Netdev List
 help / color / mirror / Atom feed
From: "Karumanchi, Vineeth" <vineeth@amd.com>
To: "Théo Lebrun" <theo.lebrun@bootlin.com>,
	"Vineeth Karumanchi" <vineeth.karumanchi@amd.com>,
	conor.dooley@microchip.com, andrew+netdev@lunn.ch,
	davem@davemloft.net, edumazet@google.com, kuba@kernel.org,
	pabeni@redhat.com
Cc: git@amd.com, netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH net] net: macb: fix NULL pointer dereference on unbind with fixed-link
Date: Thu, 3 Sep 2026 15:07:06 +0530	[thread overview]
Message-ID: <54e4c15b-e235-4c97-ba82-7e5d5a05a154@amd.com> (raw)
In-Reply-To: <DL5JESSS9JQK.2EV7XV479M0UW@bootlin.com>

Hi Théo Lebrun,

On 9/3/2026 1:57 PM, Théo Lebrun wrote:
> Hello Vineeth,
> 
> On Wed Sep 2, 2026 at 12:28 PM CEST, Vineeth Karumanchi wrote:
>> When the device tree describes a fixed-link and has no "mdio" child
>> node, macb_mii_init() returns early without allocating the MDIO bus,
>> leaving bp->mii_bus as NULL.
>>
>> Two cleanup paths then dereference this NULL bus:
>>
>> 1. On driver unbind, macb_remove() unconditionally calls
>>    mdiobus_unregister(bp->mii_bus), which oopses:
>>
>>   Unable to handle kernel NULL pointer dereference at virtual address 00000000000004a8
>>   pc : mdiobus_unregister+0x14/0xa4
>>   lr : macb_remove+0x38/0xa4
>>   Call trace:
>>    mdiobus_unregister+0x14/0xa4 (P)
>>    macb_remove+0x38/0xa4
>>    platform_remove+0x20/0x30
>>    device_release_driver_internal+0x1c8/0x224
>>    unbind_store+0xb4/0xbc
>>
>> 2. On the probe error path in macb_probe(), reached when
>>    macb_mii_init() has succeeded but a subsequent step fails, the
>>    err_out_unregister_mdio label runs the same unconditional cleanup.
>>
>> mdiobus_unregister() and mdiobus_free() do not guard against a NULL
>> bus, so guard the calls in both macb_remove() and the probe error
>> path.
> 
> Agreed on the patch! However as we are there, I looked at all
> bp->mii_bus usage and macb_phylink_connect() might crash if bp->mii_bus
> is NULL.
> 
> bp->mii_bus is NULL if
>  - no children mdio DT node AND
>  - DT declared fixed-link, see of_phy_is_fixed_link()
> 
> In macb_phylink_connect(), phy_find_first(bp->mii_bus) will be called if
>  - phylink_of_phy_connect() fails AND
>  - we have no phy-handle phandle prop
> 
> Those two overlap right?
> 

We did not observe crash in the scenario described above.
The AI-generated response below is also consistent with our observations:

A fixed-link forces phylink_of_phy_connect() to return 0 (success), so
the phy_find_first() branch is never entered.

So in macb_phylink_connect(), the fixed-link case gives dn != NULL and
ret == 0:
if (!dn || (ret && !macb_phy_handle_exists(dn))) {   /* false || (0 &&
…) → false */
        phydev = phy_find_first(bp->mii_bus);        /* not reached */

Both disjuncts are false, so phy_find_first(bp->mii_bus) is skipped.

Thanks,

> Thanks,
> 
> --
> Théo Lebrun, Bootlin
> Embedded Linux and Kernel engineering
> https://bootlin.com
> 

-- 
🙏 Vineeth


      reply	other threads:[~2026-09-03  9:37 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-02 10:28 [PATCH net] net: macb: fix NULL pointer dereference on unbind with fixed-link Vineeth Karumanchi
2026-09-03  2:17 ` Xuanqiang Luo
2026-09-03  7:23 ` Nicolai Buchwitz
2026-09-03  8:28   ` Théo Lebrun
2026-09-03  8:27 ` Théo Lebrun
2026-09-03  9:37   ` Karumanchi, Vineeth [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=54e4c15b-e235-4c97-ba82-7e5d5a05a154@amd.com \
    --to=vineeth@amd.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=conor.dooley@microchip.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=git@amd.com \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=theo.lebrun@bootlin.com \
    --cc=vineeth.karumanchi@amd.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