From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-qv1-f47.google.com (mail-qv1-f47.google.com [209.85.219.47]) (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 BEE1350AC3B for ; Fri, 4 Sep 2026 16:53:35 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.219.47 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1788540818; cv=none; b=dsR9dfOykU2N3uIE4XCRk8CiEeTfZXC2Ns20MYUrjwf68Lrjz6ceihN9x00NvSG7dFI6Ov7Xo1ctaeYanvZkrum13djU0jnR0Yiv4m7qqLhmPXXU+TaIIkKlPx/IlyYUyxOlDgwS+FLAq1gipS7JSHyksVKszdEZUwr/58rGgR4= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1788540818; c=relaxed/simple; bh=NsS/wBdgVphBehzfPZ4E+J3aHO8TnNl8akohe/pLCTY=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=A3zkUDSifDRWjao7EoLNg/e+NhDN9PX8+1vW3UEMcWZ3BjP9/uAZUM3UVWAFlZjGSWnRiXJ1iLNSLwyO9mxCSXfE06iJe2V39Pe00qcCA+RPwSxPzqMdYr5mKAsj51lfOJM7TT8iYdQM8fMtfFA0cIDP0XFt9WwZk57kScBTwPU= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=hammerspace.com; spf=pass smtp.mailfrom=hammerspace.com; dkim=pass (2048-bit key) header.d=hammerspace.com header.i=@hammerspace.com header.b=gV5Zti1u; arc=none smtp.client-ip=209.85.219.47 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=hammerspace.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=hammerspace.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=hammerspace.com header.i=@hammerspace.com header.b="gV5Zti1u" Received: by mail-qv1-f47.google.com with SMTP id 6a1803df08f44-90ce08834feso18028576d6.0 for ; Fri, 04 Sep 2026 09:53:35 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=hammerspace.com; s=google; t=1788540815; x=1789145615; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to:content-type; bh=FCUkdn4GlCBy4F6HhgzKuTpOMuxKg5g3TVFBVpDgo6w=; b=gV5Zti1usht12XqwjPlHt7OGWPC24F1N5IFdY+3acRxJfyxRabkPwsOaGG6Kn3AwLx kvGWRi0W25NF27GHWyLMLR8JzJbu7mBJ3aZnNvsjhWId54xSO50PNS+CK4bfA1a9MdXv i4ttnIc9tq7hTlg8UGfs1RIb5NMTbm52bOS34lL0bSfaRlsRyzaHYZmxuFUeQUbC45u0 2JMaD1rXAKkEoWwxQLshXByY/tBSkOc+93Sab7k+N+ZAlROENDCNJZwbX9AnntO8lmqt hcKqUryFBNaDGMxt3Uz5fF9D12GkW5WBrK0xADa84/FLFM9W4UeoJxcLL57TLG3MAhT1 v5/g== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1788540815; x=1789145615; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-gg:x-gm-message-state:from :to:cc:subject:date:message-id:reply-to:content-type; bh=FCUkdn4GlCBy4F6HhgzKuTpOMuxKg5g3TVFBVpDgo6w=; b=R8gx98D+X/4kT7OP1hJ8DBBALeGLkjb1aEcBI4hNo92LNRHRXvvkGTkRIQPL47YtNB 9XhKavHliBwS4mdJbZuq3NnwZOCbcxtvFrtTkdGIVekHHrUPomDHbrzed0tQD8Md0LKr 4AvtFvQywazcbskHMA7iBrayfJsGl5NLrarF2OQFrs3y5nmIX8xb3r+CVXreYzuF2Tv3 GwE7ti3wQLuaBdQ1bkmxAZ8g0Am8YaJj+Fb1WHLc8riYSL1ZWZpLkL6xJvsm02h7ac0h XE2ULgNsfe9hqZgntyGFJdVzkPeiRcz8NUttGXzFhMqHzlG6sUv2fQmXMI2HnPoINZbx VqGw== X-Gm-Message-State: AFuF++kmlIbdn8b80SR36Y4Ur+p5zaofalzhgrlhFEpLXLy7o6niSf3Y ZSn+JOeH2ZimvdO2fQfq2MG1DWse3FVPd8JIpl4DuOdJr1rc/l7fJSI6LO8pvmhgLI4= X-Gm-Gg: AYBFou0yqEhrLcUEqRl+eho+Q50IIAYmTe5EQt9wHA0Ktop9KqEUsYx/ik12TVrABjH yd1p4aFHg0W5Mw03WWlII2OPvz40kt1oZVkKg1zmTWTPsYvbHvnhBpGuoghPappVd5P4Ciippzb rtA6dlwoCTOhYQJAZ+zgLHOmBfqQMALcVlSq+4N7GxC93nH1U3Cr3bp+wlXvyxQHpMM99WdccOV ajz3sUFO/RHPbeM9XCsPo0eyuifGPERYLIEzxflszf+OzFc28ewgQddO/J/LnFh3DZdntYbr0Rz O4aLZgZVPXMZpbV2nrFYXFJCsVkqOL0ETC8U03/M3Zfm2NlMm2pj1+o1Qp9aeQBx5IQj9Q2XbYF XFalxFSD6aV3jcbRlBBRWjvtRw42/UzdntZjAtOEzMquHmooMkB+5b0YMseIXre/oNWtMCDtteL OUEBmfE22B2VuR29AKH7zceaVCGOF25AtUri7YyzD4pQd6OAg00M58xJck3HhIN3ls0Sa1NyNWO LRp2VtRSD6V42kwZp9rF9bf X-Received: by 2002:a05:620a:690e:b0:939:6de8:43de with SMTP id af79cd13be357-939804e9754mr677750485a.50.1788540814313; Fri, 04 Sep 2026 09:53:34 -0700 (PDT) Received: from bcodding.csb.hammerspace.com ([66.97.168.37]) by smtp.gmail.com with ESMTPSA id af79cd13be357-9397fbf7c78sm248145485a.47.2026.09.04.09.53.33 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 04 Sep 2026 09:53:33 -0700 (PDT) From: Benjamin Coddington X-Google-Original-From: Benjamin Coddington To: Trond Myklebust , Anna Schumaker Cc: linux-nfs@vger.kernel.org, Jonathan Curley , Mike Snitzer , Jeff Layton , Junrui Luo Subject: [PATCH v3 11/24] NFSv4/flexfiles: Make the pinned device node pointer RCU-managed Date: Fri, 4 Sep 2026 12:53:10 -0400 Message-ID: <99101deee644125ecc91a8cd3f9435a0d8593df0.1788530385.git.bcodding@hammerspace.com> X-Mailer: git-send-email 2.53.0 In-Reply-To: References: Precedence: bulk X-Mailing-List: linux-nfs@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Annotate mirror->dss[dss_id].mirror_ds as __rcu and convert the remaining readers, completing the preparation for re-pointing the pinned node while I/O is in flight: - ff_layout_get_mirror_ds() takes its reference under rcu_read_lock() with atomic_inc_not_zero(), retrying if it races a reset. The resolve path takes the caller's reference before publishing the node with cmpxchg(), so a concurrent reset cannot drop the last reference under the caller: one reference is held for the installed pointer and one for the caller, and the pointer's reference is released elsewhere by xchg + put. - The availability scans hold rcu_read_lock() across the walk; they only test flags on the RCU-freed node. - ff_layout_cancel_io() holds a reference across the cancel and disconnect calls. Both of its callers hold i_lock and nothing on that path sleeps, so the reference is not about blocking: it is what keeps the node alive between the RCU-protected read and the use of mirror_ds->ds. The final put is never reached here, because the mirror's own pin outlives the loop -- which matters, since that put ends in nfs_put_client() and cannot run under a spinlock. - ff_layout_mirror_prepare_stats() reads the pointer with rcu_dereference() under rcu_read_lock(). i_lock, which both callers hold, is what excludes the re-pointing walk added later in this series; it does not exclude the resolve path's cmpxchg(), which runs from I/O submission, so the read cannot claim i_lock as its update- side lock. A non-NULL pointer read there is stable regardless: the resolve path only ever installs over NULL. - ff_layout_free_mirror() tears down the last reference; no concurrency. The pointer is still only ever set once per mirror lifetime, so there is no behavior change; this commit makes the subsequent in-place re-resolve on CB_NOTIFY_DEVICEID CHANGE safe to introduce. Assisted-by: Claude:claude-fable-5 Signed-off-by: Benjamin Coddington --- fs/nfs/flexfilelayout/flexfilelayout.c | 35 ++++--- fs/nfs/flexfilelayout/flexfilelayout.h | 2 +- fs/nfs/flexfilelayout/flexfilelayoutdev.c | 110 ++++++++++++++-------- 3 files changed, 93 insertions(+), 54 deletions(-) diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index 015428fd6937..eeda084d20dd 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -313,7 +313,9 @@ static void ff_layout_free_mirror(struct nfs4_ff_layout_mirror *mirror) cred = rcu_access_pointer(mirror->dss[dss_id].rw_cred); put_cred(cred); nfs_close_local_fh(&mirror->dss[dss_id].nfl); - nfs4_ff_layout_put_deviceid(mirror->dss[dss_id].mirror_ds); + /* the last reference to the mirror is gone; no concurrency */ + nfs4_ff_layout_put_deviceid(rcu_dereference_protected( + mirror->dss[dss_id].mirror_ds, 1)); } kfree(mirror->dss); @@ -2498,22 +2500,29 @@ static void ff_layout_cancel_io(struct pnfs_layout_segment *lseg) for (idx = 0; idx < flseg->mirror_array_cnt; idx++) { mirror = flseg->mirror_array[idx]; for (dss_id = 0; dss_id < mirror->dss_count; dss_id++) { - mirror_ds = mirror->dss[dss_id].mirror_ds; - if (IS_ERR_OR_NULL(mirror_ds)) + rcu_read_lock(); + mirror_ds = rcu_dereference(mirror->dss[dss_id].mirror_ds); + if (IS_ERR_OR_NULL(mirror_ds) || + !atomic_inc_not_zero(&mirror_ds->id_node.ref)) { + rcu_read_unlock(); continue; - ds = mirror->dss[dss_id].mirror_ds->ds; + } + rcu_read_unlock(); + ds = mirror_ds->ds; if (!ds) - continue; + goto next; ds_clp = ds->ds_clp; if (!ds_clp) - continue; + goto next; clnt = ds_clp->cl_rpcclient; if (!clnt) - continue; + goto next; if (!rpc_cancel_tasks(clnt, -ECANCELED, ff_layout_match_io, lseg)) - continue; + goto next; rpc_clnt_disconnect(clnt); +next: + nfs4_ff_layout_put_deviceid(mirror_ds); } } } @@ -2975,12 +2984,13 @@ ff_layout_mirror_prepare_stats(struct pnfs_layout_hdr *lo, struct nfs4_ff_layout_ds *mirror_ds; int i = 0, dss_id; + rcu_read_lock(); list_for_each_entry(mirror, &ff_layout->mirrors, mirrors) { for (dss_id = 0; dss_id < mirror->dss_count; ++dss_id) { dss_info = &mirror->dss[dss_id]; if (i >= dev_limit) break; - mirror_ds = dss_info->mirror_ds; + mirror_ds = rcu_dereference(dss_info->mirror_ds); if (IS_ERR_OR_NULL(mirror_ds)) continue; if (!test_and_clear_bit(NFS4_FF_MIRROR_STAT_AVAIL, @@ -2990,10 +3000,8 @@ ff_layout_mirror_prepare_stats(struct pnfs_layout_hdr *lo, /* mirror refcount put in cleanup_layoutstats */ if (!refcount_inc_not_zero(&mirror->ref)) continue; - /* - * The mirror's pin holds the node while we're under - * i_lock; take a reference for the encode, put in - * ff_layout_free_layoutstats(). + /* The pin holds a reference; it is exchanged out only + * under i_lock. Put in ff_layout_free_layoutstats(). */ atomic_inc(&mirror_ds->id_node.ref); memcpy(&devinfo->dev_id, @@ -3022,6 +3030,7 @@ ff_layout_mirror_prepare_stats(struct pnfs_layout_hdr *lo, i++; } } + rcu_read_unlock(); return i; } diff --git a/fs/nfs/flexfilelayout/flexfilelayout.h b/fs/nfs/flexfilelayout/flexfilelayout.h index 8b42a98a2c4c..72b11034851a 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.h +++ b/fs/nfs/flexfilelayout/flexfilelayout.h @@ -79,7 +79,7 @@ struct nfs4_ff_layout_ds_stripe { struct nfs4_ff_layout_mirror *mirror; struct nfs4_deviceid devid; u32 efficiency; - struct nfs4_ff_layout_ds *mirror_ds; + struct nfs4_ff_layout_ds __rcu *mirror_ds; u32 fh_versions_cnt; struct nfs_fh *fh_versions; nfs4_stateid stateid; diff --git a/fs/nfs/flexfilelayout/flexfilelayoutdev.c b/fs/nfs/flexfilelayout/flexfilelayoutdev.c index ebcfba447879..5cb09e5e2138 100644 --- a/fs/nfs/flexfilelayout/flexfilelayoutdev.c +++ b/fs/nfs/flexfilelayout/flexfilelayoutdev.c @@ -321,35 +321,54 @@ ff_layout_get_mirror_ds(struct pnfs_layout_hdr *lo, struct nfs4_ff_layout_mirror *mirror, u32 dss_id) { - struct nfs4_ff_layout_ds *mirror_ds; + struct nfs4_ff_layout_ds *mirror_ds, *old; + struct nfs4_deviceid_node *node; if (mirror == NULL) return ERR_PTR(-ENODEV); - mirror_ds = mirror->dss[dss_id].mirror_ds; - if (mirror_ds == NULL) { - struct nfs4_deviceid_node *node; - +retry: + rcu_read_lock(); + mirror_ds = rcu_dereference(mirror->dss[dss_id].mirror_ds); + if (mirror_ds && !IS_ERR(mirror_ds) && + atomic_inc_not_zero(&mirror_ds->id_node.ref)) { + rcu_read_unlock(); + return mirror_ds; + } + rcu_read_unlock(); + if (IS_ERR(mirror_ds)) + return mirror_ds; + if (mirror_ds != NULL) + /* raced with a reset; the field is being re-pointed */ + goto retry; + + node = nfs4_find_get_deviceid(NFS_SERVER(lo->plh_inode), + &mirror->dss[dss_id].devid, lo->plh_lc_cred, + GFP_KERNEL); + if (node) { + mirror_ds = FF_LAYOUT_MIRROR_DS(node); + /* + * Take the caller's reference before the pointer becomes + * visible below, so a concurrent reset of the installed + * pointer cannot drop the last reference under us. + */ + atomic_inc(&node->ref); + } else { mirror_ds = ERR_PTR(-ENODEV); - node = nfs4_find_get_deviceid(NFS_SERVER(lo->plh_inode), - &mirror->dss[dss_id].devid, lo->plh_lc_cred, - GFP_KERNEL); - if (node) - mirror_ds = FF_LAYOUT_MIRROR_DS(node); - - /* check for race with another call to this function */ - if (cmpxchg(&mirror->dss[dss_id].mirror_ds, NULL, mirror_ds) && - mirror_ds != ERR_PTR(-ENODEV)) - nfs4_put_deviceid_node(node); - - mirror_ds = mirror->dss[dss_id].mirror_ds; } - if (IS_ERR(mirror_ds)) + /* check for race with another call to this function */ + old = unrcu_pointer(cmpxchg(&mirror->dss[dss_id].mirror_ds, + NULL, RCU_INITIALIZER(mirror_ds))); + if (old == NULL) return mirror_ds; - if (!atomic_inc_not_zero(&mirror_ds->id_node.ref)) - return ERR_PTR(-ENODEV); - return mirror_ds; + + /* lost the race; use the winner's node instead */ + if (node) { + nfs4_put_deviceid_node(node); + nfs4_put_deviceid_node(node); + } + goto retry; } /** @@ -573,49 +592,60 @@ unsigned int ff_layout_fetch_ds_ioerr(struct pnfs_layout_hdr *lo, static bool ff_read_layout_has_available_ds(struct pnfs_layout_segment *lseg) { struct nfs4_ff_layout_mirror *mirror; - struct nfs4_deviceid_node *devid; + struct nfs4_ff_layout_ds *mirror_ds; + bool ret = false; u32 idx, dss_id; + rcu_read_lock(); for (idx = 0; idx < FF_LAYOUT_MIRROR_COUNT(lseg); idx++) { mirror = FF_LAYOUT_COMP(lseg, idx); if (!mirror) continue; for (dss_id = 0; dss_id < mirror->dss_count; dss_id++) { - if (!mirror->dss[dss_id].mirror_ds) - return true; - if (IS_ERR(mirror->dss[dss_id].mirror_ds)) + mirror_ds = rcu_dereference(mirror->dss[dss_id].mirror_ds); + if (!mirror_ds) { + ret = true; + goto out; + } + if (IS_ERR(mirror_ds)) continue; - devid = &mirror->dss[dss_id].mirror_ds->id_node; - if (!nfs4_test_deviceid_unavailable(devid)) - return true; + if (!nfs4_test_deviceid_unavailable(&mirror_ds->id_node)) { + ret = true; + goto out; + } } } - - return false; +out: + rcu_read_unlock(); + return ret; } static bool ff_rw_layout_has_available_ds(struct pnfs_layout_segment *lseg) { struct nfs4_ff_layout_mirror *mirror; - struct nfs4_deviceid_node *devid; + struct nfs4_ff_layout_ds *mirror_ds; + bool ret = false; u32 idx, dss_id; + rcu_read_lock(); for (idx = 0; idx < FF_LAYOUT_MIRROR_COUNT(lseg); idx++) { mirror = FF_LAYOUT_COMP(lseg, idx); if (!mirror) - return false; + goto out; for (dss_id = 0; dss_id < mirror->dss_count; dss_id++) { - if (IS_ERR(mirror->dss[dss_id].mirror_ds)) - return false; - if (!mirror->dss[dss_id].mirror_ds) + mirror_ds = rcu_dereference(mirror->dss[dss_id].mirror_ds); + if (IS_ERR(mirror_ds)) + goto out; + if (!mirror_ds) continue; - devid = &mirror->dss[dss_id].mirror_ds->id_node; - if (nfs4_test_deviceid_unavailable(devid)) - return false; + if (nfs4_test_deviceid_unavailable(&mirror_ds->id_node)) + goto out; } } - - return FF_LAYOUT_MIRROR_COUNT(lseg) != 0; + ret = FF_LAYOUT_MIRROR_COUNT(lseg) != 0; +out: + rcu_read_unlock(); + return ret; } static bool ff_layout_has_available_ds(struct pnfs_layout_segment *lseg) -- 2.53.0