>>>> Nicholas Frizzell <nfrizzel(a)redhat.com> writes:
Has anyone investigated making use of multithreading for check-files
previously?
I'm not sure multithreading is all that meaningful for that script; it's
really quite simple, after all. It runs find to get a list of files,
sorts it, and diffs that against the list of files from stdin (which
gets sorted first).
I suppose the sort of stdin could theoretically be done in parallel with
the find run, but I'm not sure I see any other simple optimizations that
could be done. I guess if you have hundreds of thousands of files then
those sort calls could take some more significant amount of time, but my
understanding is that sort already parallelizes automatically (up to 8
CPUs). What is actually taking up the time here?
- J<