From: Brian Foster <bfoster@redhat.com>
To: "Darrick J. Wong" <djwong@kernel.org>
Cc: linux-xfs@vger.kernel.org
Subject: Re: [PATCH v3 1/2] xfs: set a mount flag when perag reservation is active
Date: Tue, 20 Apr 2021 12:22:48 -0400 [thread overview]
Message-ID: <YH7/2P4SHHAJjaar@bfoster> (raw)
In-Reply-To: <20210414004950.GW3957620@magnolia>
On Tue, Apr 13, 2021 at 05:49:50PM -0700, Darrick J. Wong wrote:
> On Mon, Apr 12, 2021 at 09:30:58AM -0400, Brian Foster wrote:
> > perag reservation is enabled at mount time on a per AG basis. The
> > upcoming in-core allocation btree accounting mechanism needs to know
> > when reservation is enabled and that all perag AGF contexts are
> > initialized. As a preparation step, set a flag in the mount
> > structure and unconditionally initialize the pagf on all mounts
> > where at least one reservation is active.
> >
> > Signed-off-by: Brian Foster <bfoster@redhat.com>
> > ---
> > fs/xfs/libxfs/xfs_ag_resv.c | 24 ++++++++++++++----------
> > fs/xfs/xfs_mount.h | 1 +
> > 2 files changed, 15 insertions(+), 10 deletions(-)
> >
> > diff --git a/fs/xfs/libxfs/xfs_ag_resv.c b/fs/xfs/libxfs/xfs_ag_resv.c
> > index 6c5f8d10589c..9b2fc4abad2c 100644
> > --- a/fs/xfs/libxfs/xfs_ag_resv.c
> > +++ b/fs/xfs/libxfs/xfs_ag_resv.c
> > @@ -254,6 +254,7 @@ xfs_ag_resv_init(
> > xfs_extlen_t ask;
> > xfs_extlen_t used;
> > int error = 0;
> > + bool has_resv = false;
> >
> > /* Create the metadata reservation. */
> > if (pag->pag_meta_resv.ar_asked == 0) {
> > @@ -291,6 +292,8 @@ xfs_ag_resv_init(
> > if (error)
> > goto out;
> > }
> > + if (ask)
> > + has_resv = true;
> > }
> >
> > /* Create the RMAPBT metadata reservation */
> > @@ -304,18 +307,19 @@ xfs_ag_resv_init(
> > error = __xfs_ag_resv_init(pag, XFS_AG_RESV_RMAPBT, ask, used);
> > if (error)
> > goto out;
> > + if (ask)
> > + has_resv = true;
> > }
> >
> > -#ifdef DEBUG
> > - /* need to read in the AGF for the ASSERT below to work */
> > - error = xfs_alloc_pagf_init(pag->pag_mount, tp, pag->pag_agno, 0);
> > - if (error)
> > - return error;
> > -
> > - ASSERT(xfs_perag_resv(pag, XFS_AG_RESV_METADATA)->ar_reserved +
> > - xfs_perag_resv(pag, XFS_AG_RESV_RMAPBT)->ar_reserved <=
> > - pag->pagf_freeblks + pag->pagf_flcount);
> > -#endif
> > + if (has_resv) {
> > + mp->m_has_agresv = true;
>
> If the metadata reservation succeeds but the rmapbt reservation fails
> with ENOSPC, won't we fail to set m_has_agresv true here? We don't fail
> the entire mount if ENOSPC happens, which means that there's a slight
> chance of doing the wrong thing here if all the AGs are (somehow) like
> that.
>
Yes it looks like we would skip setting the mount flag (and initializing
the pagf). I suppose we should probably just lift the out label up
before this whole branch.
Brian
> --D
>
> > + error = xfs_alloc_pagf_init(mp, tp, pag->pag_agno, 0);
> > + if (error)
> > + return error;
> > + ASSERT(xfs_perag_resv(pag, XFS_AG_RESV_METADATA)->ar_reserved +
> > + xfs_perag_resv(pag, XFS_AG_RESV_RMAPBT)->ar_reserved <=
> > + pag->pagf_freeblks + pag->pagf_flcount);
> > + }
> > out:
> > return error;
> > }
> > diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
> > index 81829d19596e..8847ffd29777 100644
> > --- a/fs/xfs/xfs_mount.h
> > +++ b/fs/xfs/xfs_mount.h
> > @@ -139,6 +139,7 @@ typedef struct xfs_mount {
> > bool m_fail_unmount;
> > bool m_finobt_nores; /* no per-AG finobt resv. */
> > bool m_update_sb; /* sb needs update in mount */
> > + bool m_has_agresv; /* perag reservations active */
> >
> > /*
> > * Bitsets of per-fs metadata that have been checked and/or are sick.
> > --
> > 2.26.3
> >
>
next prev parent reply other threads:[~2021-04-20 16:22 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-04-12 13:30 [PATCH v3 REPOST 0/2] xfs: set aside allocation btree blocks from block reservation Brian Foster
2021-04-12 13:30 ` [PATCH v3 1/2] xfs: set a mount flag when perag reservation is active Brian Foster
2021-04-14 0:49 ` Darrick J. Wong
2021-04-20 16:22 ` Brian Foster [this message]
2021-04-20 16:23 ` Brian Foster
2021-04-12 13:30 ` [PATCH v3 2/2] xfs: set aside allocation btree blocks from block reservation Brian Foster
2021-04-12 22:21 ` kernel test robot
2021-04-14 0:41 ` Darrick J. Wong
2021-04-19 0:59 ` [kbuild-all] " Chen, Rong A
2021-04-19 16:10 ` Darrick J. Wong
2021-04-14 1:00 ` Darrick J. Wong
2021-04-20 16:24 ` Brian Foster
2021-04-21 23:40 ` Darrick J. Wong
2021-04-22 12:41 ` Brian Foster
-- strict thread matches above, loose matches on Subject: below --
2021-03-18 16:17 [PATCH v3 0/2] " Brian Foster
2021-03-18 16:17 ` [PATCH v3 1/2] xfs: set a mount flag when perag reservation is active Brian Foster
2021-03-18 20:55 ` Dave Chinner
2021-03-18 22:19 ` Darrick J. Wong
2021-03-19 1:05 ` Dave Chinner
2021-03-19 1:34 ` Darrick J. Wong
2021-03-19 14:54 ` Brian Foster
2021-03-19 1:43 ` Dave Chinner
2021-03-19 1:48 ` Darrick J. Wong
2021-03-19 2:08 ` Dave Chinner
2021-03-19 14:54 ` Brian Foster
2021-03-23 22:40 ` Dave Chinner
2021-03-24 14:24 ` Brian Foster
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=YH7/2P4SHHAJjaar@bfoster \
--to=bfoster@redhat.com \
--cc=djwong@kernel.org \
--cc=linux-xfs@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).