diff options
| author | Clement Courbet <courbet@google.com> | 2019-01-30 16:02:20 +0000 |
|---|---|---|
| committer | Clement Courbet <courbet@google.com> | 2019-01-30 16:02:20 +0000 |
| commit | 362653f7afec25c71604f18856011bd02071a555 (patch) | |
| tree | 7c0fd27458db2391fed8e994e8f4d1e04d1baf1b /llvm/docs/CommandGuide | |
| parent | 4f9e3ce070c5ec7ee25c428478d1c55704a8da51 (diff) | |
| download | bcm5719-llvm-362653f7afec25c71604f18856011bd02071a555.tar.gz bcm5719-llvm-362653f7afec25c71604f18856011bd02071a555.zip | |
[llvm-exegesis] Add throughput mode.
Summary:
This just uses the latency benchmark runner on the parallel uops snippet
generator.
Fixes PR37698.
Reviewers: gchatelet
Subscribers: tschuett, RKSimon, llvm-commits
Differential Revision: https://reviews.llvm.org/D57000
llvm-svn: 352632
Diffstat (limited to 'llvm/docs/CommandGuide')
| -rw-r--r-- | llvm/docs/CommandGuide/llvm-exegesis.rst | 18 |
1 files changed, 10 insertions, 8 deletions
diff --git a/llvm/docs/CommandGuide/llvm-exegesis.rst b/llvm/docs/CommandGuide/llvm-exegesis.rst index f27db9e57ed..bbd90563005 100644 --- a/llvm/docs/CommandGuide/llvm-exegesis.rst +++ b/llvm/docs/CommandGuide/llvm-exegesis.rst @@ -10,13 +10,13 @@ DESCRIPTION ----------- :program:`llvm-exegesis` is a benchmarking tool that uses information available -in LLVM to measure host machine instruction characteristics like latency or port -decomposition. +in LLVM to measure host machine instruction characteristics like latency, +throughput, or port decomposition. Given an LLVM opcode name and a benchmarking mode, :program:`llvm-exegesis` generates a code snippet that makes execution as serial (resp. as parallel) as -possible so that we can measure the latency (resp. uop decomposition) of the -instruction. +possible so that we can measure the latency (resp. inverse throughput/uop decomposition) +of the instruction. The code snippet is jitted and executed on the host subtarget. The time taken (resp. resource usage) is measured using hardware performance counters. The result is printed out as YAML to the standard output. @@ -37,11 +37,13 @@ instruction, run: $ llvm-exegesis -mode=latency -opcode-name=ADD64rr -Measuring the uop decomposition of an instruction works similarly: +Measuring the uop decomposition or inverse throughput of an instruction works similarly: .. code-block:: bash $ llvm-exegesis -mode=uops -opcode-name=ADD64rr + $ llvm-exegesis -mode=inverse_throughput -opcode-name=ADD64rr + The output is a YAML document (the default is to write to stdout, but you can redirect the output to a file using `-benchmarks-file`): @@ -186,7 +188,7 @@ OPTIONS Specify the custom code snippet to measure. See example 2 for details. Either `opcode-index`, `opcode-name` or `snippets-file` must be set. -.. option:: -mode=[latency|uops|analysis] +.. option:: -mode=[latency|uops|inverse_throughput|analysis] Specify the run mode. @@ -197,8 +199,8 @@ OPTIONS .. option:: -benchmarks-file=</path/to/file> - File to read (`analysis` mode) or write (`latency`/`uops` modes) benchmark - results. "-" uses stdin/stdout. + File to read (`analysis` mode) or write (`latency`/`uops`/`inverse_throughput` + modes) benchmark results. "-" uses stdin/stdout. .. option:: -analysis-clusters-output-file=</path/to/file> |
