public inbox for netdev@vger.kernel.org
 help / color / mirror / Atom feed
From: Greg KH <gregkh@linuxfoundation.org>
To: jiang.kun2@zte.com.cn
Cc: andrew@lunn.ch, vivien.didelot@gmail.com, f.fainelli@gmail.com,
	olteanv@gmail.com, davem@davemloft.net, kuba@kernel.org,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	stable@vger.kernel.org, he.peilin@zte.com.cn,
	xu.xin16@zte.com.cn, fan.yu9@zte.com.cn, qiu.yutan@zte.com.cn,
	wang.yaxin@zte.com.cn, tu.qiang35@zte.com.cn,
	yang.yang29@zte.com.cn, ye.xingchen@zte.com.cn,
	zhang.yunkai@zte.com.cn
Subject: Re: [PATCH stable 5.15] net:dsa:fix the dsa_ptr null pointer dereference
Date: Mon, 30 Dec 2024 15:30:04 +0100	[thread overview]
Message-ID: <2024123054-matrix-surprise-f5c1@gregkh> (raw)
In-Reply-To: <202412261916435469rfyTVNfO8PtKWbw6X51-@zte.com.cn>

On Thu, Dec 26, 2024 at 07:16:43PM +0800, jiang.kun2@zte.com.cn wrote:
> From: Peilin He<he.peilin@zte.com.cn>
> 
> Upstream commit 6c24a03a61a2 ("net: dsa: improve shutdown sequence")
> 
> Issue
> =====
> Repeatedly accessing the DSA Ethernet controller via the ethtool command,
> followed by a system reboot, may trigger a DSA null pointer dereference,
> causing a kernel panic and preventing the system from rebooting properly.
> This can lead to data loss or denial-of-service, resulting in serious
> consequences.
> 
> The following is the panic log:
> [  172.523467] Unable to handle kernel NULL pointer dereference at virtual
> address 0000000000000020
> [  172.706923] Call trace:
> [  172.709371]  dsa_master_get_sset_count+0x24/0xa4
> [  172.714000]  ethtool_get_drvinfo+0x8c/0x210
> [  172.718193]  dev_ethtool+0x780/0x2120
> [  172.721863]  dev_ioctl+0x1b0/0x580
> [  172.725273]  sock_do_ioctl+0xc0/0x100
> [  172.728944]  sock_ioctl+0x130/0x3c0
> [  172.732440]  __arm64_sys_ioctl+0xb4/0x100
> [  172.736460]  invoke_syscall+0x50/0x120
> [  172.740219]  el0_svc_common.constprop.0+0x4c/0xf4
> [  172.744936]  do_el0_svc+0x2c/0xa0
> [  172.748257]  el0_svc+0x20/0x60
> [  172.751318]  el0t_64_sync_handler+0xe8/0x114
> [  172.755599]  el0t_64_sync+0x180/0x184
> [  172.759271] Code: a90153f3 2a0103f4 a9025bf5 f9418015 (f94012b6)
> [  172.765383] ---[ end trace 0000000000000002 ]---
> 
> Root Cause
> ==========
> Based on analysis of the Linux 5.15 stable version, the function
> dsa_master_get_sset_count() accesses members of the structure pointed
> to by cpu_dp without checking for a null pointer.  If cpu_dp is a
> null pointer, this will cause a kernel panic.
> 
> 	static int dsa_master_get_sset_count(struct net_device *dev, int sset)
> 	{
> 		struct dsa_port *cpu_dp = dev->dsa_ptr;
> 		const struct ethtool_ops *ops = cpu_dp->orig_ethtool_ops;
> 		struct dsa_switch *ds = cpu_dp->ds;
> 		...
> 	}
> 
> dev->dsa_ptr is set to NULL in the dsa_switch_shutdown() or
> dsa_master_teardown() functions. When the DSA module unloads,
> dsa_master_ethtool_teardown(dev) restores the original copy of
> the DSA device's ethtool_ops using "dev->ethtool_ops =
> cpu_dp->orig_ethtool_ops;" before setting dev->dsa_ptr to NULL.
> This ensures that ethtool_ops remains accessible after DSA unloads.
> However, dsa_switch_shutdown does not restore the original copy of
> the DSA device's ethtool_ops, potentially leading to a null pointer
> dereference of dsa_ptr and causing a system panic.  Essentially,
> when we set master->dsa_ptr to NULL, we need to ensure that
> no user ports are making requests to the DSA driver.
> 
> Solution
> ========
> The addition of the netif_device_detach() function is to ensure that
> ioctls, rtnetlinks and ethtool requests on the user ports no longer
> propagate down to the driver - we're no longer prepared to handle them.
> 
> Fixes: ee534378f005 ("net: dsa: fix panic when DSA master device unbinds on shutdown")
> Suggested-by: Vladimir Oltean <vladimir.oltean@nxp.com>
> Signed-off-by: Peilin He <he.peilin@zte.com.cn>
> Reviewed-by: xu xin <xu.xin16@zte.com.cn>
> Signed-off-by: Kun Jiang <jiang.kun2@zte.com.cn>
> Cc: Fan Yu <fan.yu9@zte.com.cn>
> Cc: Yutan Qiu <qiu.yutan@zte.com.cn>
> Cc: Yaxin Wang <wang.yaxin@zte.com.cn>
> Cc: tuqiang <tu.qiang35@zte.com.cn>
> Cc: Yang Yang <yang.yang29@zte.com.cn>
> Cc: ye xingchen <ye.xingchen@zte.com.cn>
> Cc: Yunkai Zhang <zhang.yunkai@zte.com.cn>

You dropped all the original signed-off-by lines :(

      reply	other threads:[~2024-12-30 14:30 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-12-26 11:16 [PATCH stable 5.15] net:dsa:fix the dsa_ptr null pointer dereference jiang.kun2
2024-12-30 14:30 ` Greg KH [this message]

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=2024123054-matrix-surprise-f5c1@gregkh \
    --to=gregkh@linuxfoundation.org \
    --cc=andrew@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=f.fainelli@gmail.com \
    --cc=fan.yu9@zte.com.cn \
    --cc=he.peilin@zte.com.cn \
    --cc=jiang.kun2@zte.com.cn \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=olteanv@gmail.com \
    --cc=qiu.yutan@zte.com.cn \
    --cc=stable@vger.kernel.org \
    --cc=tu.qiang35@zte.com.cn \
    --cc=vivien.didelot@gmail.com \
    --cc=wang.yaxin@zte.com.cn \
    --cc=xu.xin16@zte.com.cn \
    --cc=yang.yang29@zte.com.cn \
    --cc=ye.xingchen@zte.com.cn \
    --cc=zhang.yunkai@zte.com.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