linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Baolin Wang <baolin.wang@linux.alibaba.com>
To: Kemeng Shi <shikemeng@huaweicloud.com>,
	hughd@google.com, akpm@linux-foundation.org
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 3/5] mm/shmem: Fix potential dead loop in shmem_unuse()
Date: Wed, 14 May 2025 17:24:12 +0800	[thread overview]
Message-ID: <b7b6d5d0-2c3f-43ba-b2a9-3c9669f3f96d@linux.alibaba.com> (raw)
In-Reply-To: <20250514165047.946884-4-shikemeng@huaweicloud.com>



On 2025/5/15 00:50, Kemeng Shi wrote:
> If multi shmem_unuse() for different swap type is called concurrently,
> a dead loop could occur as following:
> shmem_unuse(typeA)               shmem_unuse(typeB)
>   mutex_lock(&shmem_swaplist_mutex)
>   list_for_each_entry_safe(info, next, ...)
>    ...
>    mutex_unlock(&shmem_swaplist_mutex)
>    /* info->swapped may drop to 0 */
>    shmem_unuse_inode(&info->vfs_inode, type)
> 
>                                    mutex_lock(&shmem_swaplist_mutex)
>                                    list_for_each_entry(info, next, ...)
>                                     if (!info->swapped)
>                                      list_del_init(&info->swaplist)
> 
>                                    ...
>                                    mutex_unlock(&shmem_swaplist_mutex)
> 
>    mutex_lock(&shmem_swaplist_mutex)
>    /* iterate with offlist entry and encounter a dead loop */
>    next = list_next_entry(info, swaplist);
>    ...
> 
> Restart the iteration if the inode is already off shmem_swaplist list
> to fix the issue.
> 
> Fixes: b56a2d8af9147 ("mm: rid swapoff of quadratic complexity")
> Signed-off-by: Kemeng Shi <shikemeng@huaweicloud.com>
> ---
>   mm/shmem.c | 3 +++
>   1 file changed, 3 insertions(+)
> 
> diff --git a/mm/shmem.c b/mm/shmem.c
> index 495e661eb8bb..0fed94c2bc09 100644
> --- a/mm/shmem.c
> +++ b/mm/shmem.c
> @@ -1505,6 +1505,7 @@ int shmem_unuse(unsigned int type)
>   		return 0;
>   
>   	mutex_lock(&shmem_swaplist_mutex);
> +start_over:
>   	list_for_each_entry_safe(info, next, &shmem_swaplist, swaplist) {
>   		if (!info->swapped) {
>   			list_del_init(&info->swaplist);
> @@ -1530,6 +1531,8 @@ int shmem_unuse(unsigned int type)

		next = list_next_entry(info, swaplist);
		if (!info->swapped)
			list_del_init(&info->swaplist);
		if (atomic_dec_and_test(&info->stop_eviction))
			wake_up_var(&info->stop_eviction);

We may still hit the list warning when calling list_del_init() for the 
off-list info->swaplist? So I hope we can add a check for the possible 
off-list:

diff --git a/mm/shmem.c b/mm/shmem.c
index 99327c30507c..f5ae5e2d6fb4 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1523,9 +1523,11 @@ int shmem_unuse(unsigned int type)
                 cond_resched();

                 mutex_lock(&shmem_swaplist_mutex);
-               next = list_next_entry(info, swaplist);
-               if (!info->swapped)
-                       list_del_init(&info->swaplist);
+               if (!list_empty(&info->swaplist)) {
+                       next = list_next_entry(info, swaplist);
+                       if (!info->swapped)
+                               list_del_init(&info->swaplist);
+               }
                 if (atomic_dec_and_test(&info->stop_eviction))
                         wake_up_var(&info->stop_eviction);
                 if (error)

>   			wake_up_var(&info->stop_eviction);
>   		if (error)
>   			break;
> +		if (list_empty(&info->swaplist))
> +			goto start_over;
>   	}
>   	mutex_unlock(&shmem_swaplist_mutex);
>   


  reply	other threads:[~2025-05-14  9:24 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-14 16:50 [PATCH 0/5] Some random fixes and cleanup to shmem Kemeng Shi
2025-05-14 16:50 ` [PATCH 1/5] mm: shmem: avoid unpaired folio_unlock() in shmem_swapin_folio() Kemeng Shi
2025-05-14  8:51   ` Baolin Wang
2025-05-14 16:50 ` [PATCH 2/5] mm: shmem: add missing shmem_unacct_size() in __shmem_file_setup() Kemeng Shi
2025-05-14  8:53   ` Baolin Wang
2025-05-14 16:50 ` [PATCH 3/5] mm/shmem: Fix potential dead loop in shmem_unuse() Kemeng Shi
2025-05-14  9:24   ` Baolin Wang [this message]
2025-05-15  1:05     ` Kemeng Shi
2025-05-15  3:59       ` Baolin Wang
2025-05-14 16:50 ` [PATCH 4/5] mm: shmem: keep inode in swaplist when failed to allocate swap entry in shmem_writepage() Kemeng Shi
2025-05-14  9:31   ` Baolin Wang
2025-05-15  1:09     ` Kemeng Shi
2025-05-14 16:50 ` [PATCH 5/5] mm/shmem: remove unneeded xa_is_value() check in shmem_unuse_swap_entries() Kemeng Shi
2025-05-14  9:31   ` Baolin Wang

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=b7b6d5d0-2c3f-43ba-b2a9-3c9669f3f96d@linux.alibaba.com \
    --to=baolin.wang@linux.alibaba.com \
    --cc=akpm@linux-foundation.org \
    --cc=hughd@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=shikemeng@huaweicloud.com \
    /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).