Remove unsed KQUEUE from usr.bin/make?

Kris Kennaway kkenn at xor.obsecurity.org
Thu Jul 7 14:57:52 PDT 2005


On 2005-07-07, Garance A Drosihn <drosih at xxxxxxx> wrote:
> At 11:01 AM -0700 7/7/05, Max Okumoto wrote:
>>
>>Yea, thats what I expect.  But I want to do the responsable thing. :-)
>>Most of the time make is just waiting for the children.  But its
>>more a question of which one will result in cleaner code.  Either
>>we go KQUEUE all the way, or we remove it.  Having it ifdef makes
>>the code harder to maintain.
>
> I'll also observe that 'make' seems to be one area where there is
> some cooperation/co-ordination between freebsd and dragonfly.  It
> would be nice to see that continue.  I don't care too much whether
> KQUEUE in make stays or goes, but I'd hate to see it stay in one
> OS and go in the other...

I use make on FreeBSD for running jobs with high levels of concurrency
(~100, or about ~300 if you count multiple concurrent make processes).
I've not benchmarked anything, but it would be a shame to remove
kqueue support because "it doesn't have a performance benefit" without
measuring it at the high end.

Kris






More information about the Users mailing list