On Thu, 2007-09-27 at 23:50 -0700, Andrew Morton wrote:

quoted text
> What we _don't_ want to happen is for other processes which are writing t=
o
quoted text
> other, non-dead devices to get collaterally blocked.  We have patches whi=
ch
quoted text
> might fix that queued for 2.6.24.  Peter?

Nasty problem, don't do that :-)

But yeah, with per BDI dirty limits we get stuck at whatever ratio that
NFS server/mount (?) has - which could be 100%. Other processes will
then work almost synchronously against their BDIs but it should work.

[ They will lower the NFS-BDI's ratio, but some fancy clipping code will
  limit the other BDIs their dirty limit to not exceed the total limit.
  And with all these NFS pages stuck, that will still be nothing. ]