Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757486AbbDVSyZ (ORCPT ); Wed, 22 Apr 2015 14:54:25 -0400 Received: from mx1.redhat.com ([209.132.183.28]:55884 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751888AbbDVSyU (ORCPT ); Wed, 22 Apr 2015 14:54:20 -0400 Message-ID: <1429728791.121496.10.camel@redhat.com> Subject: Re: ioremap_uc() followed by set_memory_wc() - burrying MTRR From: Doug Ledford To: Jason Gunthorpe Cc: "Luis R. Rodriguez" , Andy Lutomirski , mike.marciniszyn@intel.com, infinipath@intel.com, linux-rdma@vger.kernel.org, awalls@md.metrocast.net, Toshi Kani , "H. Peter Anvin" , Ingo Molnar , "linux-kernel@vger.kernel.org" , Hal Rosenstock , Sean Hefty , Suresh Siddha , Rickard Strandqvist , Roland Dreier , Juergen Gross , Mauro Carvalho Chehab , Borislav Petkov , Mel Gorman , Vlastimil Babka , Davidlohr Bueso , Dave Hansen , Jean-Christophe Plagniol-Villard , Thomas Gleixner , Ville Syrj?l? , Linux Fbdev development list , linux-media@vger.kernel.org, X86 ML , mcgrof@do-not-panic.com Date: Wed, 22 Apr 2015 14:53:11 -0400 In-Reply-To: <20150422161755.GA19500@obsidianresearch.com> References: <20150421224601.GY5622@wotan.suse.de> <20150421225732.GA17356@obsidianresearch.com> <20150421233907.GA5622@wotan.suse.de> <20150422053939.GA29609@obsidianresearch.com> <20150422152328.GB5622@wotan.suse.de> <20150422161755.GA19500@obsidianresearch.com> Organization: Red Hat, Inc. Content-Type: multipart/signed; micalg="pgp-sha1"; protocol="application/pgp-signature"; boundary="=-c/d/LKEizloNkCfqZIim" Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 4232 Lines: 106 --=-c/d/LKEizloNkCfqZIim Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Wed, 2015-04-22 at 10:17 -0600, Jason Gunthorpe wrote: > On Wed, Apr 22, 2015 at 05:23:28PM +0200, Luis R. Rodriguez wrote: > > On Tue, Apr 21, 2015 at 11:39:39PM -0600, Jason Gunthorpe wrote: > > > On Wed, Apr 22, 2015 at 01:39:07AM +0200, Luis R. Rodriguez wrote: > > > > > Mike, do you think the time is right to just remove the iPath dri= ver? > > > >=20 > > > > With PAT now being default the driver effectively won't work > > > > with write-combining on modern kernels. Even if systems are old > > > > they likely had PAT support, when upgrading kernels PAT will work > > > > but write-combing won't on ipath. > > >=20 > > > Sorry, do you mean the driver already doesn't get WC? Or do you mean > > > after some more pending patches are applied? > >=20 > > No, you have to consider the system used and the effects of calls used > > on the driver in light of this table: >=20 > So, just to be clear: >=20 > At some point Linux started setting the PAT bits during > ioremap_nocache, which overrides MTRR, and at that point the driver > became broken on all PAT capable systems? >=20 > Not only that, but we've only just noticed it now, and no user ever > complained? >=20 > So that means either no users exist, or all users are on non-PAT > systems? >=20 > This driver only works on x86-64 systems. Are there any x86-64 systems > that are not PAT capable? IIRC even the first Opteron had PAT, but my > memory is fuzzy from back then :| >=20 > > Another option in order to enable this type of checks at run time > > and still be able to build the driver on standard distributions and > > just prevent if from loading on PAT systems is to have some code in > > place which would prevent the driver from loading if PAT was > > enabled, this would enable folks to disable PAT via a kernel command > > line option, and if that was used then the driver probe would > > complete. >=20 > This seems like a reasonble option to me. At the very least we might > learn if anyone is still using these cards. >=20 > I'd also love to remove the driver if it turns out there are actually > no users. qib substantially replaces it except for a few very old > cards. To be precise, the split is that ipath powers the old HTX bus cards that only work in AMD systems, qib is all PCI-e cards. I still have a few HTX cards, but I no longer have any systems with HTX slots, so we haven't even used this driver in testing for 3 or 4 years now. And these are all old SDR cards, where the performance numbers were 800MB/s with WC enabled, 50MB/s without it. > Mike? >=20 > Jason > -- > To unsubscribe from this list: send the line "unsubscribe linux-rdma" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html --=20 Doug Ledford GPG KeyID: 0E572FDD --=-c/d/LKEizloNkCfqZIim Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAABAgAGBQJVN+4XAAoJELgmozMOVy/d974QAIbAblVZOWJ3bSplFMqY7SaJ 441MB14jdqVrdvJc8tLRLykYjy751m8lEiG63oN78va287gaAtfLJzuvQqFeUzVs +YPkD88jgVhiCM9aFV3aikgKb5kGM2TdTyHDYuQhS1v/SqBtf9xG1MU3RaU9/IjH s548v35Hrm5QQ2wK2bcl03Y70ju6JWez1hpH53sFh2k73g/Pt3p+ohRtz6Qeqcrr m1xkOxQuT0KXIkEAWvC9npm6SVZmavwDS3/a0gg6kuvzVtBPP8c9XaRYhYTkkgAG ZuHCkjCVVTqviMFw03wZbYdOrpXAvVTRgZmzJRoZVUGefOVnZaa8VmcQS/nKNQQH FKD+XcXzRUiFQPH6SUS34Xlo/bDn1FOjprkSXOnmjLbdcWvPMRhbeDyEeVRjnInq NQhATSNN0ByssHJDhgERhS6+nQqN90LCiuFKzktIxySmpv+vEFsT8vIgYsP7nRWn Xg7ZKeiGziKPmKY2yi/iSqbLd9MAhJaf8wFDG4CkfmRlrRHKxKFS/S0IvLIcpUZ/ QYWYH4owCUq/YTi098zIFRgsweKdfjVizZuFqNGd7pGpDR3U56Pd/UAzG0BlcnhU DYbq6EXmLOeW1JWt2QiKIn/mKIJG25yMiB3S4iQL/f2bNqMQiKBTFM8gWbop9dxD V07prexdr+wIaHgKFQ9k =lFcy -----END PGP SIGNATURE----- --=-c/d/LKEizloNkCfqZIim-- -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/