On Tue, Jun 17, 2008 at 11:22:40AM -0400, John Baldwin wrote: > On Tuesday 17 June 2008 11:15:20 am Maxim Sobolev wrote: > > Ed Schouten wrote: > > > * David Schultz <[EMAIL PROTECTED]> wrote: > > >> I have no objections to this, but doesn't it defeat the whole > > >> purpose to implement posix_spawn() as a library function that just > > >> calls fork/exec? > > > > > > When (if?) applications start to use posix_spawn() we may decide to move > > > it into the kernel at any time. It should be okay for now. > > > > Are there any benefits of doing it in the kernel vs. doing it via fork+exec? > > Speed. You don't have to go mark all your pages as COW or some such only to > turn around and throw the new mappings away and undo that.
The committed implementation uses vfork(). No COW is needed. On the other hand, after single-threading of the parent process have been removed from fork1(9), vfork(2) could have interesting consequences when done from the multithreaded process.
pgpyr6aOC13n2.pgp
Description: PGP signature
