llvm-capstone/lld/utils/link.yaml
Rafael Espindola 6dc954a0d9 Add a script to run various benchmarks and send the result to lnt.
Lnt is both a server and a set of script for benchmarking llvm.

I don't think it makes sense to use the scripts for lld since our
benchmarks are quite different.

The server on the other hand is very general and seems to work well
for tracking any quantities.

This patch adds a script to lld that can be used to run various
benchmarks and send the result to lnt.

The benchmarks are assumed to each be a response file in a
subdirectory. Each subdirectory can contain multiple response
files. That can be used to have a plain response.txt and a
response-icf.txt for example. The name of each benchmark is the
combination of the directory name and the "flavor": firefox-gc,
chromium-icf, etc.

For the first version the script uses perf and collects all the
metrics that a plain "perf stat" prints.

This script can then be used by a developer to test a patch or by a
bot to keep track of lld's performance.

llvm-svn: 318158
2017-11-14 16:40:30 +00:00

40 lines
786 B
YAML

format_version: '2'
name: link
run_fields:
- name: llvm_project_revision
order: true
machine_fields:
- name: hardware
- name: os
metrics:
- name: branch-misses
bigger_is_better: false
type: Real
- name: stalled-cycles-frontend
bigger_is_better: false
type: Real
- name: branches
bigger_is_better: false
type: Real
- name: context-switches
bigger_is_better: false
type: Real
- name: cpu-migrations
bigger_is_better: false
type: Real
- name: cycles
bigger_is_better: false
type: Real
- name: instructions
bigger_is_better: false
type: Real
- name: seconds-elapsed
bigger_is_better: false
type: Real
- name: page-faults
bigger_is_better: false
type: Real
- name: task-clock
bigger_is_better: false
type: Real