update to v5.99

4 files changed, 14 insertions, 4 deletions

diff --git a/readme.md b/readme.md
index 51b39a4..93805dd 100644
--- a/readme.md
+++ b/readme.md
@@ -1,6 +1,6 @@
 [![Build Status](https://travis-ci.org/herumi/xbyak.png)](https://travis-ci.org/herumi/xbyak)
 
-# Xbyak 5.98 ; JIT assembler for x86(IA32), x64(AMD64, x86-64) by C++
+# Xbyak 5.99 ; JIT assembler for x86(IA32), x64(AMD64, x86-64) by C++
 
 ## Abstract
 
@@ -16,6 +16,7 @@ Use `and_()`, `or_()`, ... instead of `and()`, `or()`.
 If you want to use them, then specify `-fno-operator-names` option to gcc/clang.
 
 ### News
+- vnni instructions such as vpdpbusd supports vex encoding.
 - (break backward compatibility) `push(byte, imm)` (resp. `push(word, imm)`) forces to cast `imm` to 8(resp. 16) bit.
 - (Windows) `#include <winsock2.h>` has been removed from xbyak.h, so add it explicitly if you need it.
 - support exception-less mode see. [Exception-less mode](#exception-less-mode)
@@ -154,6 +155,10 @@ vcvtpd2dq xmm19, [eax+32]{1to4}         --> vcvtpd2dq(xmm19, yword_b [eax+32]);
 vfpclassps k5{k3}, zword [rax+64], 5    --> vfpclassps(k5|k3, zword [rax+64], 5); // specify m512
 vfpclasspd k5{k3}, [rax+64]{1to2}, 5    --> vfpclasspd(k5|k3, xword_b [rax+64], 5); // broadcast 64-bit to 128-bit
 vfpclassps k5{k3}, [rax+64]{1to4}, 5    --> vfpclassps(k5|k3, yword_b [rax+64], 5); // broadcast 64-bit to 256-bit
+
+vpdpbusd(xm0, xm1, xm2); // default encoding is EVEX
+vpdpbusd(xm0, xm1, xm2, EvexEncoding); // same as the above
+vpdpbusd(xm0, xm1, xm2, VexEncoding); // VEX encoding
 ```
 ### Remark
 * `k1`, ..., `k7` are opmask registers.
@@ -438,6 +443,7 @@ modified new BSD License
 http://opensource.org/licenses/BSD-3-Clause
 
 ## History
+* 2020/Oct/19 ver 5.99 support VNNI instructions(Thanks to akharito)
 * 2020/Oct/17 ver 5.98 support the form of [scale * reg]
 * 2020/Sep/08 ver 5.97 replace uint32 with uint32_t etc.
 * 2020/Aug/28 ver 5.95 some constructors of register classes support constexpr if C++14 or later
diff --git a/readme.txt b/readme.txt
index 9126051..29bd659 100644
--- a/readme.txt
+++ b/readme.txt
@@ -1,5 +1,5 @@
 

-    C++用x86(IA-32), x64(AMD64, x86-64) JITアセンブラ Xbyak 5.98

+    C++用x86(IA-32), x64(AMD64, x86-64) JITアセンブラ Xbyak 5.99

 

 -----------------------------------------------------------------------------

 ◎概要

@@ -163,6 +163,9 @@ vfpclassps k5{k3}, zword [rax+64], 5    --> vfpclassps(k5|k3, zword [rax+64], 5)
 vfpclasspd k5{k3}, [rax+64]{1to2}, 5    --> vfpclasspd(k5|k3, xword_b [rax+64], 5); // broadcast 64-bit to 128-bit

 vfpclassps k5{k3}, [rax+64]{1to4}, 5    --> vfpclassps(k5|k3, xword_b [rax+64], 5); // broadcast 64-bit to 256-bit

 

+vpdpbusd(xm0, xm1, xm2); // default encoding is EVEX

+vpdpbusd(xm0, xm1, xm2, EvexEncoding); // same as the above

+vpdpbusd(xm0, xm1, xm2, VexEncoding); // VEX encoding

 

 注意

 * k1, ..., k7 は新しいopmaskレジスタです。

@@ -379,6 +382,7 @@ sample/{echo,hello}.bfは http://www.kmonos.net/alang/etc/brainfuck.php から
 -----------------------------------------------------------------------------

 ◎履歴

 

+2020/10/19 ver 5.99 VNNI命令サポート(Thanks to akharito)

 2020/10/17 ver 5.98 [scale * reg]のサポート

 2020/09/08 ver 5.97 uint32などをuint32_tに置換

 2020/08/28 ver 5.95 レジスタクラスのコンストラクタがconstexprに対応(C++14以降)

diff --git a/xbyak/xbyak.h b/xbyak/xbyak.h
index 5a529c1..14ad87a 100644
--- a/xbyak/xbyak.h
+++ b/xbyak/xbyak.h
@@ -135,7 +135,7 @@ namespace Xbyak {
 
 enum {
 	DEFAULT_MAX_CODE_SIZE = 4096,
-	VERSION = 0x5980 /* 0xABCD = A.BC(D) */
+	VERSION = 0x5990 /* 0xABCD = A.BC(D) */
 };
 
 #ifndef MIE_INTEGER_TYPE_DEFINED
diff --git a/xbyak/xbyak_mnemonic.h b/xbyak/xbyak_mnemonic.h
index 6aaa74e..ff13ed0 100644
--- a/xbyak/xbyak_mnemonic.h
+++ b/xbyak/xbyak_mnemonic.h
@@ -1,4 +1,4 @@
-const char *getVersionString() const { return "5.98"; }
+const char *getVersionString() const { return "5.99"; }
 void adc(const Operand& op, uint32_t imm) { opRM_I(op, imm, 0x10, 2); }
 void adc(const Operand& op1, const Operand& op2) { opRM_RM(op1, op2, 0x10); }
 void adcx(const Reg32e& reg, const Operand& op) { opGen(reg, op, 0xF6, 0x66, isREG32_REG32orMEM, NONE, 0x38); }