Re: cpu bound I/O behaviour in linux 5.4 (possibly others)

[Marc Lehmann <schmorp@schmorp.de>]

On Fri, Feb 14, 2020 at 12:20:46PM +0000, Filipe Manana <fdmanana@gmail.com> wrote:
> So you can't deduce that the free space cache is being used, and
> despite being the default, it was not mentioned by Marc if he's not
> using already the free space tree (-o space_cache=v2).

Today, during an idle period, I stopped the news process, waited 15
minutes for I/O tand CPU to become mostly idle, and did this:

   # mount -noremount,clear_cache,space_cache=v1 /localvol5
   # mount -noremount,clear_cache,space_cache=v2 /localvol5

Which, to my surprise, didn't make btrfs complain too much:

   [1378546.558533] BTRFS info (device dm-17): force clearing of disk cache
   [1378546.558536] BTRFS info (device dm-17): enabling disk space caching
   [1378546.558537] BTRFS info (device dm-17): disk space caching is enabled

   [1378553.868438] BTRFS info (device dm-17): enabling free space tree
   [1378553.868440] BTRFS info (device dm-17): using free space tree

I don't know if this was effective in clearing the cache, but it didn't
change the behaviour - as soon as the new process started writing files
again, it was at 100% cpu.

-- 
                The choice of a       Deliantra, the free code+content MORPG
      -----==-     _GNU_              http://www.deliantra.net
      ----==-- _       generation
      ---==---(_)__  __ ____  __      Marc Lehmann
      --==---/ / _ \/ // /\ \/ /      schmorp@schmorp.de
      -=====/_/_//_/\_,_/ /_/\_\