From: Eric Dumazet <eric.dumazet@gmail.com>
To: David Decotigny <david.decotigny@google.com>
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
"David S. Miller" <davem@davemloft.net>,
Ian Campbell <ian.campbell@citrix.com>,
Jeff Kirsher <jeffrey.t.kirsher@intel.com>,
Jiri Pirko <jpirko@redhat.com>, Joe Perches <joe@perches.com>,
Szymon Janc <szymon@janc.net.pl>
Subject: Re: [PATCH net v3 8/9] forcedeth: 64-bit stats
Date: Sat, 05 Nov 2011 08:28:18 +0100 [thread overview]
Message-ID: <1320478098.16609.9.camel@edumazet-laptop> (raw)
In-Reply-To: <4c77142ad7c6019e908884723d3c299163a55e2e.1320457247.git.david.decotigny@google.com>
Le vendredi 04 novembre 2011 à 18:53 -0700, David Decotigny a écrit :
> This converts forcedeth stats to be 64-bits. It also improves
> accounting for dropped rx frames.
>
> Tested:
> 16-way SMP x86_64 ->
> RX bytes:7244556582 (7.2 GB) TX bytes:181904254 (181.9 MB)
>
>
This changelog and patch title are misleading.
On a 32bit x86, stats are still 32bit wide after your patch.
On a 64bit x86_64, stats were already 64bit wide before your patch.
So the real thing is about not using the embedded netdevice dev->stats
structure, to reduce false sharing.
>
> Signed-off-by: David Decotigny <david.decotigny@google.com>
> ---
> drivers/net/ethernet/nvidia/forcedeth.c | 69 +++++++++++++++++++-----------
> 1 files changed, 44 insertions(+), 25 deletions(-)
>
> diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c
> index 90cdf26..08c512b 100644
> --- a/drivers/net/ethernet/nvidia/forcedeth.c
> +++ b/drivers/net/ethernet/nvidia/forcedeth.c
> @@ -799,6 +799,8 @@ struct fe_priv {
> struct timer_list stats_poll;
> u32 nic_poll_irq;
> int rx_ring_size;
> + unsigned long stats_rx_dropped;
> + unsigned long stats_rx_missed_errors;
>
> /* media detection workaround.
> * Locking: Within irq hander or disable_irq+spin_lock(&np->lock);
> @@ -821,6 +823,7 @@ struct fe_priv {
> struct nv_skb_map *tx_change_owner;
> struct nv_skb_map *tx_end_flip;
> int tx_stop;
> + unsigned long stats_tx_dropped;
>
> /* msi/msi-x fields */
> u32 msi_flags;
> @@ -1700,33 +1703,47 @@ static void nv_get_hw_stats(struct net_device *dev)
> }
>
> /*
> - * nv_get_stats: dev->get_stats function
> + * nv_get_stats: dev->ndo_get_stats64 function
> * Get latest stats value from the nic.
> * Called with read_lock(&dev_base_lock) held for read -
> * only synchronized against unregister_netdevice.
> */
> -static struct net_device_stats *nv_get_stats(struct net_device *dev)
> +static struct rtnl_link_stats64*
> +nv_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *storage)
> {
> struct fe_priv *np = netdev_priv(dev);
>
> /* If the nic supports hw counters then retrieve latest values */
> - if (np->driver_data & (DEV_HAS_STATISTICS_V1|DEV_HAS_STATISTICS_V2|DEV_HAS_STATISTICS_V3)) {
> + if (np->driver_data & (DEV_HAS_STATISTICS_V1
> + | DEV_HAS_STATISTICS_V2
> + | DEV_HAS_STATISTICS_V3)) {
> nv_get_hw_stats(dev);
>
> - /* copy to net_device stats */
> - dev->stats.tx_packets = np->estats.tx_packets;
> - dev->stats.rx_bytes = np->estats.rx_bytes;
> - dev->stats.tx_bytes = np->estats.tx_bytes;
> - dev->stats.tx_fifo_errors = np->estats.tx_fifo_errors;
> - dev->stats.tx_carrier_errors = np->estats.tx_carrier_errors;
> - dev->stats.rx_crc_errors = np->estats.rx_crc_errors;
> - dev->stats.rx_over_errors = np->estats.rx_over_errors;
> - dev->stats.rx_fifo_errors = np->estats.rx_drop_frame;
> - dev->stats.rx_errors = np->estats.rx_errors_total;
> - dev->stats.tx_errors = np->estats.tx_errors_total;
> - }
> -
> - return &dev->stats;
> + /* generic stats */
> + storage->rx_packets = np->estats.rx_packets;
> + storage->tx_packets = np->estats.tx_packets;
> + storage->rx_bytes = np->estats.rx_bytes;
> + storage->tx_bytes = np->estats.tx_bytes;
> + storage->rx_errors = np->estats.rx_errors_total;
> + storage->tx_errors = np->estats.tx_errors_total;
> + storage->rx_dropped = np->stats_rx_dropped;
> + storage->tx_dropped = np->stats_tx_dropped;
> + storage->multicast = np->estats.rx_multicast;
> +
> + /* detailed rx_errors */
> + storage->rx_length_errors = np->estats.rx_length_error;
> + storage->rx_over_errors = np->estats.rx_over_errors;
> + storage->rx_crc_errors = np->estats.rx_crc_errors;
> + storage->rx_frame_errors = np->estats.rx_frame_align_error;
> + storage->rx_fifo_errors = np->estats.rx_drop_frame;
> + storage->rx_missed_errors = np->stats_rx_missed_errors;
> +
> + /* detailed tx_errors */
> + storage->tx_carrier_errors = np->estats.tx_carrier_errors;
> + storage->tx_fifo_errors = np->estats.tx_fifo_errors;
> + }
> +
> + return storage;
> }
>
> /*
> @@ -1759,8 +1776,10 @@ static int nv_alloc_rx(struct net_device *dev)
> np->put_rx.orig = np->first_rx.orig;
> if (unlikely(np->put_rx_ctx++ == np->last_rx_ctx))
> np->put_rx_ctx = np->first_rx_ctx;
> - } else
> + } else {
> + np->stats_rx_dropped++;
> return 1;
> + }
> }
> return 0;
> }
> @@ -1791,8 +1810,10 @@ static int nv_alloc_rx_optimized(struct net_device *dev)
> np->put_rx.ex = np->first_rx.ex;
> if (unlikely(np->put_rx_ctx++ == np->last_rx_ctx))
> np->put_rx_ctx = np->first_rx_ctx;
> - } else
> + } else {
> + np->stats_rx_dropped++;
> return 1;
> + }
> }
> return 0;
> }
> @@ -1928,7 +1949,7 @@ static void nv_drain_tx(struct net_device *dev)
> np->tx_ring.ex[i].buflow = 0;
> }
> if (nv_release_txskb(np, &np->tx_skb[i]))
> - dev->stats.tx_dropped++;
> + np->stats_tx_dropped++;
> np->tx_skb[i].dma = 0;
> np->tx_skb[i].dma_len = 0;
> np->tx_skb[i].dma_single = 0;
> @@ -2651,7 +2672,7 @@ static int nv_rx_process(struct net_device *dev, int limit)
> /* the rest are hard errors */
> else {
> if (flags & NV_RX_MISSEDFRAME)
> - dev->stats.rx_missed_errors++;
> + np->stats_rx_missed_errors++;
> dev_kfree_skb(skb);
> goto next_pkt;
> }
> @@ -2694,7 +2715,6 @@ static int nv_rx_process(struct net_device *dev, int limit)
> skb_put(skb, len);
> skb->protocol = eth_type_trans(skb, dev);
> napi_gro_receive(&np->napi, skb);
> - dev->stats.rx_packets++;
> next_pkt:
> if (unlikely(np->get_rx.orig++ == np->last_rx.orig))
> np->get_rx.orig = np->first_rx.orig;
> @@ -2777,7 +2797,6 @@ static int nv_rx_process_optimized(struct net_device *dev, int limit)
> __vlan_hwaccel_put_tag(skb, vid);
> }
> napi_gro_receive(&np->napi, skb);
> - dev->stats.rx_packets++;
> } else {
> dev_kfree_skb(skb);
> }
> @@ -5199,7 +5218,7 @@ static int nv_close(struct net_device *dev)
> static const struct net_device_ops nv_netdev_ops = {
> .ndo_open = nv_open,
> .ndo_stop = nv_close,
> - .ndo_get_stats = nv_get_stats,
> + .ndo_get_stats64 = nv_get_stats64,
> .ndo_start_xmit = nv_start_xmit,
> .ndo_tx_timeout = nv_tx_timeout,
> .ndo_change_mtu = nv_change_mtu,
> @@ -5216,7 +5235,7 @@ static const struct net_device_ops nv_netdev_ops = {
> static const struct net_device_ops nv_netdev_ops_optimized = {
> .ndo_open = nv_open,
> .ndo_stop = nv_close,
> - .ndo_get_stats = nv_get_stats,
> + .ndo_get_stats64 = nv_get_stats64,
> .ndo_start_xmit = nv_start_xmit_optimized,
> .ndo_tx_timeout = nv_tx_timeout,
> .ndo_change_mtu = nv_change_mtu,
next prev parent reply other threads:[~2011-11-05 7:28 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-11-05 1:53 [PATCH net v3 0/9] forcedeth: minor fixes for stats, rmmod, sparse David Decotigny
2011-11-05 1:53 ` [PATCH net v3 1/9] forcedeth: Add messages to indicate using MSI or MSI-X David Decotigny
2011-11-05 1:53 ` [PATCH net v3 2/9] forcedeth: Acknowledge only interrupts that are being processed David Decotigny
2011-11-05 1:53 ` [PATCH net v3 3/9] forcedeth: allow to silence tx_timeout debug messages David Decotigny
2011-11-05 1:53 ` [PATCH net v3 4/9] forcedeth: expose module parameters in /sys/module David Decotigny
2011-11-05 1:53 ` [PATCH net v3 5/9] forcedeth: remove unneeded stats updates David Decotigny
2011-11-05 1:53 ` [PATCH net v3 6/9] forcedeth: Improve stats counters David Decotigny
2011-11-05 1:53 ` [PATCH net v3 7/9] forcedeth: new ethtool stat "tx_timeout" to account for tx_timeouts David Decotigny
2011-11-05 1:53 ` [PATCH net v3 8/9] forcedeth: 64-bit stats David Decotigny
2011-11-05 7:28 ` Eric Dumazet [this message]
2011-11-05 16:34 ` David Decotigny
2011-11-05 1:53 ` [PATCH net v3 9/9] forcedeth: fix a few sparse warnings (variable shadowing) David Decotigny
2011-11-05 2:27 ` [PATCH net v3 0/9] forcedeth: minor fixes for stats, rmmod, sparse David Miller
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1320478098.16609.9.camel@edumazet-laptop \
--to=eric.dumazet@gmail.com \
--cc=davem@davemloft.net \
--cc=david.decotigny@google.com \
--cc=ian.campbell@citrix.com \
--cc=jeffrey.t.kirsher@intel.com \
--cc=joe@perches.com \
--cc=jpirko@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=szymon@janc.net.pl \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox