代码拉取完成，页面将自动刷新
<?xml version="1.0"?>
<!DOCTYPE flagsdescription
SYSTEM "http://www.spec.org/dtd/cpuflags1.dtd"
>
<flagsdescription>
<!--
<filename>flags</filename>
-->
<title>Fujitsu Siemens Computers SPEC CPU2006 Flag Description</title>
<style>
<![CDATA[
body { background: white; }
]]>
</style>
<!-- HEADERs -->
<header>
<![CDATA[
<p style="text-align: center">Copyright © 2006.
Intel Corporation. All Rights Reserved.</p>
<p><h3>Intel Compilers for C++ and Fortran, Version 9.1</h3></p>
<hr />
]]>
</header>
<header class="optimization">
<![CDATA[
<p>HEADER for OPTIMIZATION</p>
]]>
</header>
<header class="portability">
<![CDATA[
<p>HEADER for PORTABILITY</p>
]]>
</header>
<header class="compiler">
<![CDATA[
<p>HEADER for COMPILER</p>
]]>
</header>
<header class="other">
<![CDATA[
<p>HEADER for OTHER</p>
]]>
</header>
<!-- /HEADERs -->
<!-- OPTIMIZATION -->
<flag name="F-Fstacksize" class="optimization" regexp="-F\d+">
<example>-F10000</example>
<![CDATA[
<p>Specifies the stack reserve amount for the program.</p>
<p>-F<n><br>
<n> is the stack reserve amount.
It can be specified as a decimal integer or by using a C-style convention
for constants (for example, -F0x1000).<br>
Default: The stack size default is chosen by the operating system.</p>
]]>
</flag>
<flag name="f-fast" class="optimization">
<example>-fast</example>
<![CDATA[
<p>Maximize speed across the entire program.</p>
<p>Turns on:</p>
<ul>
<li>-O3</li>
<li>-Qprec-div-</li>
<li>-QxP</li>
<li>-Qipo</li>
</ul>
]]>
<include flag="F-O3"/>
<include flag="F-Qprec-div-"/>
<include flag="F-QxP"/>
<include flag="F-Qipo"/>
</flag>
<flag name="F-O3" class="optimization" regexp="[-/]O3">
<example>-O3</example>
<![CDATA[
<p>Optimizes for speed. Enables high-level optimization. This level does
not guarantee higher performance. Using this option may increase the
compilation time. Impact on performance is application dependent, some
applications may not see a performance improvement.</p>
<p>The optimizations include:</p>
<ul>
<li>All optimizations done with -O2</li>
<li>loop unrolling, including instruction scheduling</li>
<li>code replication to eliminate branches</li>
<li>padding the size of certain power-of-two arrays to allow more efficient cache use.</li>
<li>When used with -Qax or -Qx, it causes the compiler to perform
more aggressive data dependency analysis than for -O2.</li>
</ul>
]]>
<include flag="F-O2"/>
</flag>
<flag name="F-Qprec-div-" class="optimization" regexp="[-/]Qprec-div-?">
<example>-Qprec-div-</example>
<![CDATA[
<p>Enables[disables] improved precision of floating-point divides.
Disabling may slightly improve speed.
Default Enabled.</p>
]]>
</flag>
<flag name="F-QxP" class="optimization" regexp="[-/]Qx(K|W|N|P|B)">
<example>-QxP</example>
<![CDATA[
<p>Generate specialized code to run exclusively on processors
supporting the extensions indicated by [codes] as described above.<br>
P stands for Pentium.</p>
]]>
</flag>
<flag name="F-Qipo" class="optimization" regexp="[-/]Qipo">
<example>-Qipo</example>
<![CDATA[
<p><b>for C and C++</b></p>
<p>Enables multi-file ip optimizations which allows inline function expansion for calls to functions defined in separate files.
The compiler decides whether to create one or more object files based on an estimate of the size of the application.
It generates one object file for small applications and two for large ones.</p>
<p><b>for Fortran</b></p>
<p>multi-file ip optimizations that includes:</p>
<ul>
<li>inline function expansion</li>
<li>interprocedural constant propagation</li>
<li>monitoring module-level static variables</li>
<li>dead code elimination</li>
<li>propagation of function characteristics</li>
<li>passing arguments in registers</li>
<li>loop-invariant code motion</li>
</ul>
]]>
</flag>
<flag name="F-O2" class="optimization" regexp="[-/]O2">
<example>-O2</example>
<![CDATA[
<p>Optimizes for speed.
<br>The -O2 option includes the following options:</p>
<ul>
<li>-Og</li>
<li>-Oi-</li>
<li>-Os</li>
<li>-Oy</li>
<li>-Ob1</li>
<li>-Gs</li>
</ul>
<p>This options defaults to ON.</p>
<p>This option also enables.</p>
<ul>
<li>inlining of intrinsics</li>
<li>Intra-file interprocedural optimizations including:</li>
<ul>
<li>inlining</li>
<li>constant propagation</li>
<li>forward substitution</li>
<li>routine attribute propagation</li>
<li>variable address-taken analysis</li>
<li>dead static function elimination</li>
<li>removal of unreferenced variables.</li>
</ul>
<li>The following performance optimizations:</li>
<ul>
<li>copy propogation.</li>
<li>dead-code elimination</li>
<li>global register allocation</li>
<li>global instruction scheduling and control speculation</li>
<li>loop unrolliing</li>
<li>optimized code selection</li>
<li>partial redundancy elimination</li>
<li>strength reduction/induction variable simplification</li>
<li>variable renaming</li>
<li>exception handling optimizations</li>
<li>tail recursions</li>
<li>peephole optimizations</li>
<li>structure assignment lowering and optimization</li>
<li>dead store elimination</li>
</ul>
</ul>
]]>
<include flag="F-Og"/>
<include flag="F-Oi"/>
<include flag="F-Os"/>
<include flag="F-Oy"/>
<include flag="F-Obn"/>
<include flag="F-Gs"/>
</flag>
<flag name="F-Qip" class="optimization" regexp="[-/]Qip">
<example>-Qip</example>
<![CDATA[
<p>Enables single-file interprocedural optimizations within a file.</p>
]]>
</flag>
<flag name="F-Qprof_gen" class="optimization" regexp="[-/]Qprof_gen">
<example>-Qprof_gen</example>
<![CDATA[
<p>Instruments a program for profiling.</p>
<p>This option instruments a program for profiling to get the execution count of each basic block.
It also creates a new static profile information file (.spi).</p>
<p>It is used in phase 1 of the Profile Guided Optimizer (PGO) to instruct the compiler
to produce code in your object files in preparation for instrumented execution.</p>
<ul>
<li>Gathers information regarding execution paths.</li>
<li>Gathers information regarding data values.</li>
<li>Does not use hardware performance counters.</li>
</ul>
]]>
</flag>
<flag name="F-Qprof_use" class="optimization" regexp="[-/]Qprof_use">
<example>-Qprof_use</example>
<![CDATA[
<p>Enables the use of profiling information during optimization.</p>
<p>This option enables the use of profiling information (including function splitting
and function grouping) during optimization.
It enables option -Qfnsplit.</p>
<p>This option instructs the compiler to produce a profile-optimized executable
and it merges available profiling output files into a pgopti.dpi file.</p>
<p>Note that there is no way to turn off'function grouping if you enable it using this option.</p>
<ul>
<li>Uses information regarding execution paths.</li>
<li>Uses information regarding data values.</li>
<li>Does not use hardware performance counters.</li>
<li>Uses techniques (like function grouping) which are not available without PGO.</li>
</ul>
]]>
<include flag="F-Qfnsplit"/>
</flag>
<flag name="F-Qfnsplit" class="optimization">
<example>-Qfnsplit</example>
<![CDATA[
<p>Enables function splitting.</p>
<p>This option enables function splitting if -Qprof-use is also specified.
Otherwise, this option has no effect.</p>
<p>It is enabled automatically if you specify -Qprof-use. If you do not specify
one of those options, the default is -Qfnsplit-, which disables
function splitting but leaves function grouping enabled.</p>
<p>To disable function splitting when you use -Qprof-use, specify -Qfnsplit-.</p>
]]>
</flag>
<flag name="F-Og" class="optimization" regexp="[-/]Og">
<example>-Og</example>
<![CDATA[
<p>Enables global optimizations.</p>
]]>
</flag>
<flag name="F-Oi" class="optimization" regexp="[-/]Oi-?">
<example>-Oi</example>
<![CDATA[
<p>Enables/disables inline expansion of intrinsic functions.</p>
<p>Default enabled</p>
]]>
</flag>
<flag name="F-Os" class="optimization" regexp="[-/]Os">
<example>-Os</example>
<![CDATA[
<p>This option enables most speed optimizations, but disables some that increase
code size for a small speed benefit.</p>
<p>Default enabled</p>
]]>
</flag>
<flag name="F-Oy" class="optimization" regexp="[-/]Oy-?">
<example>-Oy</example>
<![CDATA[
<p>Enables [disables] the use of the EBP register in optimizations.
When you disable with <i>-Oy-</i>, the EBP register is used as frame pointer.
-Oy has the effect of reducing the number of general-purpose registers by 1,
and can produce slightly less efficient code.</p>
<p>Default enabled</p>
]]>
</flag>
<flag name="F-Obn" class="optimization" regexp="[-/]Ob[1 2 3]">
<example>-Ob<n></example>
<![CDATA[
<p><b><i>n</i> = 0</b>
<br>Disables inlining of user-defined functions.
However, statement functions are always inlined</p>
<p><b><i>n</i> = 1</b>
<br>Enables inlining of functions declared with the __inline keyword.
Also enables inlining according to the C++ language</p>
<p><b><i>n</i> = 2</b>
<br>Enables inlining of any function.
However, the compiler decides which functions are inlined.
This option enables interprocedural optimizations and has the same
effect as specifying option <i>Qip</i>.</p>
<p>Default enabled with <i>n</i> = 2</p>
]]>
</flag>
<flag name="F-Gs" class="optimization" regexp="[-/]Gs<n>?">
<example>-Gs</example>
<![CDATA[
<p>Disables stack-checking for routines with n or more bytes of local
variables and compiler temporaries.</p>
<p>Default enabled with <i>n</i> = 4096.</p>
]]>
</flag>
<flag name="F-Oa" class="optimization" regexp="[-/]Oa-?">
<example>-Oa</example>
<![CDATA[
<p>Assume [not assume] no aliasing</p>
<p>Default disabled</p>
]]>
</flag>
<flag name="F-Ot" class="optimization" regexp="[-/]Ot">
<example>-Ot</example>
<![CDATA[
<p>Enables all speed optimizations.</p>
<p>Overrides -Os</p>
]]>
</flag>
<flag name="F-Ow" class="optimization" regexp="[-/]Ow-?">
<example>-Ow</example>
<![CDATA[
<p>Assume[not assume] no cross function aliasing.</p>
]]>
</flag>
<flag name="F-Gf" class="optimization" regexp="[-/]Gf">
<example>-Gf</example>
<![CDATA[
<p>Enables string-pooling optimization.</p>
]]>
</flag>
<flag name="F-Gy" class="optimization" regexp="[-/]Gy">
<example>-Gy</example>
<![CDATA[
<p>Packages functions to enable linker optimization.</p>
<p>Default enabled</p>
]]>
</flag>
<flag name="F-Qax" class="optimization" regexp="[-/]Qax[KWNPB]">
<example>-QaxP</example>
<![CDATA[
<p>Generates specialized code for processor specific codes K, W, N, P while also generating generic IA-32 code.</p>
<ul>
<li><b>K</b> = Intel Pentium III and compatible Intel processors</li>
<li><b>W</b> = Intel Pentium 4 and compatible Intel processors</li>
<li><b>N</b> = Intel Pentium 4 and compatible Intel processors.
These options also enable advanced data layout and code restructuring
optimizations to improve memory accesses for Intel processors.</li>
<li><b>P</b> = Intel Pentium 4 processor with Streaming SIMD 3 (SSE3) support.
These option also enable advanced data layout and code restructuring optimizations
to improve memory accesses for Intel processors.</li>
</ul>
]]>
</flag>
<flag name="F-Qrcd" class="optimization" regexp="[-/]Qrcd">
<example>-Qrcd</example>
<![CDATA[
<p>Enables[disables] fast conversions of floating-point to integer conversions.
This option does not guarantee that any particular rounding mode will be used.</p>
]]>
</flag>
<flag name="F-Qansi_alias" class="optimization" regexp="[-/]Qansi_alias-?">
<example>-Qansi_alias</example>
<![CDATA[
<p><b>for C and C++</b><br />
<ul>
Qansi_alias directs the compiler to assume the following:
<li>Arrays are not accessed out of bounds.</li>
<li>Pointers are not cast to non-pointer types, and vice-versa.</li>
<li>References to objects of two different scalar types cannot alias.
For example, an object of type int cannot alias with an object of type float,
or an object of type float cannot alias with an object of type double.</li>
</ul>
</p>
<p>If your program satisfies the above conditions, setting the -Qansi_alias
flag will help the compiler better optimize the program. However, if your
program does not satisfy one of the above conditions, the -Qansi_alias
flag may lead the compiler to generate incorrect code.</p>
<p><b>for Fortran</b><br />
Enables (default) or disables the compiler to assume that the program adheres to the ANSI Fortran type aliasablility rules.
For example, an object of type real cannot be accessed as an integer.
You should see the ANSI Standard for the complete set of rules.</p>
]]>
</flag>
<flag name="F-Qfp_port" class="optimization" regexp="[-/]Qfp_port">
<example>-Qfp_port</example>
<![CDATA[
<p>round fp results at assignments & casts (some speed impact)</p>
]]>
</flag>
<flag name="F-Qprefetch" class="optimization" regexp="[-/]Qprefetch">
<example>-Qprefetch</example>
<![CDATA[
<p>Enable prefetch insertion.</p>
<p>Default enabled</p>
]]>
</flag>
<flag name="F-Qunroll" class="optimization" regexp="[-/]Qunroll<n>">
<example>-Qunroll<n></example>
<![CDATA[
<p>Specifies the maximum number of times to unroll a loop. <i>n</i> = 0 disables loop unrolling.</p>
<p>Default: the compiler uses default heuristics when unrolling loops.</p>
]]>
</flag>
<flag name="F-Qoption" class="optimization" regexp="[-/]Qoption,tool,optlist">
<example>-Qoption,tool,optlist</example>
<![CDATA[
<p>
<b>-Qoption:</b> passes an option specified by optlist to a tool,
where optlist is a comma-separated list of options.<br />
<b>for C and C++</b><br />
<table>
<tr>
<th>tool:</th>
<td>Description</td>
</tr>
<tr>
<td>cpp</td>
<td>Specifies the compiler front-end preprocessor</td>
</tr>
<tr>
<td>c</td>
<td>Specifies the C++ compiler</td>
</tr>
<tr>
<td>asm</td>
<td>Specifies the assembler</td>
</tr>
<tr>
<td>link</td>
<td>Specifies the linker</td>
</tr>
<tr>
<td>oplist</td>
<td>Indicates one or more valid argument strings for the designated program.
If the argument is a command-line option, you must include the hyphen.
If the argument contains a space or tab character,
you must enclose the entire argument in quotation characters ("").
You must separate multiple ar'uments with commas</td>
</tr>
</table>
<b>for Fortran</b><br />
<table>
<tr>
<th>tool:</th>
<td>Descroption</td>
</tr>
<tr>
<td>fpp</td>
<td>Specifies the Fortran preprocessor</td>
</tr>
<tr>
<td>f</td>
<td>Specifies the Fortran compiler</td>
</tr>
<tr>
<td>asm</td>
<td>Specifies the assembler</td>
</tr>
<tr>
<td>link</td>
<td>Specifies the linker</td>
</tr>
<tr>
<td>oplist</td>
<td>Indicates one or more valid argument strings for the designated tool.
You must separate multiple arguments with commas.</td>
</tr>
</table>
</p>
<p>
<b>NOTE</b>: If <i>tool</i> is incorrectly specified,
the compiler gives an warning and the option is ignored.
For example, if -Qoption,f,... is used with the Intel C++ compiler,
the option is ignored with an warning.
</p>
<p>
-Qoption can be used with the -Qipo flag to refine IPO.
<br />The valid options that can be used for this purpose are:
<ul>
<li><i>-ip_args_in_regs = 0</i>
<br />Disables the passing of arguments in registers.</li>
<li><i>-ip_ninl_max_stats = n</i>
<br />Sets the valid max number of intermediate language statements for a function that is expanded in line.
The number n is a positive integer. The number of intermediate language statements usually
exceeds the actual number of source language statements.
The default value for n is 230. The compiler uses a larger limit for user inline functions</li>
<li><i>-ip_ninl_min_stats = n</i>
<br />Sets the valid min number of intermediate language statements for a function that is expanded in line.
The number n is a positive integer. The default values for ip_ninl_min_stats are:
<br />IA-32 compiler: ip_ninl_min_stats = 7</li>
<li><i>-ip_ninl_max_total_stats = n</i>
<br />Sets the maximum increase in size of a function, measured in intermediate language statements,
due to inlining. n is a positive integer whose default value is 2000.</li>
</ul>
</p>
]]>
</flag>
<flag name="F-Zp" class="optimization" regexp="[-/]Zp[1248]}">
<example>-Zp</example>
<![CDATA[
<p>Specifies the strictest alignment constraint for structure and union types as 1, 2. 4. 8 or 16 bytes</p>
<p>Default is 16.</p>
<p>Problem: 16 is also possible. How to write regexp?</p>
]]>
</flag>
<flag name="F-arch:SSE" class="optimization" regexp="[-/]arch:SSE">
<example>-arch:SSE</example>
<![CDATA[
<p>Enables the compiler to use SSE instructions.</p>
]]>
</flag>
<flag name="F-arch:SSE2" class="optimization" regexp="[-/]arch:SSE2">
<example>-arch:SSE2</example>
<![CDATA[
<p>Enables the compiler to use SSE2 instructions.</p>
]]>
</flag>
<flag name="F-Qpc64" class="optimization" regexp="[-/]Qpc64">
<example>-Qpc64</example>
<![CDATA[
<p>Enables floating-point significand precision control.
The value is used to round the significand to the correct number of bits.
The value must be either 32, 64 or 80.</p>
<p>Default enabled</p>
]]>
</flag>
<flag name="F-Ox" class="optimization" regexp="[-/]Ox">
<example>-Ox</example>
<![CDATA[
<p>Same as the -O2 option: enables -Gs, and -Ob1, -Og, -Oy, and -Ot.</p>
]]>
<include flag="F-Gs" />
<include flag="F-Obn" />
<include flag="F-Og" />
<include flag="F-Oy" />
<include flag="F-Ot" />
</flag>
<flag name="F-auto" class="optimization" regexp="[-/]auto">
<example>-auto</example>
<![CDATA[
<p>Determines whether local variables are put on the run-time stack.</p>
]]>
</flag>
<flag name="F-Qscalar_rep" class="optimization" regexp="[-/]Qscalar_rep-?">
<example>-Qscalar_rep-</example>
<![CDATA[
<p>Enables[disables] scalar replacement performed during loop transformations.
<br />(requires /O3).</p>
]]>
</flag>
<flag name="F-linkMultiple" class="optimization" regexp="-link -FORCE:MULTIPLE">
<example>-link -FORCE:MULTIPLE</example>
<![CDATA[
<p>Force Linking even if multiple entry names are found.</p>
]]>
</flag>
<flag name="F-shlW32M.lib" class="optimization" regexp="shlW32M.lib">
<example>shlw32m.lib</example>
<![CDATA[
<p>MicroQuill SmartHeap Library.<br>
Available from
<a href="http://www.microquill.com/">http://www.microquill.com/</a></p>
]]>
</flag>
<!-- /OPTIMIZATION -->
<!-- PORTABILITY -->
<flag name="F-TP" class="portability">
<example>-TP</example>
<![CDATA[
<p>-TP tells the compiler to process all source or unrecognized file types
as C++ source files.<br>
Default: The compiler assumes that files with the extension .c or .C
are C source files.
To handle them as C++ source files, the compiler flag -TP is needed.</p>
]]>
</flag>
<flag name="F-Qlowercase" class="portability">
<example>-Qlowercase</example>
<![CDATA[
<p>-Qlowercase causes the compiler to ignore case differences in identifiers
and to convert external names to lowercase.<br>
It is needed to specify the naming convention for mixing C and Fortran codes.</p>
]]>
</flag>
<flag name="F-assume:underscore" class="portability" regexp="-assume:underscore">
<example>-assume:underscore</example>
<![CDATA[
<p><b>-assume:[no]underscore</b><br>
Determines whether the compiler appends an underscore character
to external user-defined names.<br>
-assume:underscore is needed to specify the naming convention
for mixing C and Fortran codes.</p>
]]>
</flag>
<!-- /PORTABILITY -->
<!-- COMPILER -->
<flag name="Ficl" class="compiler" regexp="icl">
<example>icl</example>
<![CDATA[
<p> Invoke Intel C/C++ compiler.<br>
Also used to invoke linker for C/C++ programs.</p>
]]>
</flag>
<flag name="Fifort" class="compiler" regexp="ifort">
<example>icl</example>
<![CDATA[
<p> Invoke Intel Fortran compiler.<br>
Also used to invoke linker for Fortran programs
and C/Fortran mixtures.</p>
]]>
</flag>
<flag name="F-Qc99" class="compiler" regexp="[/-]Qc99-?">
<example>-Qc99</example>
<![CDATA[
<p>This option enables/disables C99 support for C programs.</p>
]]>
</flag>
<flag name="F-Qcxx-features" class="compiler" regexp="[-/]Qcxx-features">
<example>-Qcxx-features</example>
<![CDATA[
<p>This option enables standard C++ features without disabling Microsoft
features within the bounds of what is provided in the Microsoft headers and
libraries.</p>
<p>This option has the same effect as specifying -GX -GR.</p>
<p>-GX Enables C++ exception handling.<br>
-GR Enables C++ Run Time Type Information (RTTI).</p>
]]>
</flag>
<!-- /COMPILER -->
<!-- OTHER -->
<!-- currently there are no flags in the class "other" -->
<!-- /OTHER -->
</flagsdescription>
此处可能存在不合适展示的内容，页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容，可点击提交进行申诉，我们将尽快为您处理。
马建仓 AI 助手