From: "Maurizio Lombardi" <mlombard@bsdbackstore.eu>
To: "Maurizio Lombardi" <mlombard@bsdbackstore.eu>,
"zhang.guanghui@cestc.cn" <zhang.guanghui@cestc.cn>,
"chunguang.xu" <chunguang.xu@shopee.com>
Cc: "mgurtovoy" <mgurtovoy@nvidia.com>, "sagi" <sagi@grimberg.me>,
"kbusch" <kbusch@kernel.org>, "sashal" <sashal@kernel.org>,
"linux-kernel" <linux-kernel@vger.kernel.org>,
"linux-nvme" <linux-nvme@lists.infradead.org>,
"linux-block" <linux-block@vger.kernel.org>
Subject: Re: nvme-tcp: fix a possible UAF when failing to send request
Date: Wed, 12 Feb 2025 12:14:20 +0100 [thread overview]
Message-ID: <D7QF99BR5YYA.2CXBBA5RYACTY@bsdbackstore.eu> (raw)
In-Reply-To: <D7QEA7XPBML4.1R0M64OD4BH43@bsdbackstore.eu>
On Wed Feb 12, 2025 at 11:28 AM CET, Maurizio Lombardi wrote:
> On Wed Feb 12, 2025 at 10:47 AM CET, zhang.guanghui@cestc.cn wrote:
>> Hi, Thanks.
>> I will test this patch, but I am worried whether it will affect the performance.
>> Should we also consider null pointer protection?
>
> Yes, it will likely affect the performance, just check if it works.
>
> Probably it could be optimized by just protecting
> nvme_tcp_fail_request(), which AFAICT is the only function in the
> nvme_tcp_try_send() code that calls nvme_complete_rq().
Something like that, maybe, not tested:
diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c
index 841238f38fdd..488edec35a65 100644
--- a/drivers/nvme/host/tcp.c
+++ b/drivers/nvme/host/tcp.c
@@ -146,6 +146,7 @@ struct nvme_tcp_queue {
struct mutex queue_lock;
struct mutex send_mutex;
+ struct mutex poll_mutex;
struct llist_head req_list;
struct list_head send_list;
@@ -1259,7 +1260,9 @@ static int nvme_tcp_try_send(struct nvme_tcp_queue *queue)
} else if (ret < 0) {
dev_err(queue->ctrl->ctrl.device,
"failed to send request %d\n", ret);
+ mutex_lock(&queue->poll_mutex);
nvme_tcp_fail_request(queue->request);
+ mutex_unlock(&queue->poll_mutex);
nvme_tcp_done_send_req(queue);
}
out:
@@ -1397,6 +1400,7 @@ static void nvme_tcp_free_queue(struct nvme_ctrl *nctrl, int qid)
kfree(queue->pdu);
mutex_destroy(&queue->send_mutex);
mutex_destroy(&queue->queue_lock);
+ mutex_destroy(&queue->poll_mutex);
}
static int nvme_tcp_init_connection(struct nvme_tcp_queue *queue)
@@ -1710,6 +1714,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid,
init_llist_head(&queue->req_list);
INIT_LIST_HEAD(&queue->send_list);
mutex_init(&queue->send_mutex);
+ mutex_init(&queue->poll_mutex);
INIT_WORK(&queue->io_work, nvme_tcp_io_work);
if (qid > 0)
@@ -2660,7 +2665,9 @@ static int nvme_tcp_poll(struct blk_mq_hw_ctx *hctx, struct io_comp_batch *iob)
set_bit(NVME_TCP_Q_POLLING, &queue->flags);
if (sk_can_busy_loop(sk) && skb_queue_empty_lockless(&sk->sk_receive_queue))
sk_busy_loop(sk, true);
+ mutex_lock(&queue->poll_mutex);
nvme_tcp_try_recv(queue);
+ mutex_unlock(&queue->poll_mutex);
clear_bit(NVME_TCP_Q_POLLING, &queue->flags);
return queue->nr_cqe;
}
next prev parent reply other threads:[~2025-02-12 11:14 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-10 7:41 nvme-tcp: fix a possible UAF when failing to send request zhang.guanghui
2025-02-10 10:01 ` Maurizio Lombardi
2025-02-10 10:24 ` Max Gurtovoy
2025-02-10 11:16 ` zhang.guanghui
2025-02-10 11:34 ` Max Gurtovoy
2025-02-10 16:40 ` Maurizio Lombardi
[not found] ` <CAAO4dAWdsMjYMp9jdWXd_48aG0mTtVpRONqjJJ1scnc773tHzg@mail.gmail.com>
2025-02-11 8:04 ` zhang.guanghui
2025-02-12 8:11 ` Maurizio Lombardi
2025-02-12 8:23 ` Maurizio Lombardi
2025-02-12 8:52 ` Maurizio Lombardi
2025-02-12 9:47 ` zhang.guanghui
2025-02-12 10:28 ` Maurizio Lombardi
2025-02-12 11:14 ` Maurizio Lombardi [this message]
2025-02-12 11:47 ` Maurizio Lombardi
2025-02-12 15:33 ` Maurizio Lombardi
2025-02-12 16:07 ` Maurizio Lombardi
2025-02-13 2:04 ` zhang.guanghui
2025-02-17 7:46 ` Sagi Grimberg
2025-02-20 8:20 ` zhang.guanghui
2025-03-07 10:10 ` Re: nvme-tcp: fix a possible UAF when failing to send request【请注意,邮件由sagigrim@gmail.com代发】 zhang.guanghui
2025-03-11 10:52 ` Maurizio Lombardi
2025-03-13 1:48 ` zhang.guanghui
2025-03-13 7:51 ` Hannes Reinecke
2025-03-13 8:18 ` zhang.guanghui
[not found] ` <2025031316313196627826@cestc.cn>
2025-03-13 9:01 ` Maurizio Lombardi
2025-03-13 8:38 ` zhang.guanghui
2025-03-28 9:24 ` Re: nvme-tcp: fix a possible UAF when failing to send request zhang.guanghui
2025-04-01 12:11 ` Maurizio Lombardi
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=D7QF99BR5YYA.2CXBBA5RYACTY@bsdbackstore.eu \
--to=mlombard@bsdbackstore.eu \
--cc=chunguang.xu@shopee.com \
--cc=kbusch@kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=mgurtovoy@nvidia.com \
--cc=sagi@grimberg.me \
--cc=sashal@kernel.org \
--cc=zhang.guanghui@cestc.cn \
/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