Am 13.11.18 um 18:13 schrieb Simon McVittie: > On Tue, 13 Nov 2018 at 14:46:18 +0100, Michael Biebl wrote: >> On Thu, 8 Nov 2018 13:27:59 +0000 Simon McVittie <[email protected]> wrote >>> Depending on the level of testing done, number of bugs found and fixed, >>> and level of adoption of merged /usr, we could assess whether to revert >>> that change after buster is released. >> >> I guess we will only ever start finding those bugs, if we actually build >> stuff in a merged-/usr environment. > > Note that https://tests.reproducible-builds.org/debian/reproducible.html > is now doing this: build2 is done with the usrmerge package installed > (except on stretch). If the package FTBFS, that will be detected; if the > package builds with different content, it will show up as non-reproducible. > I've reported bugs in gnupg2 and libapp-options-perl for this.
How exactly did you find those packages? Do you have a list somewhere of packages which just started to fail to build reproducibly recently and you have investigated the result further then? We'd somehow need a diff between a full archive build with the merged-usr variation enabled and one where it is disabled, or do you see a another why to process the results of reproducible-builds.org in an efficient way? Regards, Michael -- Why is it that all of the instruments seeking intelligent life in the universe are pointed away from Earth?
signature.asc
Description: OpenPGP digital signature
