Hi Christophe, On Sun, Mar 13, 2016 at 10:21 PM, Christophe JAILLET <christophe.jail...@wanadoo.fr> wrote: > Hi, > > while looking at potential backport to synch and trunk, I arrived on > r1619483. > > This is a follow up of r1619444 which is a follow up of r1619383. > These 2 have been bakcported (see r1669555) > > So I was wondering if r1619483 should also have been backported at the same > time?
It's been a long time, but it seems that indeed this follow up did not make it in r1669555 :/ Recollecting, this commit uses the correct way to account for inflated bytes should there be data already available before the inflate() call (otherwise some bytes may be counted twice). I think we should merge it now, better late than never :) Regards, Yann.
