blob: 617aadc5dfe38377cd4bea9dc3adf9fa1eb4c673 [file] [log] [blame]
George Burgess IV50e73822018-10-26 20:56:03 +00001=============================================================
2How To Build Clang and LLVM with Profile-Guided Optimizations
3=============================================================
4
5Introduction
6============
7
8PGO (Profile-Guided Optimization) allows your compiler to better optimize code
9for how it actually runs. Users report that applying this to Clang and LLVM can
10decrease overall compile time by 20%.
11
12This guide walks you through how to build Clang with PGO, though it also applies
13to other subprojects, such as LLD.
14
Nico Weberca053272020-12-07 11:29:25 -050015If you want to build other software with PGO, see the `end-user documentation
16for PGO <https://clang.llvm.org/docs/UsersManual.html#profile-guided-optimization>`_.
17
18
19Using preconfigured CMake caches
20================================
21
22See https://llvm.org/docs/AdvancedBuilds.html#multi-stage-pgo
George Burgess IV50e73822018-10-26 20:56:03 +000023
24Using the script
25================
26
27We have a script at ``utils/collect_and_build_with_pgo.py``. This script is
28tested on a few Linux flavors, and requires a checkout of LLVM, Clang, and
Hans Wennborg1e5de802020-01-07 16:06:14 +010029compiler-rt. Despite the name, it performs four clean builds of Clang, so it
George Burgess IV50e73822018-10-26 20:56:03 +000030can take a while to run to completion. Please see the script's ``--help`` for
31more information on how to run it, and the different options available to you.
32If you want to get the most out of PGO for a particular use-case (e.g. compiling
33a specific large piece of software), please do read the section below on
34'benchmark' selection.
35
36Please note that this script is only tested on a few Linux distros. Patches to
37add support for other platforms, as always, are highly appreciated. :)
38
39This script also supports a ``--dry-run`` option, which causes it to print
40important commands instead of running them.
41
42
43Selecting 'benchmarks'
44======================
45
46PGO does best when the profiles gathered represent how the user plans to use the
47compiler. Notably, highly accurate profiles of llc building x86_64 code aren't
48incredibly helpful if you're going to be targeting ARM.
49
50By default, the script above does two things to get solid coverage. It:
51
52- runs all of Clang and LLVM's lit tests, and
53- uses the instrumented Clang to build Clang, LLVM, and all of the other
54 LLVM subprojects available to it.
55
56Together, these should give you:
57
58- solid coverage of building C++,
59- good coverage of building C,
60- great coverage of running optimizations,
61- great coverage of the backend for your host's architecture, and
62- some coverage of other architectures (if other arches are supported backends).
63
64Altogether, this should cover a diverse set of uses for Clang and LLVM. If you
65have very specific needs (e.g. your compiler is meant to compile a large browser
66for four different platforms, or similar), you may want to do something else.
67This is configurable in the script itself.
68
69
70Building Clang with PGO
71=======================
72
Nico Weberca053272020-12-07 11:29:25 -050073If you prefer to not use the script or the cmake cache, this briefly goes over
74how to build Clang/LLVM with PGO.
George Burgess IV50e73822018-10-26 20:56:03 +000075
76First, you should have at least LLVM, Clang, and compiler-rt checked out
77locally.
78
79Next, at a high level, you're going to need to do the following:
80
811. Build a standard Release Clang and the relevant libclang_rt.profile library
822. Build Clang using the Clang you built above, but with instrumentation
833. Use the instrumented Clang to generate profiles, which consists of two steps:
84
85 - Running the instrumented Clang/LLVM/lld/etc. on tasks that represent how
86 users will use said tools.
87 - Using a tool to convert the "raw" profiles generated above into a single,
88 final PGO profile.
89
904. Build a final release Clang (along with whatever other binaries you need)
91 using the profile collected from your benchmark
92
93In more detailed steps:
94
951. Configure a Clang build as you normally would. It's highly recommended that
96 you use the Release configuration for this, since it will be used to build
97 another Clang. Because you need Clang and supporting libraries, you'll want
98 to build the ``all`` target (e.g. ``ninja all`` or ``make -j4 all``).
99
1002. Configure a Clang build as above, but add the following CMake args:
101
102 - ``-DLLVM_BUILD_INSTRUMENTED=IR`` -- This causes us to build everything
103 with instrumentation.
104 - ``-DLLVM_BUILD_RUNTIME=No`` -- A few projects have bad interactions when
105 built with profiling, and aren't necessary to build. This flag turns them
106 off.
107 - ``-DCMAKE_C_COMPILER=/path/to/stage1/clang`` - Use the Clang we built in
108 step 1.
109 - ``-DCMAKE_CXX_COMPILER=/path/to/stage1/clang++`` - Same as above.
110
111 In this build directory, you simply need to build the ``clang`` target (and
112 whatever supporting tooling your benchmark requires).
113
1143. As mentioned above, this has two steps: gathering profile data, and then
115 massaging it into a useful form:
116
117 a. Build your benchmark using the Clang generated in step 2. The 'standard'
118 benchmark recommended is to run ``check-clang`` and ``check-llvm`` in your
119 instrumented Clang's build directory, and to do a full build of Clang/LLVM
120 using your instrumented Clang. So, create yet another build directory,
121 with the following CMake arguments:
122
123 - ``-DCMAKE_C_COMPILER=/path/to/stage2/clang`` - Use the Clang we built in
124 step 2.
125 - ``-DCMAKE_CXX_COMPILER=/path/to/stage2/clang++`` - Same as above.
126
127 If your users are fans of debug info, you may want to consider using
128 ``-DCMAKE_BUILD_TYPE=RelWithDebInfo`` instead of
129 ``-DCMAKE_BUILD_TYPE=Release``. This will grant better coverage of
130 debug info pieces of clang, but will take longer to complete and will
131 result in a much larger build directory.
132
133 It's recommended to build the ``all`` target with your instrumented Clang,
134 since more coverage is often better.
135
Hans Wennborgad84a402018-12-05 08:35:30 +0000136 b. You should now have a few ``*.profraw`` files in
George Burgess IV50e73822018-10-26 20:56:03 +0000137 ``path/to/stage2/profiles/``. You need to merge these using
138 ``llvm-profdata`` (even if you only have one! The profile merge transforms
139 profraw into actual profile data, as well). This can be done with
Hans Wennborgad84a402018-12-05 08:35:30 +0000140 ``/path/to/stage1/llvm-profdata merge
141 -output=/path/to/output/profdata.prof path/to/stage2/profiles/*.profraw``.
George Burgess IV50e73822018-10-26 20:56:03 +0000142
1434. Now, build your final, PGO-optimized Clang. To do this, you'll want to pass
144 the following additional arguments to CMake.
145
146 - ``-DLLVM_PROFDATA_FILE=/path/to/output/profdata.prof`` - Use the PGO
147 profile from the previous step.
148 - ``-DCMAKE_C_COMPILER=/path/to/stage1/clang`` - Use the Clang we built in
149 step 1.
150 - ``-DCMAKE_CXX_COMPILER=/path/to/stage1/clang++`` - Same as above.
151
152 From here, you can build whatever targets you need.
153
154 .. note::
155 You may see warnings about a mismatched profile in the build output. These
156 are generally harmless. To silence them, you can add
157 ``-DCMAKE_C_FLAGS='-Wno-backend-plugin'
158 -DCMAKE_CXX_FLAGS='-Wno-backend-plugin'`` to your CMake invocation.
159
160
161Congrats! You now have a Clang built with profile-guided optimizations, and you
162can delete all but the final build directory if you'd like.
163
164If this worked well for you and you plan on doing it often, there's a slight
165optimization that can be made: LLVM and Clang have a tool called tblgen that's
166built and run during the build process. While it's potentially nice to build
167this for coverage as part of step 3, none of your other builds should benefit
168from building it. You can pass the CMake options
169``-DCLANG_TABLEGEN=/path/to/stage1/bin/clang-tblgen
170-DLLVM_TABLEGEN=/path/to/stage1/bin/llvm-tblgen`` to steps 2 and onward to avoid
171these useless rebuilds.