From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Sun, 21 Oct 2018 14:37:45 +0200 From: Pavel Machek Subject: Re: 32-bit PTI with THP = userspace corruption Message-ID: <20181021123745.GA26042@amd> References: <20180830205527.dmemjwxfbwvkdzk2@suse.de> <20180831070722.wnulbbmillxkw7ke@suse.de> <20180911114927.gikd3uf3otxn2ekq@suse.de> <20180911121128.ikwptix6e4slvpt2@suse.de> <20180918140030.248afa21@alans-desktop> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="9amGYk9869ThD9tj" Content-Disposition: inline In-Reply-To: <20180918140030.248afa21@alans-desktop> Sender: linux-kernel-owner@vger.kernel.org To: Alan Cox Cc: Joerg Roedel , Meelis Roos , Thomas Gleixner , Linux Kernel list , linux-mm@kvack.org, Andrea Arcangeli , Linus Torvalds List-ID: --9amGYk9869ThD9tj Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue 2018-09-18 14:00:30, Alan Cox wrote: > On Tue, 11 Sep 2018 14:12:22 +0200 > Joerg Roedel wrote: >=20 > > On Tue, Sep 11, 2018 at 02:58:10PM +0300, Meelis Roos wrote: > > > The machines where I have PAE off are the ones that have less memory.= =20 > > > PAE is off just for performance reasons, not lack of PAE. PAE should = be=20 > > > present on all of my affected machines anyway and current distributio= ns=20 > > > seem to mostly assume 686 and PAE anyway for 32-bit systems. =20 > >=20 > > Right, most distributions don't even provide a non-PAE kernel for their > > users anymore. > >=20 > > How big is the performance impact of using PAE over legacy paging? >=20 > On what system. In the days of the original 36bit PAE Xeons it was around > 10% when we measured it at Red Hat, but that was long ago and as you go > newer it really ought to be vanishingly small. >=20 > There are pretty much no machines that don't support PAE and are still > even vaguely able to boot a modern Linux kernel. The oddity is the > Pentium-M but most distros shipped a hack to use PAE on the Pentium M > anyway as it seems to work fine. I do have some AMD Geode here, in form of subnotebook. Definitely newer then Pentium Ms, but no PAE... Pavel --=20 (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blo= g.html --9amGYk9869ThD9tj Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iEYEARECAAYFAlvMcxkACgkQMOfwapXb+vJ6pgCgu3iDvOnkmr4HLzuB2U2AzpSE uwgAoKVSZZZk13M9JhHUdFhHkzV6VtfE =ssuM -----END PGP SIGNATURE----- --9amGYk9869ThD9tj--