diff options
| author | Sanjay Patel <spatel@rotateright.com> | 2018-04-10 18:10:14 +0000 |
|---|---|---|
| committer | Sanjay Patel <spatel@rotateright.com> | 2018-04-10 18:10:14 +0000 |
| commit | 40ad9262bc84a601be81a7864780db8c7f0ab984 (patch) | |
| tree | d5302524b683fd34c9c7623487f22ff948bd2dfc /llvm/docs | |
| parent | 58568254bcf25145dcf763b0395d32284e620777 (diff) | |
| download | bcm5719-llvm-40ad9262bc84a601be81a7864780db8c7f0ab984.tar.gz bcm5719-llvm-40ad9262bc84a601be81a7864780db8c7f0ab984.zip | |
[llvm-mca] reorder text
On 2nd reading, putting the C example after the bit about
multiple regions makes this flow better.
llvm-svn: 329732
Diffstat (limited to 'llvm/docs')
| -rw-r--r-- | llvm/docs/CommandGuide/llvm-mca.rst | 18 |
1 files changed, 9 insertions, 9 deletions
diff --git a/llvm/docs/CommandGuide/llvm-mca.rst b/llvm/docs/CommandGuide/llvm-mca.rst index 03913076866..35642b57af4 100644 --- a/llvm/docs/CommandGuide/llvm-mca.rst +++ b/llvm/docs/CommandGuide/llvm-mca.rst @@ -36,7 +36,14 @@ substring ``LLVM-MCA-END`` marks the end of a code region. For example: ... # LLVM-MCA-END -Inline assembly directives may also be used from source code to annotate the +Multiple regions can be specified provided that they do not overlap. A code +region can have an optional description. If no user-defined region is specified, +then :program:`llvm-mca` assumes a default region which contains every +instruction in the input file. Every region is analyzed in isolation, and the +final performance report is the union of all the reports generated for every +code region. + +Inline assembly directives may be used from source code to annotate the assembly text: .. code-block:: c++ @@ -54,14 +61,7 @@ directly into llvm-mca for analysis: .. code-block:: bash - $ clang foo.cpp -O2 -target x86_64-unknown-unknown -S -o - | llvm-mca -mcpu=btver2 - -Multiple regions can be specified provided that they do not overlap. A code -region can have an optional description. If no user defined region is specified, -then :program:`llvm-mca` assumes a default region which contains every -instruction in the input file. Every region is analyzed in isolation, and the -final performance report is the union of all the reports generated for every -code region. + $ clang foo.c -O2 -target x86_64-unknown-unknown -S -o - | llvm-mca -mcpu=btver2 OPTIONS ------- |
