blob: 2c8017d14ec0ef0eaca4eabde7dfda9cfb799f5a [file] [log] [blame]
# This is just a random list of things that I think need to be done
# to rteval
1. Figure out how to add more loads. Seems that we could do it two ways:
- create a wrapper python module just like is done for kcompile and hackbench
then add the new module to a config file so rteval loads it on startup
(i.e. a plugin architecture)
- figure out some generic load wrapper so that arbitrary commands may be
used as loads.
2. Should we add a hwlatdetect option?
3. Potential new loads/measurements:
- dbench
- AMQP latencytest/perftest (over loopback)
- bonnie/bonnie++
- iozone
- KVM running guest OS
4. Add an option to run 'perf record' while a run is in progress. This is
mainly a diagnostic option to debug *why* something is performing