- zap heuristic#3, not critical
- don't rebuild the Fetch queue all the time, it's too expensive
- fix bugs in FetchQueue computation, correctly handle 1 distfile case
- create a quick path thru adjust* that handles only new paths.
- use it during LISTING while fetching files, to make more files available
quicker.
- model distfiles as FDEPENDS, DIST, so that they can get dumped.
- add -v option that dumps a bit more stuff in log files.
with these, dpb will fetch all distfiles in a not too surprising order,
and it won't consume 40% cpu while doing so...
- clean up clock handling, jobs with timer.
- pass more parameters thru state
- tweak the engine to handle distfiles
- lots of new code in Fetch for new jobs.
Deprecate: -t and -T
use -DCONNECTION_TIMEOUT=... -DDISPLAY_TIMEOUT=... instead
New feature: -DSTUCK_TIMEOUT=
kill tasks when they don't show any progress for that long.
makes for a "quick listing" of the most important ports first (e.g., groff).
document -R, even though it's not perfect yet.
flesh out all the recent parts and update the BUGS AND LIMITATIONS section.
Somewhat inefficient (deep check of package vs port signature).
More proper approach would be to tell the engine to always build, have first
step be a check for signature if the package exist, and abort if it matches:
- here we double check everything
- propagating rebuilds means non-uptodate packages give rise to non uptodate.
- if several cores are available, recheck errors, so that we're not stuck
waiting for a core to finish to start new shit.
- $logdir/stop-<machine_name> allows you to stop a given machine while
you clean it up and (possibly) restart it (e.g., won't start new jobs)
turns out we probably don't need FULLSUBDIR after all.
compute SUBDIR better, possibly with quotes.
Remove trailing / on subdirs, in case people use completion.
