The memory usage is way worse than that of tar, but I'm optimistic that we can improve things a bit in that realm. Perhaps (for example) by storing PackedString file paths, or by making the directory-reading portion of slurp lazy. In any case, 450M isn't such bad maximum memory consumption for a project the size of the kernel.
Memory use shouldn't be anywhere near that. Do you have the patch to make the test not hang on to the patch (can't remember if that's in d-u yet)?
If not, you can instead give --no-test to record.
You're right. I'm using a somewhat older version of darcs-unstable for which I happen to have a binary here at work. Giving --no-test does indeed keep the memory usage down below 10M. It also speeds up the result to less than four times slower than tar in terms of wallclock times, and maybe six or seven times the CPU time.
Two thumbs up! Great work David and Ian!!
-- Daan Leijen.
_______________________________________________ darcs-users mailing list [email protected] http://www.abridgegame.org/mailman/listinfo/darcs-users
