From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:50262) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1YBQd9-0003v6-2T for qemu-devel@nongnu.org; Wed, 14 Jan 2015 11:17:16 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1YBQd8-0001by-55 for qemu-devel@nongnu.org; Wed, 14 Jan 2015 11:17:15 -0500 Received: from mail-wi0-x232.google.com ([2a00:1450:400c:c05::232]:49115) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1YBQd7-0001bk-Qz for qemu-devel@nongnu.org; Wed, 14 Jan 2015 11:17:14 -0500 Received: by mail-wi0-f178.google.com with SMTP id z2so6509109wiv.5 for ; Wed, 14 Jan 2015 08:17:13 -0800 (PST) Date: Wed, 14 Jan 2015 16:17:10 +0000 From: Stefan Hajnoczi Message-ID: <20150114161710.GB1634@stefanha-thinkpad.redhat.com> References: <1420566495-13284-1-git-send-email-peter@lekensteyn.nl> <1420566495-13284-2-git-send-email-peter@lekensteyn.nl> <20150107131934.GB22440@stefanha-thinkpad.redhat.com> <33825683.QeFjLdu2v7@al> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="DKU6Jbt7q3WqK7+M" Content-Disposition: inline In-Reply-To: <33825683.QeFjLdu2v7@al> Subject: Re: [Qemu-devel] [PATCH v2 01/12] block/dmg: properly detect the UDIF trailer List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Wu Cc: Kevin Wolf , John Snow , qemu-devel@nongnu.org, Stefan Hajnoczi --DKU6Jbt7q3WqK7+M Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Jan 07, 2015 at 03:19:13PM +0100, Peter Wu wrote: > On Wednesday 07 January 2015 13:19:34 Stefan Hajnoczi wrote: > > On Tue, Jan 06, 2015 at 06:48:04PM +0100, Peter Wu wrote: > > > DMG files have a variable length with a UDIF trailer at the end of a > > > file. This UDIF trailer is essential as it describes the contents of > > > the image. At the moment however, the start of this trailer is almost > > > always incorrect as bdrv_getlength() returns a multiple of the block > > > size (rounded up). This results in a failure to recognize DMG files, > > > resulting in Invalid argument (EINVAL) errors. > > >=20 > > > As there is no API to retrieve the real file size, look for the magic > > > header in the last two sectors to find the start of this 512-byte UDIF > > > trailer (the "koly" block). > > >=20 > > > The resource fork offset ("info_begin") has its offset adjusted as the > > > initial value of offset does not mean "end of file" anymore, but "beg= in > > > of UDIF trailer". > > >=20 > > > Signed-off-by: Peter Wu > > > Reviewed-by: John Snow > > > --- > > > v2: added R-b, set errp as suggested by Stefan Hajnoczi > > > --- > > > block/dmg.c | 49 +++++++++++++++++++++++++++++++++++++++++++++---- > > > 1 file changed, 45 insertions(+), 4 deletions(-) > > >=20 > > > diff --git a/block/dmg.c b/block/dmg.c > > > index e455886..9183459 100644 > > > --- a/block/dmg.c > > > +++ b/block/dmg.c > > > @@ -131,6 +131,48 @@ static void update_max_chunk_size(BDRVDMGState *= s, uint32_t chunk, > > > } > > > } > > > =20 > > > +static int64_t dmg_find_koly_offset(BlockDriverState *file_bs, Error= **errp) > > > +{ > > > + int64_t length; > > > + int64_t offset =3D 0; > > > + uint8_t buffer[515]; > > > + int i, ret; > > > + > > > + /* bdrv_getlength returns a multiple of block size (512), rounde= d up. Since > > > + * dmg images can have odd sizes, try to look for the "koly" mag= ic which > > > + * marks the begin of the UDIF trailer (512 bytes). This magic c= an be found > > > + * in the last 511 bytes of the second-last sector or the first = 4 bytes of > > > + * the last sector (search space: 515 bytes) */ > > > + length =3D bdrv_getlength(file_bs); > > > + if (length < 0) { > > > + error_setg_errno(errp, -length, > > > + "Failed to get file size while reading UDIF trailer"); > > > + return length; > > > + } else if (length < 512) { > > > + error_set(errp, ERROR_CLASS_GENERIC_ERROR, > > > + "dmg file must be at least 512 bytes long"); > > > + return -EINVAL; > >=20 > > Not worth respinning but error_setg() is a shortcut for error_set(errp, > > ERROR_CLASS_GENERIC_ERROR, fmt, ...). > >=20 > > Reviewed-by: Stefan Hajnoczi >=20 > Good to know, I got a compile error when ERROR_CLASS_GENERIC_ERROR was > omitted, didn't think of using error_setg instead. >=20 > When merging, please replace the above ERROR_CLASS_GENERIC_ERROR by: >=20 > error_setg(errp, "dmg file must be at least 512 bytes long"); >=20 > (and likewise for the last error_set in the function) Done! Stefan --DKU6Jbt7q3WqK7+M Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJUtpaGAAoJEJykq7OBq3PIgfsH/idO1KFBaICBrkVqmCqhNTbW nECZM0zUzRHRJMRhCnG59UiMvyjvZY24zj38EKF9LFGniQLOvcm/4jtxQ2TAyb3G YTaIe+iDFx7VDhHeOuViN92xYLp5PqPLAwmj2ipDAAzMYe1QthdaYZVzxI5nvgGM alc1XiiQO1koxHUlvPvRb5d098zwZF5TEWA6qmm/vJj8+lfnYUElxGF4lQ4IQ0TI 1tKmUrdXLU0pZprnDFpXc7wf6Siic+FCm/4q/O80GiTT8ceXFX4LW2bH7xcJR5UO 5+PKuRtCCWC2DOZedbrAaih4MNeh/JSiIJtn2lmtMuHdxs9hH3gRM8h5OHrbhtU= =IV6Y -----END PGP SIGNATURE----- --DKU6Jbt7q3WqK7+M--