Re: [Intel-wired-lan] [PATCH iwl-next v3 3/4] igb: add AF_XDP zero-copy Rx support

From: Simon Horman <simon.horman@corigine.com>
To: Sriram Yagnaraman <sriram.yagnaraman@est.tech>
Cc: "Jesper Dangaard Brouer" <hawk@kernel.org>,
	"Daniel Borkmann" <daniel@iogearbox.net>,
	netdev@vger.kernel.org,
	"Jonathan Lemon" <jonathan.lemon@gmail.com>,
	"John Fastabend" <john.fastabend@gmail.com>,
	"Jesse Brandeburg" <jesse.brandeburg@intel.com>,
	"Alexei Starovoitov" <ast@kernel.org>,
	"Björn Töpel" <bjorn@kernel.org>,
	"Eric Dumazet" <edumazet@google.com>,
	"Tony Nguyen" <anthony.l.nguyen@intel.com>,
	"Jakub Kicinski" <kuba@kernel.org>,
	intel-wired-lan@lists.osuosl.org, bpf@vger.kernel.org,
	"Paolo Abeni" <pabeni@redhat.com>,
	"David S . Miller" <davem@davemloft.net>,
	"Magnus Karlsson" <magnus.karlsson@intel.com>
Subject: Re: [Intel-wired-lan] [PATCH iwl-next v3 3/4] igb: add AF_XDP zero-copy Rx support
Date: Sat, 15 Jul 2023 10:23:25 +0100	[thread overview]
Message-ID: <ZLJljUETnhH92kIV@corigine.com> (raw)
In-Reply-To: <20230713104717.29475-4-sriram.yagnaraman@est.tech>

On Thu, Jul 13, 2023 at 12:47:16PM +0200, Sriram Yagnaraman wrote:
> Add support for AF_XDP zero-copy receive path.
> 
> When AF_XDP zero-copy is enabled, the rx buffers are allocated from the
> xsk buff pool using igb_alloc_rx_buffers_zc.
> 
> Use xsk_pool_get_rx_frame_size to set SRRCTL rx buf size when zero-copy
> is enabled.
> 
> Signed-off-by: Sriram Yagnaraman <sriram.yagnaraman@est.tech>

...

> diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
> index 2c1e1d70bcf9..8eed3d0ab4fc 100644
> --- a/drivers/net/ethernet/intel/igb/igb_main.c
> +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> @@ -502,12 +502,14 @@ static void igb_dump(struct igb_adapter *adapter)
>  
>  		for (i = 0; i < rx_ring->count; i++) {
>  			const char *next_desc;
> -			struct igb_rx_buffer *buffer_info;
> -			buffer_info = &rx_ring->rx_buffer_info[i];
> +			struct igb_rx_buffer *buffer_info = NULL;
>  			rx_desc = IGB_RX_DESC(rx_ring, i);
>  			u0 = (struct my_u0 *)rx_desc;
>  			staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
>  
> +			if (!rx_ring->xsk_pool)
> +				buffer_info = &rx_ring->rx_buffer_info[i];
> +
>  			if (i == rx_ring->next_to_use)
>  				next_desc = " NTU";
>  			else if (i == rx_ring->next_to_clean)
> @@ -530,7 +532,7 @@ static void igb_dump(struct igb_adapter *adapter)
>  					(u64)buffer_info->dma,

Hi Sriram,

Here buffer_info is dereferenced...

>  					next_desc);
>  
> -				if (netif_msg_pktdata(adapter) &&
> +				if (netif_msg_pktdata(adapter) && buffer_info &&

And here buffer_info is checked against NULL.

This combination doesn't seem quite right.

>  				    buffer_info->dma && buffer_info->page) {
>  					print_hex_dump(KERN_INFO, "",
>  					  DUMP_PREFIX_ADDRESS,
_______________________________________________
Intel-wired-lan mailing list
Intel-wired-lan@osuosl.org
https://lists.osuosl.org/mailman/listinfo/intel-wired-lan