[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250715042959.2087-1-hdanton@sina.com>
Date: Tue, 15 Jul 2025 12:29:58 +0800
From: Hillf Danton <hdanton@...a.com>
To: syzbot <syzbot+fde6bd779f78e6e0992e@...kaller.appspotmail.com>
Cc: linux-kernel@...r.kernel.org,
syzkaller-bugs@...glegroups.com
Subject: Re: [syzbot] [bluetooth?] [bcachefs?] KASAN: slab-use-after-free Read in hci_uart_write_work
> Date: Mon, 14 Jul 2025 10:09:31 -0700 [thread overview]
> Hello,
>
> syzbot found the following issue on:
>
> HEAD commit: 3f31a806a62e Merge tag 'mm-hotfixes-stable-2025-07-11-16-1..
> git tree: upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=174b07d4580000
> kernel config: https://syzkaller.appspot.com/x/.config?x=b309c907eaab29da
> dashboard link: https://syzkaller.appspot.com/bug?extid=fde6bd779f78e6e0992e
> syz repro: https://syzkaller.appspot.com/x/repro.syz?x=127ece8c580000
#syz test
--- x/drivers/bluetooth/hci_ldisc.c
+++ y/drivers/bluetooth/hci_ldisc.c
@@ -100,13 +100,9 @@ static inline struct sk_buff *hci_uart_d
struct sk_buff *skb = hu->tx_skb;
if (!skb) {
- percpu_down_read(&hu->proto_lock);
-
if (test_bit(HCI_UART_PROTO_READY, &hu->flags) ||
test_bit(HCI_UART_PROTO_INIT, &hu->flags))
skb = hu->proto->dequeue(hu);
-
- percpu_up_read(&hu->proto_lock);
} else {
hu->tx_skb = NULL;
}
@@ -154,6 +150,11 @@ static void hci_uart_write_work(struct w
/* REVISIT: should we cope with bad skbs or ->write() returning
* and error value ?
*/
+ percpu_down_read(&hu->proto_lock);
+ if (!test_bit(HCI_UART_PROTO_READY, &hu->flags)) {
+ percpu_up_read(&hu->proto_lock);
+ return;
+ }
restart:
clear_bit(HCI_UART_TX_WAKEUP, &hu->tx_state);
@@ -179,6 +180,7 @@ restart:
if (test_bit(HCI_UART_TX_WAKEUP, &hu->tx_state))
goto restart;
+ percpu_up_read(&hu->proto_lock);
wake_up_bit(&hu->tx_state, HCI_UART_SENDING);
}
--
Powered by blists - more mailing lists