Skip to content

L
linux
Switch branch/tag
  1. 05 Dec, 2022 1 commit
    • Bartosz Taudul's avatar
      udf: Increase UDF_MAX_READ_VERSION to 0x0260 · 7868f930
      Bartosz Taudul authored 
      Some discs containing the UDF file system are unable to be mounted,
failing with the following message:

  UDF-fs: error (device sr0): udf_fill_super: minUDFReadRev=260
    (max is 250)

The UDF 2.60 specification [0] states in the section Basic Restrictions
& Requirements (page 10):

  The Minimum UDF Read Revision value shall be at most #0250 for all
  media with a UDF 2.60 file system. This indicates that a UDF 2.50
  implementation can read all UDF 2.60 media. Media that do not have a
  Metadata Partition may use a value lower than #250.

The conclusion is that the discs failing to mount were burned with a
faulty software, which didn't follow the specification. This can be
worked around by increasing UDF_MAX_READ_VERSION to 0x260, to match the
Minimum Read Revision. No other changes are required, as reading UDF
2.60 is backward compatible with UDF 2.50.

[0] http://www.osta.org/specs/pdf/udf260.pdfSigned-off-by: default avatarBartosz Taudul <wolf@nereid.pl>
Signed-off-by: default avatarJan Kara <jack@suse.cz>
      7868f930
  3. 28 Nov, 2022 2 commits
    • Rong Tao's avatar
      fs/ext2: Fix code indentation · b41b98e1
      Rong Tao authored 
      ts=4 can cause misunderstanding in code reading. It is better to replace
8 spaces with one tab.
Signed-off-by: default avatarRong Tao <rongtao@cestc.cn>
Signed-off-by: default avatarJan Kara <jack@suse.cz>
      b41b98e1
    • Al Viro's avatar
      ext2: unbugger ext2_empty_dir() · 27e714c0
      Al Viro authored 
      In 27cfa258 "ext2: fix fs corruption when trying to remove
a non-empty directory with IO error" a funny thing has happened:

-               page = ext2_get_page(inode, i, dir_has_error, &page_addr);
+               page = ext2_get_page(inode, i, 0, &page_addr);

 -               if (IS_ERR(page)) {
 -                       dir_has_error = 1;
 -                       continue;
 -               }
 +               if (IS_ERR(page))
 +                       goto not_empty;

And at not_empty: we hit ext2_put_page(page, page_addr), which does
put_page(page).  Which, unless I'm very mistaken, should oops
immediately when given ERR_PTR(-E...) as page.

OK, shit happens, insufficiently tested patches included.  But when
commit in question describes the fault-injection test that exercised
that particular failure exit...

Ow.

CC: stable@vger.kernel.org
Fixes: 27cfa258 ("ext2: fix fs corruption when trying to remove a non-empty directory with IO error")
Signed-off-by: default avatarAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: default avatarJan Kara <jack@suse.cz>
      27e714c0
  5. 21 Nov, 2022 3 commits
  7. 11 Nov, 2022 1 commit
  9. 25 Oct, 2022 1 commit
    • Jan Kara's avatar
      maintainers: Add ISOFS entry · ab7720a2
      Jan Kara authored 
      We miss ISOFS entry in MAINTAINERS file. Add it and write me as the
maintainer there since ISOFS is pretty low effort these days. Less
random patches for Andrew to merge ;-).
Signed-off-by: default avatarJan Kara <jack@suse.cz>
      ab7720a2
  11. 24 Oct, 2022 2 commits
    • Shigeru Yoshida's avatar
      udf: Avoid double brelse() in udf_rename() · c791730f
      Shigeru Yoshida authored 
      syzbot reported a warning like below [1]:

VFS: brelse: Trying to free free buffer
WARNING: CPU: 2 PID: 7301 at fs/buffer.c:1145 __brelse+0x67/0xa0
...
Call Trace:
 <TASK>
 invalidate_bh_lru+0x99/0x150
 smp_call_function_many_cond+0xe2a/0x10c0
 ? generic_remap_file_range_prep+0x50/0x50
 ? __brelse+0xa0/0xa0
 ? __mutex_lock+0x21c/0x12d0
 ? smp_call_on_cpu+0x250/0x250
 ? rcu_read_lock_sched_held+0xb/0x60
 ? lock_release+0x587/0x810
 ? __brelse+0xa0/0xa0
 ? generic_remap_file_range_prep+0x50/0x50
 on_each_cpu_cond_mask+0x3c/0x80
 blkdev_flush_mapping+0x13a/0x2f0
 blkdev_put_whole+0xd3/0xf0
 blkdev_put+0x222/0x760
 deactivate_locked_super+0x96/0x160
 deactivate_super+0xda/0x100
 cleanup_mnt+0x222/0x3d0
 task_work_run+0x149/0x240
 ? task_work_cancel+0x30/0x30
 do_exit+0xb29/0x2a40
 ? reacquire_held_locks+0x4a0/0x4a0
 ? do_raw_spin_lock+0x12a/0x2b0
 ? mm_update_next_owner+0x7c0/0x7c0
 ? rwlock_bug.part.0+0x90/0x90
 ? zap_other_threads+0x234/0x2d0
 do_group_exit+0xd0/0x2a0
 __x64_sys_exit_group+0x3a/0x50
 do_syscall_64+0x34/0xb0
 entry_SYSCALL_64_after_hwframe+0x63/0xcd

The cause of the issue is that brelse() is called on both ofibh.sbh
and ofibh.ebh by udf_find_entry() when it returns NULL.  However,
brelse() is called by udf_rename(), too.  So, b_count on buffer_head
becomes unbalanced.

This patch fixes the issue by not calling brelse() by udf_rename()
when udf_find_entry() returns NULL.

Link: https://syzkaller.appspot.com/bug?id=8297f45698159c6bca8a1f87dc983667c1a1c851 [1]
Reported-by: syzbot+7902cd7684bc35306224@syzkaller.appspotmail.com
Signed-off-by: default avatarShigeru Yoshida <syoshida@redhat.com>
Signed-off-by: default avatarJan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20221023095741.271430-1-syoshida@redhat.com
      c791730f
    • Li zeming's avatar
      fs: udf: Optimize udf_free_in_core_inode and udf_find_fileset function · 0dafb7e6
      Li zeming authored 
      These two functions perform the following optimizations.
1. Delete the type cast of foo pointer. Void * does not need to convert
the type.
2. Delete the initialization assignment of bh variable, which is
assigned first.
Signed-off-by: default avatarLi zeming <zeming@nfschina.com>
Signed-off-by: default avatarJan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20221012104235.3331-1-zeming@nfschina.com
      0dafb7e6
  13. 23 Oct, 2022 9 commits
  15. 22 Oct, 2022 21 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7