From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from xc.sipsolutions.net ([83.246.72.84]:53909 "EHLO sipsolutions.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751169AbZDPLHO (ORCPT ); Thu, 16 Apr 2009 07:07:14 -0400 Subject: Re: [RFC 0/3] mac80211 powersave improvements From: Johannes Berg To: Kalle Valo Cc: linux-wireless@vger.kernel.org In-Reply-To: <87iql4981b.fsf@litku.valot.fi> (sfid-20090416_130208_558130_BEE7A112) References: <20090415171046.336006351@sipsolutions.net> <87iql4981b.fsf@litku.valot.fi> (sfid-20090416_130208_558130_BEE7A112) Content-Type: multipart/signed; micalg="pgp-sha1"; protocol="application/pgp-signature"; boundary="=-Ev8ADcIq5d2DVSekwXpD" Date: Thu, 16 Apr 2009 13:06:40 +0200 Message-Id: <1239880000.9737.20.camel@johannes.local> (sfid-20090416_130720_703548_82711DB5) Mime-Version: 1.0 Sender: linux-wireless-owner@vger.kernel.org List-ID: --=-Ev8ADcIq5d2DVSekwXpD Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On Thu, 2009-04-16 at 14:01 +0300, Kalle Valo wrote: > > 1) improve powersave to only allow it when a single > > managed interface is active > > > > 2) disable PS when the maximum networking latency > > applications are are willing to put up with is > > smaller than the beacon interval -- in that case > > we cannot allow the AP to buffer frames >=20 > No problem with these. I'll send these along with the software beacon filtering, I think. > > 3) enable powersave by default > > > > The last part might be controversial? >=20 > I think this a bad idea. For normal users we should not enable powersave > by default, at least not yet. I believe that there would be so much > problems that we would have to revert the change eventually. We can go with the configurable one -- and default y in wireless-testing but default n in mainline for a while. Get press coverage etc. and at some point flip the switch. > > Should we have a default setting for the dynamic PS > > timeout? >=20 > Yes. Having timeout zero makes no sense in normal laptop use. So, what's a good default? Fairly long I'd say, maybe 500ms? johannes --=-Ev8ADcIq5d2DVSekwXpD Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Comment: Johannes Berg (powerbook) iQIcBAABAgAGBQJJ5xE8AAoJEKVg1VMiehFYwrwP+wfIF7HBqTwVG2I6hcWTxn1u 9dlPCcK+iyRbWjNJMrHCO7FVVEOP3rB6mIVueaiQtxvPOXyxnzFYFWGHNUr80E9C 2DaFgY+Xwvurcnd6Orlc7PBCuqxwOqH+XURd2vbdBWJbYzwQucLip1eWZ16LPPz6 qxRQpfpNy/clg0bHIEMMYBQs8aqn3n5hsGfXuCTyuXnT98JNpiPLKIvNwr81d05R WE+dnnK2WZpTsTjScRdx3DkjjIm6y1K1d7Gxjtismnv1z7MrmkZOsTNnbny811f+ 1xu6D4tCxufh6/qjEXH/Ksni466PZd+dzFKRgKj11ZL25nW5Wz106OO/up2kBeXp L0CWuZKXOV1vlq8fBlvEIEMWiRQIDJN1QRezx5kTYtq4IRB4qV1sk1lNLc4YGSGC 6EmRsu7ohWJfRI5EqIALd+4HIYysX6uX9hdzjOUQgTJHGnRhTOBUTGZwv0Ec6Sx5 /JmwVD76j8oFFVqG4T/AojGRaL0v5HyhHXismd01rgX+l2Cr2ETKdsMgks1er9Je 9wv7zjKkQbl2i/xKeES+1R8UEep+/KN8nz9gjqFQCTHZnerIyRyiY5AAvdSNlVHm D8hzkFpjxuae4qBQnWE8SNhtrhPeU8AQyITLkSAZoWpC8sB37y9/xOSLFRSRWwUp YaXpCL/8M5UR4qeHI91t =ZEtc -----END PGP SIGNATURE----- --=-Ev8ADcIq5d2DVSekwXpD--