docs/CommandGuide/llc.html - fp2-dev/platform/external/llvm - Gitiles

 <html>
 <title>LLVM: llc tool</title>

 <body bgcolor=white>

 <center><h1>LLVM: <tt>llc</tt> tool</h1></center>
 <HR>

 <h3>NAME</h3>
 <tt>llc</tt>

 <h3>SYNOPSIS</h3>
 <tt>llc [options] [filename]</tt>

 <h3>DESCRIPTION</h3>

 The <tt>llc</tt> command compiles LLVM bytecode into assembly language for a
 specified architecture.  The assembly language output can then be passed through
 a native assembler and linker to generate native code.
 <p>
 The choice of architecture for the output assembly code is determined as
 follows:

 <ul>
 	<li>
 	If the user has specified an architecture with the -m option, use that
 	architecture.
 	<p>

 	<li>
 	Examine the input LLVM bytecode file:
 	<ul>
 		<li>
 		If it specifies little endian and a pointer size of 32 bits, select the
 		x86 architecture.
 		<p>

 		<li>
 		If it specifies big endian and a pointer size of 64 bit pointers,
 		select the SparcV9 architecture.
 	</ul>
 	<p>

 	<li>
 	If <tt>llc</tt> was compiled on an architecture for which it can
 	generate code, select the architecture upon which <tt>llc</tt> was
 	compiled.
 	<p>

 	<li>
 	Print a message to the user asking him or her to specify the output
 	architecture explicitly.
 </ul>

 <p>

 If filename is not specified, or if filename is -, <tt>llc</tt> reads its input
 from standard input.  Otherwise, it will read its input from filename.
 <p>

 If the -o option is left unspecified, then <tt>llc</tt> will send its output to standard
 output if the input is from standard input.  If the -o option specifies -, then
 the output will also be sent to standard output.
 <p>

 If no -o option is specified and an input file other than - is specified, then
 <tt>llc</tt> creates the output filename as follows:

 <ul>
 	<li>
 	If the file ends in .bc, then the .bc suffix is removed, and the .s suffix
 	is appended.
 	<p>
 	<li>
 	Otherwise, the .s suffix is appended to the input filename.
 </ul>

 <h3>
 OPTIONS
 </h3>
 <ul>
 	<li>-f
 	<br>
 	Overwrite output files
 	<p>

 	<li>-m&lt;arch&gt;
 	<br>
 	Specify the architecture for which to generate assembly.  Valid
 	architectures are:

 	<dl compact>
 		<dt> x86 </dt>
 		<dd>IA-32 (Pentium and above)</dd>

 		<dt> sparcv9 </dt>
 		<dd>SPARC V9</dd>

 		<dt> c </dt>
 		<dd>Emit C code</dd>
 	</dl>
 	<p>

 	<li>-o &lt;filename&gt;
 	<br>
 	Specify the output filename.
 	<p>

 	<li>-enable-correct-eh-support
 	<br>
 	Instruct the -lowerinvoke pass to insert code for correct exception handling
   support.  This is expensive and is by default omitted for efficiency.
 	<p>

 	<li> -help
 	<br>
 	Print a summary of command line options.
 	<p>

 	<li> -stats
 	<br>
 	Print statistics.
 	<p>

 	<li> -time-passes
 	<br>
 	Record the amount of time needed for each pass and print it to standard
 	error.
 	<p>

 </ul>
 <h4>X86 Specific Options</h4>
 <ul>
 	<li>-disable-fp-elim
 	<br>
 	Disable frame pointer elimination optimization.
 	<p>

 	<li>-disable-pattern-isel
 	<br>
 	Use the 'simple' X86 instruction selector (the default).
 	<p>

 	<li>-print-machineinstrs
 	<br>
 	Print generated machine code.
 	<p>

 	<li>-regalloc=&lt;ra&gt;
 	<br>
         Specify the register allocator to use. The default is <i>local</i>.
         Valid register allocators are:
         <p>
 	<dl compact>
 		<dt> simple </dt>
 		<dd>Very simple register allocator</dd>

 		<dt> local </dt>
 		<dd>Local register allocator</dd>

 		<dt> linearscan </dt>
 		<dd>Linear scan global register allocator (experimental)</dd>
         </dl>

 	<li>-spiller=&lt;sp&gt;
 	<br>
         Specify the spiller to use for register allocators that support it.
         Currently this option is used by the linear scan register
         allocator. The default is <i>local</i>.
         Valid spillers are:
         <p>
 	<dl compact>
 		<dt> simple </dt>
 		<dd>Simple spiller</dd>

 		<dt> local </dt>
 		<dd>Local spiller</dd>
 	</dl>

 </ul>

 <h4>Sparc Specific Options</h4>
 <ul>
 	<li>-disable-peephole
 	<br>
 	Disable peephole optimization pass.
 	<p>

 	<li>-disable-sched
 	<br>
 	Disable local scheduling pass.
 	<p>

 	<li>-disable-strip
 	<br>
 	Do not strip the LLVM bytecode included in executable.
 	<p>

 	<li>-enable-maps
 	<br>
 	Emit LLVM-to-MachineCode mapping info to assembly.
 	<p>
 </ul>


 <h3>EXIT STATUS</h3>

 If <tt>llc</tt> succeeds, it will exit with 0.  Otherwise, if an error occurs,
 it will exit with a non-zero value.

 <h3>
 SEE ALSO
 </h3>
 <a href="lli.html"><tt>lli</tt></a>

 <HR>
 Maintained by the <a href="http://llvm.cs.uiuc.edu">LLVM Team</a>.
 </body>
 </html>
	<html>
	<title>LLVM: llc tool</title>

	<body bgcolor=white>

	<center><h1>LLVM: <tt>llc</tt> tool</h1></center>
	<HR>

	<h3>NAME</h3>
	<tt>llc</tt>

	<h3>SYNOPSIS</h3>
	<tt>llc [options] [filename]</tt>

	<h3>DESCRIPTION</h3>

	The <tt>llc</tt> command compiles LLVM bytecode into assembly language for a
	specified architecture. The assembly language output can then be passed through
	a native assembler and linker to generate native code.
	<p>
	The choice of architecture for the output assembly code is determined as
	follows:

	<ul>
	<li>
	If the user has specified an architecture with the -m option, use that
	architecture.
	<p>

	<li>
	Examine the input LLVM bytecode file:
	<ul>
	<li>
	If it specifies little endian and a pointer size of 32 bits, select the
	x86 architecture.
	<p>

	<li>
	If it specifies big endian and a pointer size of 64 bit pointers,
	select the SparcV9 architecture.
	</ul>
	<p>

	<li>
	If <tt>llc</tt> was compiled on an architecture for which it can
	generate code, select the architecture upon which <tt>llc</tt> was
	compiled.
	<p>

	<li>
	Print a message to the user asking him or her to specify the output
	architecture explicitly.
	</ul>

	<p>

	If filename is not specified, or if filename is -, <tt>llc</tt> reads its input
	from standard input. Otherwise, it will read its input from filename.
	<p>

	If the -o option is left unspecified, then <tt>llc</tt> will send its output to standard
	output if the input is from standard input. If the -o option specifies -, then
	the output will also be sent to standard output.
	<p>

	If no -o option is specified and an input file other than - is specified, then
	<tt>llc</tt> creates the output filename as follows:

	<ul>
	<li>
	If the file ends in .bc, then the .bc suffix is removed, and the .s suffix
	is appended.
	<p>
	<li>
	Otherwise, the .s suffix is appended to the input filename.
	</ul>

	<h3>
	OPTIONS
	</h3>
	<ul>
	<li>-f
	<br>
	Overwrite output files
	<p>

	<li>-m<arch>
	<br>
	Specify the architecture for which to generate assembly. Valid
	architectures are:

	<dl compact>
	<dt> x86 </dt>
	<dd>IA-32 (Pentium and above)</dd>

	<dt> sparcv9 </dt>
	<dd>SPARC V9</dd>

	<dt> c </dt>
	<dd>Emit C code</dd>
	</dl>
	<p>

	<li>-o <filename>
	<br>
	Specify the output filename.
	<p>

	<li>-enable-correct-eh-support
	<br>
	Instruct the -lowerinvoke pass to insert code for correct exception handling
	support. This is expensive and is by default omitted for efficiency.
	<p>

	<li> -help
	<br>
	Print a summary of command line options.
	<p>

	<li> -stats
	<br>
	Print statistics.
	<p>

	<li> -time-passes
	<br>
	Record the amount of time needed for each pass and print it to standard
	error.
	<p>

	</ul>
	<h4>X86 Specific Options</h4>
	<ul>
	<li>-disable-fp-elim
	<br>
	Disable frame pointer elimination optimization.
	<p>

	<li>-disable-pattern-isel
	<br>
	Use the 'simple' X86 instruction selector (the default).
	<p>

	<li>-print-machineinstrs
	<br>
	Print generated machine code.
	<p>

	<li>-regalloc=<ra>
	<br>
	Specify the register allocator to use. The default is <i>local</i>.
	Valid register allocators are:
	<p>
	<dl compact>
	<dt> simple </dt>
	<dd>Very simple register allocator</dd>

	<dt> local </dt>
	<dd>Local register allocator</dd>

	<dt> linearscan </dt>
	<dd>Linear scan global register allocator (experimental)</dd>
	</dl>

	<li>-spiller=<sp>
	<br>
	Specify the spiller to use for register allocators that support it.
	Currently this option is used by the linear scan register
	allocator. The default is <i>local</i>.
	Valid spillers are:
	<p>
	<dl compact>
	<dt> simple </dt>
	<dd>Simple spiller</dd>

	<dt> local </dt>
	<dd>Local spiller</dd>
	</dl>

	</ul>

	<h4>Sparc Specific Options</h4>
	<ul>
	<li>-disable-peephole
	<br>
	Disable peephole optimization pass.
	<p>

	<li>-disable-sched
	<br>
	Disable local scheduling pass.
	<p>

	<li>-disable-strip
	<br>
	Do not strip the LLVM bytecode included in executable.
	<p>

	<li>-enable-maps
	<br>
	Emit LLVM-to-MachineCode mapping info to assembly.
	<p>
	</ul>


	<h3>EXIT STATUS</h3>

	If <tt>llc</tt> succeeds, it will exit with 0. Otherwise, if an error occurs,
	it will exit with a non-zero value.

	<h3>
	SEE ALSO
	</h3>
	<a href="lli.html"><tt>lli</tt></a>

	<HR>
	Maintained by the <a href="http://llvm.cs.uiuc.edu">LLVM Team</a>.
	</body>
	</html>