First draft for reducing latency of Twalk handling by reducing the amount of thread hops, similar to previous Treaddir optimization. The performance gain is not as spectacular as on Treaddir, but there is definitely a measurable difference.
With the benchmark of patch 1, the runtime of the Twalk test was cut in half. In real world tests I measured a performance gain (i.e. running an entire guest OS, and hence mixed with all othe kinds of 9p requests) of about 2%, again measured in a mix, not concentrated on Treaddir at all. Independent of the actual performance optimization (patch 3), there are some things about Twalk handling in general which I am yet unsure about. So I'll add some of my thoughts as reply to patch 3, and depending on that I might still cleanup / reduce some of the code. Christian Schoenebeck (3): 9pfs: Twalk benchmark 9pfs: capture root stat 9pfs: reduce latency of Twalk hw/9pfs/9p.c | 128 +++++++++++++++++++++++------------ hw/9pfs/9p.h | 2 +- tests/qtest/virtio-9p-test.c | 30 ++++++++ 3 files changed, 116 insertions(+), 44 deletions(-) -- 2.20.1
