From: Zhang Yi <yi.zhang@huawei.com>
To: Theodore Ts'o <tytso@mit.edu>
Cc: Jan Kara <jack@suse.cz>, <linuxppc-dev@lists.ozlabs.org>,
Guoqing Jiang <guoqing.jiang@linux.dev>,
Sachin Sant <sachinp@linux.vnet.ibm.com>,
Ext4 Developers List <linux-ext4@vger.kernel.org>,
"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>
Subject: Re: [powerpc][5.13.0-next-20210701] Kernel crash while running ltp(chdir01) tests
Date: Sat, 3 Jul 2021 11:37:07 +0800 [thread overview]
Message-ID: <9b81eb4e-9adb-991f-31be-f5ef0092c4b3@huawei.com> (raw)
In-Reply-To: <YN+PIKV010a+j88S@mit.edu>
On 2021/7/3 6:11, Theodore Ts'o wrote:
> On Fri, Jul 02, 2021 at 12:11:54PM -0400, Theodore Ts'o wrote:
>> So it probably makes more sense to keep jbd2_journal_unregister_shrinker()
>> in jbd2_destroy_journal(), since arguably the fact that we are using a
>> shrinker is an internal implementation detail, and the users of jbd2
>> ideally shouldn't need to be expected to know they have unregister
>> jbd2's shirnkers.
>>
>> Similarly, perhaps we should be moving jbd2_journal_register_shirnker()
>> into jbd2_journal_init_common(). We can un-export the register and
>> unshrink register functions, and declare them as static functions internal
>> to fs/jbd2/journal.c.
>>
>> What do you think?
>
> Like this...
>
> commit 8f9e16badb8fda3391e03146a47c93e76680efaf
> Author: Theodore Ts'o <tytso@mit.edu>
> Date: Fri Jul 2 18:05:03 2021 -0400
>
> ext4: fix doubled call to jbd2_journal_unregister_shrinker()
>
> On Power and ARM platforms this was causing kernel crash when
> unmounting the file system, due to a percpu_counter getting destroyed
> twice.
>
> Fix this by cleaning how the jbd2 shrinker is initialized and
> uninitiazed by making it solely the responsibility of
> fs/jbd2/journal.c.
>
> Fixes: 4ba3fcdde7e3 ("jbd2,ext4: add a shrinker to release checkpointed buffers")
> Reported-by: Sachin Sant <sachinp@linux.vnet.ibm.com>
> Signed-off-by: Theodore Ts'o <tytso@mit.edu>
>
> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> index b8ff0399e171..dfa09a277b56 100644
> --- a/fs/ext4/super.c
> +++ b/fs/ext4/super.c
> @@ -1184,7 +1184,6 @@ static void ext4_put_super(struct super_block *sb)
> ext4_unregister_sysfs(sb);
>
> if (sbi->s_journal) {
> - jbd2_journal_unregister_shrinker(sbi->s_journal);
> aborted = is_journal_aborted(sbi->s_journal);
> err = jbd2_journal_destroy(sbi->s_journal);
> sbi->s_journal = NULL;
> @@ -5176,7 +5175,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
> sbi->s_ea_block_cache = NULL;
>
> if (sbi->s_journal) {
> - jbd2_journal_unregister_shrinker(sbi->s_journal);
> jbd2_journal_destroy(sbi->s_journal);
> sbi->s_journal = NULL;
> }
> @@ -5502,12 +5500,6 @@ static int ext4_load_journal(struct super_block *sb,
> ext4_commit_super(sb);
> }
>
> - err = jbd2_journal_register_shrinker(journal);
> - if (err) {
> - EXT4_SB(sb)->s_journal = NULL;
> - goto err_out;
> - }
> -
> return 0;
>
> err_out:
> diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
> index 152880c298ca..2595703aca51 100644
> --- a/fs/jbd2/journal.c
> +++ b/fs/jbd2/journal.c
> @@ -99,6 +99,8 @@ EXPORT_SYMBOL(jbd2_journal_begin_ordered_truncate);
> EXPORT_SYMBOL(jbd2_inode_cache);
>
> static int jbd2_journal_create_slab(size_t slab_size);
> +static int jbd2_journal_register_shrinker(journal_t *journal);
> +static void jbd2_journal_unregister_shrinker(journal_t *journal);
>
> #ifdef CONFIG_JBD2_DEBUG
> void __jbd2_debug(int level, const char *file, const char *func,
> @@ -2043,7 +2045,8 @@ int jbd2_journal_load(journal_t *journal)
> goto recovery_error;
>
> journal->j_flags |= JBD2_LOADED;
> - return 0;
> +
> + return jbd2_journal_register_shrinker(journal);
>
I check the ocfs2 code, if we register shrinker here, __ocfs2_recovery_thread()->
ocfs2_recover_node() seems will register and unregister a lot of unnecessary
shrinkers. It depends on the lifetime of the shrinker and the journal, because of
the jbd2_journal_destroy() destroy everything, it not a simple undo of
jbd2_load_journal(), so it's not easy to add shrinker properly. But it doesn't
seems like a real problem, just curious.
Thanks,
Yi.
> recovery_error:
> printk(KERN_WARNING "JBD2: recovery failed\n");
> @@ -2099,7 +2102,7 @@ static unsigned long jbd2_journal_shrink_count(struct shrinker *shrink,
> * Init a percpu counter to record the checkpointed buffers on the checkpoint
> * list and register a shrinker to release their journal_head.
> */
> -int jbd2_journal_register_shrinker(journal_t *journal)
> +static int jbd2_journal_register_shrinker(journal_t *journal)
> {
> int err;
>
> @@ -2122,7 +2125,6 @@ int jbd2_journal_register_shrinker(journal_t *journal)
>
> return 0;
> }
> -EXPORT_SYMBOL(jbd2_journal_register_shrinker);
>
> /**
> * jbd2_journal_unregister_shrinker()
> @@ -2130,12 +2132,13 @@ EXPORT_SYMBOL(jbd2_journal_register_shrinker);
> *
> * Unregister the checkpointed buffer shrinker and destroy the percpu counter.
> */
> -void jbd2_journal_unregister_shrinker(journal_t *journal)
> +static void jbd2_journal_unregister_shrinker(journal_t *journal)
> {
> - percpu_counter_destroy(&journal->j_jh_shrink_count);
> - unregister_shrinker(&journal->j_shrinker);
> + if (journal->j_shrinker.flags & SHRINKER_REGISTERED) {
> + percpu_counter_destroy(&journal->j_jh_shrink_count);
> + unregister_shrinker(&journal->j_shrinker);
> + }
> }
> -EXPORT_SYMBOL(jbd2_journal_unregister_shrinker);
>
> /**
> * jbd2_journal_destroy() - Release a journal_t structure.
> diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
> index 6cc035321562..632afbe4b18f 100644
> --- a/include/linux/jbd2.h
> +++ b/include/linux/jbd2.h
> @@ -1556,8 +1556,6 @@ extern int jbd2_journal_set_features
> (journal_t *, unsigned long, unsigned long, unsigned long);
> extern void jbd2_journal_clear_features
> (journal_t *, unsigned long, unsigned long, unsigned long);
> -extern int jbd2_journal_register_shrinker(journal_t *journal);
> -extern void jbd2_journal_unregister_shrinker(journal_t *journal);
> extern int jbd2_journal_load (journal_t *journal);
> extern int jbd2_journal_destroy (journal_t *);
> extern int jbd2_journal_recover (journal_t *journal);
> .
>
next prev parent reply other threads:[~2021-07-03 3:37 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-02 8:51 [powerpc][5.13.0-next-20210701] Kernel crash while running ltp(chdir01) tests Sachin Sant
2021-07-02 9:38 ` Guoqing Jiang
2021-07-02 13:13 ` Theodore Ts'o
2021-07-02 13:23 ` Zhang Yi
2021-07-02 13:52 ` Zhang Yi
2021-07-02 16:11 ` Theodore Ts'o
2021-07-02 22:11 ` Theodore Ts'o
2021-07-03 3:37 ` Zhang Yi [this message]
2021-07-03 3:52 ` Theodore Ts'o
2021-07-03 3:05 ` Zhang Yi
2021-07-03 3:35 ` Theodore Ts'o
2021-07-03 4:55 ` Zhang Yi
2021-07-04 14:04 ` Theodore Ts'o
2021-07-05 2:17 ` Zhang Yi
2021-07-05 14:50 ` [PATCH -v2] ext4: inline jbd2_journal_[un]register_shrinker() Theodore Ts'o
2021-07-05 18:29 ` Jon Hunter
2021-07-06 1:38 ` Zhang Yi
2021-07-05 9:58 ` [powerpc][5.13.0-next-20210701] Kernel crash while running ltp(chdir01) tests Jan Kara
2021-07-05 11:27 ` Sachin Sant
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=9b81eb4e-9adb-991f-31be-f5ef0092c4b3@huawei.com \
--to=yi.zhang@huawei.com \
--cc=guoqing.jiang@linux.dev \
--cc=jack@suse.cz \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=sachinp@linux.vnet.ibm.com \
--cc=tytso@mit.edu \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox