linux-kernel - Re: [PATCH net] Octeontx2-af: Fix negative array index read warning

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: <aJp-qm55O_ka7vSv@MacBook-Air.local>
Date: Mon, 11 Aug 2025 16:37:14 -0700
From: Joe Damato <joe@...a.to>
To: Chandra Mohan Sundar <chandramohan.explore@...il.com>
Cc: sgoutham@...vell.com, lcherian@...vell.com, gakula@...vell.com,
	jerinj@...vell.com, hkelam@...vell.com, sbhatta@...vell.com,
	andrew+netdev@...n.ch, davem@...emloft.net, edumazet@...gle.com,
	kuba@...nel.org, pabeni@...hat.com, shuah@...nel.org,
	netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
	linux-kernel-mentees@...ts.linux.dev
Subject: Re: [PATCH net] Octeontx2-af: Fix negative array index read warning

On Sun, Aug 10, 2025 at 11:33:27PM +0530, Chandra Mohan Sundar wrote:
> The cgx_get_cgxid function may return a negative value.
> Using this value directly as an array index triggers Coverity warnings.
> 
> Validate the returned value and handle the case gracefully.
> 
> Signed-off-by: Chandra Mohan Sundar <chandramohan.explore@...il.com>
> ---
>  drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c
> index 8375f18c8e07..b14de93a2481 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c
> @@ -3005,6 +3005,8 @@ static int cgx_print_fwdata(struct seq_file *s, int lmac_id)
>  		return -EAGAIN;
>  
>  	cgx_id = cgx_get_cgxid(cgxd);
> +	if (cgx_id < 0)
> +		return -EINVAL;
>  
>  	if (rvu->hw->lmac_per_cgx == CGX_LMACS_USX)
>  		fwdata =  &rvu->fwdata->cgx_fw_data_usx[cgx_id][lmac_id];

A couple pieces of feedback for you:

1. Since this is a fixes it needs a Fixes tag and a commit SHA that it is fixing.
2. cgx_get_cgxid is called in 3 places, so your patch would probably need to
   be expanded to fix all uses?

Overall though, did you somehow trigger this issue?

It seems like all cases where cgx_get_cgxid is used it would be extremely
difficult (maybe impossible?) for cgxd to be NULL and for it to return a
negative value.