From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga02.intel.com (mga02.intel.com [134.134.136.20]) by dpdk.org (Postfix) with ESMTP id 66147689B for ; Thu, 25 Dec 2014 01:20:16 +0100 (CET) Received: from orsmga001.jf.intel.com ([10.7.209.18]) by orsmga101.jf.intel.com with ESMTP; 24 Dec 2014 16:20:15 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.07,640,1413270000"; d="scan'208";a="628872950" Received: from kmsmsx151.gar.corp.intel.com ([172.21.73.86]) by orsmga001.jf.intel.com with ESMTP; 24 Dec 2014 16:20:13 -0800 Received: from shsmsx101.ccr.corp.intel.com (10.239.4.153) by KMSMSX151.gar.corp.intel.com (172.21.73.86) with Microsoft SMTP Server (TLS) id 14.3.195.1; Thu, 25 Dec 2014 08:20:13 +0800 Received: from shsmsx104.ccr.corp.intel.com ([169.254.5.182]) by SHSMSX101.ccr.corp.intel.com ([169.254.1.110]) with mapi id 14.03.0195.001; Thu, 25 Dec 2014 08:20:11 +0800 From: "Zhang, Helin" To: Neil Horman Thread-Topic: [dpdk-dev] [PATCH] i40e: workaround for XL710 performance Thread-Index: AQHQH4m/WQt9k4UaEEyYb8vV9XSSXJyfcDfg Date: Thu, 25 Dec 2014 00:20:11 +0000 Message-ID: References: <1419405248-14158-1-git-send-email-helin.zhang@intel.com> <20141224145506.GA23653@localhost.localdomain> In-Reply-To: <20141224145506.GA23653@localhost.localdomain> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.239.127.40] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: "dev@dpdk.org" Subject: Re: [dpdk-dev] [PATCH] i40e: workaround for XL710 performance X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 25 Dec 2014 00:20:17 -0000 Hi Neil > -----Original Message----- > From: Neil Horman [mailto:nhorman@tuxdriver.com] > Sent: Wednesday, December 24, 2014 10:55 PM > To: Zhang, Helin > Cc: dev@dpdk.org > Subject: Re: [dpdk-dev] [PATCH] i40e: workaround for XL710 performance >=20 > On Wed, Dec 24, 2014 at 03:14:08PM +0800, Helin Zhang wrote: > > on XL710, performance number is far from the expectation on recent > > firmware versions, if promiscuous mode is disabled, or promiscuous > > mode is enabled and port MAC address is equal to the packet > > destination MAC address. The fix for this issue may not be integrated > > in the following firmware version. So the workaround in software > > driver is needed. It needs to modify the initial values of 2 internal > > only registers which is the same 2 of 3 registers of it did for X710. > > Note that the workaround can be removed when it is fixed in firmware > > in the future. > > > > Signed-off-by: Helin Zhang > > --- > > lib/librte_pmd_i40e/i40e_ethdev.c | 35 > > ++++++++++++++++++++++------------- > > 1 file changed, 22 insertions(+), 13 deletions(-) > > > > diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c > > b/lib/librte_pmd_i40e/i40e_ethdev.c > > index b47a3d2..3bb75d8 100644 > > --- a/lib/librte_pmd_i40e/i40e_ethdev.c > > +++ b/lib/librte_pmd_i40e/i40e_ethdev.c > > @@ -5327,21 +5327,30 @@ i40e_debug_read_register(struct i40e_hw *hw, > > uint32_t addr, uint64_t *val) > > > > /* > > * On X710, performance number is far from the expectation on recent > > firmware > > - * versions. The fix for this issue may not be integrated in the > > following > > + * versions; on XL710, performance number is also far from the > > + expectation on > > + * recent firmware versions, if promiscuous mode is disabled, or > > + promiscuous > > + * mode is enabled and port MAC address is equal to the packet > > + destination MAC > > + * address. The fix for this issue may not be integrated in the > > + following > > * firmware version. So the workaround in software driver is needed. > > It needs > > - * to modify the initial values of 3 internal only registers. Note > > that the > > - * workaround can be removed when it is fixed in firmware in the futur= e. > > + * to modify the initial values of 3 internal only registers for > > + X710, and the > > + * same 2 internal registers for XL710. Note that the workaround can > > + be removed > > + * when it is fixed in firmware in the future. >=20 > Wouldn't it be preferable to add a firmware version check to this code so= that a > single driver can handle both cards with old and 'fixed' firmware? That = way > nothing needs to be removed and all i40e cards will have a consistent beh= avior > Neil Yes, good idea! The problem is that no firmware contains this fix till now, firmware guys e= ven cannot tell me which version will have this fix at this moment. As it reads the registers first, and compares if it is what we wanted, and = then decides if a write is needed or not. With this, removing this piece of code= is not actually needed even a fix occur in the future, though the code will be red= undant. Thank you for the comments! Regards, Helin >=20 > > */ > > -static void > > -i40e_configure_registers(struct i40e_hw *hw) -{ > > -#define I40E_GL_SWR_PRI_JOIN_MAP_0 0x26CE00 > > -#define I40E_GL_SWR_PRI_JOIN_MAP_2 0x26CE08 > > -#define I40E_GL_SWR_PM_UP_THR 0x269FBC > > + > > +/* For both X710 and XL710 */ > > #define I40E_GL_SWR_PRI_JOIN_MAP_0_VALUE 0x10000200 > > +#define I40E_GL_SWR_PRI_JOIN_MAP_0 0x26CE00 > > + > > #define I40E_GL_SWR_PRI_JOIN_MAP_2_VALUE 0x011f0200 > > +#define I40E_GL_SWR_PRI_JOIN_MAP_2 0x26CE08 > > + > > +/* For X710 only */ > > #define I40E_GL_SWR_PM_UP_THR_VALUE 0x03030303 > > +#define I40E_GL_SWR_PM_UP_THR 0x269FBC > > > > +static void > > +i40e_configure_registers(struct i40e_hw *hw) { > > static const struct { > > uint32_t addr; > > uint64_t val; > > @@ -5354,11 +5363,11 @@ i40e_configure_registers(struct i40e_hw *hw) > > uint32_t i; > > int ret; > > > > - /* Below fix is for X710 only */ > > - if (i40e_is_40G_device(hw->device_id)) > > - return; > > - > > for (i =3D 0; i < RTE_DIM(reg_table); i++) { > > + if ((i40e_is_40G_device(hw->device_id)) && > > + (reg_table[i].addr =3D=3D I40E_GL_SWR_PM_UP_THR)) > > + continue; > > + > > ret =3D i40e_debug_read_register(hw, reg_table[i].addr, ®); > > if (ret < 0) { > > PMD_DRV_LOG(ERR, "Failed to read from 0x%"PRIx32, > > -- > > 1.9.3 > > > >