Il 10/06/2014 13:48, Michael S. Tsirkin ha scritto:
Even measurable speedups in most scenarios would not matter. I don't care
if a kernel compile takes 2 minutes vs. 110 seconds (for a 10% speedup),
even though it's great that THP can speed up such a common task.
True. But I am not sure why would such a user play with vhost-user at all.
That one seems to mostly be about using aggressive polling
to drive down guest to guest latency.
But then there is so much more you have to do to get the performance
you're looking for, including using GB hugepages which needs hugetlbfs
anyway.
Anyhow, since there is a warning and the behavior is the same as 2.0 the
question is moot, I think. Renaming memory-backend-file to
memory-backend-hugetlbfs would suggest that there is a regression, which
is not the case.
Paolo