Linux-Fsdevel Archive on lore.kernel.org
 help / color / Atom feed
From: Jan Kara <jack@suse.cz>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Al Viro <viro@ZenIV.linux.org.uk>,
	<linux-fsdevel@vger.kernel.org>, <linux-mm@kvack.org>,
	Jan Kara <jack@suse.cz>
Subject: [PATCH] vfs: Avoid softlockups in drop_pagecache_sb()
Date: Mon, 14 Jan 2019 09:53:43 +0100
Message-ID: <20190114085343.15011-1-jack@suse.cz> (raw)

When superblock has lots of inodes without any pagecache (like is the
case for /proc), drop_pagecache_sb() will iterate through all of them
without dropping sb->s_inode_list_lock which can lead to softlockups
(one of our customers hit this).

Fix the problem by going to the slow path and doing cond_resched() in
case the process needs rescheduling.

Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/drop_caches.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

Andrew, can you please merge this patch? Thanks!

diff --git a/fs/drop_caches.c b/fs/drop_caches.c
index 82377017130f..d31b6c72b476 100644
--- a/fs/drop_caches.c
+++ b/fs/drop_caches.c
@@ -21,8 +21,13 @@ static void drop_pagecache_sb(struct super_block *sb, void *unused)
 	spin_lock(&sb->s_inode_list_lock);
 	list_for_each_entry(inode, &sb->s_inodes, i_sb_list) {
 		spin_lock(&inode->i_lock);
+		/*
+		 * We must skip inodes in unusual state. We may also skip
+		 * inodes without pages but we deliberately won't in case
+		 * we need to reschedule to avoid softlockups.
+		 */
 		if ((inode->i_state & (I_FREEING|I_WILL_FREE|I_NEW)) ||
-		    (inode->i_mapping->nrpages == 0)) {
+		    (inode->i_mapping->nrpages == 0 && !need_resched())) {
 			spin_unlock(&inode->i_lock);
 			continue;
 		}
@@ -30,6 +35,7 @@ static void drop_pagecache_sb(struct super_block *sb, void *unused)
 		spin_unlock(&inode->i_lock);
 		spin_unlock(&sb->s_inode_list_lock);
 
+		cond_resched();
 		invalidate_mapping_pages(inode->i_mapping, 0, -1);
 		iput(toput_inode);
 		toput_inode = inode;
-- 
2.16.4


             reply index

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-14  8:53 Jan Kara [this message]
2019-01-30  0:56 ` Andrew Morton
2019-01-30 10:07   ` Jan Kara
  -- strict thread matches above, loose matches on Subject: below --
2018-12-06 16:11 Jan Kara
2019-01-08 15:16 ` Michal Hocko

Reply instructions:

You may reply publically 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=20190114085343.15011-1-jack@suse.cz \
    --to=jack@suse.cz \
    --cc=akpm@linux-foundation.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=viro@ZenIV.linux.org.uk \
    /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

Linux-Fsdevel Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-fsdevel/0 linux-fsdevel/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-fsdevel linux-fsdevel/ https://lore.kernel.org/linux-fsdevel \
		linux-fsdevel@vger.kernel.org
	public-inbox-index linux-fsdevel

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-fsdevel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git