From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:32841) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XXZEV-00088w-Dq for qemu-devel@nongnu.org; Fri, 26 Sep 2014 13:23:07 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1XXZEQ-00007i-Rh for qemu-devel@nongnu.org; Fri, 26 Sep 2014 13:23:03 -0400 Received: from mx1.redhat.com ([209.132.183.28]:29982) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XXZEQ-00005b-B9 for qemu-devel@nongnu.org; Fri, 26 Sep 2014 13:22:58 -0400 Message-ID: <5425A0E8.2060405@redhat.com> Date: Fri, 26 Sep 2014 11:22:48 -0600 From: Eric Blake MIME-Version: 1.0 References: <1411622627-22110-1-git-send-email-tony@bakeyournoodle.com> <1411686852-26487-1-git-send-email-tony@bakeyournoodle.com> <1411686852-26487-2-git-send-email-tony@bakeyournoodle.com> In-Reply-To: <1411686852-26487-2-git-send-email-tony@bakeyournoodle.com> Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="Ct6b08FSvm2DEFV2wbCgvts834nqrU2p0" Subject: Re: [Qemu-devel] [PATCH v2 2/2] block/raw-posix: use seek_hole ahead of fiemap List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Tony Breeds , qemu-devel@nongnu.org Cc: Kevin Wolf , =?UTF-8?B?UMOhZHJhaWcgQnJhZHk=?= , Markus Armbruster , Stefan Hajnoczi , Max Reitz This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --Ct6b08FSvm2DEFV2wbCgvts834nqrU2p0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable On 09/25/2014 05:14 PM, Tony Breeds wrote: > try_fiemap() uses FIEMAP_FLAG_SYNC which has a significant performance > impact. >=20 > Prefer seek_hole() over fiemap() to avoid this impact where possible. > seek_hole is more widely used and, arguably, has potential to be > optimised in the kernel. It also has the potential to be less precise than fiemap on older kernels (any filesystem that supports fiemap but has not yet wired up seek_hole will get the kernel's default fallback of claiming no holes at all), but that problem will fade as more people use modern kernels. This matches what coreutils does, so I like it on that grounds. >=20 > Reported-By: Michael Steffens > Signed-off-by: Tony Breeds > Cc: Kevin Wolf > Cc: Markus Armbruster > Cc: Stefan Hajnoczi > Cc: Max Reitz > Cc: P=C3=A1draig Brady > Cc: Eric Blake > --- > Changes since v1: > - split in to 2 patches > - tried to make the commit messages better Thanks; it is indeed better. >=20 > block/raw-posix.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) Reviewed-by: Eric Blake --=20 Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org --Ct6b08FSvm2DEFV2wbCgvts834nqrU2p0 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Public key at http://people.redhat.com/eblake/eblake.gpg iQEcBAEBCAAGBQJUJaDoAAoJEKeha0olJ0Nq44kH/RbuEE7JR25SQKhOfmDj3AWW u3e8fiTcdQwTsfgj87/0F2RM723ruyNX7uEDdyrG5UqUh61t7OFznl95B/BLKeXM GW8Tb6MLt7AB5z8LGenWyDRXo2JI3DqBopwPhnBW6DL0bizmBsPp7xHHzXpqkFwL FDNt7ciV04RL67OzRKgCoSOjsOMrDjZBu3O03qxH63fIIyfIdsVovB6Jj9Cd445z wEN5USYO7/HXDcUa1akhARETtbfbAmBacByfIfu9yaSIDv0Y3ewaUj0eKnWo0qNp FZtVU/MTPTqa/rkB7AXpEZVzknkgUhAOU71rkRXntv+jWWII5QuaAKZ0ZzNMVDk= =3trG -----END PGP SIGNATURE----- --Ct6b08FSvm2DEFV2wbCgvts834nqrU2p0--