Re: [PATCH bpf-next v9 5/5] igc: Add launch time support to XDP ZC

[Maciej Fijalkowski]

On Thu, Feb 06, 2025 at 02:04:08PM +0800, Song Yoong Siang wrote:
> Enable Launch Time Control (LTC) support for XDP zero copy via XDP Tx
> metadata framework.
> 
> This patch has been tested with tools/testing/selftests/bpf/xdp_hw_metadata
> on Intel I225-LM Ethernet controller. Below are the test steps and result.
> 
> Test 1: Send a single packet with the launch time set to 1 s in the future.
> 
> Test steps:
> 1. On the DUT, start the xdp_hw_metadata selftest application:
>    $ sudo ./xdp_hw_metadata enp2s0 -l 1000000000 -L 1
> 
> 2. On the Link Partner, send a UDP packet with VLAN priority 1 to port 9091
>    of the DUT.
> 
> Result:
> When the launch time is set to 1 s in the future, the delta between the
> launch time and the transmit hardware timestamp is 0.016 us, as shown in
> printout of the xdp_hw_metadata application below.
>   0x562ff5dc8880: rx_desc[4]->addr=84110 addr=84110 comp_addr=84110 EoP
>   rx_hash: 0xE343384 with RSS type:0x1
>   HW RX-time:   1734578015467548904 (sec:1734578015.4675)
>                 delta to User RX-time sec:0.0002 (183.103 usec)
>   XDP RX-time:   1734578015467651698 (sec:1734578015.4677)
>                  delta to User RX-time sec:0.0001 (80.309 usec)
>   No rx_vlan_tci or rx_vlan_proto, err=-95
>   0x562ff5dc8880: ping-pong with csum=561c (want c7dd)
>                   csum_start=34 csum_offset=6
>   HW RX-time:   1734578015467548904 (sec:1734578015.4675)
>                 delta to HW Launch-time sec:1.0000 (1000000.000 usec)
>   0x562ff5dc8880: complete tx idx=4 addr=4018
>   HW Launch-time:   1734578016467548904 (sec:1734578016.4675)
>                     delta to HW TX-complete-time sec:0.0000 (0.016 usec)
>   HW TX-complete-time:   1734578016467548920 (sec:1734578016.4675)
>                          delta to User TX-complete-time sec:0.0000
>                          (32.546 usec)
>   XDP RX-time:   1734578015467651698 (sec:1734578015.4677)
>                  delta to User TX-complete-time sec:0.9999
>                  (999929.768 usec)
>   HW RX-time:   1734578015467548904 (sec:1734578015.4675)
>                 delta to HW TX-complete-time sec:1.0000 (1000000.016 usec)
>   0x562ff5dc8880: complete rx idx=132 addr=84110
> 
> Test 2: Send 1000 packets with a 10 ms interval and the launch time set to
>         500 us in the future.
> 
> Test steps:
> 1. On the DUT, start the xdp_hw_metadata selftest application:
>    $ sudo chrt -f 99 ./xdp_hw_metadata enp2s0 -l 500000 -L 1 > \
>      /dev/shm/result.log
> 
> 2. On the Link Partner, send 1000 UDP packets with a 10 ms interval and
>    VLAN priority 1 to port 9091 of the DUT.
> 
> Result:
> When the launch time is set to 500 us in the future, the average delta
> between the launch time and the transmit hardware timestamp is 0.016 us,
> as shown in the analysis of /dev/shm/result.log below. The XDP launch time
> works correctly in sending 1000 packets continuously.
>   Min delta: 0.005 us
>   Avr delta: 0.016 us
>   Max delta: 0.031 us
>   Total packets forwarded: 1000
> 
> Signed-off-by: Song Yoong Siang <yoong.siang.song@xxxxxxxxx>

Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@xxxxxxxxx>

also with one nit. Thanks!

> ---
>  drivers/net/ethernet/intel/igc/igc.h      |  1 +
>  drivers/net/ethernet/intel/igc/igc_main.c | 57 ++++++++++++++++++++++-
>  2 files changed, 56 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
> index b8111ad9a9a8..cd1d7b6c1782 100644
> --- a/drivers/net/ethernet/intel/igc/igc.h
> +++ b/drivers/net/ethernet/intel/igc/igc.h
> @@ -579,6 +579,7 @@ struct igc_metadata_request {
>  	struct xsk_tx_metadata *meta;
>  	struct igc_ring *tx_ring;
>  	u32 cmd_type;
> +	u16 used_desc;
>  };
>  
>  struct igc_q_vector {
> diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
> index 3df608601a4b..f239f744247d 100644
> --- a/drivers/net/ethernet/intel/igc/igc_main.c
> +++ b/drivers/net/ethernet/intel/igc/igc_main.c
> @@ -2973,9 +2973,44 @@ static u64 igc_xsk_fill_timestamp(void *_priv)
>  	return *(u64 *)_priv;
>  }
>  
> +static void igc_xsk_request_launch_time(u64 launch_time, void *_priv)
> +{
> +	struct igc_metadata_request *meta_req = _priv;
> +	struct igc_ring *tx_ring = meta_req->tx_ring;
> +	__le32 launch_time_offset;
> +	bool insert_empty = false;
> +	bool first_flag = false;
> +
> +	if (!tx_ring->launchtime_enable)
> +		return;
> +
> +	launch_time_offset = igc_tx_launchtime(tx_ring,
> +					       ns_to_ktime(launch_time),
> +					       &first_flag, &insert_empty);
> +	if (insert_empty) {
> +		/* Disregard the launch time request if the required empty frame
> +		 * fails to be inserted.
> +		 */
> +		if (igc_insert_empty_frame(tx_ring))
> +			return;
> +
> +		meta_req->tx_buffer =
> +			&tx_ring->tx_buffer_info[tx_ring->next_to_use];
> +		/* Inserting an empty packet requires two descriptors:
> +		 * one data descriptor and one context descriptor.
> +		 */
> +		meta_req->used_desc += 2;

tiny nit: you could accumulate the amount of used descs in local variable
and update the igc_metadata_request::used_desc once.

> +	}
> +
> +	/* Use one context descriptor to specify launch time and first flag. */
> +	igc_tx_ctxtdesc(tx_ring, launch_time_offset, first_flag, 0, 0, 0);
> +	meta_req->used_desc += 1;
> +}
> +
>  const struct xsk_tx_metadata_ops igc_xsk_tx_metadata_ops = {
>  	.tmo_request_timestamp		= igc_xsk_request_timestamp,
>  	.tmo_fill_timestamp		= igc_xsk_fill_timestamp,
> +	.tmo_request_launch_time	= igc_xsk_request_launch_time,
>  };
>  
>  static void igc_xdp_xmit_zc(struct igc_ring *ring)
> @@ -2998,7 +3033,13 @@ static void igc_xdp_xmit_zc(struct igc_ring *ring)
>  	ntu = ring->next_to_use;
>  	budget = igc_desc_unused(ring);
>  
> -	while (xsk_tx_peek_desc(pool, &xdp_desc) && budget--) {
> +	/* Packets with launch time require one data descriptor and one context
> +	 * descriptor. When the launch time falls into the next Qbv cycle, we
> +	 * may need to insert an empty packet, which requires two more
> +	 * descriptors. Therefore, to be safe, we always ensure we have at least
> +	 * 4 descriptors available.
> +	 */
> +	while (xsk_tx_peek_desc(pool, &xdp_desc) && budget >= 4) {
>  		struct igc_metadata_request meta_req;
>  		struct xsk_tx_metadata *meta = NULL;
>  		struct igc_tx_buffer *bi;
> @@ -3019,9 +3060,19 @@ static void igc_xdp_xmit_zc(struct igc_ring *ring)
>  		meta_req.tx_ring = ring;
>  		meta_req.tx_buffer = bi;
>  		meta_req.meta = meta;
> +		meta_req.used_desc = 0;
>  		xsk_tx_metadata_request(meta, &igc_xsk_tx_metadata_ops,
>  					&meta_req);
>  
> +		/* xsk_tx_metadata_request() may have updated next_to_use */
> +		ntu = ring->next_to_use;
> +
> +		/* xsk_tx_metadata_request() may have updated Tx buffer info */
> +		bi = meta_req.tx_buffer;
> +
> +		/* xsk_tx_metadata_request() may use a few descriptors */
> +		budget -= meta_req.used_desc;
> +
>  		tx_desc = IGC_TX_DESC(ring, ntu);
>  		tx_desc->read.cmd_type_len = cpu_to_le32(meta_req.cmd_type);
>  		tx_desc->read.olinfo_status = cpu_to_le32(olinfo_status);
> @@ -3039,9 +3090,11 @@ static void igc_xdp_xmit_zc(struct igc_ring *ring)
>  		ntu++;
>  		if (ntu == ring->count)
>  			ntu = 0;
> +
> +		ring->next_to_use = ntu;
> +		budget--;
>  	}
>  
> -	ring->next_to_use = ntu;
>  	if (tx_desc) {
>  		igc_flush_tx_descriptors(ring);
>  		xsk_tx_release(pool);
> -- 
> 2.34.1
>