svn commit: r217830 - head/share/man/man9
jhb at freebsd.org
Wed Jan 26 22:05:41 UTC 2011
On Wednesday, January 26, 2011 4:14:15 pm mdf at freebsd.org wrote:
> On Wed, Jan 26, 2011 at 1:10 PM, Robert N. M. Watson
> <rwatson at freebsd.org> wrote:
> > On 26 Jan 2011, at 18:29, mdf at FreeBSD.org wrote:
> >>> I suppose an important question is now often we see this actually
> >> I don't believe we've ever seen a memory failure relating to sysctls
> >> at Isilon and we've been using the equivalent of this code for a few
> >> years. Our machines aren't low memory but they are under memory
> >> pressure sometimes.
> > The kinds of cases I worry about are things like the tcp connection
monitoring sysctls. Most systems have a dozen, hundred, or a thousand
connections. Some have half a million or a million. If we switched to
requiring wiring every page needed to store that list, it would do terrible
things to the system. So really what I have in mind is: either we handle cases
like that well, or we put in a clear warning and have obvious failure modes to
catch the cases where it didn't work out. In practice, I think we would not
want to switch the tcpcb/inpcb sysctl for this reason, but as people say "ah,
this is convenient" we need to make sure it's handled well, and easy to debug
problems when they do arise.
> But I think that problem exists today using sysctl for output, since
> it's non-iterative. In fact, it's often worse today, because in
> addition to the user-space buffer that needs to be large enough to
> hold the output, the kernel needs to malloc(9) a buffer to hold it
> before doing the one SYSCTL_OUT at the end that most routines I've
> seen use.
Not always. I think in the case of the inpcb's what happens is that we hold
enough references on objects that we can drop the locks while calling
SYSCTL_OUT() without requiring us to 1) allocate a full duplicate of the
buffer in KVM, or 2) wire the entire userland buffer.
More information about the svn-src-head