Andrew Gallatin wrote:
Randall Stewart writes:
> Andrew Gallatin wrote:
> > Randall Stewart writes:
> > >nmbclusters = 1024 + maxusers * 64;
> > > +nmbjumbop = 100 + (maxusers * 4);
> >
> > The limit on page-size jumbos seems far too small. Since the socket
> > buffe
Randall Stewart writes:
> Andrew Gallatin wrote:
> > Randall Stewart writes:
> > > nmbclusters = 1024 + maxusers * 64;
> > > +nmbjumbop = 100 + (maxusers * 4);
> >
> > The limit on page-size jumbos seems far too small. Since the socket
> > buffer code now uses page-siz
Andrew Gallatin wrote:
Randall Stewart writes:
> nmbclusters = 1024 + maxusers * 64;
> +nmbjumbop = 100 + (maxusers * 4);
The limit on page-size jumbos seems far too small. Since the socket
buffer code now uses page-sized jumbos, I'd expect to see its limit be
the same as nmbclust
Randall Stewart writes:
> nmbclusters = 1024 + maxusers * 64;
> +nmbjumbop = 100 + (maxusers * 4);
The limit on page-size jumbos seems far too small. Since the socket
buffer code now uses page-sized jumbos, I'd expect to see its limit be
the same as nmbclusters.
Drew
_
On Fri, Jan 19, 2007 at 09:17:07AM -0500, Randall Stewart wrote:
> All:
>
> George and I have been discussing making the attached change
> to limit kernel large cluster growth.
>
> Note I also have added a fix so that if the limit
> is 0 (unlimited), then you cannot change the limit
> (back downw
