From patchwork Tue Apr 2 12:13:06 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Heng Qi X-Patchwork-Id: 13613839 X-Patchwork-Delegate: kuba@kernel.org Received: from out30-131.freemail.mail.aliyun.com (out30-131.freemail.mail.aliyun.com [115.124.30.131]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 6317277F13 for ; Tue, 2 Apr 2024 12:13:18 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=115.124.30.131 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1712060003; cv=none; b=hSqvYWP0nLZmjGCDhmPdtUtHi7zJAkF8jG9JRYnk9AsUn/mBhgdMsRj3bxemGkO7Jp/Eaejlyzan+auv/WtBzKnXOygo+smp65DkYoV5ofiJhX0G7L5/PnNP9cTwAc5He1VyKndnMRC3fDIPFMkMFz7x6cGzuuoGhtgDFHoofhw= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1712060003; c=relaxed/simple; bh=LhtzJbjDVK9ON304Gj4v9CicwhZJnV3aUT6obJ1+mdM=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References; b=ucTS+5Ibj1L5lFirGHicTzhTel67n14kLENszvhpF+Jy/EIP9/9r8wkMKKvN8auxt6JrNKilS3rMc/bHD9mJBW/n/he2eHQZYMbRGtxuhj9N28KwBgYFBJQ8Yh4JPoYnnTAbhx4DfjZ+lTkzRnD+eWqaVAuptsYINbZOvKrgJlM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.alibaba.com; spf=pass smtp.mailfrom=linux.alibaba.com; dkim=pass (1024-bit key) header.d=linux.alibaba.com header.i=@linux.alibaba.com header.b=pW6NPZuv; arc=none smtp.client-ip=115.124.30.131 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.alibaba.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.alibaba.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux.alibaba.com header.i=@linux.alibaba.com header.b="pW6NPZuv" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.alibaba.com; s=default; t=1712059991; h=From:To:Subject:Date:Message-Id; bh=OvCDIe9Kyu//4m9HJerX8WPjKpeEj4dvMKEylqAWFe0=; b=pW6NPZuvEC+5jR+WR5e4S+Iu+qKsnwf3nU8trRVdzyHZoBQN7zsK5PxCMxJG2rMLwFBnqMeGt/qtl9IaibXcLbTrlOLmU0RMgUoAbYG9RObupS5kDrvuoEiC2/LUJtiWSvdiH53/Ut0ZKVNoMhQDAx1iEEnWAakkisVyr4A46Dw= X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R931e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=ay29a033018046051;MF=hengqi@linux.alibaba.com;NM=1;PH=DS;RN=9;SR=0;TI=SMTPD_---0W3oLGdI_1712059990; Received: from localhost(mailfrom:hengqi@linux.alibaba.com fp:SMTPD_---0W3oLGdI_1712059990) by smtp.aliyun-inc.com; Tue, 02 Apr 2024 20:13:11 +0800 From: Heng Qi To: netdev@vger.kernel.org, virtualization@lists.linux.dev Cc: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Jason Wang , "Michael S. Tsirkin" , Xuan Zhuo Subject: [PATCH net-next v3 1/3] ethtool: provide customized dim profile management Date: Tue, 2 Apr 2024 20:13:06 +0800 Message-Id: <1712059988-7705-2-git-send-email-hengqi@linux.alibaba.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: <1712059988-7705-1-git-send-email-hengqi@linux.alibaba.com> References: <1712059988-7705-1-git-send-email-hengqi@linux.alibaba.com> Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: X-Patchwork-Delegate: kuba@kernel.org The NetDIM library, currently leveraged by an array of NICs, delivers excellent acceleration benefits. Nevertheless, NICs vary significantly in their dim profile list prerequisites. Specifically, virtio-net backends may present diverse sw or hw device implementation, making a one-size-fits-all parameter list impractical. On Alibaba Cloud, the virtio DPU's performance under the default DIM profile falls short of expectations, partly due to a mismatch in parameter configuration. I also noticed that ice/idpf/ena and other NICs have customized profilelist or placed some restrictions on dim capabilities. Motivated by this, I tried adding new params for "ethtool -C" that provides a per-device control to modify and access a device's interrupt parameters. Usage ======== 1. Query the currently customized list of the device $ ethtool -c ethx ... rx-eqe-profs: {.usec = 1, .pkts = 256, .comps = 0,}, {.usec = 8, .pkts = 256, .comps = 0,}, {.usec = 64, .pkts = 256, .comps = 0,}, {.usec = 128, .pkts = 256, .comps = 0,}, {.usec = 256, .pkts = 256, .comps = 0,} rx-cqe-profs: n/a tx-eqe-profs: n/a tx-cqe-profs: n/a 2. Tune $ ethtool -C ethx rx-eqe-profs 1,1,0_2,2,0_3,3,0_4,4,0_5,5,0 $ ethtool -c ethx ... rx-eqe-profs: {.usec = 1, .pkts = 1, .comps = 0,}, {.usec = 2, .pkts = 2, .comps = 0,}, {.usec = 3, .pkts = 3, .comps = 0,}, {.usec = 4, .pkts = 4, .comps = 0,}, {.usec = 5, .pkts = 5, .comps = 0,} rx-cqe-profs: n/a tx-eqe-profs: n/a tx-cqe-profs: n/a 3. Hint If the device does not support some type of customized dim profiles, the corresponding "n/a" will display. Signed-off-by: Heng Qi --- Documentation/netlink/specs/ethtool.yaml | 29 +++++ Documentation/networking/ethtool-netlink.rst | 8 ++ include/linux/dim.h | 7 ++ include/linux/ethtool.h | 16 ++- include/uapi/linux/ethtool_netlink.h | 24 ++++ lib/dim/net_dim.c | 6 - net/ethtool/coalesce.c | 159 ++++++++++++++++++++++++++- 7 files changed, 240 insertions(+), 9 deletions(-) diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml index 197208f4..dcfdca9 100644 --- a/Documentation/netlink/specs/ethtool.yaml +++ b/Documentation/netlink/specs/ethtool.yaml @@ -408,6 +408,18 @@ attribute-sets: - name: combined-count type: u32 + - + name: profile + attributes: + - + name: usec + type: u16 + - + name: pkts + type: u16 + - + name: comps + type: u16 - name: coalesce @@ -497,6 +509,23 @@ attribute-sets: - name: tx-aggr-time-usecs type: u32 + - + name: rx-eqe-profs + type: nest + nested-attributes: profile + - + name: rx-cqe-profs + type: nest + nested-attributes: profile + - + name: tx-eqe-profs + type: nest + nested-attributes: profile + - + name: tx-cqe-profs + type: nest + nested-attributes: profile + - name: pause-stat attributes: diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst index d583d9a..66480e1 100644 --- a/Documentation/networking/ethtool-netlink.rst +++ b/Documentation/networking/ethtool-netlink.rst @@ -1033,6 +1033,10 @@ Kernel response contents: ``ETHTOOL_A_COALESCE_TX_AGGR_MAX_BYTES`` u32 max aggr size, Tx ``ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES`` u32 max aggr packets, Tx ``ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS`` u32 time (us), aggr, Tx + ``ETHTOOL_A_COALESCE_RX_EQE_PROFILE`` nested profile of DIM EQE, Rx + ``ETHTOOL_A_COALESCE_RX_CQE_PROFILE`` nested profile of DIM CQE, Rx + ``ETHTOOL_A_COALESCE_TX_EQE_PROFILE`` nested profile of DIM EQE, Tx + ``ETHTOOL_A_COALESCE_TX_CQE_PROFILE`` nested profile of DIM CQE, Tx =========================================== ====== ======================= Attributes are only included in reply if their value is not zero or the @@ -1098,6 +1102,10 @@ Request contents: ``ETHTOOL_A_COALESCE_TX_AGGR_MAX_BYTES`` u32 max aggr size, Tx ``ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES`` u32 max aggr packets, Tx ``ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS`` u32 time (us), aggr, Tx + ``ETHTOOL_A_COALESCE_RX_EQE_PROFILE`` nested profile of DIM EQE, Rx + ``ETHTOOL_A_COALESCE_RX_CQE_PROFILE`` nested profile of DIM CQE, Rx + ``ETHTOOL_A_COALESCE_TX_EQE_PROFILE`` nested profile of DIM EQE, Tx + ``ETHTOOL_A_COALESCE_TX_CQE_PROFILE`` nested profile of DIM CQE, Tx =========================================== ====== ======================= Request is rejected if it attributes declared as unsupported by driver (i.e. diff --git a/include/linux/dim.h b/include/linux/dim.h index f343bc9..43398f5 100644 --- a/include/linux/dim.h +++ b/include/linux/dim.h @@ -10,6 +10,13 @@ #include #include +/* Number of DIM profiles and period mode. */ +#define NET_DIM_PARAMS_NUM_PROFILES 5 +#define NET_DIM_DEFAULT_RX_CQ_PKTS_FROM_EQE 256 +#define NET_DIM_DEFAULT_TX_CQ_PKTS_FROM_EQE 128 +#define NET_DIM_DEF_PROFILE_CQE 1 +#define NET_DIM_DEF_PROFILE_EQE 1 + /* * Number of events between DIM iterations. * Causes a moderation of the algorithm run. diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index 9901e56..0fd81e8 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -18,6 +18,7 @@ #include #include #include +#include struct compat_ethtool_rx_flow_spec { u32 flow_type; @@ -238,6 +239,10 @@ struct kernel_ethtool_coalesce { u32 tx_aggr_max_bytes; u32 tx_aggr_max_frames; u32 tx_aggr_time_usecs; + struct dim_cq_moder rx_eqe_profs[NET_DIM_PARAMS_NUM_PROFILES]; + struct dim_cq_moder rx_cqe_profs[NET_DIM_PARAMS_NUM_PROFILES]; + struct dim_cq_moder tx_eqe_profs[NET_DIM_PARAMS_NUM_PROFILES]; + struct dim_cq_moder tx_cqe_profs[NET_DIM_PARAMS_NUM_PROFILES]; }; /** @@ -284,7 +289,11 @@ bool ethtool_convert_link_mode_to_legacy_u32(u32 *legacy_u32, #define ETHTOOL_COALESCE_TX_AGGR_MAX_BYTES BIT(24) #define ETHTOOL_COALESCE_TX_AGGR_MAX_FRAMES BIT(25) #define ETHTOOL_COALESCE_TX_AGGR_TIME_USECS BIT(26) -#define ETHTOOL_COALESCE_ALL_PARAMS GENMASK(26, 0) +#define ETHTOOL_COALESCE_RX_EQE_PROFILE BIT(27) +#define ETHTOOL_COALESCE_RX_CQE_PROFILE BIT(28) +#define ETHTOOL_COALESCE_TX_EQE_PROFILE BIT(29) +#define ETHTOOL_COALESCE_TX_CQE_PROFILE BIT(30) +#define ETHTOOL_COALESCE_ALL_PARAMS GENMASK(30, 0) #define ETHTOOL_COALESCE_USECS \ (ETHTOOL_COALESCE_RX_USECS | ETHTOOL_COALESCE_TX_USECS) @@ -316,6 +325,11 @@ bool ethtool_convert_link_mode_to_legacy_u32(u32 *legacy_u32, (ETHTOOL_COALESCE_TX_AGGR_MAX_BYTES | \ ETHTOOL_COALESCE_TX_AGGR_MAX_FRAMES | \ ETHTOOL_COALESCE_TX_AGGR_TIME_USECS) +#define ETHTOOL_COALESCE_PROFILE \ + (ETHTOOL_COALESCE_RX_EQE_PROFILE | \ + ETHTOOL_COALESCE_RX_CQE_PROFILE | \ + ETHTOOL_COALESCE_TX_EQE_PROFILE | \ + ETHTOOL_COALESCE_TX_CQE_PROFILE) #define ETHTOOL_STAT_NOT_SET (~0ULL) diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h index 3f89074..7fc3466 100644 --- a/include/uapi/linux/ethtool_netlink.h +++ b/include/uapi/linux/ethtool_netlink.h @@ -416,12 +416,36 @@ enum { ETHTOOL_A_COALESCE_TX_AGGR_MAX_BYTES, /* u32 */ ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES, /* u32 */ ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS, /* u32 */ + ETHTOOL_A_COALESCE_RX_EQE_PROFILE, /* nest - _A_MODERATIONS_MODERATION */ + ETHTOOL_A_COALESCE_RX_CQE_PROFILE, /* nest - _A_MODERATIONS_MODERATION */ + ETHTOOL_A_COALESCE_TX_EQE_PROFILE, /* nest - _A_MODERATIONS_MODERATION */ + ETHTOOL_A_COALESCE_TX_CQE_PROFILE, /* nest - _A_MODERATIONS_MODERATION */ /* add new constants above here */ __ETHTOOL_A_COALESCE_CNT, ETHTOOL_A_COALESCE_MAX = (__ETHTOOL_A_COALESCE_CNT - 1) }; +enum { + ETHTOOL_A_MODERATIONS_UNSPEC, + ETHTOOL_A_MODERATIONS_MODERATION, /* nest, _A_MODERATION_* */ + + /* add new constants above here */ + __ETHTOOL_A_MODERATIONS_CNT, + ETHTOOL_A_MODERATIONS_MAX = (__ETHTOOL_A_MODERATIONS_CNT - 1) +}; + +enum { + ETHTOOL_A_MODERATION_UNSPEC, + ETHTOOL_A_MODERATION_USEC, /* u16 */ + ETHTOOL_A_MODERATION_PKTS, /* u16 */ + ETHTOOL_A_MODERATION_COMPS, /* u16 */ + + /* add new constants above here */ + __ETHTOOL_A_MODERATION_CNT, + ETHTOOL_A_MODERATION_MAX = (__ETHTOOL_A_MODERATION_CNT - 1) +}; + /* PAUSE */ enum { diff --git a/lib/dim/net_dim.c b/lib/dim/net_dim.c index 4e32f7a..67d5beb 100644 --- a/lib/dim/net_dim.c +++ b/lib/dim/net_dim.c @@ -11,12 +11,6 @@ * There are different set of profiles for RX/TX CQs. * Each profile size must be of NET_DIM_PARAMS_NUM_PROFILES */ -#define NET_DIM_PARAMS_NUM_PROFILES 5 -#define NET_DIM_DEFAULT_RX_CQ_PKTS_FROM_EQE 256 -#define NET_DIM_DEFAULT_TX_CQ_PKTS_FROM_EQE 128 -#define NET_DIM_DEF_PROFILE_CQE 1 -#define NET_DIM_DEF_PROFILE_EQE 1 - #define NET_DIM_RX_EQE_PROFILES { \ {.usec = 1, .pkts = NET_DIM_DEFAULT_RX_CQ_PKTS_FROM_EQE,}, \ {.usec = 8, .pkts = NET_DIM_DEFAULT_RX_CQ_PKTS_FROM_EQE,}, \ diff --git a/net/ethtool/coalesce.c b/net/ethtool/coalesce.c index 83112c1..333dcd1 100644 --- a/net/ethtool/coalesce.c +++ b/net/ethtool/coalesce.c @@ -51,6 +51,10 @@ static u32 attr_to_mask(unsigned int attr_type) __CHECK_SUPPORTED_OFFSET(COALESCE_TX_USECS_HIGH); __CHECK_SUPPORTED_OFFSET(COALESCE_TX_MAX_FRAMES_HIGH); __CHECK_SUPPORTED_OFFSET(COALESCE_RATE_SAMPLE_INTERVAL); +__CHECK_SUPPORTED_OFFSET(COALESCE_RX_EQE_PROFILE); +__CHECK_SUPPORTED_OFFSET(COALESCE_RX_CQE_PROFILE); +__CHECK_SUPPORTED_OFFSET(COALESCE_TX_EQE_PROFILE); +__CHECK_SUPPORTED_OFFSET(COALESCE_TX_CQE_PROFILE); const struct nla_policy ethnl_coalesce_get_policy[] = { [ETHTOOL_A_COALESCE_HEADER] = @@ -82,6 +86,13 @@ static int coalesce_prepare_data(const struct ethnl_req_info *req_base, static int coalesce_reply_size(const struct ethnl_req_info *req_base, const struct ethnl_reply_data *reply_base) { + int modersz = nla_total_size(0) + /* _MODERATIONS_MODERATION, nest */ + nla_total_size(sizeof(u16)) + /* _MODERATION_USEC */ + nla_total_size(sizeof(u16)) + /* _MODERATION_PKTS */ + nla_total_size(sizeof(u16)); /* _MODERATION_COMPS */ + int total_modersz = nla_total_size(0) + /* _{R,T}X_{E,C}QE_PROFILE, nest */ + modersz * NET_DIM_PARAMS_NUM_PROFILES; + return nla_total_size(sizeof(u32)) + /* _RX_USECS */ nla_total_size(sizeof(u32)) + /* _RX_MAX_FRAMES */ nla_total_size(sizeof(u32)) + /* _RX_USECS_IRQ */ @@ -108,7 +119,8 @@ static int coalesce_reply_size(const struct ethnl_req_info *req_base, nla_total_size(sizeof(u8)) + /* _USE_CQE_MODE_RX */ nla_total_size(sizeof(u32)) + /* _TX_AGGR_MAX_BYTES */ nla_total_size(sizeof(u32)) + /* _TX_AGGR_MAX_FRAMES */ - nla_total_size(sizeof(u32)); /* _TX_AGGR_TIME_USECS */ + nla_total_size(sizeof(u32)) + /* _TX_AGGR_TIME_USECS */ + total_modersz * 4; /* _{R,T}X_{E,C}QE_PROFILE */ } static bool coalesce_put_u32(struct sk_buff *skb, u16 attr_type, u32 val, @@ -127,6 +139,61 @@ static bool coalesce_put_bool(struct sk_buff *skb, u16 attr_type, u32 val, return nla_put_u8(skb, attr_type, !!val); } +/** + * coalesce_put_profile - fill reply with a nla nest with four child nla nests. + * @skb: socket buffer the message is stored in + * @attr_type: nest attr type ETHTOOL_A_COALESCE_*X_*QE_PROFILE + * @profs: data passed to userspace + * @supported_params: modifiable parameters supported by the driver + * + * Put a dim profile nest attribute. Refer to ETHTOOL_A_MODERATIONS_MODERATION. + */ +static bool coalesce_put_profile(struct sk_buff *skb, u16 attr_type, + const struct dim_cq_moder *profs, + u32 supported_params) +{ + struct nlattr *profile_attr, *moder_attr; + bool valid = false; + int i; + + for (i = 0; i < NET_DIM_PARAMS_NUM_PROFILES; i++) { + if (profs[i].usec || profs[i].pkts || profs[i].comps) { + valid = true; + break; + } + } + + if (!valid || !(supported_params & attr_to_mask(attr_type))) + return false; + + profile_attr = nla_nest_start(skb, attr_type); + if (!profile_attr) + return -EMSGSIZE; + + for (i = 0; i < NET_DIM_PARAMS_NUM_PROFILES; i++) { + moder_attr = nla_nest_start(skb, ETHTOOL_A_MODERATIONS_MODERATION); + if (!moder_attr) + goto nla_cancel_profile; + + if (nla_put_u16(skb, ETHTOOL_A_MODERATION_USEC, profs[i].usec) || + nla_put_u16(skb, ETHTOOL_A_MODERATION_PKTS, profs[i].pkts) || + nla_put_u16(skb, ETHTOOL_A_MODERATION_COMPS, profs[i].comps)) + goto nla_cancel_moder; + + nla_nest_end(skb, moder_attr); + } + + nla_nest_end(skb, profile_attr); + + return 0; + +nla_cancel_moder: + nla_nest_cancel(skb, moder_attr); +nla_cancel_profile: + nla_nest_cancel(skb, profile_attr); + return -EMSGSIZE; +} + static int coalesce_fill_reply(struct sk_buff *skb, const struct ethnl_req_info *req_base, const struct ethnl_reply_data *reply_base) @@ -189,7 +256,15 @@ static int coalesce_fill_reply(struct sk_buff *skb, coalesce_put_u32(skb, ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES, kcoal->tx_aggr_max_frames, supported) || coalesce_put_u32(skb, ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS, - kcoal->tx_aggr_time_usecs, supported)) + kcoal->tx_aggr_time_usecs, supported) || + coalesce_put_profile(skb, ETHTOOL_A_COALESCE_RX_EQE_PROFILE, + kcoal->rx_eqe_profs, supported) || + coalesce_put_profile(skb, ETHTOOL_A_COALESCE_RX_CQE_PROFILE, + kcoal->rx_cqe_profs, supported) || + coalesce_put_profile(skb, ETHTOOL_A_COALESCE_TX_EQE_PROFILE, + kcoal->tx_eqe_profs, supported) || + coalesce_put_profile(skb, ETHTOOL_A_COALESCE_TX_CQE_PROFILE, + kcoal->tx_cqe_profs, supported)) return -EMSGSIZE; return 0; @@ -227,6 +302,16 @@ static int coalesce_fill_reply(struct sk_buff *skb, [ETHTOOL_A_COALESCE_TX_AGGR_MAX_BYTES] = { .type = NLA_U32 }, [ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES] = { .type = NLA_U32 }, [ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS] = { .type = NLA_U32 }, + [ETHTOOL_A_COALESCE_RX_EQE_PROFILE] = { .type = NLA_NESTED }, + [ETHTOOL_A_COALESCE_RX_CQE_PROFILE] = { .type = NLA_NESTED }, + [ETHTOOL_A_COALESCE_TX_EQE_PROFILE] = { .type = NLA_NESTED }, + [ETHTOOL_A_COALESCE_TX_CQE_PROFILE] = { .type = NLA_NESTED }, +}; + +static const struct nla_policy coalesce_set_profile_policy[] = { + [ETHTOOL_A_MODERATION_USEC] = {.type = NLA_U16}, + [ETHTOOL_A_MODERATION_PKTS] = {.type = NLA_U16}, + [ETHTOOL_A_MODERATION_COMPS] = {.type = NLA_U16}, }; static int @@ -253,6 +338,68 @@ static int coalesce_fill_reply(struct sk_buff *skb, return 1; } +/** + * ethnl_update_profile - get a nla nest with four child nla nests from userspace. + * @dst: data get from the driver and modified by ethnl_update_profile. + * @nests: nest attr ETHTOOL_A_COALESCE_*X_*QE_PROFILE to set driver's profile. + * @mod: whether the data is modified + * + * Layout of nests: + * Nested ETHTOOL_A_COALESCE_*X_*QE_PROFILE attr + * Nested ETHTOOL_A_MODERATIONS_MODERATION attr + * ETHTOOL_A_MODERATION_USEC attr + * ETHTOOL_A_MODERATION_PKTS attr + * ETHTOOL_A_MODERATION_COMPS attr + * ... + * Nested ETHTOOL_A_MODERATIONS_MODERATION attr + * ETHTOOL_A_MODERATION_USEC attr + * ETHTOOL_A_MODERATION_PKTS attr + * ETHTOOL_A_MODERATION_COMPS attr + */ +static inline void ethnl_update_profile(struct dim_cq_moder *dst, + const struct nlattr *nests, + bool *mod, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb_moder[ARRAY_SIZE(coalesce_set_profile_policy)]; + struct dim_cq_moder profs[NET_DIM_PARAMS_NUM_PROFILES]; + struct nlattr *nest; + int ret, rem, i = 0; + + if (!nests) + return; + + nla_for_each_nested(nest, nests, rem) { + if (WARN_ONCE(nla_type(nest) != ETHTOOL_A_MODERATIONS_MODERATION, + "unexpected nest attrtype %u\n", nla_type(nest))) + return; + + ret = nla_parse_nested(tb_moder, + ARRAY_SIZE(coalesce_set_profile_policy) - 1, + nest, coalesce_set_profile_policy, + extack); + if (ret || + !tb_moder[ETHTOOL_A_MODERATION_USEC] || + !tb_moder[ETHTOOL_A_MODERATION_PKTS] || + !tb_moder[ETHTOOL_A_MODERATION_COMPS]) { + NL_SET_ERR_MSG(extack, "wrong ETHTOOL_A_MODERATION_* attribute\n"); + return; + } + + profs[i].usec = nla_get_u16(tb_moder[ETHTOOL_A_MODERATION_USEC]); + profs[i].pkts = nla_get_u16(tb_moder[ETHTOOL_A_MODERATION_PKTS]); + profs[i].comps = nla_get_u16(tb_moder[ETHTOOL_A_MODERATION_COMPS]); + + if (dst[i].usec != profs[i].usec || dst[i].pkts != profs[i].pkts || + dst[i].comps != profs[i].comps) + *mod = true; + + i++; + } + + memcpy(dst, profs, sizeof(profs)); +} + static int __ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info, bool *dual_change) @@ -316,6 +463,14 @@ static int coalesce_fill_reply(struct sk_buff *skb, tb[ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES], &mod); ethnl_update_u32(&kernel_coalesce.tx_aggr_time_usecs, tb[ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS], &mod); + ethnl_update_profile(kernel_coalesce.rx_eqe_profs, + tb[ETHTOOL_A_COALESCE_RX_EQE_PROFILE], &mod, info->extack); + ethnl_update_profile(kernel_coalesce.rx_cqe_profs, + tb[ETHTOOL_A_COALESCE_RX_CQE_PROFILE], &mod, info->extack); + ethnl_update_profile(kernel_coalesce.tx_eqe_profs, + tb[ETHTOOL_A_COALESCE_TX_EQE_PROFILE], &mod, info->extack); + ethnl_update_profile(kernel_coalesce.tx_cqe_profs, + tb[ETHTOOL_A_COALESCE_TX_CQE_PROFILE], &mod, info->extack); /* Update operation modes */ ethnl_update_bool32(&coalesce.use_adaptive_rx_coalesce, From patchwork Tue Apr 2 12:13:07 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Heng Qi X-Patchwork-Id: 13613836 X-Patchwork-Delegate: kuba@kernel.org Received: from out30-132.freemail.mail.aliyun.com (out30-132.freemail.mail.aliyun.com [115.124.30.132]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 0C28956458 for ; Tue, 2 Apr 2024 12:13:14 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=115.124.30.132 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1712059997; cv=none; b=JayfD6YPoS1DhBIKp7j4FD6hJSIE07kaYCfq6M48AzUfsST9dJOwecM+6Ejbnt6RPacSLo91OqGTGHWJBbwXtTRGU9tOOTUbiJyPaCEapb7JlFw/5qPpD1Q8TE3QyhJPNTdBhXxe6fSgIyCXe6N5pZLSNtUHndQWmDKFV4fbcpQ= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1712059997; c=relaxed/simple; bh=XgtvABkp8d+PXjqRPm7pwOMp+j4Ay747QRHYC+YIVk0=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References; b=D7g/wZMnN75odT/yhNa8B2ks0VZwE5KP4rVqfSoGWJA+OV0o+MxGE/N1a4r1x4T+iEw71ksyu3vVnKvIBrVNmq+9Ps7FMnuyiPiu0aax3A/k8YXnNVfxtUWQZnGalyzcVl1i/+ZBsDMizRI13/2K0W3MsJa+FxheO5O/RWs+cJc= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.alibaba.com; spf=pass smtp.mailfrom=linux.alibaba.com; dkim=pass (1024-bit key) header.d=linux.alibaba.com header.i=@linux.alibaba.com header.b=U4Vdhh+Y; arc=none smtp.client-ip=115.124.30.132 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.alibaba.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.alibaba.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux.alibaba.com header.i=@linux.alibaba.com header.b="U4Vdhh+Y" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.alibaba.com; s=default; t=1712059992; h=From:To:Subject:Date:Message-Id; bh=7XnhnARxovlo7ns+6unt3hFoBOlpEe2/Uc5vve9T1V8=; b=U4Vdhh+YUn8/6mpfBbE8kDdropmxjwl5YDmd4LzYJvNbiuwGDFV6lrXNj1l+dUH+A270X+Y80e/XVNsps5n90hM/2+ZoUjD7zx8isgRDkeyEnC9ok9OEYN/ONyrxpdsCq0Qw4YfWKvmloZDBTi9433ZU2duQ2ptSFA4s40zEKEw= X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R941e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=ay29a033018046060;MF=hengqi@linux.alibaba.com;NM=1;PH=DS;RN=9;SR=0;TI=SMTPD_---0W3oKXPh_1712059991; Received: from localhost(mailfrom:hengqi@linux.alibaba.com fp:SMTPD_---0W3oKXPh_1712059991) by smtp.aliyun-inc.com; Tue, 02 Apr 2024 20:13:12 +0800 From: Heng Qi To: netdev@vger.kernel.org, virtualization@lists.linux.dev Cc: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Jason Wang , "Michael S. Tsirkin" , Xuan Zhuo Subject: [PATCH net-next v3 2/3] virtio-net: refactor dim initialization/destruction Date: Tue, 2 Apr 2024 20:13:07 +0800 Message-Id: <1712059988-7705-3-git-send-email-hengqi@linux.alibaba.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: <1712059988-7705-1-git-send-email-hengqi@linux.alibaba.com> References: <1712059988-7705-1-git-send-email-hengqi@linux.alibaba.com> Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: X-Patchwork-Delegate: kuba@kernel.org Extract the initialization and destruction actions of dim for use in the next patch. Signed-off-by: Heng Qi --- drivers/net/virtio_net.c | 37 ++++++++++++++++++++++++++----------- 1 file changed, 26 insertions(+), 11 deletions(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index e709d44..5c56fdc 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -2278,6 +2278,13 @@ static int virtnet_enable_queue_pair(struct virtnet_info *vi, int qp_index) return err; } +static void virtnet_dim_clean(struct virtnet_info *vi, + int start_qnum, int end_qnum) +{ + for (; start_qnum <= end_qnum; start_qnum++) + cancel_work_sync(&vi->rq[start_qnum].dim.work); +} + static int virtnet_open(struct net_device *dev) { struct virtnet_info *vi = netdev_priv(dev); @@ -2301,11 +2308,9 @@ static int virtnet_open(struct net_device *dev) err_enable_qp: disable_delayed_refill(vi); cancel_delayed_work_sync(&vi->refill); - - for (i--; i >= 0; i--) { + virtnet_dim_clean(vi, 0, i); + for (i--; i >= 0; i--) virtnet_disable_queue_pair(vi, i); - cancel_work_sync(&vi->rq[i].dim.work); - } return err; } @@ -2470,7 +2475,7 @@ static int virtnet_rx_resize(struct virtnet_info *vi, if (running) { napi_disable(&rq->napi); - cancel_work_sync(&rq->dim.work); + virtnet_dim_clean(vi, qindex, qindex); } err = virtqueue_resize(rq->vq, ring_num, virtnet_rq_unmap_free_buf); @@ -2720,10 +2725,9 @@ static int virtnet_close(struct net_device *dev) /* Make sure refill_work doesn't re-enable napi! */ cancel_delayed_work_sync(&vi->refill); - for (i = 0; i < vi->max_queue_pairs; i++) { + virtnet_dim_clean(vi, 0, vi->max_queue_pairs - 1); + for (i = 0; i < vi->max_queue_pairs; i++) virtnet_disable_queue_pair(vi, i); - cancel_work_sync(&vi->rq[i].dim.work); - } return 0; } @@ -4422,6 +4426,19 @@ static int virtnet_find_vqs(struct virtnet_info *vi) return ret; } +static void virtnet_dim_init(struct virtnet_info *vi) +{ + int i; + + if (!virtio_has_feature(vi->vdev, VIRTIO_NET_F_VQ_NOTF_COAL)) + return; + + for (i = 0; i < vi->max_queue_pairs; i++) { + INIT_WORK(&vi->rq[i].dim.work, virtnet_rx_dim_work); + vi->rq[i].dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; + } +} + static int virtnet_alloc_queues(struct virtnet_info *vi) { int i; @@ -4441,6 +4458,7 @@ static int virtnet_alloc_queues(struct virtnet_info *vi) goto err_rq; INIT_DELAYED_WORK(&vi->refill, refill_work); + virtnet_dim_init(vi); for (i = 0; i < vi->max_queue_pairs; i++) { vi->rq[i].pages = NULL; netif_napi_add_weight(vi->dev, &vi->rq[i].napi, virtnet_poll, @@ -4449,9 +4467,6 @@ static int virtnet_alloc_queues(struct virtnet_info *vi) virtnet_poll_tx, napi_tx ? napi_weight : 0); - INIT_WORK(&vi->rq[i].dim.work, virtnet_rx_dim_work); - vi->rq[i].dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; - sg_init_table(vi->rq[i].sg, ARRAY_SIZE(vi->rq[i].sg)); ewma_pkt_len_init(&vi->rq[i].mrg_avg_pkt_len); sg_init_table(vi->sq[i].sg, ARRAY_SIZE(vi->sq[i].sg)); From patchwork Tue Apr 2 12:13:08 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Heng Qi X-Patchwork-Id: 13613837 X-Patchwork-Delegate: kuba@kernel.org Received: from out30-97.freemail.mail.aliyun.com (out30-97.freemail.mail.aliyun.com [115.124.30.97]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id CD1B965BD1 for ; Tue, 2 Apr 2024 12:13:15 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=115.124.30.97 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1712059998; cv=none; b=bZuUzCUb0EDvYaLxxPAgcrptlMKjPwqXpt5t1+GEQHoqU/+JJmYl9sEFCl3FOwz3tEd8bir3A++YYoIUZaL5khFREe0NMvz4cPFCOCLfVXqF8MkhxBvMH2Cje+TFxM3nDLv3d0NEZFIF9IqBLJ0ZvyK4rTsmVpOI0l5oUwzCmuM= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1712059998; c=relaxed/simple; bh=KzaA75a97JDAUOkqPD+hLaUrTTpBbFMyYuxVTdoH+iI=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References; b=RFxJY264CdivcCAYrHiKAE+wZpIx95NbYQ8Xo8U9sJq6AkTmIqTeW8Pk6R/A6u1444LBDkPlMagw3C7c98dmcxa1aMBcm7my/UOYI4mFZxAR6F3RephMWqIpguKnqQfk2RKDZKfWkLc0tVjVGDrdDcPt3J9vbi0cp1DOrpj8+fM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.alibaba.com; spf=pass smtp.mailfrom=linux.alibaba.com; dkim=pass (1024-bit key) header.d=linux.alibaba.com header.i=@linux.alibaba.com header.b=moHXC+Ln; arc=none smtp.client-ip=115.124.30.97 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.alibaba.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.alibaba.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux.alibaba.com header.i=@linux.alibaba.com header.b="moHXC+Ln" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.alibaba.com; s=default; t=1712059993; h=From:To:Subject:Date:Message-Id; bh=ZSZO3ZY+29/8eAwC19tnjxV20F7cJUXQbjJvawTvOcM=; b=moHXC+LnaOLPZm1UsOtF6eK+KZs7oeGNwhZiPBhBKS/03kOorvAzpeDbJoN36PfoltkeRtyYJW54CEp6UqY4c2GV13tnU6ltykqCv2NtixxIkjzGALY6QCHGUGP3epBznyPWZmQk7Q/2JrLhpOu/sPkjU9VCrOxJ3hKXgLtJSzo= X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R491e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=ay29a033018046049;MF=hengqi@linux.alibaba.com;NM=1;PH=DS;RN=9;SR=0;TI=SMTPD_---0W3oSL1B_1712059992; Received: from localhost(mailfrom:hengqi@linux.alibaba.com fp:SMTPD_---0W3oSL1B_1712059992) by smtp.aliyun-inc.com; Tue, 02 Apr 2024 20:13:13 +0800 From: Heng Qi To: netdev@vger.kernel.org, virtualization@lists.linux.dev Cc: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Jason Wang , "Michael S. Tsirkin" , Xuan Zhuo Subject: [PATCH net-next v3 3/3] virtio-net: support dim profile fine-tuning Date: Tue, 2 Apr 2024 20:13:08 +0800 Message-Id: <1712059988-7705-4-git-send-email-hengqi@linux.alibaba.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: <1712059988-7705-1-git-send-email-hengqi@linux.alibaba.com> References: <1712059988-7705-1-git-send-email-hengqi@linux.alibaba.com> Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: X-Patchwork-Delegate: kuba@kernel.org Virtio-net has different types of back-end device implementations. In order to effectively optimize the dim library's gains for different device implementations, let's use the new interface params to fine-tune the profile list. Signed-off-by: Heng Qi --- drivers/net/virtio_net.c | 42 ++++++++++++++++++++++++++++++++++++++++-- include/linux/virtio_net.h | 11 +++++++++++ 2 files changed, 51 insertions(+), 2 deletions(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 5c56fdc..732214b 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -330,6 +330,12 @@ struct virtnet_info { struct virtnet_interrupt_coalesce intr_coal_tx; struct virtnet_interrupt_coalesce intr_coal_rx; + /* DIM profile list for different queue mode */ + struct dim_cq_moder rx_eqe_conf[NET_DIM_PARAMS_NUM_PROFILES]; + struct dim_cq_moder rx_cqe_conf[NET_DIM_PARAMS_NUM_PROFILES]; + struct dim_cq_moder tx_eqe_conf[NET_DIM_PARAMS_NUM_PROFILES]; + struct dim_cq_moder tx_cqe_conf[NET_DIM_PARAMS_NUM_PROFILES]; + unsigned long guest_offloads; unsigned long guest_offloads_capable; @@ -3588,7 +3594,7 @@ static void virtnet_rx_dim_work(struct work_struct *work) if (!rq->dim_enabled) continue; - update_moder = net_dim_get_rx_moderation(dim->mode, dim->profile_ix); + update_moder = vi->rx_eqe_conf[dim->profile_ix]; if (update_moder.usec != rq->intr_coal.max_usecs || update_moder.pkts != rq->intr_coal.max_packets) { err = virtnet_send_rx_ctrl_coal_vq_cmd(vi, qnum, @@ -3631,6 +3637,27 @@ static int virtnet_should_update_vq_weight(int dev_flags, int weight, return 0; } +static int virtnet_update_profile(struct virtnet_info *vi, + struct kernel_ethtool_coalesce *kc) +{ + int i; + + if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_VQ_NOTF_COAL)) { + for (i = 0; i < NET_DIM_PARAMS_NUM_PROFILES; i++) + if (kc->rx_eqe_profs[i].comps) + return -EINVAL; + } else { + return -EOPNOTSUPP; + } + + for (i = 0; i < NET_DIM_PARAMS_NUM_PROFILES; i++) { + vi->rx_eqe_conf[i].usec = kc->rx_eqe_profs[i].usec; + vi->rx_eqe_conf[i].pkts = kc->rx_eqe_profs[i].pkts; + } + + return 0; +} + static int virtnet_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec, struct kernel_ethtool_coalesce *kernel_coal, @@ -3657,6 +3684,10 @@ static int virtnet_set_coalesce(struct net_device *dev, } } + ret = virtnet_update_profile(vi, kernel_coal); + if (ret) + return ret; + if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_NOTF_COAL)) ret = virtnet_send_notf_coal_cmds(vi, ec); else @@ -3693,6 +3724,10 @@ static int virtnet_get_coalesce(struct net_device *dev, ec->tx_max_coalesced_frames = 1; } + if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_VQ_NOTF_COAL)) + memcpy(kernel_coal->rx_eqe_profs, vi->rx_eqe_conf, + sizeof(vi->rx_eqe_conf)); + return 0; } @@ -3872,7 +3907,8 @@ static int virtnet_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info) static const struct ethtool_ops virtnet_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_MAX_FRAMES | - ETHTOOL_COALESCE_USECS | ETHTOOL_COALESCE_USE_ADAPTIVE_RX, + ETHTOOL_COALESCE_USECS | ETHTOOL_COALESCE_USE_ADAPTIVE_RX | + ETHTOOL_COALESCE_RX_EQE_PROFILE, .get_drvinfo = virtnet_get_drvinfo, .get_link = ethtool_op_get_link, .get_ringparam = virtnet_get_ringparam, @@ -4437,6 +4473,8 @@ static void virtnet_dim_init(struct virtnet_info *vi) INIT_WORK(&vi->rq[i].dim.work, virtnet_rx_dim_work); vi->rq[i].dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; } + + memcpy(vi->rx_eqe_conf, rx_eqe_conf, sizeof(vi->rx_eqe_conf)); } static int virtnet_alloc_queues(struct virtnet_info *vi) diff --git a/include/linux/virtio_net.h b/include/linux/virtio_net.h index 4dfa9b6..82aa4cc 100644 --- a/include/linux/virtio_net.h +++ b/include/linux/virtio_net.h @@ -6,9 +6,20 @@ #include #include #include +#include #include #include +/* This is copied from NET_DIM_RX_EQE_PROFILES in DIM library */ +#define VIRTNET_DIM_RX_PKTS 256 +static const struct dim_cq_moder rx_eqe_conf[] = { + {.usec = 1, .pkts = VIRTNET_DIM_RX_PKTS,}, + {.usec = 8, .pkts = VIRTNET_DIM_RX_PKTS,}, + {.usec = 64, .pkts = VIRTNET_DIM_RX_PKTS,}, + {.usec = 128, .pkts = VIRTNET_DIM_RX_PKTS,}, + {.usec = 256, .pkts = VIRTNET_DIM_RX_PKTS,} +}; + static inline bool virtio_net_hdr_match_proto(__be16 protocol, __u8 gso_type) { switch (gso_type & ~VIRTIO_NET_HDR_GSO_ECN) {