Received-SPF: pass (google.com: domain of linux-kernel+bounces-95772-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Date: Thu, 7 Mar 2024 14:51:46 +0000
From: Simon Horman <horms@kernel.org>
To: Radha Mohan Chintakuntla <radhac@marvell.com>
Cc: davem@davemloft.net, edumazet@google.com, kuba@kernel.org,
	pabeni@redhat.com, netdev@vger.kernel.org,
	linux-kernel@vger.kernel.org, sgoutham@marvell.com,
	lcherian@marvell.com, gakula@marvell.com, hkelam@marvell.com,
	jerinj@marvell.com, sbhatta@marvell.com
Subject: Re: [PATCH] octeontx2-af: Increase maximum BPID channels
Message-ID: <20240307145146.GB576211@kernel.org>
References: <20240306213806.431830-1-radhac@marvell.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20240306213806.431830-1-radhac@marvell.com>

On Wed, Mar 06, 2024 at 01:38:06PM -0800, Radha Mohan Chintakuntla wrote:
> Any NIX interface type can have maximum 256 channels. So increased the
> backpressure ID count to 256 so that it can cover cn9k and cn10k SoCs that
> have different NIX interface types with varied maximum channels.
> 
> Signed-off-by: Radha Mohan Chintakuntla <radhac@marvell.com>
> ---
>  drivers/net/ethernet/marvell/octeontx2/af/mbox.h | 6 ++----
>  1 file changed, 2 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
> index d5c4f810da61..223a2e39172c 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
> @@ -1207,10 +1207,8 @@ struct nix_bp_cfg_req {
>  	/* bpid_per_chan = 1 assigns separate bp id for each channel */
>  };
>  
> -/* PF can be mapped to either CGX or LBK interface,
> - * so maximum 64 channels are possible.
> - */
> -#define NIX_MAX_BPID_CHAN	64
> +/* Maximum channels any single NIX interface can have */
> +#define NIX_MAX_BPID_CHAN	256
>  struct nix_bp_cfg_rsp {
>  	struct mbox_msghdr hdr;
>  	u16	chan_bpid[NIX_MAX_BPID_CHAN]; /* Channel and bpid mapping */

Hi Radha,

looking over this, I am curious to know how out-of bounds access
to chan_bpid is prevented. The bounds seems to be the
the number of PF or VF rings. Which I assume is derived from
the HW. But if so, what if the HW reports more than NIX_MAX_BPID_CHAN
rings?

On a different note, struct includes the following field:

	u16 bpid[NIX_MAX_BPID_CHAN];

But here the index used seems to be

1. VLAN priority (which has maximum value of 8) if DCB is used
2. 0 otherwise

So perhaps fewer elements are needed?

Apologies in advance if I'm on the wrong track here.