public inbox for linux-nfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: Sagi Grimberg <sagi@grimberg.me>
Cc: Christoph Hellwig <hch@lst.de>, Chuck Lever <cel@kernel.org>,
	Trond Myklebust <trondmy@kernel.org>,
	Anna Schumaker <anna@kernel.org>,
	linux-nfs@vger.kernel.org
Subject: Re: add a LRU for delegations
Date: Thu, 8 Jan 2026 14:46:35 +0100	[thread overview]
Message-ID: <20260108134635.GA8624@lst.de> (raw)
In-Reply-To: <ea31c230-1ace-4721-872e-2313b497214f@grimberg.me>

On Wed, Jan 07, 2026 at 09:22:49PM +0200, Sagi Grimberg wrote:
> I think that there is merit for tracking file usage frequency and 
> accounting it for deleg return
> policy, and I don't necessarily think that it would not be worth the 
> overhead (compared to sending rpcs to the server)...
> but I agree that its a much heavier lift (it can always be done 
> incrementally to this patchset).

There's a reason the rest of the kernel uses fairly simple LRU (or
rather modified LRU / clock like) eviction policies like this, in
that the sampling is quite overboard.   I'm not going to keep you
from looking into this, but I don't think it's worthwhile.

> And, in genenral, I think that the server is in a much better position to 
> solicit preemptive delegation recalls as opposed
> to the client voluntarily return delegations when crossing a somewhat 
> arbitrary delegation_watermark.

Yes.  For that we'd need working RECALL_ANY support, though.  And
the way that works in the spec where the value is a number to keep
instead of a number to returns makes it pretty weird unfortunately.


  reply	other threads:[~2026-01-08 13:46 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-07  7:26 add a LRU for delegations Christoph Hellwig
2026-01-07  7:26 ` [PATCH 01/24] NFS: remove __nfs_client_for_each_server Christoph Hellwig
2026-01-07  7:26 ` [PATCH 02/24] NFS: remove nfs_client_mark_return_unused_delegation_types Christoph Hellwig
2026-01-07  7:26 ` [PATCH 03/24] NFS: remove nfs_client_mark_return_all_delegations Christoph Hellwig
2026-01-07  7:26 ` [PATCH 04/24] NFS: remove the NULL inode check in nfs4_inode_return_delegation_on_close Christoph Hellwig
2026-01-07  7:26 ` [PATCH 05/24] NFS: remove nfs_inode_detach_delegation Christoph Hellwig
2026-01-07  7:26 ` [PATCH 06/24] NFS: remove nfs_start_delegation_return Christoph Hellwig
2026-01-07  7:26 ` [PATCH 07/24] NFS: assert rcu_read_lock is held in nfs_start_delegation_return_locked Christoph Hellwig
2026-01-07  7:26 ` [PATCH 08/24] NFS: drop the _locked postfix from nfs_start_delegation_return Christoph Hellwig
2026-01-07  7:27 ` [PATCH 09/24] NFS: remove NFS_DELEGATION_INODE_FREEING Christoph Hellwig
2026-01-07  7:27 ` [PATCH 10/24] NFS: open code nfs_delegation_need_return Christoph Hellwig
2026-01-07  7:27 ` [PATCH 11/24] NFS: remove nfs_free_delegation Christoph Hellwig
2026-01-07  7:27 ` [PATCH 12/24] NFS: rewrite nfs_delegations_present in terms of nr_active_delegations Christoph Hellwig
2026-01-07  7:27 ` [PATCH 13/24] NFS: move delegation lookup into can_open_delegated Christoph Hellwig
2026-01-07  7:27 ` [PATCH 14/24] NFS: return bool from nfs_detach_delegation{,_locked} Christoph Hellwig
2026-01-07  7:27 ` [PATCH 15/24] NFS: move the deleg_cur check out of nfs_detach_delegation_locked Christoph Hellwig
2026-01-07  7:27 ` [PATCH 16/24] NFS: simplify the detached delegation check in update_open_stateid Christoph Hellwig
2026-01-07  7:27 ` [PATCH 17/24] NFS: take a delegation reference in nfs4_get_valid_delegation Christoph Hellwig
2026-01-07  7:27 ` [PATCH 18/24] NFS: don't consume a delegation reference in nfs_end_delegation_return Christoph Hellwig
2026-01-07  7:27 ` [PATCH 19/24] NFS: use refcount_inc_not_zero nfs_start_delegation_return Christoph Hellwig
2026-01-07  7:27 ` [PATCH 20/24] NFS: use a local RCU critical section in nfs_start_delegation_return Christoph Hellwig
2026-01-07  7:27 ` [PATCH 21/24] NFS: reformat nfs_mark_delegation_revoked Christoph Hellwig
2026-01-07  7:27 ` [PATCH 22/24] NFS: add a separate delegation return list Christoph Hellwig
2026-01-25 13:30   ` Chris Mason
2026-01-26  6:06     ` Christoph Hellwig
2026-01-07  7:27 ` [PATCH 23/24] NFS: return delegations from the end of a LRU when over the watermark Christoph Hellwig
2026-01-07  7:27 ` [PATCH 24/24] NFS: make nfs_mark_return_unreferenced_delegations less aggressive Christoph Hellwig
2026-01-07 15:18 ` add a LRU for delegations Chuck Lever
2026-01-07 16:22   ` Christoph Hellwig
2026-01-07 19:22     ` Sagi Grimberg
2026-01-08 13:46       ` Christoph Hellwig [this message]
2026-01-09  0:13         ` Sagi Grimberg
2026-01-08 14:34       ` Chuck Lever
2026-01-08 15:35         ` Christoph Hellwig
2026-01-09  0:30         ` Sagi Grimberg
2026-01-09  1:07           ` Trond Myklebust
2026-01-11  7:46             ` Sagi Grimberg
2026-01-15 16:24 ` Christoph Hellwig
2026-01-21 20:42   ` Anna Schumaker
  -- strict thread matches above, loose matches on Subject: below --
2025-12-18  5:56 Christoph Hellwig
2025-12-18  5:50 Christoph Hellwig

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=20260108134635.GA8624@lst.de \
    --to=hch@lst.de \
    --cc=anna@kernel.org \
    --cc=cel@kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=sagi@grimberg.me \
    --cc=trondmy@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