Devicetree
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Loic Poulain" <loic.poulain@oss.qualcomm.com>
Cc: robh@kernel.org, krzk+dt@kernel.org, conor+dt@kernel.org,
	devicetree@vger.kernel.org
Subject: Re: [PATCH v2 5/8] net: of_net: Add of_get_nvmem_eui48() helper for EUI-48 lookup
Date: Fri, 08 May 2026 18:20:25 +0000	[thread overview]
Message-ID: <20260508182026.22146C2BCB0@smtp.kernel.org> (raw)
In-Reply-To: <20260507-block-as-nvmem-v2-5-bf17edd5134e@oss.qualcomm.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] of_get_mac_address_nvmem() corrupts the output buffer on failure, causing network drivers to use invalid MAC addresses.
--

commit dfaa504d49a935a1aa8c630022d2ce8a6b303cce
Author: Loic Poulain <loic.poulain@oss.qualcomm.com>

net: of_net: Add of_get_nvmem_eui48() helper for EUI-48 lookup

Factor out the common NVMEM EUI-48 retrieval logic from
of_get_mac_address_nvmem() into a new of_get_nvmem_eui48() helper that
accepts the NVMEM cell name as a parameter.

> diff --git a/net/core/of_net.c b/net/core/of_net.c
> index 93ea425b9248a..79b289de0f16a 100644
> --- a/net/core/of_net.c
> +++ b/net/core/of_net.c
[ ... ]
> @@ -75,27 +72,53 @@ int of_get_mac_address_nvmem(struct device_node *np, u8 *addr)
>  		return ret;
>  	}
>  
> -	cell = of_nvmem_cell_get(np, "mac-address");
> +	ret = of_get_nvmem_eui48(np, "mac-address", addr);
> +	if (ret)
> +		return ret;
> +
> +	if (!is_valid_ether_addr(addr))
> +		return -EINVAL;

Does this change inadvertently clobber the caller's output buffer when an
invalid MAC address is read?

Before this patch, of_get_mac_address_nvmem() only modified the output
buffer if the NVMEM-provided MAC address passed is_valid_ether_addr().

Now, it calls of_get_nvmem_eui48() which unconditionally copies any non-zero
6-byte value into the caller's buffer, and only then checks
is_valid_ether_addr(addr). If the MAC is invalid (e.g. multicast or
broadcast), it returns -EINVAL but leaves the invalid MAC in the caller's
buffer.

This can break existing callers that rely on the buffer remaining unmodified
on error. For example, in drivers/net/ethernet/freescale/enetc/enetc_pf_common.c,
enetc_setup_mac_address() initializes the buffer to zeroes, ignores non-defer
errors, and falls back to a hardware or random MAC if the buffer is still
all-zeroes. Because the buffer is clobbered with a non-zero invalid MAC,
the fallback is bypassed and enetc_set_si_hw_addr() programs the hardware
with the invalid MAC.

Similarly, in drivers/net/ethernet/renesas/sh_eth.c, read_mac_address()
checks if the buffer is non-zero to decide whether to read the MAC from
hardware registers. Clobbering the buffer causes the driver to skip the
valid hardware MAC and instead fall back to a random MAC.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260507-block-as-nvmem-v2-0-bf17edd5134e@oss.qualcomm.com?part=5

  reply	other threads:[~2026-05-08 18:20 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-07 15:24 [PATCH v2 0/8] Support for block device NVMEM providers Loic Poulain
2026-05-07 15:24 ` [PATCH v2 1/8] dt-bindings: mmc: Document support for nvmem-layout Loic Poulain
2026-05-13 22:42   ` Rob Herring
2026-05-07 15:24 ` [PATCH v2 2/8] dt-bindings: net: wireless: qcom,ath10k: Add NVMEM MAC address cell Loic Poulain
2026-05-08 18:20   ` sashiko-bot
2026-05-07 15:24 ` [PATCH v2 3/8] dt-bindings: bluetooth: qcom: Add NVMEM BD " Loic Poulain
2026-05-07 15:24 ` [PATCH v2 4/8] block: implement NVMEM provider Loic Poulain
2026-05-08 18:20   ` sashiko-bot
2026-05-11 11:27   ` Bartosz Golaszewski
2026-05-07 15:24 ` [PATCH v2 5/8] net: of_net: Add of_get_nvmem_eui48() helper for EUI-48 lookup Loic Poulain
2026-05-08 18:20   ` sashiko-bot [this message]
2026-05-11 10:36   ` Bartosz Golaszewski
2026-05-07 15:24 ` [PATCH v2 6/8] Bluetooth: hci_sync: Add NVMEM-backed BD address retrieval Loic Poulain
2026-05-08 18:20   ` sashiko-bot
2026-05-07 15:24 ` [PATCH v2 7/8] Bluetooth: qca: Set NVMEM BD address quirks when address is invalid Loic Poulain
2026-05-11 11:29   ` Bartosz Golaszewski
2026-05-07 15:24 ` [PATCH v2 8/8] arm64: dts: qcom: arduino-imola: Describe NVMEM layout for WiFi/BT addresses Loic Poulain
2026-05-11 11:30   ` Bartosz Golaszewski

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=20260508182026.22146C2BCB0@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=krzk+dt@kernel.org \
    --cc=loic.poulain@oss.qualcomm.com \
    --cc=robh@kernel.org \
    --cc=sashiko@lists.linux.dev \
    /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