From: Josef Bacik <josef@toxicpanda.com>
To: dsterba@suse.cz, linux-btrfs@vger.kernel.org, kernel-team@fb.com
Subject: Re: [PATCH 4/4] btrfs: do not create raid sysfs entries under any locks
Date: Tue, 8 Sep 2020 08:52:19 -0400 [thread overview]
Message-ID: <ed683b7e-bd67-1f27-58ac-a979d8c97f90@toxicpanda.com> (raw)
In-Reply-To: <20200908124003.GE28318@twin.jikos.cz>
On 9/8/20 8:40 AM, David Sterba wrote:
> On Tue, Sep 01, 2020 at 05:40:38PM -0400, Josef Bacik wrote:
>> }
>>
>> list_for_each_entry(space_info, &info->space_info, list) {
>> + int i;
>> +
>> + for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) {
>> + if (list_empty(&space_info->block_groups[i]))
>> + continue;
>> + cache = list_first_entry(&space_info->block_groups[i],
>> + struct btrfs_block_group,
>> + list);
>> + btrfs_sysfs_add_block_group_type(cache);
>> + }
>> +
>
> I had the previous version of the patch pass fstests, with no lockdep
> warnings and then realized it's not the v2 that depends on the 3rd patch
> removing RCU from this list traversal.
>
> btrfs_sysfs_add_block_group_type does not seem to be conflicting RCU in
> this loop so now I'm undecided if v1 is ok or if we really need v2, sice
> the patch 3 removing RCU seems suspicious.
>
It conflicts with RCU here because we could sleep, that's why I had to remove it.
Alternatively we could just loop a second time through the space_info's outside
of the RCU to do the btrfs_sysfs_add_block_group_type(), and I can drop the RCU
removal patch altogether. It's up to you, but we still need v2 because the
problem still exists without it. Thanks,
Josef
next prev parent reply other threads:[~2020-09-08 17:35 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-01 21:40 [PATCH 0/4][v2] Lockdep fixes Josef Bacik
2020-09-01 21:40 ` [PATCH 1/4] btrfs: fix lockdep splat in add_missing_dev Josef Bacik
2020-09-02 6:23 ` Anand Jain
2020-09-03 11:17 ` David Sterba
2020-09-01 21:40 ` [PATCH 2/4] btrfs: init sysfs for devices outside of the chunk_mutex Josef Bacik
2020-09-02 6:21 ` Anand Jain
2020-09-02 17:45 ` David Sterba
2020-09-03 11:41 ` Anand Jain
2020-09-03 11:42 ` Anand Jain
2020-09-03 11:18 ` David Sterba
2020-09-01 21:40 ` [PATCH 3/4] btrfs: kill the rcu protection for fs_info->space_info Josef Bacik
2020-09-02 8:04 ` Nikolay Borisov
2020-09-02 10:32 ` David Sterba
2020-09-01 21:40 ` [PATCH 4/4] btrfs: do not create raid sysfs entries under any locks Josef Bacik
2020-09-08 12:40 ` David Sterba
2020-09-08 12:52 ` Josef Bacik [this message]
2020-09-04 14:20 ` [PATCH 0/4][v2] Lockdep fixes David Sterba
2020-09-07 13:05 ` David Sterba
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=ed683b7e-bd67-1f27-58ac-a979d8c97f90@toxicpanda.com \
--to=josef@toxicpanda.com \
--cc=dsterba@suse.cz \
--cc=kernel-team@fb.com \
--cc=linux-btrfs@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).