All of lore.kernel.org
 help / color / mirror / Atom feed
From: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
To: syzbot <syzbot+ece2915262061d6e0ac1@syzkaller.appspotmail.com>,
	syzkaller-bugs@googlegroups.com,
	Mel Gorman <mgorman@techsingularity.net>,
	"Huang, Ying" <ying.huang@intel.com>,
	Vlastimil Babka <vbabka@suse.cz>,
	Andrew Morton <akpm@linux-foundation.org>,
	Alexander Potapenko <glider@google.com>,
	Andrey Konovalov <andreyknvl@gmail.com>,
	Dmitry Vyukov <dvyukov@google.com>,
	Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: kasan-dev <kasan-dev@googlegroups.com>, linux-mm <linux-mm@kvack.org>
Subject: [PATCH] lib/stackdepot: stackdepot: don't use __GFP_KSWAPD_RECLAIM from __stack_depot_save() if atomic context
Date: Sat, 20 May 2023 20:33:04 +0900	[thread overview]
Message-ID: <ca8e3803-4757-358e-dcf2-4824213a9d2c@I-love.SAKURA.ne.jp> (raw)
In-Reply-To: <ecba318b-7452-92d0-4a2f-2f6c9255f771@I-love.SAKURA.ne.jp>

syzbot is reporting lockdep warning in __stack_depot_save(), for
wakeup_kswapd() from wake_all_kswapds() from __alloc_pages_slowpath()
calls wakeup_kcompactd() when __GFP_KSWAPD_RECLAIM is set and
__GFP_DIRECT_RECLAIM is not set (i.e. GFP_ATOMIC).

Since __stack_depot_save() might be called with arbitrary locks held,
__stack_depot_save() should not wake kswapd which in turn wakes kcompactd.

Reported-by: syzbot <syzbot+ece2915262061d6e0ac1@syzkaller.appspotmail.com>
Closes: https://syzkaller.appspot.com/bug?extid=ece2915262061d6e0ac1
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Fixes: cd11016e5f52 ("mm, kasan: stackdepot implementation. Enable stackdepot for SLAB")
---
 lib/stackdepot.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/lib/stackdepot.c b/lib/stackdepot.c
index 2f5aa851834e..5c331a80b87a 100644
--- a/lib/stackdepot.c
+++ b/lib/stackdepot.c
@@ -405,7 +405,10 @@ depot_stack_handle_t __stack_depot_save(unsigned long *entries,
 		 * contexts and I/O.
 		 */
 		alloc_flags &= ~GFP_ZONEMASK;
-		alloc_flags &= (GFP_ATOMIC | GFP_KERNEL);
+		if (!(alloc_flags & __GFP_DIRECT_RECLAIM))
+			alloc_flags &= __GFP_HIGH;
+		else
+			alloc_flags &= GFP_KERNEL;
 		alloc_flags |= __GFP_NOWARN;
 		page = alloc_pages(alloc_flags, DEPOT_POOL_ORDER);
 		if (page)
-- 
2.18.4



  reply	other threads:[~2023-05-20 11:33 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-20  8:26 [syzbot] [kernel?] possible deadlock in scheduler_tick (2) syzbot
2023-05-20 11:02 ` Tetsuo Handa
2023-05-20 11:33   ` Tetsuo Handa [this message]
2023-05-20 13:14     ` [PATCH] lib/stackdepot: stackdepot: don't use __GFP_KSWAPD_RECLAIM from __stack_depot_save() if atomic context Tetsuo Handa
2023-05-20 22:44       ` Tetsuo Handa
2023-05-22  2:13         ` Huang, Ying
2023-05-22  2:47           ` Tetsuo Handa
2023-05-22  3:07             ` Huang, Ying
2023-05-22 11:33               ` Tetsuo Handa
2023-05-23  0:07                 ` Huang, Ying
2023-05-23  0:45                   ` Tetsuo Handa
2023-05-23  1:10                     ` Huang, Ying
2023-05-24 12:09             ` Michal Hocko
2023-05-27 15:25     ` [PATCH] kasan,kmsan: remove __GFP_KSWAPD_RECLAIM usage from kasan/kmsan Tetsuo Handa
2023-05-29  1:07       ` Huang, Ying
2023-05-31 13:31         ` Alexander Potapenko
2023-06-09 22:31           ` Andrew Morton
     [not found]             ` <19d6c965-a9cf-16a5-6537-a02823d67c0a@I-love.SAKURA.ne.jp>
2023-06-12  1:30               ` [PATCH v3] lib/stackdepot: fix gfp flags manipulation in __stack_depot_save() Huang, Ying
2023-06-21 12:56               ` Alexander Potapenko
2023-06-21 14:07                 ` Tetsuo Handa
2023-06-21 14:42                   ` Alexander Potapenko
2023-06-21 14:54                     ` Tetsuo Handa
2023-06-21 15:37             ` [PATCH] kasan,kmsan: remove __GFP_KSWAPD_RECLAIM usage from kasan/kmsan Alexander Potapenko
2023-05-27 21:01 ` [syzbot] [ntfs3?] possible deadlock in scheduler_tick (2) syzbot

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=ca8e3803-4757-358e-dcf2-4824213a9d2c@I-love.SAKURA.ne.jp \
    --to=penguin-kernel@i-love.sakura.ne.jp \
    --cc=akpm@linux-foundation.org \
    --cc=andreyknvl@gmail.com \
    --cc=dvyukov@google.com \
    --cc=glider@google.com \
    --cc=kasan-dev@googlegroups.com \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@techsingularity.net \
    --cc=ryabinin.a.a@gmail.com \
    --cc=syzbot+ece2915262061d6e0ac1@syzkaller.appspotmail.com \
    --cc=syzkaller-bugs@googlegroups.com \
    --cc=vbabka@suse.cz \
    --cc=ying.huang@intel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.