Date: Fri, 16 Jul 2010 17:16:54 +1000 From: Peter Jeremy <peterjeremy@acm.org> To: Alan Cox <alc@FreeBSD.org> Cc: svn-src-head@FreeBSD.org, svn-src-all@FreeBSD.org, src-committers@FreeBSD.org Subject: Re: svn commit: r210124 - head/sys/amd64/amd64 Message-ID: <20100716071654.GA26367@server.vk2pj.dyndns.org> In-Reply-To: <201007151625.o6FGPp0D038539@svn.freebsd.org> References: <201007151625.o6FGPp0D038539@svn.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--T4sUOijqQbZv57TR Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On 2010-Jul-15 16:25:51 +0000, Alan Cox <alc@FreeBSD.org> wrote: >Log: > Optimize pmap_remove()'s handling of PG_G mappings. Specifically, =2E.. > calls pmap_invalidate_page(). Altogether, these changes eliminate 53% > of the TLB shootdowns for a "buildworld" on a ZFS file system. On > FFS, the reduction is 3%. I have no feeling for the cost of a TLB shootdown. What is the likely impact on buildworld on a ZFS filesystem? Barely noticable or several percent faster? --=20 Peter Jeremy --T4sUOijqQbZv57TR Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.15 (FreeBSD) iEYEARECAAYFAkxAB2YACgkQ/opHv/APuIejKACfapzvWvOpsRTr3n0UyJnJXKg1 GxcAn31m+gXHB2nAtV0tpVa4sSt/bNl6 =3Q7E -----END PGP SIGNATURE----- --T4sUOijqQbZv57TR--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20100716071654.GA26367>