On Sat, 2019-04-20 at 22:55 +0200, Gilles Filippini wrote: > Gilles Filippini a écrit le 20/04/2019 à 19:22 : > > Adam D. Barratt a écrit le 20/04/2019 à 12:41 : > > > On Sat, 2019-04-20 at 10:43 +0100, Adam D. Barratt wrote: > > > > On Tue, 2019-04-16 at 19:37 +0200, Gilles Filippini wrote: > > > > > Adam D. Barratt a écrit le 14/04/2019 à 22:48 : > > > > > > Control: tags -1 + confirmed > > > > > > > > > > > > On Fri, 2019-02-22 at 17:29 +0100, Gilles Filippini wrote: > > > > > > > Because RC bug #922453 affects Stretch as well, I propose > > > > > > > a > > > > > > > stable > > > > > > > update > > > > > > > for cernlib. Debdiff attached. > > > > > > > > > > > > cernlib (20061220+dfsg3-4.3+deb9u1) stable-proposed- > > > > > > updates; > > > > > > urgency=medium > > > > > > > > > > > > Please use "stretch" rather than "stable-proposed-updates" > > > > > > and go > > > > > > ahead. > > > > > > > > > > Done. > > > > > > > > Unfortunately, the arm64 build appears to be hitting #863152 in > > > > binutils. I'm not sure if there's anything that can be done to > > > > work > > > > around this in cernlib, but noting it here in any case, as it > > > > means > > > > that cernlib won't be getting into 9.9. > > > > > > For the record, if a fix for cernlib can get sorted before the > > > 9.9 > > > window closes later this weekend (e.g. simply disabling PIE on > > > arm64 > > > may be sufficient) then getting the updates in should be OK. > > > > The cernlib build system is far from easy to understand and patch. > > I > > don't think I'll be able to find a solution in time for 9.9. > > I might have been over pessimistic. I found a way to force -no-pie > when linking fortran executables. Fix tested on an arm64 porter box. > Debdiff attached.
Great, thanks. Please feel free to upload the new version. Regards, Adam
