[Rpm-maint] [rpm-software-management/rpm] Please support smaller build-ids (#950)

Frank Ch. Eigler notifications at github.com
Mon Nov 25 17:16:23 UTC 2019


>  Note that there is precedent for using only 64 bits. The dwarf standard (http://www.dwarfstd.org/doc/DWARF5.pdf section 7.32 "Type Signature Computation") uses the last 8 bytes of md5 for type signatures.

I see your point, but it is likely that the number of binaries in an OS distribution is orders of magnitude greater than the number of distinct type declarations / signatures of its source code.  And AIUI, the type signature code is meant for use primarily for indexing within or amongst a small set of related dwarf files already, i.e., another level of buildid type indexing would already have been applied to tell them apart.

This is not to say that rpm per se should be enforcing distro-scale concerns - but if it doesn't, some other piece of distro releng tooling must.  An 8-byte (16-hex-char) code is likely too small within a single distro/arch, never mind globally.  (I'm collecting some buildid population figures for a slice of fedora via debuginfod as we speak.)

-- 
You are receiving this because you commented.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/950#issuecomment-558253495
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rpm.org/pipermail/rpm-maint/attachments/20191125/54bd0b03/attachment.html>


More information about the Rpm-maint mailing list