1 llc - LLVM static compiler
2 ==========================
7 :program:`llc` [*options*] [*filename*]
12 The :program:`llc` command compiles LLVM source inputs into assembly language
13 for a specified architecture. The assembly language output can then be passed
14 through a native assembler and linker to generate a native executable.
16 The choice of architecture for the output assembly code is automatically
17 determined from the input file, unless the :option:`-march` option is used to
23 If ``filename`` is "``-``" or omitted, :program:`llc` reads from standard input.
24 Otherwise, it will from ``filename``. Inputs can be in either the LLVM assembly
25 language format (``.ll``) or the LLVM bitcode format (``.bc``).
27 If the :option:`-o` option is omitted, then :program:`llc` will send its output
28 to standard output if the input is from standard input. If the :option:`-o`
29 option specifies "``-``", then the output will also be sent to standard output.
31 If no :option:`-o` option is specified and an input file other than "``-``" is
32 specified, then :program:`llc` creates the output filename by taking the input
33 filename, removing any existing ``.bc`` extension, and adding a ``.s`` suffix.
35 Other :program:`llc` options are described below.
42 Print a summary of command line options.
46 Generate code at different optimization levels. These correspond to the
47 ``-O0``, ``-O1``, ``-O2``, and ``-O3`` optimization levels used by
50 .. option:: -mtriple=<target triple>
52 Override the target triple specified in the input file with the specified
55 .. option:: -march=<arch>
57 Specify the architecture for which to generate assembly, overriding the target
58 encoded in the input file. See the output of ``llc -help`` for a list of
59 valid architectures. By default this is inferred from the target triple or
60 autodetected to the current architecture.
62 .. option:: -mcpu=<cpuname>
64 Specify a specific chip in the current architecture to generate code for.
65 By default this is inferred from the target triple and autodetected to
66 the current architecture. For a list of available CPUs, use:
70 llvm-as < /dev/null | llc -march=xyz -mcpu=help
72 .. option:: -filetype=<output file type>
74 Specify what kind of output ``llc`` should generated. Options are: ``asm``
75 for textual assembly ( ``'.s'``), ``obj`` for native object files (``'.o'``)
76 and ``null`` for not emitting anything (for performance testing).
78 Note that not all targets support all options.
80 .. option:: -mattr=a1,+a2,-a3,...
82 Override or control specific attributes of the target, such as whether SIMD
83 operations are enabled or not. The default set of attributes is set by the
84 current CPU. For a list of available attributes, use:
88 llvm-as < /dev/null | llc -march=xyz -mattr=help
90 .. option:: --frame-pointer
92 Specify effect of frame pointer elimination optimization (all,non-leaf,none).
94 .. option:: --disable-excess-fp-precision
96 Disable optimizations that may produce excess precision for floating point.
97 Note that this option can dramatically slow down code on some systems
100 .. option:: --enable-no-infs-fp-math
102 Enable optimizations that assume no Inf values.
104 .. option:: --enable-no-nans-fp-math
106 Enable optimizations that assume no NAN values.
108 .. option:: --enable-unsafe-fp-math
110 Enable optimizations that make unsafe assumptions about IEEE math (e.g. that
111 addition is associative) or may not work for all input ranges. These
112 optimizations allow the code generator to make use of some instructions which
113 would otherwise not be usable (such as ``fsin`` on X86).
117 Print statistics recorded by code-generation passes.
119 .. option:: --time-passes
121 Record the amount of time needed for each pass and print a report to standard
124 .. option:: --load=<dso_path>
126 Dynamically load ``dso_path`` (a path to a dynamically shared object) that
127 implements an LLVM target. This will permit the target name to be used with
128 the :option:`-march` option so that code can be generated for that target.
130 .. option:: -meabi=[default|gnu|4|5]
132 Specify which EABI version should conform to. Valid EABI versions are *gnu*,
133 *4* and *5*. Default value (*default*) depends on the triple.
135 .. option:: -stack-size-section
137 Emit the .stack_sizes section which contains stack size metadata. The section
138 contains an array of pairs of function symbol values (pointer size) and stack
139 sizes (unsigned LEB128). The stack size values only include the space allocated
140 in the function prologue. Functions with dynamic stack allocations are not
144 Tuning/Configuration Options
145 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
147 .. option:: --print-machineinstrs
149 Print generated machine code between compilation phases (useful for debugging).
151 .. option:: --regalloc=<allocator>
153 Specify the register allocator to use.
154 Valid register allocators are:
158 Basic register allocator.
162 Fast register allocator. It is the default for unoptimized code.
166 Greedy register allocator. It is the default for optimized code.
170 Register allocator based on 'Partitioned Boolean Quadratic Programming'.
172 .. option:: --spiller=<spiller>
174 Specify the spiller to use for register allocators that support it. Currently
175 this option is used only by the linear scan register allocator. The default
176 ``spiller`` is *local*. Valid spillers are:
186 Intel IA-32-specific Options
187 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
189 .. option:: --x86-asm-syntax=[att|intel]
191 Specify whether to emit assembly code in AT&T syntax (the default) or Intel
197 If :program:`llc` succeeds, it will exit with 0. Otherwise, if an error
198 occurs, it will exit with a non-zero value.