From mboxrd@z Thu Jan 1 00:00:00 1970 From: ethan zhao Subject: Re: [PATCH NET-NEXT V2] netxen: fix ethtool rx_dropped information in ethtool get_ethtool_stats() Date: Thu, 10 Jul 2014 09:31:31 +0800 Message-ID: <53BDECF3.9030703@oracle.com> References: <1404827830-22990-1-git-send-email-ethan.zhao@oracle.com> <53BCABD3.3050207@oracle.com> <31AFFC7280259C4184970ABA9AFE8B93E074D13E@avmb3.qlogic.org> <53BD2B42.9000008@oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Manish Chopra , David Miller , "balbi@ti.com" , netdev , linux-kernel , "sriharsha.devdas@oracle.com" , "ethan.kernel@gmail.com" To: vaughan Return-path: In-Reply-To: <53BD2B42.9000008@oracle.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On 2014/7/9 19:45, vaughan wrote: > On 07/09/2014 07:32 PM, Manish Chopra wrote: >>> -----Original Message----- >>> From: netdev-owner@vger.kernel.org [mailto:netdev-owner@vger.kernel= =2Eorg] >>> On Behalf Of ethan zhao >>> Sent: Wednesday, July 09, 2014 8:11 AM >>> To: David Miller >>> Cc: balbi@ti.com; netdev; linux-kernel; sriharsha.devdas@oracle.com= ; >>> ethan.kernel@gmail.com; vaughan >>> Subject: Re: [PATCH NET-NEXT V2] netxen: fix ethtool rx_dropped inf= ormation in >>> ethtool get_ethtool_stats() >>> >>> David, >>> >>> Please help to review and confirm this patch. >>> >>> Thanks, >>> Ethan >>> >>> On 2014/7/8 21:57, Ethan Zhao wrote: >>>> netxen driver has implemented netxen_nic_get_ethtool_stats() >>>> interface, but it doesn't collect stats.rxdropped in driver, so we >>>> will get different rx_dropped statistic information while using if= config and >>> ethtool. >>>> this patch fills stats.rxdropped field with data from net core wit= h >>>> dev_get_stats() just as ixgbe driver did for some of its stats. >>>> >>>> V2: only fix the stats.rxdropped field. >>>> >>>> Tested with last netxen 4.0.82 >>>> Compiled with net-next branch 3.16-rc2 >>>> >>>> Signed-off-by: Ethan Zhao >>>> Tested-by: Sriharsha Yadagudde >>>> --- >>>> .../ethernet/qlogic/netxen/netxen_nic_ethtool.c | 34 +++++= ++++++++++++- >>> -- >>>> 1 files changed, 29 insertions(+), 5 deletions(-) >>>> >>>> diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool= =2Ec >>>> b/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c >>>> index 4ca2c19..49e6a1b 100644 >>>> --- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c >>>> +++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c >>>> @@ -33,22 +33,30 @@ >>>> #include "netxen_nic.h" >>>> #include "netxen_nic_hw.h" >>>> >>>> +enum {NETDEV_STATS, NETXEN_STATS}; >>>> + >>>> struct netxen_nic_stats { >>>> char stat_string[ETH_GSTRING_LEN]; >>>> + int type; >>>> int sizeof_stat; >>>> int stat_offset; >>>> }; >>>> >>>> -#define NETXEN_NIC_STAT(m) sizeof(((struct netxen_adapter *)0)->m= ), \ >>>> +#define NETXEN_NIC_STAT(m) NETXEN_STATS, \ >>>> + sizeof(((struct netxen_adapter *)0)->m), \ >>>> offsetof(struct netxen_adapter, m) >>>> >>>> +#define NETXEN_NETDEV_STAT(m) NETDEV_STATS, \ >>>> + sizeof(((struct rtnl_link_stats64 *)0)->m), \ >>>> + offsetof(struct rtnl_link_stats64, m) >>>> + >>>> #define NETXEN_NIC_PORT_WINDOW 0x10000 >>>> #define NETXEN_NIC_INVALID_DATA 0xDEADBEEF >>>> >>>> static const struct netxen_nic_stats netxen_nic_gstrings_stats[= ] =3D { >>>> {"xmit_called", NETXEN_NIC_STAT(stats.xmitcalled)}, >>>> {"xmit_finished", NETXEN_NIC_STAT(stats.xmitfinished)}, >>>> - {"rx_dropped", NETXEN_NIC_STAT(stats.rxdropped)}, >>>> + {"rx_dropped", NETXEN_NETDEV_STAT(rx_dropped)}, >>>> {"tx_dropped", NETXEN_NIC_STAT(stats.txdropped)}, >>>> {"csummed", NETXEN_NIC_STAT(stats.csummed)}, >>>> {"rx_pkts", NETXEN_NIC_STAT(stats.rx_pkts)}, @@ -679,11 +687,2= 7 @@ >>>> netxen_nic_get_ethtool_stats(struct net_device *dev, >>>> { >>>> struct netxen_adapter *adapter =3D netdev_priv(dev); >>>> int index; >>>> + struct rtnl_link_stats64 temp; >>>> + const struct rtnl_link_stats64 *net_stats; >>>> + char *p =3D NULL; >>>> >>>> + net_stats =3D dev_get_stats(dev, &temp); >>>> for (index =3D 0; index < NETXEN_NIC_STATS_LEN; index++) { >>>> - char *p =3D >>>> - (char *)adapter + >>>> - netxen_nic_gstrings_stats[index].stat_offset; >>>> + >>>> + switch (netxen_nic_gstrings_stats[index].type) { >>>> + case NETDEV_STATS: >>>> + p =3D (char *)net_stats + >>>> + netxen_nic_gstrings_stats[index].stat_offset; >>>> + break; >>>> + case NETXEN_STATS: >>>> + p =3D (char *)adapter + >>>> + netxen_nic_gstrings_stats[index].stat_offset; >>>> + break; >>>> + default: >>>> + data[index] =3D 0; >>>> + continue; >>>> + } >>>> + >>>> data[index] =3D >>>> (netxen_nic_gstrings_stats[index].sizeof_stat =3D=3D >>>> sizeof(u64)) ? *(u64 *) p : *(u32 *) p; >> Ethan, I can't download this patch as I don=E2=80=99t see it in netd= ev patchwork. Just looking at the diff I think there is style issue wit= h macro definition of NETXEN_NIC_STAT and NETXEN_NETDEV_STAT. >> Enclose macro definition in parentheses as they are containing compl= ex values. > It's not an error as checkpatch.pl indicates. As the definition of > netxen_nic_stats shows, > > struct netxen_nic_stats { > char stat_string[ETH_GSTRING_LEN]; > + int type; > int sizeof_stat; > int stat_offset; > }; > > This macros consists of multiple elements, which shouldn't be enclose= d > in parentheses. Agree, I am sure it is error of checkpatch.pl. Thanks, Ethan >> Send this patch targeted for net instead of net-next being a bug fix= =2E Please also run checkpatch.pl to correct any style warning/errors i= n the patch. >> >> Thanks!! >>