From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-yw1-f175.google.com (mail-yw1-f175.google.com [209.85.128.175]) (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 2A614411FB9 for ; Wed, 22 Jul 2026 17:20:20 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.128.175 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784740823; cv=none; b=Br9adRIro0bx1zlpVdD1IRP/dY60TD3BsaiJl8nqFF1ubt2OBz/A0CT3NfhvMngwq2pjN6RDpVESLBKtYzklYLb7VqDrZc2OZZp5EEjR6SzYyg2ICCYlCj//TEfcPGagrT1VyeAUWFO2FK9MV6RbX0x9ztl9BuDAoDp9jBKjv6I= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784740823; c=relaxed/simple; bh=hSwoM5fYECzpJjInQHh4DeaEIkR/oeowe2f3PU+yB5Q=; h=Message-ID:Subject:From:To:Cc:Date:In-Reply-To:References: Content-Type:MIME-Version; b=HGsU+T99/apAkC49N0UHlIKHeh8DaNdKdv3pPRB/vSIDe6J2LT58RCVHlyoC5FO1Alyid4Ga1nrsikl2pbP3mseAurH6mqNn88DICcuSQLsMkUo06nAxqk95E2CFgKx+naAyWNEpH2jyM0AO2qH+lcZxNxB71P0UoCs+bE8o1J0= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=dubeyko.com; spf=pass smtp.mailfrom=dubeyko.com; dkim=pass (2048-bit key) header.d=dubeyko-com.20251104.gappssmtp.com header.i=@dubeyko-com.20251104.gappssmtp.com header.b=JIB9zQYh; arc=none smtp.client-ip=209.85.128.175 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=dubeyko.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=dubeyko.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=dubeyko-com.20251104.gappssmtp.com header.i=@dubeyko-com.20251104.gappssmtp.com header.b="JIB9zQYh" Received: by mail-yw1-f175.google.com with SMTP id 00721157ae682-80dc4a68e4aso100401717b3.0 for ; Wed, 22 Jul 2026 10:20:20 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=dubeyko-com.20251104.gappssmtp.com; s=20251104; t=1784740820; x=1785345620; darn=vger.kernel.org; h=mime-version:user-agent:content-transfer-encoding:content-type :autocrypt:references:in-reply-to:date:cc:to:from:subject:message-id :from:to:cc:subject:date:message-id:reply-to:content-type; bh=MSzUPgIh+xa7YWqApOSZw6hzwFlIazSNiDaMddWGUUs=; b=JIB9zQYh/Rb+9OiqcvXVLuLbVb8EeopEpxqG1dm9oR1GczhbN0MpzC4+xTn12D3zG0 k+jhLefuCNxi6MwdXgQ4J+YOlXurHZI+S0Qe9UAZAdswVx96g+N0WGZZCod96hfSojy7 vKpQPwKjagaApsem43vfweSHV2vs463h8aMjyaSV1CWEoPZI+EvO4TcObuz4HNzx0Rlr jB31zZobNcHNUNc7sBqkDcQU42CPu53cZWv+92T7ECSB5uHuReZdwNnOa7fLjRVoMV13 72DFzEcbSbMrYAus7L/c4z1qSycIicBaSS866IZFUe+s3U4dIeYoNdLtDE4BkJX267+B AKxA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1784740820; x=1785345620; h=mime-version:user-agent:content-transfer-encoding:content-type :autocrypt:references:in-reply-to:date:cc:to:from:subject:message-id :x-gm-gg:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to:content-type; bh=MSzUPgIh+xa7YWqApOSZw6hzwFlIazSNiDaMddWGUUs=; b=ONebrvQuYUgwYgZQVu8f8VhNHM9CQdFzZ3Bp1O2wBWHB9Kbrg9ViwODUw8Iw+LkjhG 9KtlIha/zGI5UnEzbErnnCtePjLNGFoBcW0r1iKADvRWhwNv8/1r5bdPPnnwAGw/3nRx LrSH3KQevhAM94fx3dVDUZC/FTIVfF+FZZZ1WC6DRP/3oxIdwp+uRQ55V/n4oKRvqTG+ n0DZoueia/F90KfdMJnkwD4sHy7+5J2+PFfIE/s9NESQCuxN8W5qYSJdXn+L5Zqqeg8z CzDThUSMo+uoF/+xGAhvtRbPY19DnK8nRYe1xPyGGjmdzWSvHSdUb12ST7kIQmA75HUu Cb0w== X-Forwarded-Encrypted: i=1; AHgh+Rr0wqzDh5ALnrHHSNWP/wYtc636cOSJ/RI1uyP4XJfpBZPupnm32Ar8NGU26RXHdjwuICFDji4KBwV9ML0=@vger.kernel.org X-Gm-Message-State: AOJu0YzwAqCfR0JYi4Sp3K/8SEcN3/d1PbjG0cN5Jy1hm6fHIhjOAKIk epL9xhiGszB26Puw/0Uzyx9q8yASM1NfWWzsEQMVsz7iKG8rN8xRtl5doAF6k+9vYZw= X-Gm-Gg: AR+sD12H8rP575kLu/XX7r1ANtn+CrK4ijvprdRoZDvSQ+MxVtdhQhyibwNGaAkmmNq NdP3YIh1y6ddV+HxqnQ2wROsL4H+uVEiFtZYtZjaRGgMxzBE/1t1dGhovrzGp0nLHpKdvf15TuM 5Dd1uaK5zT1GOgC6FOLobzT0E8b1uEwSem9y3LQ35tKfH9WEI3vecG6WnnS9a3AvF+YmNkp5X4A IXukt5nWlKgHzEgmvIGVxVnjTmRObq7TrSwDxdzN7jog6lBBuoJvdyzOX85v4Kc5ZgupIVjdgaq ojmBaJQGyG2mSC3jkzJfRgEirHrSMTLS9bJNmzYNtf246iTaWJTwClVYBko7eNMtNz2nToByYud ZjqtpMEg+5TnwozKs2EoJ5zALrQLbeIHEWa5uhfXIKgwOsAIOl6zRBm6tTiiAXUVIzGB+9vVeWS zGikPScmy2E1SQpiCOMgmJUtZiYhipIYVrGqpJ3+ZsMg6gXYcyYe2F95MuFSMZE8h8Fa725gD6/ Ajh6KyMHn+AvGzGpRp7wIC86b+nt7glBsnlZ6tz2f5HKwg6+4tq0x8b1DVd8MImXkqVZNU= X-Received: by 2002:a05:690c:360e:b0:81d:2194:6c4a with SMTP id 00721157ae682-81ef2610887mr78447727b3.47.1784740819725; Wed, 22 Jul 2026 10:20:19 -0700 (PDT) Received: from ?IPv6:2600:1700:6476:1430:12d8:c69b:ed00:1fe4? ([2600:1700:6476:1430:12d8:c69b:ed00:1fe4]) by smtp.gmail.com with ESMTPSA id 00721157ae682-81f33d9fd46sm16658927b3.28.2026.07.22.10.20.18 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 22 Jul 2026 10:20:19 -0700 (PDT) Message-ID: Subject: Re: [PATCH] ceph: avoid fs reclaim while using `current->journal_info` From: Viacheslav Dubeyko To: Max Kellermann , idryomov@gmail.com, amarkuze@redhat.com, xiubo.li@clyso.com, ceph-devel@vger.kernel.org, linux-kernel@vger.kernel.org Cc: stable@vger.kernel.org Date: Wed, 22 Jul 2026 10:20:17 -0700 In-Reply-To: <20260722114931.3557110-1-max.kellermann@ionos.com> References: <20260722114931.3557110-1-max.kellermann@ionos.com> Autocrypt: addr=slava@dubeyko.com; prefer-encrypt=mutual; keydata=mQINBGgaTLYBEADaJc/WqWTeunGetXyyGJ5Za7b23M/ozuDCWCp+yWUa2GqQKH40dxRIR zshgOmAue7t9RQJU9lxZ4ZHWbi1Hzz85+0omefEdAKFmxTO6+CYV0g/sapU0wPJws3sC2Pbda9/eJ ZcvScAX2n/PlhpTnzJKf3JkHh3nM1ACO3jzSe2/muSQJvqMLG2D71ccekr1RyUh8V+OZdrPtfkDam V6GOT6IvyE+d+55fzmo20nJKecvbyvdikWwZvjjCENsG9qOf3TcCJ9DDYwjyYe1To8b+mQM9nHcxp jUsUuH074BhISFwt99/htZdSgp4csiGeXr8f9BEotRB6+kjMBHaiJ6B7BIlDmlffyR4f3oR/5hxgy dvIxMocqyc03xVyM6tA4ZrshKkwDgZIFEKkx37ec22ZJczNwGywKQW2TGXUTZVbdooiG4tXbRBLxe ga/NTZ52ZdEkSxAUGw/l0y0InTtdDIWvfUT+WXtQcEPRBE6HHhoeFehLzWL/o7w5Hog+0hXhNjqte fzKpI2fWmYzoIb6ueNmE/8sP9fWXo6Av9m8B5hRvF/hVWfEysr/2LSqN+xjt9NEbg8WNRMLy/Y0MS p5fgf9pmGF78waFiBvgZIQNuQnHrM+0BmYOhR0JKoHjt7r5wLyNiKFc8b7xXndyCDYfniO3ljbr0j tXWRGxx4to6FwARAQABtCZWaWFjaGVzbGF2IER1YmV5a28gPHNsYXZhQGR1YmV5a28uY29tPokCVw QTAQoAQQIbAQUJA8JnAAULCQgHAgYVCgkICwIEFgIDAQIeAQIXgBYhBFXDC2tnzsoLQtrbBDlc2cL fhEB1BQJoGl5PAhkBAAoJEDlc2cLfhEB17DsP/jy/Dx19MtxWOniPqpQf2s65enkDZuMIQ94jSg7B F2qTKIbNR9SmsczjyjC+/J7m7WZRmcqnwFYMOyNfh12aF2WhjT7p5xEAbvfGVYwUpUrg/lcacdT0D Yk61GGc5ZB89OAWHLr0FJjI54bd7kn7E/JRQF4dqNsxU8qcPXQ0wLHxTHUPZu/w5Zu/cO+lQ3H0Pj pSEGaTAh+tBYGSvQ4YPYBcV8+qjTxzeNwkw4ARza8EjTwWKP2jWAfA/ay4VobRfqNQ2zLoo84qDtN Uxe0zPE2wobIXELWkbuW/6hoQFPpMlJWz+mbvVms57NAA1HO8F5c1SLFaJ6dN0AQbxrHi45/cQXla 9hSEOJjxcEnJG/ZmcomYHFneM9K1p1K6HcGajiY2BFWkVet9vuHygkLWXVYZ0lr1paLFR52S7T+cf 6dkxOqu1ZiRegvFoyzBUzlLh/elgp3tWUfG2VmJD3lGpB3m5ZhwQ3rFpK8A7cKzgKjwPp61Me0o9z HX53THoG+QG+o0nnIKK7M8+coToTSyznYoq9C3eKeM/J97x9+h9tbizaeUQvWzQOgG8myUJ5u5Dr4 6tv9KXrOJy0iy/dcyreMYV5lwODaFfOeA4Lbnn5vRn9OjuMg1PFhCi3yMI4lA4umXFw0V2/OI5rgW BQELhfvW6mxkihkl6KLZX8m1zcHitCpWaWFjaGVzbGF2IER1YmV5a28gPFNsYXZhLkR1YmV5a29Aa WJtLmNvbT6JAlQEEwEKAD4WIQRVwwtrZ87KC0La2wQ5XNnC34RAdQUCaBpd7AIbAQUJA8JnAAULCQ gHAgYVCgkICwIEFgIDAQIeAQIXgAAKCRA5XNnC34RAdYjFEACiWBEybMt1xjRbEgaZ3UP5i2bSway DwYDvgWW5EbRP7JcqOcZ2vkJwrK3gsqC3FKpjOPh7ecE0I4vrabH1Qobe2N8B2Y396z24mGnkTBbb 16Uz3PC93nFN1BA0wuOjlr1/oOTy5gBY563vybhnXPfSEUcXRd28jI7z8tRyzXh2tL8ZLdv1u4vQ8 E0O7lVJ55p9yGxbwgb5vXU4T2irqRKLxRvU80rZIXoEM7zLf5r7RaRxgwjTKdu6rYMUOfoyEQQZTD 4Xg9YE/X8pZzcbYFs4IlscyK6cXU0pjwr2ssjearOLLDJ7ygvfOiOuCZL+6zHRunLwq2JH/RmwuLV mWWSbgosZD6c5+wu6DxV15y7zZaR3NFPOR5ErpCFUorKzBO1nA4dwOAbNym9OGkhRgLAyxwpea0V0 ZlStfp0kfVaSZYo7PXd8Bbtyjali0niBjPpEVZdgtVUpBlPr97jBYZ+L5GF3hd6WJFbEYgj+5Af7C UjbX9DHweGQ/tdXWRnJHRzorxzjOS3003ddRnPtQDDN3Z/XzdAZwQAs0RqqXrTeeJrLppFUbAP+HZ TyOLVJcAAlVQROoq8PbM3ZKIaOygjj6Yw0emJi1D9OsN2UKjoe4W185vamFWX4Ba41jmCPrYJWAWH fAMjjkInIPg7RLGs8FiwxfcpkILP0YbVWHiNAabQoVmlhY2hlc2xhdiBEdWJleWtvIDx2ZHViZXlr b0BrZXJuZWwub3JnPokCVAQTAQoAPhYhBFXDC2tnzsoLQtrbBDlc2cLfhEB1BQJoVemuAhsBBQkDw mcABQsJCAcCBhUKCQgLAgQWAgMBAh4BAheAAAoJEDlc2cLfhEB1GRwP/1scX5HO9Sk7dRicLD/fxo ipwEs+UbeA0/TM8OQfdRI4C/tFBYbQCR7lD05dfq8VsYLEyrgeLqP/iRhabLky8LTaEdwoAqPDc/O 9HRffx/faJZqkKc1dZryjqS6b8NExhKOVWmDqN357+Cl/H4hT9wnvjCj1YEqXIxSd/2Pc8+yw/KRC AP7jtRzXHcc/49Lpz/NU5irScusxy2GLKa5o/13jFK3F1fWX1wsOJF8NlTx3rLtBy4GWHITwkBmu8 zI4qcJGp7eudI0l4xmIKKQWanEhVdzBm5UnfyLIa7gQ2T48UbxJlWnMhLxMPrxgtC4Kos1G3zovEy Ep+fJN7D1pwN9aR36jVKvRsX7V4leIDWGzCdfw1FGWkMUfrRwgIl6i3wgqcCP6r9YSWVQYXdmwdMu 1RFLC44iF9340S0hw9+30yGP8TWwd1mm8V/+zsdDAFAoAwisi5QLLkQnEsJSgLzJ9daAsE8KjMthv hUWHdpiUSjyCpigT+KPl9YunZhyrC1jZXERCDPCQVYgaPt+Xbhdjcem/ykv8UVIDAGVXjuk4OW8la nf8SP+uxkTTDKcPHOa5rYRaeNj7T/NClRSd4z6aV3F6pKEJnEGvv/DFMXtSHlbylhyiGKN2Amd0b4 9jg+DW85oNN7q2UYzYuPwkHsFFq5iyF1QggiwYYTpoVXsw Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable User-Agent: Evolution 3.60.1 (by Flathub.org) Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 On Wed, 2026-07-22 at 13:49 +0200, Max Kellermann wrote: > handle_reply() stores a `ceph_mds_request` pointer in > `current->journal_info` while filling the inode and dentry cache from > an MDS reply. >=20 > An allocation in this section can enter direct reclaim and prune > dentries from another filesystem.=C2=A0 If this dirties an ext4 inode, > ext4 > starts a JBD2 transaction.=C2=A0 JBD2 interprets the Ceph request in > `current->journal_info` as a journal handle and dereferences the > request's `r_tid` as `h_transaction`, causing a kernel crash, e.g.: >=20 > =C2=A0Unable to handle kernel paging request at virtual address > 00000000077b4818 > =C2=A0[...] > =C2=A0Internal error: Oops: 0000000096000004 [#1]=C2=A0 SMP > =C2=A0Modules linked in: > =C2=A0CPU: 6 UID: 0 PID: 2699135 Comm: kworker/6:3 Tainted: G=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 > W=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 6.18.38-i3 = #1113 NONE > =C2=A0[...] > =C2=A0Workqueue: ceph-msgr ceph_con_workfn > =C2=A0pstate: 80400009 (Nzcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=3D--) > =C2=A0pc : jbd2__journal_start+0x2c/0x208 > =C2=A0lr : __ext4_journal_start_sb+0x100/0x178 > =C2=A0[...] > =C2=A0Call trace: > =C2=A0 jbd2__journal_start+0x2c/0x208 (P) > =C2=A0 __ext4_journal_start_sb+0x100/0x178 > =C2=A0 ext4_dirty_inode+0x3c/0x90 > =C2=A0 __mark_inode_dirty+0x58/0x400 > =C2=A0 iput.part.0+0x2b0/0x370 > =C2=A0 iput+0x18/0x30 > =C2=A0 dentry_unlink_inode+0xc0/0x158 > =C2=A0 __dentry_kill+0x80/0x250 > =C2=A0 shrink_dentry_list+0x90/0x130 > =C2=A0 prune_dcache_sb+0x60/0x98 > =C2=A0 super_cache_scan+0xe8/0x190 > =C2=A0 do_shrink_slab+0x174/0x388 > =C2=A0 shrink_slab+0xd8/0x4c0 > =C2=A0 shrink_node+0x31c/0x908 > =C2=A0 do_try_to_free_pages+0xd0/0x508 > =C2=A0 try_to_free_pages+0x11c/0x238 > =C2=A0 __alloc_frozen_pages_noprof+0x4d0/0xdd0 > =C2=A0 __folio_alloc_noprof+0x18/0x70 > =C2=A0 __filemap_get_folio+0x248/0x440 > =C2=A0 ceph_readdir_prepopulate+0x570/0x9e8 > =C2=A0 mds_dispatch+0x1424/0x1ba0 > =C2=A0 ceph_con_process_message+0x74/0xa0 > =C2=A0 ceph_con_v1_try_read+0x3a0/0x1510 > =C2=A0 ceph_con_workfn+0x260/0x460 >=20 > Enter a scoped NOFS allocation context and leave it after clearing > `journal_info`.=C2=A0 This prevents filesystem reclaim from recursing int= o > another filesystem while the field contains Ceph-private data. >=20 > Fixes: 315f24088048 ("ceph: fix security xattr deadlock") > Cc: stable@vger.kernel.org > Signed-off-by: Max Kellermann > --- > =C2=A0fs/ceph/mds_client.c | 11 +++++++++++ > =C2=A01 file changed, 11 insertions(+) >=20 > diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c > index 853bf698b356..3c692ad02c85 100644 > --- a/fs/ceph/mds_client.c > +++ b/fs/ceph/mds_client.c > @@ -6,6 +6,7 @@ > =C2=A0#include > =C2=A0#include > =C2=A0#include > +#include > =C2=A0#include > =C2=A0#include > =C2=A0#include > @@ -4015,6 +4016,7 @@ static void handle_reply(struct > ceph_mds_session *session, struct ceph_msg *msg) > =C2=A0 struct ceph_mds_reply_head *head =3D msg->front.iov_base; > =C2=A0 struct ceph_mds_reply_info_parsed *rinfo;=C2=A0 /* parsed reply > info */ > =C2=A0 struct ceph_snap_realm *realm; > + unsigned int nofs_flags; > =C2=A0 u64 tid; > =C2=A0 int err, result; > =C2=A0 int mds =3D session->s_mds; > @@ -4158,6 +4160,14 @@ static void handle_reply(struct > ceph_mds_session *session, struct ceph_msg *msg) > =C2=A0 > =C2=A0 /* insert trace into our cache */ > =C2=A0 mutex_lock(&req->r_fill_mutex); > + > + /* disable fs reclaim while we are using current- > >journal_info > + * for our own purposes, or else shrinkers of other > + * filesystems might dereference this pointer as a different > + * type > + */ > + nofs_flags =3D memalloc_nofs_save(); > + > =C2=A0 current->journal_info =3D req; > =C2=A0 err =3D ceph_fill_trace(mdsc->fsc->sb, req); > =C2=A0 if (err =3D=3D 0) { > @@ -4166,6 +4176,7 @@ static void handle_reply(struct > ceph_mds_session *session, struct ceph_msg *msg) > =C2=A0 err =3D ceph_readdir_prepopulate(req, req- > >r_session); > =C2=A0 } > =C2=A0 current->journal_info =3D NULL; > + memalloc_nofs_restore(nofs_flags); > =C2=A0 mutex_unlock(&req->r_fill_mutex); > =C2=A0 > =C2=A0 up_read(&mdsc->snap_rwsem); Makes sense. Reviewed-by: Viacheslav Dubeyko Thanks, Slava.