Netdev List
 help / color / mirror / Atom feed
From: Xinsheng Zhu <xinsheng.zhu.ny@gmail.com>
To: netdev@vger.kernel.org
Cc: linux-usb@vger.kernel.org, linux-kernel@vger.kernel.org,
	andrew+netdev@lunn.ch, davem@davemloft.net, edumazet@kernel.org,
	kuba@kernel.org, pabeni@redhat.com
Subject: [PATCH net] net: usb: r8153_ecm: reject short register reads
Date: Mon,  5 Oct 2026 01:59:43 -0400	[thread overview]
Message-ID: <20261005055943.35180-1-xinsheng.zhu.ny@gmail.com> (raw)

pla_read_word() and pla_write_word() read four bytes into an
uninitialized stack variable, but only check usbnet_read_cmd() for
negative return values. The helper can also return a non-negative
length smaller than requested, leaving part or all of tmp
uninitialized.

pla_read_word() may then derive the register value from uninitialized
bytes. pla_write_word() may include uninitialized bytes in the buffer
sent to the device during its read-modify-write operation.

Require the return value to equal sizeof(tmp) before using the buffer.
Return -ENODATA for short reads while preserving negative errors from
usbnet_read_cmd().

Found by manual code inspection with LLM assistance. The issue has not
been reproduced at runtime.

Fixes: c1aedf015ebd ("net/usb/r8153_ecm: support ECM mode for RTL8153")
Signed-off-by: Xinsheng Zhu <xinsheng.zhu.ny@gmail.com>
---
 drivers/net/usb/r8153_ecm.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/usb/r8153_ecm.c b/drivers/net/usb/r8153_ecm.c
index 8d860dacdf49..0988a3444c55 100644
--- a/drivers/net/usb/r8153_ecm.c
+++ b/drivers/net/usb/r8153_ecm.c
@@ -23,8 +23,10 @@ static int pla_read_word(struct usbnet *dev, u16 index)
 
 	ret = usbnet_read_cmd(dev, RTL8152_REQ_GET_REGS, RTL8152_REQT_READ, index,
 			      MCU_TYPE_PLA | byen, &tmp, sizeof(tmp));
-	if (ret < 0)
+	if (ret != sizeof(tmp)) {
+		ret = ret < 0 ? ret : -ENODATA;
 		goto out;
+	}
 
 	ret = __le32_to_cpu(tmp);
 	ret >>= (shift * 8);
@@ -55,8 +57,10 @@ static int pla_write_word(struct usbnet *dev, u16 index, u32 data)
 	ret = usbnet_read_cmd(dev, RTL8152_REQ_GET_REGS, RTL8152_REQT_READ, index,
 			      MCU_TYPE_PLA | byen, &tmp, sizeof(tmp));
 
-	if (ret < 0)
+	if (ret != sizeof(tmp)) {
+		ret = ret < 0 ? ret : -ENODATA;
 		goto out;
+	}
 
 	data |= __le32_to_cpu(tmp) & ~mask;
 	tmp = __cpu_to_le32(data);

base-commit: 6dc989ea46b96ce170840174b4a38c4a387fb005
-- 
2.54.0 (Apple Git-157)


             reply	other threads:[~2026-10-05  6:00 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-05  5:59 Xinsheng Zhu [this message]
2026-10-08  2:30 ` [PATCH net] net: usb: r8153_ecm: reject short register reads patchwork-bot+netdevbpf

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=20261005055943.35180-1-xinsheng.zhu.ny@gmail.com \
    --to=xinsheng.zhu.ny@gmail.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=edumazet@kernel.org \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.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