All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	patches@lists.linux.dev,
	Kent Overstreet <kent.overstreet@linux.dev>
Subject: [PATCH 6.15 18/34] bcachefs: Fix subvol to missing root repair
Date: Sat,  7 Jun 2025 12:07:59 +0200	[thread overview]
Message-ID: <20250607100720.433460280@linuxfoundation.org> (raw)
In-Reply-To: <20250607100719.711372213@linuxfoundation.org>

6.15-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Kent Overstreet <kent.overstreet@linux.dev>

commit 29cc6fb7c068c773049d3bde14b939033893eff4 upstream.

We had a bug where the root inode of a subvolume was erronously deleted:
bch2_evict_inode() called bch2_inode_rm(), meaning the VFS inode's
i_nlink was somehow set to 0 when it shouldn't have - the inode in the
btree indicated it clearly was not unlinked.

This has been addressed with additional safety checks in
bch2_inode_rm() - pulling in the safety checks we already were doing
when deleting unlinked inodes in recovery - but the really disastrous
bug was in check_subvols(), which on finding a dangling subvol (subvol
with a missing root inode) would delete the subvolume.

I assume this bug dates from early check_directory_structure() code,
which originally handled subvolumes and normal paths - the idea being
that still live contents of the subvolume would get reattached
somewhere.

But that's incorrect, and disastrously so; deleting a subvolume triggers
deleting the snapshot ID it points to, deleting the entire contents.

The correct way to repair is to recreate the root inode if it's missing;
then any contents will get reattached under that subvolume's lost+found.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 fs/bcachefs/subvolume.c |   19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

--- a/fs/bcachefs/subvolume.c
+++ b/fs/bcachefs/subvolume.c
@@ -6,6 +6,7 @@
 #include "errcode.h"
 #include "error.h"
 #include "fs.h"
+#include "inode.h"
 #include "recovery_passes.h"
 #include "snapshot.h"
 #include "subvolume.h"
@@ -113,10 +114,20 @@ static int check_subvol(struct btree_tra
 			     "subvolume %llu points to missing subvolume root %llu:%u",
 			     k.k->p.offset, le64_to_cpu(subvol.v->inode),
 			     le32_to_cpu(subvol.v->snapshot))) {
-			ret = bch2_subvolume_delete(trans, iter->pos.offset);
-			bch_err_msg(c, ret, "deleting subvolume %llu", iter->pos.offset);
-			ret = ret ?: -BCH_ERR_transaction_restart_nested;
-			goto err;
+			/*
+			 * Recreate - any contents that are still disconnected
+			 * will then get reattached under lost+found
+			 */
+			bch2_inode_init_early(c, &inode);
+			bch2_inode_init_late(&inode, bch2_current_time(c),
+					     0, 0, S_IFDIR|0700, 0, NULL);
+			inode.bi_inum			= le64_to_cpu(subvol.v->inode);
+			inode.bi_snapshot		= le32_to_cpu(subvol.v->snapshot);
+			inode.bi_subvol			= k.k->p.offset;
+			inode.bi_parent_subvol		= le32_to_cpu(subvol.v->fs_path_parent);
+			ret = __bch2_fsck_write_inode(trans, &inode);
+			if (ret)
+				goto err;
 		}
 	} else {
 		goto err;



  parent reply	other threads:[~2025-06-07 10:10 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-07 10:07 [PATCH 6.15 00/34] 6.15.2-rc1 review Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 01/34] tracing: Fix compilation warning on arm32 Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 02/34] Revert "x86/smp: Eliminate mwait_play_dead_cpuid_hint()" Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 03/34] ACPICA: Introduce ACPI_NONSTRING Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 04/34] ACPICA: Apply ACPI_NONSTRING Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 05/34] ACPICA: Apply ACPI_NONSTRING in more places Greg Kroah-Hartman
2025-06-12  5:31   ` Collin Funk
2025-06-12  7:23     ` Jiri Slaby
2025-06-12  8:15       ` Ahmed Salem
2025-06-13  2:17         ` Collin Funk
2025-06-13  2:12       ` Collin Funk
2025-06-07 10:07 ` [PATCH 6.15 06/34] pinctrl: armada-37xx: use correct OUTPUT_VAL register for GPIOs > 31 Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 07/34] pinctrl: armada-37xx: set GPIO output value before setting direction Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 08/34] clk: samsung: correct clock summary for hsi1 block Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 09/34] acpi-cpufreq: Fix nominal_freq units to KHz in get_max_boost_ratio() Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 10/34] Documentation: ACPI: Use all-string data node references Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 11/34] pinctrl: mediatek: eint: Fix invalid pointer dereference for v1 platforms Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 12/34] rtc: Make rtc_time64_to_tm() support dates before 1970 Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 13/34] rtc: Fix offset calculation for .start_secs < 0 Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 14/34] bcachefs: Kill un-reverted directory i_size code Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 15/34] bcachefs: Repair code for directory i_size Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 16/34] bcachefs: delete dead code from may_delete_deleted_inode() Greg Kroah-Hartman
2025-06-07 10:07 ` [PATCH 6.15 17/34] bcachefs: Run may_delete_deleted_inode() checks in bch2_inode_rm() Greg Kroah-Hartman
2025-06-07 10:07 ` Greg Kroah-Hartman [this message]
2025-06-07 10:08 ` [PATCH 6.15 19/34] usb: quirks: Add NO_LPM quirk for SanDisk Extreme 55AE Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 20/34] usb: storage: Ignore UAS driver for SanDisk 3.2 Gen2 storage device Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 21/34] USB: serial: pl2303: add new chip PL2303GC-Q20 and PL2303GT-2AB Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 22/34] usb: typec: ucsi: fix Clang -Wsign-conversion warning Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 23/34] Bluetooth: hci_qca: move the SoC type check to the right place Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 24/34] serial: jsm: fix NPE during jsm_uart_port_init Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 25/34] nvmem: rmem: select CONFIG_CRC32 Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 26/34] usb: usbtmc: Fix timeout value in get_stb Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 27/34] binder: fix use-after-free in binderfs_evict_inode() Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 28/34] binder: fix yet another UAF in binder_devices Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 29/34] thunderbolt: Do not double dequeue a configuration request Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 30/34] dt-bindings: pwm: adi,axi-pwmgen: Fix clocks Greg Kroah-Hartman
2025-06-10 14:12   ` David Lechner
2025-06-07 10:08 ` [PATCH 6.15 31/34] dt-bindings: remoteproc: qcom,sm8150-pas: Add missing SC8180X compatible Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 32/34] dt-bindings: usb: cypress,hx3: Add support for all variants Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 33/34] dt-bindings: phy: imx8mq-usb: fix fsl,phy-tx-vboost-level-microvolt property Greg Kroah-Hartman
2025-06-07 10:08 ` [PATCH 6.15 34/34] Revert "drm/amd/display: more liberal vmin/vmax update for freesync" Greg Kroah-Hartman
2025-06-07 11:42 ` [PATCH 6.15 00/34] 6.15.2-rc1 review Christian Heusel
2025-06-07 11:50 ` Ronald Warsow
2025-06-08  5:26   ` Luna Jernberg
2025-06-07 20:19 ` Florian Fainelli
2025-06-08  5:13 ` Takeshi Ogasawara
2025-06-08  5:38 ` Ron Economos
2025-06-08  6:53 ` Naresh Kamboju
2025-06-08 17:40 ` Peter Schneider
2025-06-08 22:31 ` Mark Brown
2025-06-09 12:48 ` Jon Hunter
2025-06-12 20:30 ` Miguel Ojeda

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=20250607100720.433460280@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=kent.overstreet@linux.dev \
    --cc=patches@lists.linux.dev \
    --cc=stable@vger.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.