Date: Sat, 5 Mar 2005 14:54:23 -0500 From: David Schultz <das@FreeBSD.ORG> To: Peter Jeremy <PeterJeremy@optushome.com.au> Cc: cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/sys/kern kern_sig.c Message-ID: <20050305195422.GA42679@VARK.MIT.EDU> In-Reply-To: <20050305015636.GC4394@cirb503493.alcatel.com.au> References: <422710DD.1070203@freebsd.org> <422719E0.10703@samsco.org> <42279B6D.1000005@freebsd.org> <20050304182629.GA39457@VARK.MIT.EDU> <4228AB27.3020204@samsco.org> <4228B504.7000302@elischer.org> <20050304212329.GA40395@VARK.MIT.EDU> <20050304231620.GA4394@cirb503493.alcatel.com.au> <20050304234704.GA41005@VARK.MIT.EDU> <20050305015636.GC4394@cirb503493.alcatel.com.au>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Mar 05, 2005, Peter Jeremy wrote: > On Fri, 2005-Mar-04 18:47:04 -0500, David Schultz wrote: > >vm.swap_enabled *does* control kstack swapping only. > > Not according to it's documentation: "Enable entire process swapout". > And based on a quick check of the code I can't see anything in the > code path that restricts vm.swap_enabled to kstacks only. ``Entire process'' used to mean the U-area and kstack, I think, but we don't have a U-area anymore, so vm.swap_enabled basically just controls the kstack swapping. The only extra thing the swapper does that the sysctl turns off is deactivate some pages in processes that have already had their kstack swapped out or have exceeded their rlimits. But it just unmaps those pages and makes them more likely to be paged out; it doesn't actually push them out of core. > That said, the option "NO_SWAPPING" does state that it controls > swapping of stack pages only but appears to implement a superset of > vm.swap_enabled. No, I believe NO_SWAPPING is just the compile-time version of vm.swap_enabled.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050305195422.GA42679>