📄 csdn_文档中心_快速初始化内存(2).htm
字号:
lang=EN-US><FONT size=3>00401A4A<SPAN
style="mso-spacerun: yes"> </SPAN>movdqa<SPAN
style="mso-spacerun: yes">
</SPAN>xmmword ptr [edx+10h],xmm0</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401A4F<SPAN
style="mso-spacerun: yes"> </SPAN>movdqa<SPAN
style="mso-spacerun: yes">
</SPAN>xmmword ptr [edx+20h],xmm0</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401A54<SPAN
style="mso-spacerun: yes"> </SPAN>movdqa<SPAN
style="mso-spacerun: yes">
</SPAN>xmmword ptr [edx+30h],xmm0</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401A59<SPAN
style="mso-spacerun: yes"> </SPAN>movdqa<SPAN
style="mso-spacerun: yes">
</SPAN>xmmword ptr [edx+40h],xmm0</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401A5E<SPAN
style="mso-spacerun: yes"> </SPAN>movdqa<SPAN
style="mso-spacerun: yes">
</SPAN>xmmword ptr [edx+50h],xmm0</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401A63<SPAN
style="mso-spacerun: yes"> </SPAN>movdqa<SPAN
style="mso-spacerun: yes">
</SPAN>xmmword ptr [edx+60h],xmm0</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401A68<SPAN
style="mso-spacerun: yes"> </SPAN>movdqa<SPAN
style="mso-spacerun: yes">
</SPAN>xmmword ptr [edx+70h],xmm0</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401A6D<SPAN
style="mso-spacerun: yes"> </SPAN>add<SPAN
style="mso-spacerun: yes">
</SPAN>edx,80h</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401A73<SPAN
style="mso-spacerun: yes"> </SPAN>cmp<SPAN
style="mso-spacerun: yes">
</SPAN>ecx,80h</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401A79<SPAN
style="mso-spacerun: yes"> </SPAN>jge<SPAN
style="mso-spacerun: yes">
</SPAN>___intel_new_memset+750h (00401a40)</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><o:p><FONT size=3> </FONT></o:p></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><FONT size=3><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">可见</SPAN><SPAN
lang=EN-US>Intel</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">的实现使用了</SPAN><SPAN
lang=EN-US>SSE2</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">的</SPAN><SPAN
lang=EN-US>128</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">位</SPAN><SPAN
lang=EN-US>xmm</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">寄存器，并且为了促使指令并行，放置了</SPAN><SPAN
lang=EN-US>8</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">条复制指令，这样每个循环能够复制</SPAN><SPAN
lang=EN-US>128</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">×</SPAN><SPAN
lang=EN-US>8=512 Bit</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">。</SPAN></FONT></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><FONT size=3><B
style="mso-bidi-font-weight: normal"><SPAN
lang=EN-US>MSC</SPAN></B><B
style="mso-bidi-font-weight: normal"><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">版本：</SPAN>
<SPAN lang=EN-US><o:p></o:p></SPAN></B></FONT></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>42:<SPAN
style="mso-spacerun: yes">
</SPAN>for (j=0; j< LoopTimes; j++)</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>43:<SPAN
style="mso-spacerun: yes">
</SPAN>{</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>44:<SPAN
style="mso-spacerun: yes">
</SPAN>memset(lpByte,1,SIZE);</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>0040103B<SPAN
style="mso-spacerun: yes"> </SPAN>mov<SPAN
style="mso-spacerun: yes">
</SPAN><SPAN
style="mso-spacerun: yes"> </SPAN>ecx,1900000h</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401040<SPAN
style="mso-spacerun: yes"> </SPAN>mov<SPAN
style="mso-spacerun: yes">
</SPAN>eax,1010101h</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401045<SPAN
style="mso-spacerun: yes"> </SPAN>mov<SPAN
style="mso-spacerun: yes">
</SPAN>edi,ebx</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401047<SPAN
style="mso-spacerun: yes"> </SPAN>dec<SPAN
style="mso-spacerun: yes">
</SPAN>edx</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>00401048<SPAN
style="mso-spacerun: yes"> </SPAN>rep stos<SPAN
style="mso-spacerun: yes"> </SPAN>dword ptr
[edi]</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><FONT size=3>0040104A<SPAN
style="mso-spacerun: yes"> </SPAN>jne<SPAN
style="mso-spacerun: yes">
</SPAN>threadfunc+3Bh (0040103b)</FONT></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><o:p><FONT size=3> </FONT></o:p></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><FONT size=3><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">如果是</SPAN><SPAN
lang=EN-US>Debug</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">版本，因为微软提供了</SPAN><SPAN
lang=EN-US>CRT</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">的</SPAN><SPAN
lang=EN-US>sourcecode</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">，可以跟踪其汇编实现，在</SPAN><SPAN
lang=EN-US>Release</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">版本，优化结果为把函数调用展开，但因为这里的实现只使用了普通的</SPAN><SPAN
lang=EN-US>386</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">指令按</SPAN><SPAN
lang=EN-US>DWORD</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">传送数据，所以在性能上会有如此大的差异。</SPAN><SPAN
lang=EN-US><o:p></o:p></SPAN></FONT></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN lang=EN-US
style="FONT-FAMILY: Arial; FONT-SIZE: 9pt; mso-bidi-font-size: 10.0pt"><o:p> </o:p></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><FONT size=3><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">另外，如果测试代码中的</SPAN><SPAN
lang=EN-US
style="COLOR: #a000a0; FONT-SIZE: 12pt; mso-font-kerning: 0pt">SIZE</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">定义为较小的值，比如</SPAN><SPAN
lang=EN-US>1024*128</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">，在</SPAN><SPAN
lang=EN-US>L2 cache</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">为</SPAN><SPAN
lang=EN-US>512K</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">的</SPAN><SPAN
lang=EN-US>P4</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">上，两种方法的执行结果相差不大，由此可见</SPAN><SPAN
lang=EN-US>Cache</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">对局部访问的促进作用。</SPAN><SPAN
lang=EN-US><o:p></o:p></SPAN></FONT></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN
lang=EN-US><o:p><FONT size=3> </FONT></o:p></SPAN></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><FONT size=3><B
style="mso-bidi-font-weight: normal"><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">说明：</SPAN></B><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">以上例子在</SPAN><SPAN
lang=EN-US>VC6</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">和</SPAN><SPAN
lang=EN-US>VC7</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">中的结果相同。如果直接用</SPAN><SPAN
lang=EN-US>Intel</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">编译器编译，直接使用</SPAN><SPAN
lang=EN-US>memset</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">即可，</SPAN><SPAN
lang=EN-US>Intel</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">编译器在遇到</SPAN><SPAN
lang=EN-US>memset</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">时实际上编译为</SPAN><SPAN
lang=EN-US>__VEC_memset</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">，然后链接到</SPAN><SPAN
lang=EN-US>Intel</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">的</SPAN><SPAN
lang=EN-US>Runtime</SPAN><SPAN
style="FONT-FAMILY: 宋体; mso-ascii-font-family: 'Times New Roman'; mso-hansi-font-family: 'Times New Roman'">库上。</SPAN></FONT><SPAN
lang=EN-US><o:p></o:p></SPAN></P><BR></TD></TR></TBODY></TABLE></TD></TR></TBODY></TABLE><BR>
<TABLE align=center bgColor=#006699 border=0 cellPadding=0 cellSpacing=0
width=770>
<TBODY>
<TR bgColor=#006699>
<TD align=middle bgColor=#006699 id=white><FONT
color=#ffffff>对该文的评论</FONT></TD>
<TD align=middle>
<SCRIPT src="CSDN_文档中心_快速初始化内存(2).files/readnum.htm"></SCRIPT>
</TD></TR></TBODY></TABLE><BR>
<DIV align=center>
<TABLE align=center bgColor=#cccccc border=0 cellPadding=2 cellSpacing=1
width=770>
<TBODY>
<TR>
<TH bgColor=#006699 id=white><FONT
color=#ffffff>我要评论</FONT></TH></TR></TBODY></TABLE></DIV>
<DIV align=center>
<TABLE border=0 width=770>
<TBODY>
<TR>
<TD>你没有登陆，无法发表评论。 请先<A
href="http://www.csdn.net/member/login.asp?from=/Develop/read_article.asp?id=27632">登陆</A>
<A
href="http://www.csdn.net/expert/zc.asp">我要注册</A><BR></TD></TR></TBODY></TABLE></DIV><BR>
<HR noShade SIZE=1 width=770>
<TABLE border=0 cellPadding=0 cellSpacing=0 width=500>
<TBODY>
<TR align=middle>
<TD height=10 vAlign=bottom><A
href="http://www.csdn.net/intro/intro.asp?id=2">网站简介</A> - <A
href="http://www.csdn.net/intro/intro.asp?id=5">广告服务</A> - <A
href="http://www.csdn.net/map/map.shtm">网站地图</A> - <A
href="http://www.csdn.net/help/help.asp">帮助信息</A> - <A
href="http://www.csdn.net/intro/intro.asp?id=2">联系方式</A> - <A
href="http://www.csdn.net/english">English</A> </TD>
<TD align=middle rowSpan=3><A
href="http://www.hd315.gov.cn/beian/view.asp?bianhao=010202001032100010"><IMG
border=0 height=48 src="CSDN_文档中心_快速初始化内存(2).files/biaoshi.gif"
width=40></A></TD></TR>
<TR align=middle>
<TD vAlign=top>百联美达美公司 版权所有 京ICP证020026号</TD></TR>
<TR align=middle>
<TD vAlign=top><FONT face=Verdana>Copyright © CSDN.net, Inc. All rights
reserved</FONT></TD></TR>
<TR>
<TD height=15></TD>
<TD></TD></TR></TBODY></TABLE></DIV>
<DIV></DIV><!--内容结束//--><!--结束//--></BODY></HTML>
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -