Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Thu, 23 May 2019 13:43:49 -0400
From:   Johannes Weiner <hannes@cmpxchg.org>
To:     Matthew Wilcox <willy@infradead.org>
Cc:     Andrew Morton <akpm@linux-foundation.org>, linux-mm@kvack.org,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        kernel-team@fb.com
Subject: xarray breaks thrashing detection and cgroup isolation
Message-ID: <20190523174349.GA10939@cmpxchg.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.11.4 (2019-03-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hello,

I noticed that recent upstream kernels don't account the xarray nodes
of the page cache to the allocating cgroup, like we used to do for the
radix tree nodes.

This results in broken isolation for cgrouped apps, allowing them to
escape their containment and harm other cgroups and the system with an
excessive build-up of nonresident information.

It also breaks thrashing/refault detection because the page cache
lives in a different domain than the xarray nodes, and so the shadow
shrinker can reclaim nonresident information way too early when there
isn't much cache in the root cgroup.

This appears to be the culprit:

commit a28334862993b5c6a8766f6963ee69048403817c
Author: Matthew Wilcox <willy@infradead.org>
Date:   Tue Dec 5 19:04:20 2017 -0500

    page cache: Finish XArray conversion
    
    With no more radix tree API users left, we can drop the GFP flags
    and use xa_init() instead of INIT_RADIX_TREE().
    
    Signed-off-by: Matthew Wilcox <willy@infradead.org>

diff --git a/fs/inode.c b/fs/inode.c
index 42f6d25f32a5..9b808986d440 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -349,7 +349,7 @@ EXPORT_SYMBOL(inc_nlink);
 
 static void __address_space_init_once(struct address_space *mapping)
 {
-       INIT_RADIX_TREE(&mapping->i_pages, GFP_ATOMIC | __GFP_ACCOUNT);
+       xa_init_flags(&mapping->i_pages, XA_FLAGS_LOCK_IRQ);
        init_rwsem(&mapping->i_mmap_rwsem);
        INIT_LIST_HEAD(&mapping->private_list);
        spin_lock_init(&mapping->private_lock);

It fairly blatantly drops __GFP_ACCOUNT.

I'm not quite sure how to fix this, since the xarray code doesn't seem
to have per-tree gfp flags anymore like the radix tree did. We cannot
add SLAB_ACCOUNT to the radix_tree_node_cachep slab cache. And the
xarray api doesn't seem to really support gfp flags, either (xas_nomem
does, but the optimistic internal allocations have fixed gfp flags).