cluster-devel.redhat.com archive mirror
 help / color / mirror / Atom feed
From: Andreas Gruenbacher <agruenba@redhat.com>
To: cluster-devel.redhat.com
Subject: [Cluster-devel] [PATCH v2 2/6] gfs2: Clean up gfs2_adjust_reservation
Date: Mon, 3 Dec 2018 15:58:26 +0100	[thread overview]
Message-ID: <CAHc6FU7PpF18kPJbitsKM7Y_OTXn+1heDbsoHdHC5a4iUYt4zg@mail.gmail.com> (raw)
In-Reply-To: <936623100.38540989.1543847558772.JavaMail.zimbra@redhat.com>

On Mon, 3 Dec 2018 at 15:32, Bob Peterson <rpeterso@redhat.com> wrote:
>
> ----- Original Message -----
> > Unconditionally call gfs2_adjust_reservation in gfs2_alloc_blocks.  Move
> > the code for updating rd_free and rd_free_clone from gfs2_alloc_blocks
> > into gfs2_adjust_reservation.
> >
> > Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
> > ---
> >  fs/gfs2/rgrp.c | 26 ++++++++++++++++----------
> >  1 file changed, 16 insertions(+), 10 deletions(-)
> >
> > diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
> > index bbcf8b8b2597f..733e21cd4cf25 100644
> > --- a/fs/gfs2/rgrp.c
> > +++ b/fs/gfs2/rgrp.c
> (snip)
> > -     if (gfs2_rs_active(&ip->i_res))
> > -             gfs2_adjust_reservation(ip, &rbm, *nblocks);
> > +     error = gfs2_adjust_reservation(ip, &rbm, *nblocks);
>
> Wait. At what point do we require a reservation for all block allocations?
> In today's code, a reservation is not required, so the check for
> "if (gfs2_rs_active(&ip->i_res))" is necessary because you can't adjust a
> reservation that's not active (in the tree)...being within the tree
> determines whether the reservation is active or not.
>
> Yes, I'm looking forward to the day when we require a reservation for all
> block allocations as per my original version (this needs to happen to allow
> rgrp sharing between processes.) Perhaps a future patch will add the
> requirement, but at this point, there's no such requirement, which means
> this patch will probably break during a git bisect, no?

gfs2_adjust_reservation already checks if there is an active
reservation, so the duplicate check here is unnecessary.

> We only get into this situation when all bitmaps are nearly full or fragmented
> and therefore no reservation will fit the minimum requirements, but we still
> need to allow small one-shot block allocations. So we need to do lots of
> full-bitmap testing. Or else we need to add this patch only after the
> patch that makes reservations mandatory.
>
> Regards,
> Bob Peterson

Andreas



  reply	other threads:[~2018-12-03 14:58 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-12-01 11:10 [Cluster-devel] [PATCH v2 0/6] gfs2: Prepare for resource group glock sharing Andreas Gruenbacher
2018-12-01 11:10 ` [Cluster-devel] [PATCH v2 1/6] gfs2: Only use struct gfs2_rbm for bitmap manipulations Andreas Gruenbacher
2018-12-01 11:10 ` [Cluster-devel] [PATCH v2 2/6] gfs2: Clean up gfs2_adjust_reservation Andreas Gruenbacher
2018-12-03 14:32   ` Bob Peterson
2018-12-03 14:58     ` Andreas Gruenbacher [this message]
2018-12-04 22:02   ` Andreas Gruenbacher
2018-12-01 11:10 ` [Cluster-devel] [PATCH v2 3/6] gfs2: Add per-reservation reserved block accounting Andreas Gruenbacher
2018-12-01 11:10 ` [Cluster-devel] [PATCH v2 4/6] gfs2: Add local resource group locking Andreas Gruenbacher
2018-12-01 11:10 ` [Cluster-devel] [PATCH v2 5/6] gfs2: Allow node-wide exclusive glock sharing Andreas Gruenbacher
2018-12-01 11:10 ` [Cluster-devel] [PATCH v2 6/6] gfs2: Introduce resource group sharing Andreas Gruenbacher

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=CAHc6FU7PpF18kPJbitsKM7Y_OTXn+1heDbsoHdHC5a4iUYt4zg@mail.gmail.com \
    --to=agruenba@redhat.com \
    /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).