Requsting help: reproducibility regressions in pwgen and e2fsprogs?

[Theodore Ts'o]

Hi there, I've recently uploaded updates to sid for two packages I
maintain, pwgen and e2fsprogs.  I uploaded them source-only, and as
far as I know I don't _think_ I've made any changes that should have
broken reproducibility --- especially with pwgen.

Yet both seem to be broken, and with pwgen, it looks like it's being
the debugging pathname has somehow leaked into the ELF / debugging
information:

https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/diffoscope-results/pwgen.html
https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/diffoscope-results/e2fsprogs.html

Could someone help me by taking a quick peak at the diffoscope results
(especially pwgen, since I think that will be the simpler one), and
let me know what I should do to fix things?

Many thanks!!

					- Ted