hardware_TrimIntegrity: Compare read speed before/after trim.

Use FIO to measure 4K random read QD32 performance before / after
trim in hardware_TrimIntegrity. Device that doing trim properly
should report some speed difference because we don't need to
access NAND when read from trimmed LBA.

TEST=Tested in falco.
Tested passed. Result shows that median read latency is 4x better.

Change-Id: I76fe8a27c49e8eacc782e5c2cbcadc8e24bad7bd
Reviewed-on: https://chromium-review.googlesource.com/183677
Tested-by: Puthikorn Voravootivat <puthik@chromium.org>
Reviewed-by: Grant Grundler <grundler@chromium.org>
Commit-Queue: Puthikorn Voravootivat <puthik@chromium.org>
2 files changed