665 void evpmovwb(Address dst, KRegister mask, XMMRegister src, int vector_len); 1666 1667 void vpmovzxwd(XMMRegister dst, XMMRegister src, int vector_len); 1668 1669 void evpmovdb(Address dst, XMMRegister src, int vector_len); 1670 1671 #ifndef _LP64 // no 32bit push/pop on amd64 1672 void popl(Address dst); 1673 #endif 1674 1675 #ifdef _LP64 1676 void popq(Address dst); 1677 #endif 1678 1679 void popcntl(Register dst, Address src); 1680 void popcntl(Register dst, Register src); 1681 1682 void vpopcntd(XMMRegister dst, XMMRegister src, int vector_len); 1683 1684 #ifdef _LP64 1685 void popcntq(Register dst, Address src); 1686 void popcntq(Register dst, Register src); 1687 #endif 1688 1689 // Prefetches (SSE, SSE2, 3DNOW only) 1690 | 1651 void pinsrb(XMMRegister dst, Address src, int imm8); 1652 // SSE 2 insert 1653 void pinsrw(XMMRegister dst, Register src, int imm8); 1654 void pinsrw(XMMRegister dst, Address src, int imm8); 1655 1656 // SSE4.1 packed move 1657 void pmovzxbw(XMMRegister dst, XMMRegister src); 1658 void pmovzxbw(XMMRegister dst, Address src); 1659 1660 void vpmovzxbw( XMMRegister dst, Address src, int vector_len); 1661 void vpmovzxbw(XMMRegister dst, XMMRegister src, int vector_len); 1662 void evpmovzxbw(XMMRegister dst, KRegister mask, Address src, int vector_len); 1663 1664 void evpmovwb(Address dst, XMMRegister src, int vector_len); 1665 void evpmovwb(Address dst, KRegister mask, XMMRegister src, int vector_len); 1666 1667 void vpmovzxwd(XMMRegister dst, XMMRegister src, int vector_len); 1668 1669 void evpmovdb(Address dst, XMMRegister src, int vector_len); 1670 1671 // Multiply add 1672 void pmaddwd(XMMRegister dst, XMMRegister src); 1673 void vpmaddwd(XMMRegister dst, XMMRegister nds, XMMRegister src, int vector_len); 1674 // Multiply add accumulate 1675 void evpdpwssd(XMMRegister dst, XMMRegister nds, XMMRegister src, int vector_len); 1676 1677 #ifndef _LP64 // no 32bit push/pop on amd64 1678 void popl(Address dst); 1679 #endif 1680 1681 #ifdef _LP64 1682 void popq(Address dst); 1683 #endif 1684 1685 void popcntl(Register dst, Address src); 1686 void popcntl(Register dst, Register src); 1687 1688 void vpopcntd(XMMRegister dst, XMMRegister src, int vector_len); 1689 1690 #ifdef _LP64 1691 void popcntq(Register dst, Address src); 1692 void popcntq(Register dst, Register src); 1693 #endif 1694 1695 // Prefetches (SSE, SSE2, 3DNOW only) 1696 |