netdev - Re: [net-next,v20,3/7] net: mtip: Add buffers management functions to the L2 switch driver

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20260128022554.4151421-1-kuba@kernel.org>
Date: Tue, 27 Jan 2026 18:25:54 -0800
From: Jakub Kicinski <kuba@...nel.org>
To: lukasz.majewski@...lbox.org
Cc: Jakub Kicinski <kuba@...nel.org>,
	festevam@...il.com,
	linux-arm-kernel@...ts.infradead.org,
	s.hauer@...gutronix.de,
	kernel@...gutronix.de,
	horms@...nel.org,
	shawnguo@...nel.org,
	imx@...ts.linux.dev,
	linux-kernel@...r.kernel.org,
	conor+dt@...nel.org,
	devicetree@...r.kernel.org,
	wahrenst@....net,
	pabeni@...hat.com,
	davem@...emloft.net,
	netdev@...r.kernel.org,
	richardcochran@...il.com,
	edumazet@...gle.com,
	andrew+netdev@...n.ch,
	robh@...nel.org,
	krzk+dt@...nel.org,
	andrew@...n.ch
Subject: Re: [net-next,v20,3/7] net: mtip: Add buffers management functions to the L2 switch driver

This is an AI-generated review of your patch. The human sending this
email has considered the AI review valid, or at least plausible.

For local reproduction steps see https://netdev-ai.bots.linux.dev/ai-local.html
---
net: mtip: Add buffers management functions to the L2 switch driver

This patch adds buffer management functions for the MTIP L2 switch
driver, implementing mtip_free_buffers(), mtip_create_page_pool(), and
mtip_alloc_buffers() for RX page pool and TX bounce buffer handling.

> diff --git a/drivers/net/ethernet/freescale/mtipsw/mtipl2sw.c b/drivers/net/ethernet/freescale/mtipsw/mtipl2sw.c
> index 03c5163d6508..9c21a3d29c1c 100644
> --- a/drivers/net/ethernet/freescale/mtipsw/mtipl2sw.c
> +++ b/drivers/net/ethernet/freescale/mtipsw/mtipl2sw.c
> @@ -888,11 +888,96 @@ static void mtip_get_drvinfo(struct net_device *dev,
>
>  static void mtip_free_buffers(struct net_device *dev)
>  {
> +	struct mtip_ndev_priv *priv = netdev_priv(dev);
> +	struct switch_enet_private *fep = priv->fep;
> +	int i;
> +
> +	for (i = 0; i < RX_RING_SIZE; i++) {
> +		page_pool_put_full_page(fep->page_pool,
> +					fep->page[i], false);

Can fep->page[i] be NULL here?  When mtip_alloc_buffers() fails partway
through the RX allocation loop, it calls mtip_free_buffers() via the
error path.  At that point, only some fep->page[] entries have been
allocated - the rest remain NULL from the kzalloc initialization.

The call chain page_pool_put_full_page()->page_pool_put_netmem()->
page_pool_unref_and_test()->page_pool_unref_netmem()->
netmem_get_pp_ref_count_ref() will attempt to access &NULL->pp_ref_count
when passed a NULL page, causing a NULL pointer dereference.

Should there be a NULL check before calling page_pool_put_full_page()?

> +		fep->page[i] = NULL;
> +	}
> +
> +	page_pool_destroy(fep->page_pool);
> +	fep->page_pool = NULL;
> +
> +	for (i = 0; i < TX_RING_SIZE; i++)
> +		kfree(fep->tx_bounce[i]);
>  }

[ ... ]

> +static int mtip_alloc_buffers(struct net_device *dev)
> +{
> +	struct mtip_ndev_priv *priv = netdev_priv(dev);
> +	struct switch_enet_private *fep = priv->fep;
> +	struct cbd_t *bdp;
> +	struct page *page;
> +	int i, ret;
> +
> +	ret = mtip_create_page_pool(fep, RX_RING_SIZE);
> +	if (ret < 0) {
> +		dev_err(&fep->pdev->dev, "Failed to create page pool\n");
> +		return ret;
> +	}
> +
> +	bdp = fep->rx_bd_base;
> +	for (i = 0; i < RX_RING_SIZE; i++) {
> +		page = page_pool_dev_alloc_pages(fep->page_pool);
> +		if (!page) {
> +			dev_err(&fep->pdev->dev,
> +				"Failed to allocate page for rx buffer\n");
> +			goto err;
> +		}

[ ... ]

> + err:
> +	mtip_free_buffers(dev);

When this error path is taken, fep->page[] may contain NULL entries for
pages that were never allocated.

> +	return -ENOMEM;
>  }