From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from userp2130.oracle.com ([156.151.31.86]:55260 "EHLO userp2130.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1729572AbfEVQmi (ORCPT ); Wed, 22 May 2019 12:42:38 -0400 Date: Wed, 22 May 2019 09:42:30 -0700 From: "Darrick J. Wong" Subject: Re: [PATCH 11/12] mkfs: validate start and end of aligned logs Message-ID: <20190522164230.GE5141@magnolia> References: <155839420081.68606.4573219764134939943.stgit@magnolia> <155839427473.68606.3900005341580158051.stgit@magnolia> <03cd588a-f4b0-0c79-4862-31ae0e5cd3dd@sandeen.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <03cd588a-f4b0-0c79-4862-31ae0e5cd3dd@sandeen.net> Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Eric Sandeen Cc: linux-xfs@vger.kernel.org On Tue, May 21, 2019 at 02:24:40PM -0500, Eric Sandeen wrote: > On 5/20/19 6:17 PM, Darrick J. Wong wrote: > > From: Darrick J. Wong > > > > Validate that the start and end of the log stay within a single AG if > > we adjust either end to align to stripe units. > > > > Signed-off-by: Darrick J. Wong > > --- > > mkfs/xfs_mkfs.c | 15 ++++++++++++++- > > 1 file changed, 14 insertions(+), 1 deletion(-) > > > > > > diff --git a/mkfs/xfs_mkfs.c b/mkfs/xfs_mkfs.c > > index 5b66074d..8f84536e 100644 > > --- a/mkfs/xfs_mkfs.c > > +++ b/mkfs/xfs_mkfs.c > > @@ -3044,15 +3044,28 @@ align_internal_log( > > struct xfs_mount *mp, > > int sunit) > > { > > + uint64_t logend; > > + > > /* round up log start if necessary */ > > if ((cfg->logstart % sunit) != 0) > > cfg->logstart = ((cfg->logstart + (sunit - 1)) / sunit) * sunit; > > > > + /* if our log start rounds into the next AG we're done */ > > /* If our log start overlaps the next AG's metadata, fail */ Ok. > > + if (!xfs_verify_fsbno(mp, cfg->logstart)) { > > + fprintf(stderr, > > +_("Due to stripe alignment, the internal log start (%lld) cannot be aligned\n" > > + "within an allocation group.\n"), > > + (long long) cfg->logstart); > > Hm, should it suggest what should be modified to try again ...? But what should be modified, exactly? -d su=0,sw=0? > > + usage(); > > + } > > + > > /* round up/down the log size now */ > > align_log_size(cfg, sunit); > > > > /* check the aligned log still fits in an AG. */ > > - if (cfg->logblocks > cfg->agsize - XFS_FSB_TO_AGBNO(mp, cfg->logstart)) { > > + logend = cfg->logstart + cfg->logblocks - 1; > > + if (XFS_FSB_TO_AGNO(mp, cfg->logstart) != XFS_FSB_TO_AGNO(mp, logend) || > > + !xfs_verify_fsbno(mp, logend)) { > > this xfs_verify_fsbno is probably redundant but can't hurt? Will respin patch. --D > -Eric > > > fprintf(stderr, > > _("Due to stripe alignment, the internal log size (%lld) is too large.\n" > > "Must fit within an allocation group.\n"), > >