ext4: don't remount read-only with errors=continue on reboot
authorJan Kara <jack@suse.cz>
Fri, 27 Nov 2020 11:33:54 +0000 (12:33 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 30 Dec 2020 10:54:13 +0000 (11:54 +0100)
commit b08070eca9e247f60ab39d79b2c25d274750441f upstream.

ext4_handle_error() with errors=continue mount option can accidentally
remount the filesystem read-only when the system is rebooting. Fix that.

Fixes: 1dc1097ff60e ("ext4: avoid panic during forced reboot")
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Andreas Dilger <adilger@dilger.ca>
Cc: stable@kernel.org
Link: https://lore.kernel.org/r/20201127113405.26867-2-jack@suse.cz
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fs/ext4/super.c

index 9447204..2b08b16 100644 (file)
@@ -666,19 +666,17 @@ static bool system_going_down(void)
 
 static void ext4_handle_error(struct super_block *sb)
 {
+       journal_t *journal = EXT4_SB(sb)->s_journal;
+
        if (test_opt(sb, WARN_ON_ERROR))
                WARN_ON_ONCE(1);
 
-       if (sb_rdonly(sb))
+       if (sb_rdonly(sb) || test_opt(sb, ERRORS_CONT))
                return;
 
-       if (!test_opt(sb, ERRORS_CONT)) {
-               journal_t *journal = EXT4_SB(sb)->s_journal;
-
-               ext4_set_mount_flag(sb, EXT4_MF_FS_ABORTED);
-               if (journal)
-                       jbd2_journal_abort(journal, -EIO);
-       }
+       ext4_set_mount_flag(sb, EXT4_MF_FS_ABORTED);
+       if (journal)
+               jbd2_journal_abort(journal, -EIO);
        /*
         * We force ERRORS_RO behavior when system is rebooting. Otherwise we
         * could panic during 'reboot -f' as the underlying device got already