Date: Wed, 31 May 2006 11:41:31 -0700 From: Brooks Davis <brooks@one-eyed-alien.net> To: emulation@freebsd.org, rdivacky@freebsd.org Subject: linuxolator: sched_[sg]etaffinity Message-ID: <20060531184131.GD20044@odin.ac.hmc.edu>
next in thread | raw e-mail | index | archive | help
--yudcn1FV7Hsu/q59 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable This is mostly for the archives, but seeing the list of unimplemented system calls reminded me. If we do end up implementing the CPU affinity functions at some point, there are a number of landmines. In particular, there are at least three differnet APIs for them depending on the vendor. Presumably we'll want to implement what ever Red Hat does, but this is something to watch out for. The Open MPI project actually wrote a library for Linux so one binary can access all the APIs, it's rather evil. http://www.open-mpi.org/software/plpa/ -- Brooks --=20 Any statement of the form "X is the one, true Y" is FALSE. PGP fingerprint 655D 519C 26A7 82E7 2529 9BF0 5D8E 8BE9 F238 1AD4 --yudcn1FV7Hsu/q59 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQFEfeNbXY6L6fI4GtQRAuleAJ9Bw5c/PeC1IhvnWFCFktcq1MWt5gCbBrbS 2QOJANAWNB3HtAa30gYEjwg= =yR76 -----END PGP SIGNATURE----- --yudcn1FV7Hsu/q59--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060531184131.GD20044>