Thanks for the replies; my apologies for the delay in repying back.... Based on the replies, I was able to track the problem down. The cdb "patch" is incompatable with the latest version of portage. Removing that from /etc/portage/modules allows portage to work.
That said, there is still the bug of using the filesystem as a database. It now takes about thirty minutes for any emerge run to finish reading in /var/cache/edb/dep and most of /var/db/pkg. It is imperative that portage use a single file for those databases. (One per db is fine, in case the above is ambiguous.) This is the same problem that git ran into, leading to pack files, or that cnews and inn hit, leading to the rotating spool files. It may work reasonably well on a headless server, with only a few packages installed. But on a general purpose workstation with many packages installed it falls over hard. Incidently, the above 30 minute minimum run time is even when running with --nodeps. I'll also be posting a bug report about unmerging. Portage is stat(2)ing *every* file under a directory that might need removal rather than just checking whether there are any entries in the dir. Running stat(2) on everything in /usr/lib, /usr/bin, /usr/share/doc and similar completely thrashes the dirent cache. It is even worse than running ldconfig when no changes were made in dirs that ldconfig monitors. Thanks for the work; I hope to see performance improvements this year. -JimC -- James Cloos <[EMAIL PROTECTED]> OpenPGP: 1024D/ED7DAEA6 -- gentoo-portage-dev@gentoo.org mailing list
