tech-toolchain archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: Sets, subsets, syspkgs, and MK*
Date: Wed, 16 Dec 2009 16:39:33 +0000
From: Quentin Garnier <cube%cubidou.net@localhost>
Message-ID: <20091216163933.GH15236%taryn.cubidou.net@localhost>
[An aside to this thread ...]
| But a lot of work has been done lately to get reproducible builds,
yes, and I meant to say it at the time but I was running way behind on
my list reading, and it wouldn't have been a timely reply - now enough
time has passed that maybe I can get away with this comment...
That work was done in (slightly) the wrong way - largely it consisted
in moving all of the dates out of the binary files. That was a mistake.
What should have been done was to put a constant date into the files,
the date when the release was announced (or last built before being
announced, or anything else that takes the release engineer's fancy).
That is, rather than 0, stick a constant there - as long as the constant
isn't lost, it is just as reproducible, but also means that we can easily
make a different set of binaries (for example, a binary patch could
update the date to the patch release date in files that are patched,
so tracking patch installation is much cleaner).
kre
- References:
- Re: Sets, subsets, syspkgs, and MK*
- Re: Sets, subsets, syspkgs, and MK*
- Re: Sets, subsets, syspkgs, and MK*
- Re: Sets, subsets, syspkgs, and MK*
- Re: Sets, subsets, syspkgs, and MK*
- Re: Sets, subsets, syspkgs, and MK*
- Re: Sets, subsets, syspkgs, and MK*
- Re: Sets, subsets, syspkgs, and MK*
- Re: Sets, subsets, syspkgs, and MK*
- Re: Sets, subsets, syspkgs, and MK*
Home |
Main Index |
Thread Index |
Old Index