Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
Message-ID: <b201a3ed-5698-4e91-adc9-34c938e43668@pensando.io>
Date:   Wed, 30 Mar 2022 09:13:21 -0700
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:91.0)
 Gecko/20100101 Thunderbird/91.7.0
Subject: Re: [PATCH net] ixgbe: ensure IPsec VF<->PF compatibility
Content-Language: en-US
To:     Leon Romanovsky <leon@kernel.org>,
        "David S . Miller" <davem@davemloft.net>,
        Jakub Kicinski <kuba@kernel.org>
Cc:     Leon Romanovsky <leonro@nvidia.com>,
        intel-wired-lan@lists.osuosl.org,
        Jeff Kirsher <jeffrey.t.kirsher@intel.com>,
        Jesse Brandeburg <jesse.brandeburg@intel.com>,
        linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
        Paolo Abeni <pabeni@redhat.com>, Raed Salem <raeds@nvidia.com>,
        Shannon Nelson <shannon.nelson@oracle.com>,
        Tony Nguyen <anthony.l.nguyen@intel.com>,
        Steffen Klassert <steffen.klassert@secunet.com>
References: <3702fad8a016170947da5f3c521a9251cf0f4a22.1648637865.git.leonro@nvidia.com>
From:   Shannon Nelson <snelson@pensando.io>
In-Reply-To: <3702fad8a016170947da5f3c521a9251cf0f4a22.1648637865.git.leonro@nvidia.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 3/30/22 4:01 AM, Leon Romanovsky wrote:
> From: Leon Romanovsky <leonro@nvidia.com>
>
> The VF driver can forward any IPsec flags and such makes the function
> is not extendable and prone to backward/forward incompatibility.
>
> If new software runs on VF, it won't know that PF configured something
> completely different as it "knows" only XFRM_OFFLOAD_INBOUND flag.
>
> Fixes: eda0333ac293 ("ixgbe: add VF IPsec management")
> Reviewed-by: Raed Salem <raeds@nvidia.com>
> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
> ---
> There is no simple fix for this VF/PF incompatibility as long as FW
> doesn't filter/decline unsupported options when convey mailbox from VF
> to PF.
> ---
>   drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c | 4 +++-
>   1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c
> index e596e1a9fc75..236f244e3f65 100644
> --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c
> +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c
> @@ -903,7 +903,9 @@ int ixgbe_ipsec_vf_add_sa(struct ixgbe_adapter *adapter, u32 *msgbuf, u32 vf)
>   	/* Tx IPsec offload doesn't seem to work on this
>   	 * device, so block these requests for now.
>   	 */
> -	if (!(sam->flags & XFRM_OFFLOAD_INBOUND)) {
> +	sam->flags = sam->flags & ~XFRM_OFFLOAD_IPV6;
> +	if (!(sam->flags & XFRM_OFFLOAD_INBOUND) ||
> +	    sam->flags & ~XFRM_OFFLOAD_INBOUND) {

So after stripping the IPV6 flag, you're checking to be sure that 
INBOUND is the only flag enabled, right?
Could you use
     if (sam->flags != XFRM_OFFLOAD_INBOUND) {
instead?

sln

>   		err = -EOPNOTSUPP;
>   		goto err_out;
>   	}