<?xml version="1.0"?>
<!DOCTYPE flagsdescription
    SYSTEM "http://www.spec.org/dtd/cpuflags2.dtd"
>

<!-- ... You'll want a nice wide window when reading or editing this .................................................... -->

<!-- This flags file is a descendant of an example: $SPEC/Docs/flags-advanced.xml and is also used as the flags file for Sun
CPU2006 submissions.  In order to serve as an example, there are more details in some of the comments than might otherwise be
strictly necessary, and there are definitely more flags than are actually going to be used in the first submission.  --> 

<flagsdescription>

<filename>Fujitsu-M12-1</filename>

<title>Fujitsu M12-1 platform flags file</title>

<style>
<![CDATA[
ul { margin-top:.2em; }
table.lightborder        { border-style:solid; border-width:0px 0px 1px; border-color:#c0c0c0; }
td.lightborder           { vertical-align:top; padding:2px .3em 2px .2em; border-style:solid; border-width:1px 0px 0px 0px; border-color:#c0c0c0; }
th.lightborder           { vertical-align:top; text-align:left; font-family:monospace; font-weight:normal; padding:2px .3em 2px .2em;border-style:solid; border-width:1px 0px 0px 0px; border-color:#c0c0c0; }

]]>
</style>

<submit_command>
<![CDATA[
<p><b><kbd>submit=echo 'pbind -b...' > dobmk; sh dobmk</kbd></b> (SPEC tools, Unix shell)
<br /> When running multiple copies of benchmarks, the SPEC config file feature <kbd><b>submit</b></kbd> is often used to
cause individual jobs to be bound to specific processors.  If so, the specific command may be found in the config file; here
is a brief guide to understanding that command: </p>
   <ul>
   <li><kbd><b>submit=</b></kbd> causes the SPEC tools to use this line when submitting jobs.</li>
   <li><kbd><b>echo ...&gt; dobmk</b></kbd> causes the generated commands to be written to a file, namely
       <kbd>dobmk</kbd>.</li>
   <li><kbd><b>pbind -b</b></kbd> causes this copy's processes to be bound to the CPU specified by the expression that
       follows it.  See the config file used in the submission for the exact syntax, which tends to be cumbersome because of
       the need to carefully quote parts of the expression.  When all expressions are evaluated, each core ends up with
       exactly one copy of each benchmark.  The pbind expression may include various elements from the SPEC toolset and from
       standard Unix commands, such as:
       <ul>
       <li><kbd><b>$BIND</b></kbd>: a reference to a value from the bind line, a line of the form 
          "<tt>bind&nbsp;=&nbsp;n&nbsp;n&nbsp;n&nbsp;n</tt>", where each "n" is a processor number.  See
          <a href="http://www.spec.org/cpu2006/docs/config.html#bind">http://www.spec.org/cpu2006/docs/config.html#bind</a>  
          for details on this feature.</li>
       <li><kbd><b>$$</b></kbd>: the current process id</li>
       <li><kbd><b>$SPECCOPYNUM</b></kbd>: the SPEC tools-assigned number for this copy of the benchmark.</li>
       <li><kbd><b>psrinfo</b></kbd>: find out what processors are available</li>
       <li><kbd><b>grep on-line</b></kbd>: search the <kbd>psrinfo</kbd> output for information regarding on-line cpus</li>
       <li><kbd><b>expr</b></kbd>: Calculate simple arithmetic expressions.  For example, the effect of binding jobs to a
           (quote-resolved) expression such as: <br /><tt>expr ( $SPECCOPYNUM / 4 ) * 8 + ($SPECCOPYNUM % 4 ) )</tt>
           <br />would be to send the jobs to processors whose numbers are:
           <br /><tt>0,1,2,3, 8,9,10,11, 16,17,18,19 ...</tt></li>
       <li><kbd><b>awk...print \$1</b></kbd>: Pick out the line corresponding to this copy of the benchmark and use the CPU
           number mentioned at the start of this line.</li>
       </ul></li>
   <li><kbd><b>sh dobmk</b></kbd> actually runs the benchmark.</li>
   </ul>
]]>
</submit_command>

<sw_environment>
<![CDATA[
<p><b><kbd>LD_LIBRARY_PATH=&lt;directories&gt;</kbd></b> 
<br />LD_LIBRARY_PATH controls the search order for both the compile-time and run-time linkers.  Usually, it can be
defaulted; but testers may sometimes choose to explicitly set it (as documented in the notes in the submission), in order to
ensure that the correct versions of libraries are picked up.  </p>

<p><b><kbd>MADV=access_lwp</kbd></b> and <b><kbd>LD_PRELOAD=madv.so.1</kbd></b> 
<br />When the madv.so.1 shared object is present in the LD_PRELOAD list, it is possible to provide advice to the system
about how memory is likely to be accessed.  The advice present in MADV applies to all processes and their descendants.  A
commonly used value is <kbd>access_lwp</kbd>, which means that when memory is allocated, the next process to touch it will be
the primary user.  Examples of other possible values include <kbd>sequential</kbd>, for memory that is used only once and
then no longer needed and <kbd>acces_many</kbd> when many processes will be sharing data.</p>

<p><b><kbd>MPSSHEAP=&lt;size&gt;</kbd></b>, <b><kbd>MPSSSTACK=&lt;size&gt;</kbd></b>, and
<b><kbd>LD_PRELOAD=mpss.so.1</kbd></b> 
<br />When these variables are set, the mpss.so.1 shared object will set the preferred page size for new processes, and their
descendants, to the requested sizes for the heap and stack.</p>

]]>
</sw_environment>

<os_tuning>
<![CDATA[

<p><b>Platform settings</b></p>

<p>One or more of the following settings may have been applied to the testbed.  If so, the "Platform 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><kbd>autoup=&lt;n&gt;</kbd></b> (Unix /etc/system)
<br />When the file system flush daemon <tt>fsflush</tt> runs, it writes to disk all modified file buffers that are more
than <kbd>n</kbd> seconds old.  </p>

<p><b><kbd>bufhwm=&lt;n&gt;</kbd></b> (Unix /etc/system)
<br />Sets the upper limit of the file system buffer cache.  The units for bufhwm are in kilobytes.
</p>

<p><b><kbd>cpu_bringup_set=&lt;n&gt;</kbd></b> (Unix /etc/system)
<br /> Specifies which processors are enabled at boot time.  <kbd><b>&lt;n&gt;</b></kbd> represents a bitmap of the
processors to be brought online.  </p>

<p><b><kbd>segmap_percent=&lt;n&gt;</kbd></b> (Unix /etc/system)
<br />This value controls the size of the segmap cache as a percent of total memory.  Set this value to help keep the file system cache from consuming memory unnecessarily.
</p>

<p><b><kbd>STACKSIZE=&lt;n&gt;</kbd></b> (Unix environment variable)
<br /> Set the size of the stack (temporary storage area) for each slave thread of a multithreaded program.  </p>


<p><b><kbd>psrset -c &lt;n&gt;</kbd></b> (Unix, superuser commands)
<br /> Creates a new processor set and displays the new processor set ID.
</p>

<p><b><kbd>psrset -e &lt;n&gt;</kbd></b> (Unix, superuser commands)
<br /> Executes a command (with optional arguments) in the specified processor set.
The command process and any child processes are executed only by processors in the processor set.</p>

<p><b><kbd>svcadm disable webconsole</kbd></b> (Unix, superuser commands)
<br />Turns off the Sun Web Console, a browser-based interface that performs systems management.
If it is enabled, system administrators can manage systems, devices and services from remote systems.</p>

<p><b><kbd>svcadm disable ldmd</kbd></b> (Unix, superuser commands)
<br />Turns off the Logical Domains Manager, create and manage logical domains, as well as map logical domains to physical resources.
</p>

<p><b><kbd>ts_dispatch_extended=&lt;n&gt;</kbd></b> (Unix /etc/system)
<br />Controls which dispatch table is loaded upon boot.  A value of 1 loads the large system table, a value of 0 loads the regular system table.
</p>

<p><b><kbd>tune_t_fsflushr=&lt;n&gt;</kbd></b> (Unix /etc/system)
<br /> Controls the number of seconds between runs of the file system flush daemon, <tt>fsflush</tt>.  </p>

<p><b><kbd>doiflush=&lt;n&gt;</kbd></b> (Unix /etc/system)
<br /> Controls whether file system metadata syncs will be executed during <tt>fsflush</tt> invocations.  </p>

<p><b><kbd>dopageflush=&lt;n&gt;</kbd></b> (Unix /etc/system)
<br /> Controls whether memory is examined for modified pages during <tt>fsflush</tt> invocations.  </p>

<p><b><kbd>zfs:zfs_arc_max=&lt;n&gt;</kbd></b> (Unix /etc/system)
<br /> Determines the maximum size of the ZFS Adaptive Replacement Cache (ARC).  </p>


]]>
</os_tuning>

<firmware>
<![CDATA[
<p><b>Firmware / BIOS / Microcode settings</b></p>

<p><b><kbd>setcod</kbd></b> (eXtended System Control Facility, XSCF)
<br /> <tt>setcod</tt> is a command to set the CPU Activation to be used in the physical partition (PPAR).  </p>
<ul>
   <li><kbd><b>-p &lt;n&gt;</b></kbd>Specifies PPAR-ID. Depending on the system configuration, you can specify an integer 0 to 15. </li>
   <li><kbd><b>-s cpu</b></kbd>Sets the allocation of the CPU core resource used in PPAR. </li>
</ul>

]]>
</firmware>

</flagsdescription>