linux-kernel - Re: [PATCH][next] staging: wfx: check for memory allocation failures from wfx_alloc

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <20200103112641.GH3911@kadam>
Date:   Fri, 3 Jan 2020 14:26:41 +0300
From:   Dan Carpenter <dan.carpenter@...cle.com>
To:     Colin King <colin.king@...onical.com>
Cc:     Jérôme Pouiller <jerome.pouiller@...abs.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        devel@...verdev.osuosl.org, kernel-janitors@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH][next] staging: wfx: check for memory allocation failures
 from wfx_alloc_hif

On Sat, Dec 21, 2019 at 12:15:43AM +0000, Colin King wrote:
> From: Colin Ian King <colin.king@...onical.com>
> 
> Currently calls to wfx_alloc_hif are not checking for a null return
> when a memory allocation fails and this leads to null pointer
> dereferencing issues.  Fix this by adding null pointer checks and
> returning passing down -ENOMEM errors where necessary. The error
> checking in the current driver is a bit sparse, so this may need
> some extra attention later if required.
> 
> Fixes: f95a29d40782 ("staging: wfx: add HIF commands helpers")
> Signed-off-by: Colin Ian King <colin.king@...onical.com>
> ---
>  drivers/staging/wfx/hif_tx.c |  6 ++++++
>  drivers/staging/wfx/sta.c    | 13 +++++++------
>  2 files changed, 13 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/staging/wfx/hif_tx.c b/drivers/staging/wfx/hif_tx.c
> index 8a34a52dd5b9..d8e159670eae 100644
> --- a/drivers/staging/wfx/hif_tx.c
> +++ b/drivers/staging/wfx/hif_tx.c
> @@ -366,6 +366,9 @@ int hif_set_edca_queue_params(struct wfx_vif *wvif, u16 queue,
>  	struct hif_req_edca_queue_params *body = wfx_alloc_hif(sizeof(*body),
>  							       &hif);
>  

I hate allocations in declaration block.  It's way more likely to have
a bug like this where it's missing the NULL check.

> +	if (!body)
> +		return -ENOMEM;
> +
>  	WARN_ON(arg->aifs > 255);
>  	body->aifsn = arg->aifs;
>  	body->cw_min = cpu_to_le16(arg->cw_min);
> @@ -390,6 +393,9 @@ int hif_set_pm(struct wfx_vif *wvif, bool ps, int dynamic_ps_timeout)
>  	struct hif_msg *hif;
>  	struct hif_req_set_pm_mode *body = wfx_alloc_hif(sizeof(*body), &hif);
>  
> +	if (!body)
> +		return -ENOMEM;
> +
>  	if (ps) {
>  		body->pm_mode.enter_psm = 1;
>  		// Firmware does not support more than 128ms
> diff --git a/drivers/staging/wfx/sta.c b/drivers/staging/wfx/sta.c
> index 9a61478d98f8..c08d691fe870 100644
> --- a/drivers/staging/wfx/sta.c
> +++ b/drivers/staging/wfx/sta.c
> @@ -316,6 +316,7 @@ int wfx_conf_tx(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
>  {
>  	struct wfx_dev *wdev = hw->priv;
>  	struct wfx_vif *wvif = (struct wfx_vif *) vif->drv_priv;
> +	int ret = 0;
>  
>  	WARN_ON(queue >= hw->queues);
>  
> @@ -326,10 +327,10 @@ int wfx_conf_tx(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
>  	if (wvif->vif->type == NL80211_IFTYPE_STATION) {
>  		hif_set_uapsd_info(wvif, wvif->uapsd_mask);
>  		if (wvif->setbssparams_done && wvif->state == WFX_STATE_STA)
> -			wfx_update_pm(wvif);
> +			ret = wfx_update_pm(wvif);
>  	}
>  	mutex_unlock(&wdev->conf_mutex);
> -	return 0;
> +	return ret;
>  }
>  
>  int wfx_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
> @@ -1322,7 +1323,7 @@ int wfx_config(struct ieee80211_hw *hw, u32 changed)
>  	if (changed & IEEE80211_CONF_CHANGE_PS) {
>  		wvif = NULL;
>  		while ((wvif = wvif_iterate(wdev, wvif)) != NULL)
> -			wfx_update_pm(wvif);
> +			ret = wfx_update_pm(wvif);

We reset "ret" on every iteration through the loop and only use the
last value.  Probably we should break out of the loop on failure.

>  		wvif = wdev_to_wvif(wdev, 0);
>  	}
>  

regards,
dan carpenter