<?xml version="1.0"?>
<!DOCTYPE flagsdescription SYSTEM "http://www.spec.org/dtd/cpuflags1.dtd">
<flagsdescription>

<!-- filename to begin with "Fujitsu.RX900.ic11.1-linux64.xml" -->
<filename>Fujitsu.RX900.ic11.1-linux64.xml</filename>

<title> Fujitsu RX900 CPU2006 Flags Description </title>
<header>
<![CDATA[
<p style="text-align: left; color: red; font-size: larger; background-color: black">
 Copyright &copy; 2006 Intel Corporation.  All Rights Reserved.</p>
]]>
</header>

<platform_settings>
 <![CDATA[ 
	 <p><b>Platform settings</b></p>

         <p>One or more of the following settings may have been set.  If so, the "General Notes" section of the
         report will say so; and you can read below to find out more about what these settings mean.</p>
   
         <p><b>KMP_STACKSIZE </b></p>
         <p>
         Specify stack size to be allocated for each thread. 
	 </p>

         <p><b>KMP_AFFINITY </b></p>
         <p>
         KMP_AFFINITY  =  &lt; physical | logical &gt;, starting-core-id <br/>
         specifies the static mapping of user threads to physical cores. For example, 
         if you have a system configured with 8 cores, OMP_NUM_THREADS=8 and 
         KMP_AFFINITY=physical,0 then thread 0 will mapped to core 0, thread 1 will be mapped to core 1, and 
         so on in a round-robin fashion.   <br/> </p>

         <p>
         KMP_AFFINITY = granularity=fine,scatter <br/>
         The value for the environment variable KMP_AFFINITY affects how the threads from an auto-parallelized program are scheduled across processors. <br/>
         Specifying granularity=fine selects the finest granularity level, causes each OpenMP thread to be bound to a single thread context. <br/>
         This ensures that there is only one thread per core on cores supporting HyperThreading Technology<br/>
         Specifying scatter distributes the threads as evenly as possible across the entire system. <br/> 
         Hence a combination of these two options, will spread the threads evenly across sockets, with one thread per physical core. <br/>
         </p>

         <p><b>OMP_NUM_THREADS </b></p>
         <p>
	  Sets the maximum number of threads to use for OpenMP* parallel regions if no 
          other value is specified in the application. This environment variable 
          applies to both -openmp and -parallel (Linux and Mac OS X) or /Qopenmp and /Qparallel (Windows).
          Example syntax on a Linux system with 8 cores:
          export OMP_NUM_THREADS=8
          </p>

	 <p><b>Hardware Prefetch:</b></p> 
	 <p>
	 This BIOS option allows the enabling/disabling of a processor mechanism to                 
	 prefetch data into the cache according to a pattern-recognition algorithm.
	 </p>
	 <p>                
	 In some cases, setting this option to Disabled may improve
	 performance. Users should only disable this option 
	 after performing application benchmarking to verify improved
	 performance in their environment.
	 </p>

	 <p><b>Adjacent Sector Prefetch:</b></p> 
	 <p>
	 This BIOS option allows the enabling/disabling of a processor mechanism to                 
	 fetch the adjacent cache line within a 128-byte sector that contains 
	 the data needed due to a cache line miss.
	 </p>
	 <p>                
	 In some cases, setting this option to Disabled may improve
	 performance. Users should only disable this option 
	 after performing application benchmarking to verify improved
	 performance in their environment.
	 </p>
         

	 <p><b>ulimit -s &lt;n&gt; </b></p>
	 <p> Sets the stack size to <b>n</b> kbytes, or <b>unlimited</b> to allow the stack size 
         to grow without limit.
	 </p>


	 <p><b>echo 1 > /proc/sys/vm/zone_reclaim_mode</b></p>

	<p>Zone_reclaim_mode allows someone to set more or less aggressive approaches to
	reclaim memory when a zone runs out of memory. If it is set to zero then no
	zone reclaim occurs. Allocations will be satisfied from other zones / nodes
	in the system.</p>
	
	<p>This is value ORed together of</p>
	
	<p>1       = Zone reclaim on<br />
	2       = Zone reclaim writes dirty pages out<br />
	4       = Zone reclaim swaps pages<br /></p>
	
	<p>zone_reclaim_mode is set during bootup to 1 if it is determined that pages
	from remote zones will cause a measurable performance reduction. The
	page allocator will then reclaim easily reusable pages (those page
	cache pages that are currently not used) before allocating off node pages.</p>
	
	<p>It may be beneficial to switch off zone reclaim if the system is
	used for a file server and all of memory should be used for caching files
	from disk. In that case the caching effect is more important than
	data locality.</p>

	<p>Allowing zone reclaim to write out pages stops processes that are
	writing large amounts of data from dirtying pages on other nodes. Zone
	reclaim will write out dirty pages if a zone fills up and so effectively
	throttle the process. This may decrease the performance of a single process
	since it cannot use all of system memory to buffer the outgoing writes
	anymore but it preserve the memory on other nodes so that the performance
	of other processes running on other nodes will not be affected.</p>

	<p>Allowing regular swap effectively restricts allocations to the local
	node unless explicitly overridden by memory policies or cpuset
	configurations.</p>



	 <p><b>mkdir /dev/cpuset; mount -t cpuset none /dev/cpuset; echo 1 > /dev/cpuset/memory_spread_page</b></p>

        <p> There are two boolean flag files per cpuset that control where the 
        kernel allocates pages for the file system buffers and related in
        kernel data structures.  They are called 'memory_spread_page' and
        'memory_spread_slab'.
         If the per-cpuset boolean flag file 'memory_spread_page' is set, then
        the kernel will spread the file system buffers (page cache) evenly
        over all the nodes that the faulting task is allowed to use, instead
        of preferring to put those pages on the node where the task is running.
         If the per-cpuset boolean flag file 'memory_spread_slab' is set,
        then the kernel will spread some file system related slab caches,
        such as for inodes and dentries evenly over all the nodes that the
        faulting task is allowed to use, instead of preferring to put those
        pages on the node where the task is running.
        The setting of these flags does not affect anonymous data segment or
        stack segment pages of a task.</p>
        
        <p> By default, both kinds of memory spreading are off, and memory
        pages are allocated on the node local to where the task is running,
        except perhaps as modified by the tasks NUMA mempolicy or cpuset
        configuration, so long as sufficient free memory pages are available.
        When new cpusets are created, they inherit the memory spread settings
        of their parent.
        Setting memory spreading causes allocations for the affected page
        or slab caches to ignore the tasks NUMA mempolicy and be spread
        instead.    Tasks using mbind() or set_mempolicy() calls to set NUMA
        mempolicies will not notice any change in these calls as a result of
        their containing tasks memory spread settings.  If memory spreading
        is turned off, then the currently specified NUMA mempolicy once again
        applies to memory page allocations.
        Both 'memory_spread_page' and 'memory_spread_slab' are boolean flag
        files.  By default they contain "0", meaning that the feature is off
        for that cpuset.  If a "1" is written to that file, then that turns
        the named feature on.</p>


         <p><b>Using numactl to bind processes and memory to cores</b></p>
         <p>For multi-copy runs or single copy runs on systems with multiple sockets, it is advantageous to bind a process to a particular core.  Otherwise, the OS may arbitrarily move your process from one core to another.  This can affect performance.  To help, SPEC allows the use of a "submit" command where users can specify a utility to use to bind processes.  We have found the utility 'numactl' to be the best choice.</p>
         <p>numactl runs processes with a specific NUMA scheduling or memory placement policy.  The policy is set for a command and inherited by all of its children.  The numactl flag "--physcpubind" specifies which core(s) to bind the process. "-l" instructs numactl to keep a process memory on the local node while "-m" specifies which node(s) to place a process memory.  For full details on using numactl, please refer to your Linux documentation, 'man numactl'</p>


  ]]> 
</platform_settings>

<!--
******************************************************************************************************
* Compilers
******************************************************************************************************
-->

<flag name="intel_icc_64bit" class="compiler" regexp="icc.* -m64(?=\s|$)">

   <![CDATA[
      <p>Invoke the Intel C compiler 11.1 for Intel 64 applications </p>
      <p> You need binutils 2.16.91.0.7 or later with this compiler to support new instructions on Intel Core 2 processors </p>
   ]]>

</flag>

<flag name="intel_icpc_64bit" class="compiler" regexp="icpc.* -m64(?=\s|$)">

   <![CDATA[
      <p>Invoke the Intel C++ compiler 11.1 for Intel 64 applications </p>
      <p> You need binutils 2.16.91.0.7 or later with this compiler to support new instructions on Intel Core 2 processors </p>
   ]]>

</flag>

<flag name="intel_ifort_64bit" class="compiler" regexp="(?=\s|^)ifort.* -m64(?=\s|$)">
   <![CDATA[
      <p> Invoke the Intel Fortran compiler 11.1 for Intel 64 applications.</p>
      <p> You need binutils 2.16.91.0.7 or later with this compiler to support new instructions on Intel Core 2 processors </p>
   ]]>

</flag>

<flag name="intel_icc_32bit" class="compiler" regexp="icc.* -m32(?=\s|$)">

   <![CDATA[
      <p>Invoke the Intel C compiler 11.1 for IA32 applications </p>
      <p> You need binutils 2.16.91.0.7 or later with this compiler to support new instructions on Intel Core 2 processors </p>
   ]]>

</flag>

<flag name="intel_icpc_32bit" class="compiler" regexp="(?=\s|^)icpc.* -m32(?=\s|$)">

   <![CDATA[
      <p> Invoke the Intel C++ compiler 11.1 for IA32 applications.</p>
      <p> You need binutils 2.16.91.0.7 or later with this compiler to support new instructions on Intel Core 2 processors </p>
   ]]>

</flag>

<flag name="intel_ifort_32bit" class="compiler" regexp="ifort.* -m32(?=\s|$)">
   <![CDATA[
      <p> Invoke the Intel Fortran compiler 11.1 for IA32 applications.</p>
      <p> You need binutils 2.16.91.0.7 or later with this compiler to support new instructions on Intel Core 2 processors </p>
   ]]>

</flag>

<!--
******************************************************************************************************
* Portability
******************************************************************************************************
-->
<flag name="lowercase_routine_name" class="portability" regexp="(?:/\S+/)?-Qlowercase(?=\s|$)">

   <![CDATA[
      <p>For mixed-language benchmarks, tell the compiler to convert routine names to 
      lowercase for compatibility</p>
   ]]>

</flag>

<flag name="add-underscore_to_routine_name" class="portability" regexp="(?:/\S+/)?\/assume\:underscore(?=\s|$)">

   <![CDATA[
      <p>For mixed-language benchmarks, tell the compiler to assume that routine 
      names end with an underscore</p>
   ]]>

</flag>

<flag name="assume_cplusplus_sources" class="portability" regexp="(?:/\S+/)?-TP(?=\s|$)">
      Tell the compiler to treat source files as C++ regardless of the file extension
</flag>

<flag name="f-nofor_main" class="portability" regexp="-nofor_main(?=\s|$)">
   <![CDATA[
<p>This option specifies that the main program is not written in Fortran. 
It is a link-time option that prevents the compiler from linking for_main.o 
into applications. </p>

<p>For example, if the main program is written in C and calls a Fortran subprogram, 
specify -nofor-main when compiling the program with the ifort command. 
If you omit this option, the main program must be a Fortran program.</p>
   ]]> 
</flag>

<!--
******************************************************************************************************
* Optimizations
******************************************************************************************************
-->
<flag name="f-O1" class="optimization" regexp="-O1(?=\s|$)">
      
   <![CDATA[
      <p>Enables optimizations for speed and disables some optimizations that increase code size and affect speed. <br />
         To limit code size, this option: </p>
         <ul>
         <li> Enables global optimization; this includes data-flow analysis, 
           code motion, strength reduction and test replacement, split-lifetime
           analysis, and instruction scheduling. </li>
         <li> Disables intrinsic recognition and intrinsics inlining. </li>
         </ul>
  
         <p> The O1 option may improve performance for applications with very large 
         code size, many branches, and execution time not dominated by code within loops. </p>
         
         -O1 sets the following options:<br />
         -funroll-loops0, -fno-builtin, -mno-ieee-fp, -fomit-framepointer, -ffunction-sections, -ftz
   ]]>

   <include flag="f-funroll-loops"/>
   <include flag="f-fno-builtin"/>
   <include flag="f-mno-ieee-fp"/>
   <include flag="f-fomit-framepointer"/>
   <include flag="f-ffunction-sections"/>
   <include flag="f-ftz"/>
   
</flag>

<flag name="f-O2" class="optimization" regexp="-O2(?=\s|$)">

   <![CDATA[
      <p>Enables optimizations for speed. This is the generally recommended 
      optimization level. This option also enables: <br />
      - Inlining of intrinsics<br />
      - Intra-file interprocedural optimizations, which include: <br />
        - inlining<br />
        - constant propagation<br />
        - forward substitution<br />
        - routine attribute propagation<br />
        - variable address-taken analysis<br />
        - dead static function elimination<br />
        - removal of unreferenced variables<br />
      - The following capabilities for performance gain: <br /> 
        - constant propagation<br />
        - copy propagation<br />
        - dead-code elimination<br />
        - global register allocation<br />
        - global instruction scheduling and control speculation<br />
        - loop unrolling<br />
        - optimized code selection<br />
        - partial redundancy elimination<br />
        - strength reduction/induction variable simplification<br />
        - variable renaming<br />
        - exception handling optimizations<br />
        - tail recursions<br />
        - peephole optimizations<br />
        - structure assignment lowering and optimizations<br />
        - dead store elimination<br />
      </p>

   ]]> 

   <include flag="f-O1"/> 
</flag>

<flag name="f-O3" class="optimization" regexp="-O3(?=\s|$)">

   <![CDATA[
      <p>Enables O2 optimizations plus more aggressive optimizations, 
         such as prefetching, scalar replacement, and loop and memory 
         access transformations. Enables optimizations for maximum speed,   
         such as: </p>
         <ul>
         <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>
         </ul> <br/>
         <p>
         On IA-32 and Intel EM64T processors, when O3 is used with options 
         -ax or -x (Linux) or with options /Qax or /Qx (Windows), the compiler 
         performs more aggressive data dependency analysis than for O2, which 
         may result in longer compilation times. 
         The O3 optimizations may not cause higher performance unless loop and 
         memory access transformations take place. The optimizations may slow 
         down code in some cases compared to O2 optimizations.  
         The O3 option is recommended for applications that have loops that heavily 
         use floating-point calculations and process large data sets. </p>
   ]]> 

   <include flag="f-O2"/>

</flag>

<flag name="f-funroll-loops" class="optimization" regexp="-funroll-loops\d+(?=\s|$)">
      Tells the compiler the maximum number of times to unroll loops. For example -funroll-loops0 would
      disable unrolling of loops.  
</flag>



<flag name="f-fno-builtin" class="optimization" regexp="-fno-builtin(?=\s|$)">
 -fno-builtin disables inline expansion for all intrinsic functions.
</flag>

<flag name="f-mno-ieee-fp" class="optimization" regexp="-f-mno-ieee-fp(?=\s|$)">
 This option trades off floating-point precision for speed by removing 
the restriction to conform to the IEEE standard. 
</flag>

<flag name="f-fomit-framepointer" class="optimization" regexp="-fomit-framepointer(?=\s|$)">
  EBP is used as a general-purpose register in optimizations.       
</flag>

<flag name="f-ffunction-sections" class="optimization" regexp="-ffunction-sections(?=\s|$)">
  Places each function in its own COMDAT section.     
</flag>

<flag name="f-ftz" class="optimization" regexp="-ftz(?=\s|$)">
 Flushes denormal results to zero.      
</flag>

<flag name="f-unroll" class="optimization" regexp="-unroll(\d+)(?=\s|$)">

<example>-unroll&lt;n&gt;</example> 
This option sets the maximum number of times a loop can be unrolled, to $1.

<ex_replacement> n. For example, -unroll1 will unroll loops just once. To disable loop unrolling, use -unroll0. </ex_replacement>

</flag>

<flag name="f-par-schedule" class="optimization" regexp="-par-schedule-static\=(\d+)(?=\s|$)">
   <![CDATA[
<p>The -par-schedule option lets you specify a scheduling algorithm or a tuning method for loop iterations.<br /> 
It specifies how iterations are to be divided among the threads of the team. This option affects performance <br />
tuning and can provide better performance during auto-parallelization.</p>
   ]]> 

<example>-par-schedule-static=n</example>
   <![CDATA[
-par-schedule-static=n tells the compiler to divide iterations into contiguous pieces (chunks) of size n. <br />
The chunks are assigned to threads in the team in a round-robin fashion in the order of the thread number. <br />
Note that the last chunk to be assigned may have a smaller number of iterations. If n is not specified, <br />
the iteration space is divided into chunks that are approximately equal in size, and each thread is assigned at most one chunk.<br />
   ]]> 

<ex_replacement> n. For example, -par-schedule-static=32768 will split iterations into chunks of size 32768. </ex_replacement>

</flag>

<flag name="f-ip" class="optimization" regexp="-ip(?=\s|$)">
This option enables additional interprocedural optimizations for single 
file compilation. These optimizations are a subset of full intra-file 
interprocedural optimizations. One of these optimizations enables the 
compiler to perform inline function expansion for calls to functions 
defined within the current source file.
</flag>

<flag name="f-ipo" class="optimization" regexp="-ipo(?=\s|$)">
   <![CDATA[
      <p>Multi-file ip optimizations that includes:<br />
       - inline function expansion<br />
       - interprocedural constant propagation<br />
       - dead code elimination<br />
       - propagation of function characteristics<br />
       - passing arguments in registers<br />
       - loop-invariant code motion</p>
   ]]> 
</flag>

<flag name="f-auto-ilp32" class="optimization" regexp="-auto-ilp32(?=\s|$)">
   <![CDATA[
<p>This option instructs the compiler to analyze and transform the program so that 
64-bit pointers are shrunk to 32-bit pointers, and 64-bit longs (on Linux) are 
shrunk into 32-bit longs wherever it is legal and safe to do so. 
In order for this option to be effective the compiler must be able to optimize using 
the -ipo/-Qipo option and must be able to analyze all library/external calls the program makes. </p>

<p>This option requires that the size of the program executable never exceeds 2^32 bytes and all 
data values can be represented within 32 bits. If the program can run correctly in a 32-bit system, 
these requirements are implicitly satisfied. If the program violates these size restrictions, 
unpredictable behavior might occur.</p>
   ]]> 
</flag>

<flag name="f-disablescalarrep" class="optimization" regexp="-scalar-rep-">
   <![CDATA[
<p> -scalar-rep enables scalar replacement performed during loop transformation. 
To use this option, you must also specify O3. -scalar-rep- disables this optimization. </p>
   ]]> 
</flag>


<flag name="f-no-alias" class="optimization" regexp="-fno-alias(?=\s|$)">
   <![CDATA[
<p> This options tells the compiler to assume no aliasing in the program. </p>
   ]]> 
</flag>


<flag name="f-fast" class="optimization" regexp="-fast(?=\s|$)">

   <![CDATA[
      <p>The -fast option enhances execution speed across the entire program 
      by including the following options that can improve run-time performance:</p>

      <p style="text-indent: -45px;margin-left: 45px">
      -O3&nbsp;&nbsp;&nbsp;(maximum speed and high-level optimizations)</p>
      <p style="text-indent: -45px;margin-left: 45px"> 
      -ipo&nbsp;(enables interprocedural optimizations across files)</p>
      <p style="text-indent: -45px;margin-left: 45px">  
      -xSSSE3&nbsp;&nbsp;(generate code specialized for Intel(R) Core(TM)2 Duo processors, Intel(R) Core(TM)2 Quad processors 
                      and  Intel(R) Xeon(R) processors with SSSE3)</p>
      <p style="text-indent: -45px;margin-left: 45px"> 
      -static&nbsp;
             Statically link in libraries at link time</p>
      <p style="text-indent: -45px;margin-left: 45px"> 
      -no-prec-div&nbsp;(disable -prec-div)
             where -prec-div improves precision of FP divides (some speed impact)</p>
   
      <p>To override one of the options set by /fast, specify that option after the 
      -fast option on the command line. The exception is the xT or QxT option
      which can't be overridden. The options set by /fast may change from 
      release to release.</p>
   ]]> 

   <include flag="f-O3"/>
   <include flag="f-ipo"/>
   <include flag="f-xSSSE3"/>
   <include flag="f-static"/>
   <include flag="f-no-prec-div"/>
</flag>

<flag name="f-static" class="optimization" regexp="-static(?=\s|$)">
         	 Compiler option to statically link in libraries at link time
</flag>

<flag name="f-xSSE42" class="optimization" regexp="-xSSE4.2(?=\s|$)">

   <![CDATA[

      <p>Code is optimized for Intel(R) processors with support for SSE 4.2 instructions.
      The resulting code may contain unconditional use of features that are not supported
      on other processors. This option also enables new optimizations in addition to
      Intel processor-specific optimizations including advanced data layout and code
      restructuring optimizations to improve memory accesses for Intel processors.</p>

      <p> Do not use this option if you are executing a program on a processor that
      is not an Intel processor. If you use this option on a non-compatible processor
      to compile the main program (in Fortran) or the function main() in C/C++, the
      program will display a fatal run-time error if they are executed on unsupported
      processors. </p>

   ]]>

</flag>

<flag name="f-xSSE41" class="optimization" regexp="-xSSE4.1(?=\s|$)">

   <![CDATA[

      <p>Code is optimized for Intel(R) processors with support for SSE 4.1 instructions.
      The resulting code may contain unconditional use of features that are not supported
      on other processors. This option also enables new optimizations in addition to
      Intel processor-specific optimizations including advanced data layout and code
      restructuring optimizations to improve memory accesses for Intel processors.</p>

      <p> Do not use this option if you are executing a program on a processor that
      is not an Intel processor. If you use this option on a non-compatible processor
      to compile the main program (in Fortran) or the function main() in C/C++, the
      program will display a fatal run-time error if they are executed on unsupported
      processors. </p>

   ]]>

</flag>

<flag name="f-xSSE3_ATOM" class="optimization" regexp="-xSSE3_ATOM(?=\s|$)">

   <![CDATA[

      <p>Code is optimized for Intel(R) Atom(TM) processors.
      The resulting code may contain unconditional use of features that are not supported
      on other processors. This option also enables new optimizations in addition to
      Intel processor-specific optimizations including advanced data layout and code
      restructuring optimizations to improve memory accesses for Intel processors.</p>

      <p> Do not use this option if you are executing a program on a processor that
      is not an Intel processor. If you use this option on a non-compatible processor
      to compile the main program (in Fortran) or the function main() in C/C++, the
      program will display a fatal run-time error if they are executed on unsupported
      processors. </p>

   ]]>

</flag>


<flag name="f-xSSSE3" class="optimization" regexp="-xSSSE3(?=\s|$)">

   <![CDATA[

      <p>Code is optimized for Intel(R) processors with support for SSSE3 instructions.
      The resulting code may contain unconditional use of features that are not supported
      on other processors. This option also enables new optimizations in addition to
      Intel processor-specific optimizations including advanced data layout and code
      restructuring optimizations to improve memory accesses for Intel processors.</p>

      <p> Do not use this option if you are executing a program on a processor that
      is not an Intel processor. If you use this option on a non-compatible processor
      to compile the main program (in Fortran) or the function main() in C/C++, the
      program will display a fatal run-time error if they are executed on unsupported
      processors. </p>

   ]]>

</flag>

<flag name="f-QxB" class="optimization" regexp="-QxB(?=\s|$)">

    <![CDATA[       
      <p>Code is optimized for Intel Pentium M and compatible Intel processors. The 
      resulting code may contain unconditional use of features that are not supported 
      on other processors. This option also enables new optimizations in addition to 
      Intel processor-specific optimizations including advanced data layout and code 
      restructuring optimizations to improve memory accesses for Intel processors.</p>
     
      <p> Do not use this option if you are executing a program on a processor that 
      is not an Intel processor. If you use this option on a non-compatible processor 
      to compile the main program (in Fortran) or the function main() in C/C++, the 
      program will display a fatal run-time error if they are executed on unsupported 
      processors. </p>

    ]]> 
  
</flag>

<flag name="f-QxW" class="optimization" regexp="-QxW(?=\s|$)">

    <![CDATA[       
      <p>Code is optimized for Intel Pentium 4 and compatible Intel processors; 
      this is the default for Intel?EM64T systems. The resulting code may contain 
      unconditional use of features that are not supported on other processors. </p>
     
    ]]> 
  
</flag>

<flag name="f-parallel" class="optimization" regexp="-parallel(?=\s|$)" parallel="yes">

    <![CDATA[       
      <p>Tells the auto-parallelizer to generate multithreaded code for loops that can be safely executed in parallel.
      To use this option, you must also specify option O2 or O3. The default numbers of threads spawned is equal to 
      the number of processors detected in the system where the binary is compiled. This can be changed by setting the 
      environment variable OMP_NUM_THREADS </p>
     
    ]]> 
  
</flag>

<flag name="f-libguide.lib" class="optimization" regexp="libguide.lib(?=\s|$)">

    <![CDATA[       
      <p>The use of -Qparallel to generate auto-parallelized code requires support libraries that are 
      dynamically linked by default. Specifying libguide.lib on the link line, statically links in 
      libguide.lib to allow auto-parallelized binaries to work on systems which do not have the dynamic version 
      of this library installed.</p>
     
    ]]> 
  
</flag>

<flag name="f-libguide40.lib" class="optimization" regexp="libguide40.lib(?=\s|$)">

    <![CDATA[       
      <p>The use of -Qparallel to generate auto-parallelized code requires spport libraries that are 
      dynamically linked by default. Specifying libguide40.lib on the link line, statically links in 
      libguide40.lib to allow auto-parallelized binaries to work on systems which do not have the 
      dynamic version of this library installed.</p>
     
    ]]> 
  
</flag>



<flag name="f-archSSE2" class="optimization" regexp="-arch\:SSE2(?=\s|$)">

  <![CDATA[

      <p> Optimizes for Intel Pentium 4 and compatible processors with Streaming SIMD Extensions 2 (SSE2).</p>

  ]]>

</flag>


<flag name="f-no-prec-div" class="optimization" regexp="-no-prec.div">
(disable/enable[default] -prec-div)
                
   <![CDATA[
      <p>-no-prec-div enables optimizations that give slightly less precise results 
         than full IEEE division. </p>

      <p>When you specify -no-prec-div along with some optimizations, such as 
         -xN and -xB (Linux) or /QxN and /QxB (Windows), 
         the compiler may change floating-point division computations into 
         multiplication by the reciprocal of the denominator. 
         For example, A/B is computed as A * (1/B) to improve the speed of the 
         computation.</p>
      <p>However, sometimes the value produced by this transformation is 
         not as accurate as full IEEE division. When it is important to have fully 
         precise IEEE division, do not use -no-prec-div. 

         This will enable the default -prec-div and the result will be more accurate, 
         with some loss of performance.</p>
   ]]> 

</flag>

<flag name="prof_gen" class="optimization" regexp="-prof-gen(?=\s|$)">

   <![CDATA[
      <p>Instrument program for profiling for the first phase of
      two-phase profile guided optimization. This instrumentation gathers information
      about a program's execution paths and data values but does not gather
      information from hardware performance counters. The profile instrumentation
      also gathers data for optimizations which are unique to profile-feedback
      optimization.</p>
   ]]> 

</flag>

<flag name="prof_use" class="optimization" regexp="-prof-use(?=\s|$)">

   <![CDATA[
      <p>Instructs the compiler to produce a profile-optimized 
      executable and merges available dynamic information (.dyn) 
      files into a pgopti.dpi file. If you perform multiple 
      executions of the instrumented program, -prof-use merges 
      the dynamic information files again and overwrites the 
      previous pgopti.dpi file.<br />
      Without any other options, the current directory is 
      searched for .dyn files</p>
   ]]> 
      
</flag>

<flag name="link_force_multiple1" class="optimization" regexp="-Wl\,-z\,muldefs(?=\s|$)">

   <![CDATA[
      <p>Enable SmartHeap and/or other library usage by forcing the linker to 
      ignore multiple definitions if present</p>
   ]]> 
 
</flag>

<flag name="SmartHeap64" class="optimization" regexp="-L\S+\s+-lsmartheap64(?=\s|$)">

   <![CDATA[
      <p>MicroQuill SmartHeap Library V8.1 (64-bit) available from http://www.microquill.com/</p>
   ]]> 

</flag>

<flag name="SmartHeap" class="optimization" regexp="-L\S+\s+-lsmartheap(?=\s|$)">

   <![CDATA[
      <p>MicroQuill SmartHeap Library V8.1 (32-bit) available from http://www.microquill.com/</p>
   ]]> 

</flag>

<flag name="set_stack_space" class="optimization" regexp="(?:/\S+/)?/F\d*">
      set the stack reserve amount specified to the linker 
</flag>

<flag name="f-ansi-alias" class="optimization" regexp="-ansi-alias(?=\s|$)">
   <![CDATA[
      <p>Enable use of ANSI aliasing rules in optimizations. This  option tells the compiler to assume that the program
      adheres to ISO C Standard aliasability rules. </p>
      <p>If your program adheres to these rules, then  this  option allows  the  compiler to optimize more aggressively. <br />
      If it doesn't adhere to these rules, then it can cause the compiler to generate incorrect code.</p>
   ]]> 
</flag>

<flag name="f-opt-prefetch" class="optimization" regexp="-opt-prefetch(?=\s|$)">
      Enable/disable(DEFAULT) the compiler to generate prefetch instructions to prefetch data. 
</flag>

<flag name="f-inline-calloc" class="optimization" regexp="-inline-calloc(?=\s|$)">
      Directs the compiler to inline calloc() calls as malloc()/memset()
</flag>


<flag name="f-opt-malloc-options" class="optimization" regexp="-opt-malloc-options=3(?=\s|$)">
   <![CDATA[
      <p>The compiler adds setup code in the C/C++/Fortran main function to enable optimal malloc algorithms:</p>
      <ul>
      <li> n=0: Default, no changes to the malloc options. No call to mallopt() is made. </li>
      <li> n=1: M_MMAP_MAX=2 and M_TRIM_THRESHOLD=0x10000000. Call mallopt with the two settings. </li>
      <li> n=2: M_MMAP_MAX=2 and M_TRIM_THRESHOLD=0x40000000. Call mallopt with these two settings. </li>
      <li> n=3: M_MMAP_MAX=0 and M_TRIM_THRESHOLD=-1.    Call mallopt with these two settings. This 
      will cause use of sbrk() calls instead of mmap() calls to get memory from the system. </li>
      </ul>
	  <p> The two parameters, M_MMAP_MAX and M_TRIM_THRESHOLD, are described below </p>

      <p>Function: int mallopt (int param, int value) When calling mallopt, the param argument 
      specifies the parameter to be set, and value the new value to be set. Possible choices 
      for param, as defined in malloc.h, are: </p>
      <ul>
      <li> M_TRIM_THRESHOLD This is the minimum size (in bytes) of the top-most, releasable chunk 
      that will cause sbrk to be called with a negative argument in order to return memory 
      to the system. </li>
      <li> M_TOP_PAD This parameter determines the amount of extra memory to obtain from the system 
      when a call to sbrk is required. It also specifies the number of bytes to retain when 
      shrinking the heap by calling sbrk with a negative argument. This provides the necessary 
      hysteresis in heap size such that excessive amounts of system calls can be avoided. </li>
      <li> M_MMAP_THRESHOLD All chunks larger than this value are allocated outside the normal heap, 
      using the mmap system call. This way it is guaranteed that the memory for these chunks 
      can be returned to the system on free. Note that requests smaller than this threshold 
      might still be allocated via mmap. </li>
	<li> M_MMAP_MAX The maximum number of chunks to allocate 
      with mmap. Setting this to zero disables all use of mmap. </li>
      </ul>
      
   ]]>
</flag>


<flag name="f-vec-guard-write" class="optimization" regexp="-vec-guard-write(?=\s|$)">
      Enables cache/bandwidth optimization for stores under conditionals (within vector loops)
      This option tells the compiler to perform a conditional check in a vectorized loop. 
      This checking avoids unnecessary stores and may improve performance by conserving bandwidth.
</flag>

<flag name="f-par-runtime-control" class="optimization" regexp="-par-runtime-control(?=\s|$)" parallel="yes">
      Enable compiler to generate runtime control code for effective automatic parallelization.
      
      This option generates code to perform run-time checks for loops that have symbolic loop bounds. 
      If the granularity of a loop is greater than the parallelization threshold, the loop will be 
      executed in parallel. If you do not specify this option, the compiler may not parallelize loops 
      with symbolic loop bounds if the compile-time granularity estimation of a loop cannot ensure 
      it is beneficial to parallelize the loop.
</flag>

<flag name="f-opt-ra-region-strategy-block" class="optimization" regexp="-opt-ra-region-strategy.block(?=\s|$)">
   <![CDATA[
      <p>Select the method that the register allocator uses to partition
      each routine into regions</p>
      <ul>
      <li>routine - one region per routine</li>
      <li>block - one region per block</li>
      <li>trace - one region per trace</li> 
      <li>loop - one region per loop</li>
      <li>default - compiler selects best option</li>
      </ul>
   ]]>
</flag>

<flag name="f-opt-ra-region-strategy-routine" class="optimization" regexp="-opt-ra-region-strategy.routine(?=\s|$)">
   <![CDATA[
      <p>Select the method that the register allocator uses to partition
      each routine into regions</p>
      <ul>
      <li>routine - one region per routine</li>
      <li>block - one region per block</li>
      <li>trace - one region per trace</li> 
      <li>loop - one region per loop</li>
      <li>default - compiler selects best option</li>
      </ul>
   ]]>
</flag>


<flag name="f-opt-multi-version-aggressive" class="optimization" regexp="-opt-multi-version-aggressive(?=\s|$)">
      Multi-versioning is used for generating different versions of the loop based on  run time dependence testing,  
      alignment and checking for short/long trip counts.  If this option is turned on, it will trigger more versioning 
      at the expense of creating more overhead to check for pointer aliasing and scalar replacement.
</flag>

<flag name="f-auto" class="optimization" regexp="-auto(?=\s|$)">
      Make all local variables AUTOMATIC. Same as -automatic
</flag>

<flag name="f-unroll-aggressive" class="optimization" regexp="-unroll-aggressive(?=\s|$)">
      Enables more aggressive unrolling heuristics
</flag>

<flag name="f-opt-streaming-stores-always" class="optimization" regexp="-opt-streaming-stores.always(?=\s|$)">
   <![CDATA[
      <p>Specifies whether streaming stores are generated:</p>
      <p>always - enables generation of streaming stores under the assumption that the application is memory bound</p>
      <p>auto   - compiler decides when streaming stores are used (DEFAULT)</p>
      <p>never  - disables generation of streaming stores</p>
   ]]> 
</flag>


<flag name="f-Oi-" class="optimization" regexp="-Oi-">
      Disables inline expansion of all intrinsic functions. 
</flag>

<flag name="f-Op-" class="optimization" regexp="-Op-(?=\s|$)">

   <![CDATA[
      <p>Disables conformance to the ANSI C and IEEE 754 standards for 
      floating-point arithmetic.</p>
   ]]> 
 
</flag>

<flag name="f-Oy" class="optimization" regexp="-Oy(?=\s|$)">
      Allows use of EBP as a general-purpose register in optimizations.  
</flag>

<flag name="f-Os" class="optimization" regexp="-Os(?=\s|$)">

   <![CDATA[
      <p>This option enables most speed optimizations, but disables some 
      that increase code size for a small speed benefit.</p>
   ]]> 
  
</flag>

<flag name="f-Og" class="optimization" regexp="-Og(?=\s|$)">
      This option enables global optimizations.  
</flag>

<flag name="f-Ob_n" class="optimization" regexp="-Ob(0|1|2)(?=\s|$)">

   <![CDATA[
      <p>Specifies the level of inline function expansion.</p>
         
         <p style="text-indent: -45px;margin-left: 45px">
         Ob0 - Disables inlining of user-defined functions. Note that 
               statement functions are always inlined.</p>
         <p style="text-indent: -45px;margin-left: 45px">
         Ob1 - Enables inlining when an inline keyword or an inline 
               attribute is specified. Also enables inlining according 
               to the C++ language.</p>
         <p style="text-indent: -45px;margin-left: 45px">
         Ob2 - Enables inlining of any function at the compiler's 
               discretion. </p>
   ]]> 
  
</flag>

<flag name="f-Gy" class="optimization" regexp="-Gy(?=\s|$)">

   <![CDATA[
      <p>This option tells the compiler to separate functions into COMDATs 
      for the linker.</p>
   ]]> 
     
</flag>

<flag name="f-GF" class="optimization" regexp="-GF(?=\s|$)">
      This option enables read only string-pooling optimization.   
</flag>

<flag name="f-Gf" class="optimization" regexp="-Gf(?=\s|$)">
      This option enables read/write string-pooling optimization.    
</flag>

<flag name="f-Gs" class="optimization" regexp="-Gs(?=\s|$)">

   <![CDATA[
      <p>This option disables stack-checking for routines with 4096 bytes 
      of local variables and compiler temporaries.</p>
   ]]> 
     
</flag>


</flagsdescription>