Date: Mon, 23 Apr 2007 22:14:34 -0700 From: John-Mark Gurney <gurney_j@resnet.uoregon.edu> To: David G Lawrence <dg@dglawrence.com> Cc: Tim Kientzle <kientzle@freebsd.org>, current@freebsd.org, "Jesper B. Rosenkilde" <jbr@humppa.dk> Subject: Re: Suggestions on Avoiding syscall Overhead Message-ID: <20070424051433.GT73385@funkthat.com> In-Reply-To: <20070424042102.GI38475@tnn.dglawrence.com> References: <f126fae00704221639l68095de1ye7ce9ba3d921bf20@mail.gmail.com> <20070423113400.GC28587@gw.humppa.dk> <462CD251.9060105@freebsd.org> <20070423161711.GV39474@elvis.mu.org> <462D821F.6030707@freebsd.org> <20070424042102.GI38475@tnn.dglawrence.com>
next in thread | previous in thread | raw e-mail | index | archive | help
David Greenman wrote this message on Mon, Apr 23, 2007 at 21:21 -0700:
> every process for process-specific data, however - the cost of allocating
> it, initializing it, etc, for every fork() would be a pessimization in
> most cases, I think.
You could always do something similar to a COW, where once it is
accessed, it is allocated and filled w/ the specific information
necessary... If it never gets used, it never gets allocated...
--
John-Mark Gurney Voice: +1 415 225 5579
"All that I will do, has been done, All that I have, has not."
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070424051433.GT73385>