--mojUlQ0s9EVzWg2t
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Hello,

On Thu, Nov 28, 2002 at 09:52:15AM +0000, Richard Earnshaw wrote:
> > I've increased PMAP_STATIC_L1S to 128 and still get one or three per ev=
ening.
>=20
> That's quite a lot of active processes (though not necessarily=20
> unreasonable). How much RAM do you have?

64 MB.

> > Wouldn't the wiser strategy be:
> >=20
> > 1. try if we have a free one on the l1pt free list
> > 2. if this fails, try to allocate a new one
> > 3. if this still fails, go to the static list?
> >=20
> > This way, we only go to the static list if memory is fragmented or if we
> > run out of memory...
> >=20
> > Would this fail during bootstrap? As the machine in question is my prod=
uction
> > server, I don't want to waste time trying stupid things...
> >=20
> > The only alternative to make this pmap stable would be to use a VAX-like
> > 16kB VM pages with 4KB mmu pages strategy, which naturally ensures that=
 we
> > never run out of aligned 16kB blocks.
>=20
> It might make things a little more stable, though it would probably only=
=20
> delay the onset of resource starvation slightly; and it would mean that=
=20
> normally the static pool of L1S will be dead memory -- which argues for i=
t=20
> being a *much* lower number.

I agree here.

	-is

--mojUlQ0s9EVzWg2t
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: 2.6.i

iQEVAgUBPeYeQDCn4om+4LhpAQE9zggAlnx77Gy2NDtxHOwdkLHRrzqCz8ixBMry
JGm0UWSseBK5yleLcQUm8Ik8eUmpdQdu79uwGGZVt5i7FrdruQvWakVgx86NjkkC
tql3oU7oQWn2AIzStpyovfGub6pSf5jxesQDbKnE//+VMpQdpx8JoO8b2GoJl8/o
ljh4epBJOiaWdKMY2Ci2qqxKnzHNph3jiLpcs8fMt6REzk62q1gdl9uZtAKSWlaS
PL9eZF48hhGybyfzX13zlB8Wz0l2mVICZAqDBHBjWKQmgpM8h7Bc8LyNXrFI0n9F
Qe7JgbjCyZWJj83Kdfz9Sdn/7OEaPepscVcyOHkbMHjmploimvC9RQ==
=OcvW
-----END PGP SIGNATURE-----

--mojUlQ0s9EVzWg2t--