netdev - Re: [PATCH net-next v2 1/3] net: libwx: support hardware statistics

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: <66543766-9a68-b122-f33e-8640c402a0c2@huawei.com>
Date: Wed, 27 Sep 2023 20:38:37 +0800
From: Yunsheng Lin <linyunsheng@...wei.com>
To: Jiawen Wu <jiawenwu@...stnetic.com>, <netdev@...r.kernel.org>,
	<davem@...emloft.net>, <edumazet@...gle.com>, <kuba@...nel.org>,
	<pabeni@...hat.com>, <andrew@...n.ch>
CC: <mengyuanlou@...-swift.com>
Subject: Re: [PATCH net-next v2 1/3] net: libwx: support hardware statistics

On 2023/9/21 11:30, Jiawen Wu wrote:
> +void wx_update_stats(struct wx *wx)
> +{
> +	struct net_device_stats *net_stats = &wx->netdev->stats;
> +	struct wx_hw_stats *hwstats = &wx->stats;
> +
> +	u64 non_eop_descs = 0, alloc_rx_buff_failed = 0;
> +	u64 hw_csum_rx_good = 0, hw_csum_rx_error = 0;
> +	u64 restart_queue = 0, tx_busy = 0;
> +	u64 packets = 0, bytes = 0;
> +	u32 i;
> +
> +	/* gather some stats to the wx struct that are per queue */
> +	for (i = 0; i < wx->num_rx_queues; i++) {
> +		struct wx_ring *rx_ring = wx->rx_ring[i];
> +
> +		non_eop_descs += rx_ring->rx_stats.non_eop_descs;
> +		alloc_rx_buff_failed += rx_ring->rx_stats.alloc_rx_buff_failed;
> +		hw_csum_rx_good += rx_ring->rx_stats.csum_good_cnt;
> +		hw_csum_rx_error += rx_ring->rx_stats.csum_err;
> +		packets += rx_ring->stats.packets;
> +		bytes += rx_ring->stats.bytes;
> +	}
> +	wx->non_eop_descs = non_eop_descs;
> +	wx->alloc_rx_buff_failed = alloc_rx_buff_failed;
> +	wx->hw_csum_rx_error = hw_csum_rx_error;
> +	wx->hw_csum_rx_good = hw_csum_rx_good;
> +	net_stats->rx_bytes = bytes;
> +	net_stats->rx_packets = packets;
> +
> +	bytes = 0;
> +	packets = 0;
> +	for (i = 0; i < wx->num_tx_queues; i++) {
> +		struct wx_ring *tx_ring = wx->tx_ring[i];
> +
> +		restart_queue += tx_ring->tx_stats.restart_queue;
> +		tx_busy += tx_ring->tx_stats.tx_busy;

Is the ring->syncp needed to protect the fetching here from
the updating in wx_clean_tx_irq() and wx_xmit_frame_ring()?

> +		packets += tx_ring->stats.packets;
> +		bytes += tx_ring->stats.bytes;
> +	}
> +	wx->restart_queue = restart_queue;
> +	wx->tx_busy = tx_busy;
> +	net_stats->tx_packets = packets;
> +	net_stats->tx_bytes = bytes;
> +
> +	hwstats->gprc += rd32(wx, WX_RDM_PKT_CNT);
> +	hwstats->gptc += rd32(wx, WX_TDM_PKT_CNT);
> +	hwstats->gorc += rd64(wx, WX_RDM_BYTE_CNT_LSB);
> +	hwstats->gotc += rd64(wx, WX_TDM_BYTE_CNT_LSB);
> +	hwstats->tpr += rd64(wx, WX_RX_FRAME_CNT_GOOD_BAD_L);
> +	hwstats->tpt += rd64(wx, WX_TX_FRAME_CNT_GOOD_BAD_L);
> +	hwstats->bprc += rd64(wx, WX_RX_BC_FRAMES_GOOD_L);
> +	hwstats->bptc += rd64(wx, WX_TX_BC_FRAMES_GOOD_L);
> +	hwstats->mprc += rd64(wx, WX_RX_MC_FRAMES_GOOD_L);
> +	hwstats->mptc += rd64(wx, WX_TX_MC_FRAMES_GOOD_L);
> +	hwstats->roc += rd32(wx, WX_RX_OVERSIZE_FRAMES_GOOD);
> +	hwstats->ruc += rd32(wx, WX_RX_UNDERSIZE_FRAMES_GOOD);
> +	hwstats->lxonoffrxc += rd32(wx, WX_MAC_LXONOFFRXC);
> +	hwstats->lxontxc += rd32(wx, WX_RDB_LXONTXC);
> +	hwstats->lxofftxc += rd32(wx, WX_RDB_LXOFFTXC);
> +	hwstats->o2bgptc += rd32(wx, WX_TDM_OS2BMC_CNT);
> +	hwstats->b2ospc += rd32(wx, WX_MNG_BMC2OS_CNT);
> +	hwstats->o2bspc += rd32(wx, WX_MNG_OS2BMC_CNT);
> +	hwstats->b2ogprc += rd32(wx, WX_RDM_BMC2OS_CNT);
> +	hwstats->rdmdrop += rd32(wx, WX_RDM_DRP_PKT);
> +	hwstats->crcerrs += rd64(wx, WX_RX_CRC_ERROR_FRAMES_L);
> +	hwstats->rlec += rd64(wx, WX_RX_LEN_ERROR_FRAMES_L);
> +
> +	net_stats->multicast = 0;
> +	for (i = 0; i < wx->mac.max_rx_queues; i++)
> +		net_stats->multicast += rd32(wx, WX_PX_MPRC(i));
> +	/* Rx Errors */
> +	net_stats->rx_errors = hwstats->crcerrs + hwstats->rlec;
> +	net_stats->rx_length_errors = hwstats->rlec;
> +	net_stats->rx_crc_errors = hwstats->crcerrs;
> +}
> +EXPORT_SYMBOL(wx_update_stats);

...

> @@ -877,9 +879,11 @@ static bool wx_clean_tx_irq(struct wx_q_vector *q_vector,
>  
>  		if (__netif_subqueue_stopped(tx_ring->netdev,
>  					     tx_ring->queue_index) &&
> -		    netif_running(tx_ring->netdev))
> +		    netif_running(tx_ring->netdev)) {
>  			netif_wake_subqueue(tx_ring->netdev,
>  					    tx_ring->queue_index);
> +			++tx_ring->tx_stats.restart_queue;
> +		}
>  	}
>  
>  	return !!budget;
> @@ -956,6 +960,7 @@ static int wx_maybe_stop_tx(struct wx_ring *tx_ring, u16 size)
>  
>  	/* A reprieve! - use start_queue because it doesn't call schedule */
>  	netif_start_subqueue(tx_ring->netdev, tx_ring->queue_index);
> +	++tx_ring->tx_stats.restart_queue;
>  
>  	return 0;
>  }
> @@ -1533,8 +1538,10 @@ static netdev_tx_t wx_xmit_frame_ring(struct sk_buff *skb,
>  		count += TXD_USE_COUNT(skb_frag_size(&skb_shinfo(skb)->
>  						     frags[f]));
>  
> -	if (wx_maybe_stop_tx(tx_ring, count + 3))
> +	if (wx_maybe_stop_tx(tx_ring, count + 3)) {
> +		tx_ring->tx_stats.tx_busy++;
>  		return NETDEV_TX_BUSY;
> +	}
>