From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vlad Zolotarov Subject: Re: [PATCH net-next v9 6/7] ixgbevf: Add RSS Key query code Date: Mon, 30 Mar 2015 18:31:15 +0300 Message-ID: <55196C43.3020605@cloudius-systems.com> References: <1427645497-8339-1-git-send-email-vladz@cloudius-systems.com> <1427645497-8339-7-git-send-email-vladz@cloudius-systems.com> <551876DA.1000607@gmail.com> <551943CE.4050108@cloudius-systems.com> <551966C4.2090301@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Cc: jeffrey.t.kirsher@intel.com, avi@cloudius-systems.com, gleb@cloudius-systems.com To: Alexander Duyck , Alexander Duyck , netdev@vger.kernel.org Return-path: Received: from mail-wi0-f176.google.com ([209.85.212.176]:36633 "EHLO mail-wi0-f176.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752881AbbC3PbT (ORCPT ); Mon, 30 Mar 2015 11:31:19 -0400 Received: by wibg7 with SMTP id g7so107556603wib.1 for ; Mon, 30 Mar 2015 08:31:18 -0700 (PDT) In-Reply-To: <551966C4.2090301@redhat.com> Sender: netdev-owner@vger.kernel.org List-ID: On 03/30/15 18:07, Alexander Duyck wrote: > > On 03/30/2015 05:38 AM, Vlad Zolotarov wrote: >> >> >> On 03/30/15 01:04, Alexander Duyck wrote: >>> On 03/29/2015 09:11 AM, Vlad Zolotarov wrote: >>>> Add the ixgbevf_get_rss_key() function that queries the PF for an >>>> RSS Random Key >>>> using a new VF-PF channel IXGBE_VF_GET_RSS_KEY command. >>>> >>>> This patch adds the support for 82599 and x540 devices only. >>>> Support for other >>>> devices will be added later. >>>> >>>> Signed-off-by: Vlad Zolotarov >>>> --- >>>> New in v9: >>>> - Reduce the support to 82599 and x540 devices only. >>>> - Added IXGBEVF_RSS_HASH_KEY_SIZE macro. >>>> >>>> New in v8: >>>> - Protect a mailbox access. >>>> >>>> New in v6: >>>> - Return a proper return code when an operation is blocked by PF. >>>> >>>> New in v2: >>>> - Added a more detailed patch description. >>>> >>>> New in v1 (compared to RFC): >>>> - Use "if-else" statement instead of a "switch-case" for a >>>> single option case >>>> (in ixgbevf_get_rss_key()). >>>> --- >>>> drivers/net/ethernet/intel/ixgbevf/ixgbevf.h | 1 + >>>> drivers/net/ethernet/intel/ixgbevf/mbx.h | 1 + >>>> drivers/net/ethernet/intel/ixgbevf/vf.c | 66 >>>> ++++++++++++++++++++++++++++ >>>> drivers/net/ethernet/intel/ixgbevf/vf.h | 1 + >>>> 4 files changed, 69 insertions(+) >>>> >>>> diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h >>>> b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h >>>> index bc939a1..6771ae3 100644 >>>> --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h >>>> +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h >>>> @@ -145,6 +145,7 @@ struct ixgbevf_ring { >>>> #define MAX_RX_QUEUES IXGBE_VF_MAX_RX_QUEUES >>>> #define MAX_TX_QUEUES IXGBE_VF_MAX_TX_QUEUES >>>> #define IXGBEVF_MAX_RSS_QUEUES 2 >>>> +#define IXGBEVF_RSS_HASH_KEY_SIZE 40 >>>> #define IXGBEVF_DEFAULT_TXD 1024 >>>> #define IXGBEVF_DEFAULT_RXD 512 >>>> diff --git a/drivers/net/ethernet/intel/ixgbevf/mbx.h >>>> b/drivers/net/ethernet/intel/ixgbevf/mbx.h >>>> index 66e138b..82f44e0 100644 >>>> --- a/drivers/net/ethernet/intel/ixgbevf/mbx.h >>>> +++ b/drivers/net/ethernet/intel/ixgbevf/mbx.h >>>> @@ -110,6 +110,7 @@ enum ixgbe_pfvf_api_rev { >>>> /* mailbox API, version 1.2 VF requests */ >>>> #define IXGBE_VF_GET_RETA 0x0a /* VF request for RETA */ >>>> +#define IXGBE_VF_GET_RSS_KEY 0x0b /* get RSS hash key */ >>>> /* length of permanent address message returned from PF */ >>>> #define IXGBE_VF_PERMADDR_MSG_LEN 4 >>>> diff --git a/drivers/net/ethernet/intel/ixgbevf/vf.c >>>> b/drivers/net/ethernet/intel/ixgbevf/vf.c >>>> index 2676c0b..ec68145 100644 >>>> --- a/drivers/net/ethernet/intel/ixgbevf/vf.c >>>> +++ b/drivers/net/ethernet/intel/ixgbevf/vf.c >>>> @@ -301,6 +301,72 @@ static inline int >>>> ixgbevf_get_reta_locked(struct ixgbe_hw *hw, u32 *msgbuf, >>>> return 0; >>>> } >>>> +static inline int ixgbevf_get_rss_key_locked(struct ixgbe_hw >>>> *hw, u8 *rss_key) >>>> +{ >>>> + int err; >>>> + u32 msgbuf[IXGBE_VFMAILBOX_SIZE]; >>>> + >>>> + /* We currently support the RSS Random Key retrieval for 82599 >>>> and x540 >>>> + * devices only. >>>> + * >>>> + * Thus return an error if API doesn't support RSS Random Key >>>> retrieval >>>> + * or if the operation is not supported for this device type. >>>> + */ >>>> + if (hw->api_version != ixgbe_mbox_api_12 || >>>> + hw->mac.type >= ixgbe_mac_X550_vf) >>>> + return -EPERM; >>>> + >>> The return type here should be not supported, or IXGBE_NOT_IMPLEMENTED, >>> not no permissions. >> >> I think a standard EOPNOTSUPP would be a much better choice. >> Why do u have this ixgbe-specific return values in addition to >> standard POSIX ones anyway? > > It has to do with being OS agnostic. If you want to port this to BSD, > Linux, DPDK, etc, then you can't always rely on all the error return > values being there so there is a tendency to implement your own. As > with the other function you might be best off just returning > IXGBE_ERR_MBX since the x550 and API_12 don't support the RSS messages. Returning the IXGBE_ERR_MBX from the ethtool callback would be the most confusing since it is defined to -100 which corresponds to ENETDOWN value in Linux. This would result in the most confusing ethtool error message. > > - Alex