[Virtio-fs] [PATCH 1/2] virtio-fs: try hard to do inline reclaim

Vivek Goyal vgoyal at redhat.com
Thu Aug 15 12:45:24 UTC 2019


On Wed, Aug 14, 2019 at 01:51:14PM -0700, Liu Bo wrote:

[..]
> > Also, as of now inline reclaims are enabled only for writes. (no reads, no
> > mmap). So this should not help with the randread improvements you posted
> > in second patch.
> >
> 
> I explained it in another email under the thread.

Going forward, can you please generate patches and test results on top of
latest virtio-fs code I am putting on github/gitlab. If you test it on
your internal kernels where you are carrying your own patches, these
results don't help.

> 
> > Does this work without deadlocks with cache_size=2M.
> 
> Well, with or without these 2 changes, on cache_size=2M kernel build end up with
> waiting for dmap "forever", at least lockdep didn't report deadlocks.

Have you tried the latest kernel and does that hang too? Or it is the
hang with your kernel?

https://github.com/rhvgoyal/linux/commits/vivek-5.3-aug-14-2019

With above kernel I tested with dmap=1 and it works for me. I ran
"blogbench -d /mnt/virtio-fs/test-dir/" and that works.

In last few weeks I have worked on sorting out various issues with
hangs/deadlocks with dmap=1. 

Thanks
Vivek




More information about the Virtio-fs mailing list