netdev - Re: [PATCH v0] qed/qed_dev: guard against a possible division by zero

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

Message-ID: <Y+TVVuLgF+V7iTO1@corigine.com>
Date:   Thu, 9 Feb 2023 12:13:26 +0100
From:   Simon Horman <simon.horman@...igine.com>
To:     Daniil Tatianin <d-tatianin@...dex-team.ru>
Cc:     Ariel Elior <aelior@...vell.com>,
        Manish Chopra <manishc@...vell.com>,
        "David S. Miller" <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>,
        Yuval Mintz <Yuval.Mintz@...gic.com>, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v0] qed/qed_dev: guard against a possible division by zero

On Thu, Feb 09, 2023 at 01:38:13PM +0300, Daniil Tatianin wrote:
> Previously we would divide total_left_rate by zero if num_vports
> happened to be 1 because non_requested_count is calculated as
> num_vports - req_count. Guard against this by explicitly checking for
> zero when doing the division.
> 
> Found by Linux Verification Center (linuxtesting.org) with the SVACE
> static analysis tool.
> 
> Fixes: bcd197c81f63 ("qed: Add vport WFQ configuration APIs")
> Signed-off-by: Daniil Tatianin <d-tatianin@...dex-team.ru>
> ---
>  drivers/net/ethernet/qlogic/qed/qed_dev.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
> index d61cd32ec3b6..90927f68c459 100644
> --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
> +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
> @@ -5123,7 +5123,7 @@ static int qed_init_wfq_param(struct qed_hwfn *p_hwfn,
>  
>  	total_left_rate	= min_pf_rate - total_req_min_rate;
>  
> -	left_rate_per_vp = total_left_rate / non_requested_count;
> +	left_rate_per_vp = total_left_rate / (non_requested_count ?: 1);

I don't know if num_vports can be 1.
But if it is then I agree that the above will be a divide by zero.

I do, however, wonder if it would be better to either:

* Treat this case as invalid and return with -EINVAL if num_vports is 1; or

* Skip both the calculation immediately above and the code
  in the if condition below, which is the only place where
  the calculated value is used, if num_vports is 1.
  I don't think the if clause makes much sense if num_vports is one.

>  	if (left_rate_per_vp <  min_pf_rate / QED_WFQ_UNIT) {
>  		DP_VERBOSE(p_hwfn, NETIF_MSG_LINK,
>  			   "Non WFQ configured vports rate [%d Mbps] is less than one percent of configured PF min rate[%d Mbps]\n",
> -- 
> 2.25.1
>