From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-wm1-f47.google.com (mail-wm1-f47.google.com [209.85.128.47]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id C7B3643F4B3 for ; Tue, 7 Jul 2026 21:42:34 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.128.47 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1783460556; cv=none; b=C042HQwiMBYLFGOjIBcfYlq/I16BhfRJwZaeQtd7Jzo6hXyZhj368ZodBtq9PcyIo8zDFc6U6/JJZXtlAjDjtfXnyZOqQc+iu/o0kQcEDmr0nKQINdKwldgafzZRohkXO8dm60+sQA/RP3DbKILx+BXmOrgY6SjTFy1Y8qnsHUk= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1783460556; c=relaxed/simple; bh=zSnKGAXUYKrxSN8e3ZX9H2W/sQ/pbShhAYN+qxSOEeo=; h=From:To:Cc:Subject:Date:Message-ID:MIME-Version; b=Ag6hiBdwuyg/e1Lhefwl6Jy+h83ykISDPLIYebMfywIZrigXhmeoAkrG5IUMRVv4mbqgJOly6cH8qPL3pwR/uNJGHQt/+rkjsiVMlTKhF4QgawU2oeAdGtXQ3DyMS1lXbV4oSwLmm2cQ42LhMKsN2308/k8qhGvL6Snj6e1J610= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=ionos.com; spf=pass smtp.mailfrom=ionos.com; dkim=pass (2048-bit key) header.d=ionos.com header.i=@ionos.com header.b=Ra57FTmI; arc=none smtp.client-ip=209.85.128.47 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=ionos.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=ionos.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=ionos.com header.i=@ionos.com header.b="Ra57FTmI" Received: by mail-wm1-f47.google.com with SMTP id 5b1f17b1804b1-493e4ccccc2so3775125e9.2 for ; Tue, 07 Jul 2026 14:42:34 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ionos.com; s=google; t=1783460553; x=1784065353; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:message-id:date:subject:cc :to:from:from:to:cc:subject:date:message-id:reply-to; bh=fr34X1DQSGmN+02DaqdMQrFw+Hk4XAU6EmNPh+kwNxk=; b=Ra57FTmI8/mCYM7vVM7NidSEkmDQ3yF3jAqLfkbF/3Ji2a1dWV2BI1SzhQkIcUjHRd VN9U8ytSX4pyxT7wI6BikqXoE7zYeP2hce69K0FgFuO5iGQi648m9OYX+16UTtqlyHMs rh3jlSqvx82uDkdHUXm+HBfFPpyFjl93Juj7ewBXpf4/WGn0QaTs/5M5+w7LhUwo4iFW wEAXx9hIAnXHrLRH3uTgUFFWEQvVj/xrxj1dA5XbJUgegjt7N6qnL5vrn0nBAHDhwRsg u4hhUESFKu7Mj4Lj2AnVUVMyjrSYy1C70WTN2rU1GyUHcX5M0LK7ubSJWF+R6HI/cNcQ F8gA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1783460553; x=1784065353; h=content-transfer-encoding:mime-version:message-id:date:subject:cc :to:from:x-gm-gg:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to; bh=fr34X1DQSGmN+02DaqdMQrFw+Hk4XAU6EmNPh+kwNxk=; b=dg7RTlbGTtJ5gjREqK3ay+LK48ic2AYb56a3sSoS+mKQogIVnxtWmMYNLN53r6xfFu JzroBup1OkBuGu6oshzExb1ln8mYVQWZ+vw17OhOWVgb3AEuZTu5KHiWfltqt7JTuDyq R3eClKoEpSjz4VTNQtjHxLtixII4HOnDbK9L3Z1wjmaAH5yjIGZ+pMqtwd3KDbkaFkiZ hUpxv3Hef9sfvKDhS8Q3eQGrU/EoFuOG5KI85Av+crJxD/01uP1xiEjGMVvfsV9rD/2v XYudYhg40eywWePOv/DKjPZJxvma+vYln8ZSSm+IVVrJQABLXbzEntZ140s+5G+Ci0f+ YZgQ== X-Forwarded-Encrypted: i=1; AHgh+RrPq2NA325aUBtDSH9Di2Bd8/mgQFs1OQVLFUGXyzPqclK1SuOari+0i59bRculyISShjdp+8hBRau0@vger.kernel.org X-Gm-Message-State: AOJu0YxSHNMNUeUgB4MOl111wj+PQV4qZPi5DXY3VMgV9JFFMnLFVBAj YOdzwVmk0C5vVGo5aNnyG0un5i5nF6kkL6BH/+j3l5N5yJqSIicUZvfAxHM4wcxKQEk= X-Gm-Gg: AfdE7ck+CKsky/pUlGeqmJM8NE8vBqLAVa9Iuwr48sKPxQ89ou80pzGIsQByW2g/v5r K4Zs0LPX64wgWxB2wPjN7Fy60cWCKy9mfCyApM3GowpXVW6T+iGeK6lyQqKYKAC6L1H2zTNHFBK IIj7ZJEJtunpXQZ6HybUiGa595aFX+iqwd+TKPexueZafd9ALcvythLz+Z1tGzXW2LIKKAt585h OZExubnwfjlyB/jTO67xr9NBPqdRdHjb0LBDXGibwsRkaE+sH0hqm9GG7BSrU5Bk3rwbjDfCnFl Dy+passYMEqdy4zmRDWjBtebs9/fjYOFP9eyRZYWxyWAEEtl5EiX7PN5GAYIhnwisKDOMnVL1Kt cCjFmJ6Xs0/3A1BOLubmzDfT1ct0x1amOUV4ZX2A16d/7dGrSgJJiA9eSq6UsbrgZgAOX6mpyYB QfMmjN4K4gqMGcWqcAtD90mgHvOGBXCTQDLWmTWAS8dM2U8y1mUSkqRltxFcsl5+8I1+pfQBk3k KIWfyjK2KpbVOxa X-Received: by 2002:a05:600c:548d:b0:493:c478:8744 with SMTP id 5b1f17b1804b1-493df0644a7mr78150655e9.18.1783460553100; Tue, 07 Jul 2026 14:42:33 -0700 (PDT) Received: from raven.intern.cm-ag (p200300dc6f45eb00023064fffe740809.dip0.t-ipconnect.de. [2003:dc:6f45:eb00:230:64ff:fe74:809]) by smtp.gmail.com with ESMTPSA id ffacd0b85a97d-47a9e4d843csm36076615f8f.14.2026.07.07.14.42.32 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 07 Jul 2026 14:42:32 -0700 (PDT) From: Max Kellermann To: idryomov@gmail.com, amarkuze@redhat.com, ceph-devel@vger.kernel.org, linux-kernel@vger.kernel.org Cc: Max Kellermann , stable@vger.kernel.org Subject: [PATCH] fs/ceph/dir: do not repeat ceph_trim_dentries() if no progress possible Date: Tue, 7 Jul 2026 23:42:28 +0200 Message-ID: <20260707214228.10769-1-max.kellermann@ionos.com> X-Mailer: git-send-email 2.47.3 Precedence: bulk X-Mailing-List: ceph-devel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit ceph_cap_reclaim_work() re-queues itself for as long as ceph_trim_dentries() returns -EAGAIN, which happens whenever a lease walk exhausts its `nr_to_scan` budget. This creates a busy loop that consumes CPU without making any progress when there is nothing to reclaim: with no cap pressure (`count==0`) and every scanned lease still valid, each pass runs the full scan budget down to zero and returns `-EAGAIN`, only to be queued again immediately. The dir-lease walk made this worse. When `expire_dir_lease` is `false` (i.e. we have no intention of reclaiming dir leases), __dir_lease_check() returned `TOUCH` for every valid lease. `TOUCH` moves the dentry to the tail of the list and resets `di->time` via __dentry_dir_lease_touch(), so a walk over N valid leases pointlessly rewrote the list, refreshed the timestamps (preventing them from ever aging out) and always drained `nr_to_scan`, guaranteeing the `-EAGAIN` requeue. Fix this in three steps: - Return `KEEP` instead of `TOUCH` when `expire_dir_lease` is `false`. If we are not going to reclaim the lease, leave it in place instead of churning the list and resetting its timestamp; the walk then terminates naturally (or via `STOP` at the first fresh lease). - Only return `-EAGAIN` from the first (dentry-lease) walk when something was actually freed. A full batch that frees nothing means retrying the same list immediately is futile; fall through to the dir-lease walk instead. - After both walks, bail out with success (0) when nothing was freed and there is no cap pressure (`count==0`). There is no reason to keep retrying when we are not over the cap limit and made no progress. Under real cap pressure (`count>0`) the reclaim path is unchanged and still retries via `-EAGAIN`. Without this patch, I saw 500 ceph_trim_dentries() calls per second on our web servers. This is very visible in `/proc/lock_stat` (5 minute capture): class name con-bounces contentions waittime-min waittime-max waittime-total waittime-avg acq-bounces acquisitions holdtime-min holdtime-max holdtime-total holdtime-avg &mdsc->dentry_list_lock: 126180 128218 0.04 8063.44 15986965.20 124.69 1573354 5296812 0.04 8291.28 74164526.48 14.00 ----------------------- &mdsc->dentry_list_lock 111736 [<000000007b11e319>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 2631 [<0000000050597999>] __dentry_leases_walk+0x64/0x2c8 &mdsc->dentry_list_lock 3878 [<00000000c0022f62>] __ceph_dentry_lease_touch+0x5c/0xa8 &mdsc->dentry_list_lock 9973 [<000000002f27cb6f>] __dentry_lease_unlist+0x50/0xa0 ----------------------- &mdsc->dentry_list_lock 123621 [<0000000050597999>] __dentry_leases_walk+0x64/0x2c8 &mdsc->dentry_list_lock 1822 [<000000007b11e319>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 2720 [<000000002f27cb6f>] __dentry_lease_unlist+0x50/0xa0 &mdsc->dentry_list_lock 55 [<00000000c0022f62>] __ceph_dentry_lease_touch+0x5c/0xa8 With this patch: class name con-bounces contentions waittime-min waittime-max waittime-total waittime-avg acq-bounces acquisitions holdtime-min holdtime-max holdtime-total holdtime-avg &mdsc->dentry_list_lock: 1203 1215 0.16 408.88 33082.88 27.23 4320501 7357389 0.04 500.64 1961578.00 0.27 ----------------------- &mdsc->dentry_list_lock 1029 [<000000003c9aea8a>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 169 [<000000002038c577>] __dentry_lease_unlist+0x50/0xa0 &mdsc->dentry_list_lock 16 [<00000000c991106d>] __ceph_dentry_lease_touch+0x5c/0xa8 &mdsc->dentry_list_lock 1 [<00000000612fe15f>] __dentry_leases_walk+0x64/0x2c8 ----------------------- &mdsc->dentry_list_lock 158 [<000000002038c577>] __dentry_lease_unlist+0x50/0xa0 &mdsc->dentry_list_lock 858 [<000000003c9aea8a>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 182 [<00000000612fe15f>] __dentry_leases_walk+0x64/0x2c8 &mdsc->dentry_list_lock 17 [<00000000c991106d>] __ceph_dentry_lease_touch+0x5c/0xa8 __dentry_leases_walk() is almost gone. The total wait time is reduced by a factor of 483. That will give some latency gains to ceph_readdir(). Cc: stable@vger.kernel.org Fixes: 37c4efc1ddf9 ("ceph: periodically trim stale dentries") Signed-off-by: Max Kellermann --- fs/ceph/dir.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 27ce9e55e947..45cdb611fe2a 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -1759,11 +1759,11 @@ static int __dir_lease_check(const struct dentry *dentry, if (ret > 0) { if (time_before(jiffies, di->time + lwc->dir_lease_ttl)) return STOP; + if (!lwc->expire_dir_lease) + return KEEP; /* Move dentry to tail of dir lease list if we don't want * to delete it. So dentries in the list are checked in a * round robin manner */ - if (!lwc->expire_dir_lease) - return TOUCH; if (dentry->d_lockref.count > 0 || (di->flags & CEPH_DENTRY_REFERENCED)) return TOUCH; @@ -1790,7 +1790,7 @@ int ceph_trim_dentries(struct ceph_mds_client *mdsc) lwc.dir_lease = false; lwc.nr_to_scan = CEPH_CAPS_PER_RELEASE * 2; freed = __dentry_leases_walk(mdsc, &lwc); - if (!lwc.nr_to_scan) /* more invalid leases */ + if (freed > 0 && !lwc.nr_to_scan) /* more invalid leases */ return -EAGAIN; if (lwc.nr_to_scan < CEPH_CAPS_PER_RELEASE) @@ -1800,6 +1800,10 @@ int ceph_trim_dentries(struct ceph_mds_client *mdsc) lwc.expire_dir_lease = freed < count; lwc.dir_lease_ttl = mdsc->fsc->mount_options->caps_wanted_delay_max * HZ; freed +=__dentry_leases_walk(mdsc, &lwc); + if (freed == 0 && count == 0) + /* no progress possible currently, retry futile */ + return 0; + if (!lwc.nr_to_scan) /* more to check */ return -EAGAIN; -- 2.47.3