llvm-project/llvm/docs/CommandGuide/llvm-opt-report.rst
Tobias Stadler 1f34d68c4f
[Remarks] Remove yaml-strtab format (#144527)
Background: The yaml-strtab format looks just like the yaml format,
except that the values in the key/value pairs of the remarks are
deduplicated and replaced by indices into a string table (see removed
test cases for examples). The motivation behind this format was to
reduce size of the remarks files. However, it was quickly superseded by
the bitstream format.

Therefore, remove the yaml-strtab format, as it doesn't have a good
usecase anymore:
  - It isn't particularly efficient
  - It isn't human-readable
  - It isn't straightforward to parse in external tools that can't use the
remarks library. We don't even support it in opt-viewer.

llvm-remarkutil is also missing options to parse/convert yaml-strtab, so
the chance that anyone is actually using this format is low.
2025-06-18 14:25:41 +01:00

121 lines
3.1 KiB
ReStructuredText

llvm-opt-report - generate optimization report from YAML
========================================================
.. program:: llvm-opt-report
SYNOPSIS
--------
:program:`llvm-opt-report` [*options*] [input]
DESCRIPTION
-----------
:program:`llvm-opt-report` is a tool to generate an optimization report from YAML optimization record files.
You need to create an input YAML optimization record file before running :program:`llvm-opt-report`.
It provides information on the execution time, memory usage, and other details of each optimization pass.
.. code-block:: console
$ clang -c foo.c -o foo.o -O3 -fsave-optimization-record
Then, you create a report using the :program:`llvm-opt-report` command with the YAML optimization record file :file:`foo.opt.yaml` as input.
.. code-block:: console
$ llvm-opt-report foo.opt.yaml -o foo.lst
foo.lst is the generated optimization report.
.. code-block::
< foo.c
1 | void bar();
2 | void foo() { bar(); }
3 |
4 | void Test(int *res, int *c, int *d, int *p, int n) {
5 | int i;
6 |
7 | #pragma clang loop vectorize(assume_safety)
8 V4,1 | for (i = 0; i < 1600; i++) {
9 | res[i] = (p[i] == 0) ? res[i] : res[i] + d[i];
10 | }
11 |
12 U16 | for (i = 0; i < 16; i++) {
13 | res[i] = (p[i] == 0) ? res[i] : res[i] + d[i];
14 | }
15 |
16 I | foo();
17 |
18 | foo(); bar(); foo();
I | ^
I | ^
19 | }
20 |
Symbols printed on the left side of the program indicate what kind of optimization was performed.
The meanings of the symbols are as follows:
- I: The function is inlined.
- U: The loop is unrolled. The following number indicates the unroll factor.
- V: The loop is vectorized. The following numbers indicate the vector length and the interleave factor.
.. note::
If a specific line of code is output twice, it means that the same optimization pass was applied to that
line of code twice, and the pass was able to further optimize the code on the second iteration.
OPTIONS
-------
If ``input`` is "``-``" or omitted, :program:`llvm-opt-report` reads from standard
input. Otherwise, it will read from the specified filename.
If the :option:`-o` option is omitted, then :program:`llvm-opt-report` will send its output
to standard output. If the :option:`-o` option specifies "``-``", then the output will also
be sent to standard output.
.. option:: --help
Display available options.
.. option:: --version
Display the version of this program.
.. option:: --format=<string>
The format of the optimization record file.
The Argument is one of the following:
- yaml
- bitstream
.. option:: --no-demangle
Do not demangle function names.
.. option:: -o=<string>
Output file.
.. option:: -r=<string>
Root for relative input paths.
.. option:: -s
Do not include vectorization factors, etc.
EXIT STATUS
-----------
:program:`llvm-opt-report` returns 0 on success. Otherwise, an error message is printed
to standard error, and the tool returns 1.