From patchwork Fri Dec 6 00:12:08 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ahmed Zaki X-Patchwork-Id: 13896121 X-Patchwork-Delegate: kuba@kernel.org Received: from mgamail.intel.com (mgamail.intel.com [198.175.65.12]) (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 75E57A927 for ; Fri, 6 Dec 2024 00:12:20 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=198.175.65.12 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1733443942; cv=none; b=FSyRbq0WmWZQ8Jdl3lxU5hJ1OJ/r5udz7i/6EzAfd3JDsxyLo/K4JYLl0Et4FX2Gkjq4yU/n6ycu4eONiPDA1ZKENqEkO/yPd7OyPOaMOfLRPaHOxwmkMjdxxykZzm4oHTCtYuiipoHYW6J4rXrlNEVlsnh7nNvyZk8ALD+Sn0g= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1733443942; c=relaxed/simple; bh=Y0bK+z9quarTgGyece9m9dp3o0lhPnL0tt2CAtvKZgw=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=tUdzKdM6lFotDUNKkrb/Pilz+kBF+Wmg7a3IjXB0+BjFqLuYVi/tov/B9W1tNWQSOlY2EraUYw+k8LxaWQuBBWiuRnkF9EUCODHGWTMaUZFFmeCQkPR+UYQqOm8zXblpotW+90pUyIf43LNF+7fXmIwYMdetMAQbe/8jnkguCIo= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=intel.com; spf=pass smtp.mailfrom=intel.com; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b=TDZ+mTpq; arc=none smtp.client-ip=198.175.65.12 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=intel.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=intel.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b="TDZ+mTpq" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1733443940; x=1764979940; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=Y0bK+z9quarTgGyece9m9dp3o0lhPnL0tt2CAtvKZgw=; b=TDZ+mTpqvC4HZVxU48xrewxSe17W0/yTNqyE93lev4jyBHcS7ahQYIu4 0VRPIVTdw3rMtuph81ZkdxMTIWe6JiVIV/6Qh13gPiMfs+b+qtaQQoDMA +p8laTHPeST/7csDJRtQIRngGCM/s8+O3O4XzSySboWGOkb6O1MiOlMGl 3B6epZ3Ludqb3xS9h/NNukay198WSCw/+vs/imrN7AsZxREpRwKg0Eq2A M879CSuvx8PNu0o2LEMDsvPGo45yt5HTkRRtBfHQKaejHwIV8Y7PGcy4H ilm7vJU76rIAAs9Oka4JF6bdcckTmc1cEReo758K+FEHnKsltt+auQoUv w==; X-CSE-ConnectionGUID: DVdPCxEPRbinBJVvRWW3Lg== X-CSE-MsgGUID: OhadLSDMTiyr3/xhF9mvqg== X-IronPort-AV: E=McAfee;i="6700,10204,11277"; a="45162752" X-IronPort-AV: E=Sophos;i="6.12,211,1728975600"; d="scan'208";a="45162752" Received: from fmviesa005.fm.intel.com ([10.60.135.145]) by orvoesa104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 05 Dec 2024 16:12:20 -0800 X-CSE-ConnectionGUID: PezlXTMJQNeCngxy3fSxsA== X-CSE-MsgGUID: B8EGLPjnSKWaQ0FJB1Vbyw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.12,211,1728975600"; d="scan'208";a="98694929" Received: from ibganev-mobl.amr.corp.intel.com (HELO azaki-desk1.intel.com) ([10.125.108.131]) by fmviesa005-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 05 Dec 2024 16:12:18 -0800 From: Ahmed Zaki To: netdev@vger.kernel.org Cc: Ahmed Zaki , Jakub Kicinski Subject: [PATCH RFC net-next 1/2] net: napi: add CPU affinity to napi->config Date: Thu, 5 Dec 2024 17:12:08 -0700 Message-ID: <20241206001209.213168-2-ahmed.zaki@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20241206001209.213168-1-ahmed.zaki@intel.com> References: <20241206001209.213168-1-ahmed.zaki@intel.com> Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Patchwork-Delegate: kuba@kernel.org X-Patchwork-State: RFC A common task for most drivers is to remember the user's CPU affinity to its IRQs. On each netdev reset, the driver must then re-assign the user's setting to the IRQs. Add CPU affinity mask to napi->config. To delegate the CPU affinity management to the core, drivers must: 1 - add a persistent napi config: netif_napi_add_config() 2 - bind an IRQ to the napi instance: netif_napi_set_irq() the core will then make sure to use re-assign affinity to the napi's IRQ. Suggested-by: Jakub Kicinski Signed-off-by: Ahmed Zaki --- include/linux/netdevice.h | 22 ++++++++++++++++++++++ net/core/dev.c | 7 ++++++- 2 files changed, 28 insertions(+), 1 deletion(-) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ecc686409161..8660de791a1a 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -350,6 +350,7 @@ struct napi_config { u64 gro_flush_timeout; u64 irq_suspend_timeout; u32 defer_hard_irqs; + cpumask_t affinity_mask; unsigned int napi_id; }; @@ -393,6 +394,7 @@ struct napi_struct { int irq; int index; struct napi_config *config; + struct irq_affinity_notify affinity_notify; }; enum { @@ -2666,10 +2668,30 @@ static inline void *netdev_priv(const struct net_device *dev) void netif_queue_set_napi(struct net_device *dev, unsigned int queue_index, enum netdev_queue_type type, struct napi_struct *napi); +static inline void +netif_napi_affinity_notify(struct irq_affinity_notify *notify, + const cpumask_t *mask) +{ + struct napi_struct *napi = + container_of(notify, struct napi_struct, affinity_notify); + + if (napi->config) + cpumask_copy(&napi->config->affinity_mask, mask); +} + +static inline void +netif_napi_affinity_release(struct kref __always_unused *ref) {} static inline void netif_napi_set_irq(struct napi_struct *napi, int irq) { napi->irq = irq; + + if (irq > 0 && napi->config) { + napi->affinity_notify.notify = netif_napi_affinity_notify; + napi->affinity_notify.release = netif_napi_affinity_release; + irq_set_affinity_notifier(irq, &napi->affinity_notify); + irq_set_affinity(irq, &napi->config->affinity_mask); + } } /* Default NAPI poll() weight diff --git a/net/core/dev.c b/net/core/dev.c index 13d00fc10f55..d58779d57994 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6843,6 +6843,8 @@ void __netif_napi_del(struct napi_struct *napi) return; if (napi->config) { + if (napi->irq > 0) + irq_set_affinity_notifier(napi->irq, NULL); napi->index = -1; napi->config = NULL; } @@ -11184,7 +11186,7 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name, { struct net_device *dev; size_t napi_config_sz; - unsigned int maxqs; + unsigned int maxqs, i; BUG_ON(strlen(name) >= sizeof(dev->name)); @@ -11280,6 +11282,9 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name, dev->napi_config = kvzalloc(napi_config_sz, GFP_KERNEL_ACCOUNT); if (!dev->napi_config) goto free_all; + for (i = 0; i < maxqs; i++) + cpumask_copy(&dev->napi_config[i].affinity_mask, + cpu_online_mask); strscpy(dev->name, name); dev->name_assign_type = name_assign_type; From patchwork Fri Dec 6 00:12:09 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ahmed Zaki X-Patchwork-Id: 13896122 X-Patchwork-Delegate: kuba@kernel.org Received: from mgamail.intel.com (mgamail.intel.com [198.175.65.12]) (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 0E0F4B652 for ; Fri, 6 Dec 2024 00:12:22 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=198.175.65.12 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1733443943; cv=none; b=STb6vj8ALuRbYab6Q5LTpWLqmvFVMWJi4ju5uUuo+MHYp8QDOXSyQi93X3wjBTpOHUlgWKT762f78MCR7XLCpwqdl3bQQooL7qsMnTH+DmHOW+5S3mWH/b0kiLxokwnt+MTdX8ddTi3Dv8W3Xca5nR5sga9DXbvbFmrcTnymr6g= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1733443943; c=relaxed/simple; bh=7ER5lWILzNrhbb1zBqrLw42qLhuTtYo0d5eiZ2qDScs=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=htRiS+ZblAAs3Gi55C6dC5qOA+lArn+IZNqfbKSD5+FMZZ2Zju1JX9r/SeMPRymRhE9fMXJSVb1T4VG/RY4fd9/fKbnipn0qomoOtkMLHGqc/eoEcBMtUVPOaVF3W6CM1r8DXMLS5WBLpiGi4u/bY/Sj0OUtqiyMIpwTY2wefJY= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=intel.com; spf=pass smtp.mailfrom=intel.com; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b=kiv9jji+; arc=none smtp.client-ip=198.175.65.12 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=intel.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=intel.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b="kiv9jji+" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1733443942; x=1764979942; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=7ER5lWILzNrhbb1zBqrLw42qLhuTtYo0d5eiZ2qDScs=; b=kiv9jji+zqIeCeWONefGUxpuZpWhOxuLZQLTEdm630x4fPz4CzZYUdIz P65R/uTaW0ATTjeefOeN9q4XZen1ChsoUR4TF/vTPwCh0fLdVd2VFkP3I NM7ssc59dJ+YlU4faBEAxaYIY0Ri3+qgHfis35arHQOJmOTBgHzoc9TrB 50nbTZHFELsT+A+v9TBRbrcUTr8oGoA/z3F8D5FGUlxBmfFM5JAHybfEc I1Smes9Oyiw0/nXVOfSRadMYeDaEz3igajzBnSM9DxHFrplTlwyrvvSGO p6RApkCVAMYGW1NQyGgu7GibhrPPeb6LwDG5YPKHUlRDUpjncEXXwoMEW g==; X-CSE-ConnectionGUID: c0kVnJFARyypZbss1ygEBg== X-CSE-MsgGUID: uZ8Qm3GyRuS2mUWRFvJ7Hw== X-IronPort-AV: E=McAfee;i="6700,10204,11277"; a="45162756" X-IronPort-AV: E=Sophos;i="6.12,211,1728975600"; d="scan'208";a="45162756" Received: from fmviesa005.fm.intel.com ([10.60.135.145]) by orvoesa104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 05 Dec 2024 16:12:22 -0800 X-CSE-ConnectionGUID: /2nalrKPQ0Spz9mYDjF9/g== X-CSE-MsgGUID: 63pIbFhbQLqtuIwvi6+GBw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.12,211,1728975600"; d="scan'208";a="98694936" Received: from ibganev-mobl.amr.corp.intel.com (HELO azaki-desk1.intel.com) ([10.125.108.131]) by fmviesa005-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 05 Dec 2024 16:12:20 -0800 From: Ahmed Zaki To: netdev@vger.kernel.org Cc: Ahmed Zaki , intel-wired-lan@lists.osuosl.org Subject: [PATCH RFC net-next 2/2] idpf: use napi's irq affinity Date: Thu, 5 Dec 2024 17:12:09 -0700 Message-ID: <20241206001209.213168-3-ahmed.zaki@intel.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20241206001209.213168-1-ahmed.zaki@intel.com> References: <20241206001209.213168-1-ahmed.zaki@intel.com> Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Patchwork-Delegate: kuba@kernel.org X-Patchwork-State: RFC Delete the driver CPU affinity info and use the core's napi config instead. Cc: intel-wired-lan@lists.osuosl.org Signed-off-by: Ahmed Zaki --- drivers/net/ethernet/intel/idpf/idpf_txrx.c | 18 ++++-------------- drivers/net/ethernet/intel/idpf/idpf_txrx.h | 6 ++---- 2 files changed, 6 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c index da2a5becf62f..ffa75a98cdc6 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c @@ -3553,8 +3553,6 @@ void idpf_vport_intr_rel(struct idpf_vport *vport) q_vector->tx = NULL; kfree(q_vector->rx); q_vector->rx = NULL; - - free_cpumask_var(q_vector->affinity_mask); } kfree(vport->q_vectors); @@ -3581,8 +3579,6 @@ static void idpf_vport_intr_rel_irq(struct idpf_vport *vport) vidx = vport->q_vector_idxs[vector]; irq_num = adapter->msix_entries[vidx].vector; - /* clear the affinity_mask in the IRQ descriptor */ - irq_set_affinity_hint(irq_num, NULL); kfree(free_irq(irq_num, q_vector)); } } @@ -3761,8 +3757,6 @@ static int idpf_vport_intr_req_irq(struct idpf_vport *vport) "Request_irq failed, error: %d\n", err); goto free_q_irqs; } - /* assign the mask for this irq */ - irq_set_affinity_hint(irq_num, q_vector->affinity_mask); } return 0; @@ -4183,12 +4177,11 @@ static void idpf_vport_intr_napi_add_all(struct idpf_vport *vport) for (v_idx = 0; v_idx < vport->num_q_vectors; v_idx++) { struct idpf_q_vector *q_vector = &vport->q_vectors[v_idx]; + int irq_num = vport->adapter->msix_entries[v_idx].vector; - netif_napi_add(vport->netdev, &q_vector->napi, napi_poll); - - /* only set affinity_mask if the CPU is online */ - if (cpu_online(v_idx)) - cpumask_set_cpu(v_idx, q_vector->affinity_mask); + netif_napi_add_config(vport->netdev, &q_vector->napi, + napi_poll, v_idx); + netif_napi_set_irq(&q_vector->napi, irq_num); } } @@ -4232,9 +4225,6 @@ int idpf_vport_intr_alloc(struct idpf_vport *vport) q_vector->rx_intr_mode = IDPF_ITR_DYNAMIC; q_vector->rx_itr_idx = VIRTCHNL2_ITR_IDX_0; - if (!zalloc_cpumask_var(&q_vector->affinity_mask, GFP_KERNEL)) - goto error; - q_vector->tx = kcalloc(txqs_per_vector, sizeof(*q_vector->tx), GFP_KERNEL); if (!q_vector->tx) diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h index 9c1fe84108ed..5efb3402b378 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h @@ -397,7 +397,6 @@ struct idpf_intr_reg { * @rx_intr_mode: Dynamic ITR or not * @rx_itr_idx: RX ITR index * @v_idx: Vector index - * @affinity_mask: CPU affinity mask */ struct idpf_q_vector { __cacheline_group_begin_aligned(read_mostly); @@ -434,13 +433,12 @@ struct idpf_q_vector { __cacheline_group_begin_aligned(cold); u16 v_idx; - cpumask_var_t affinity_mask; __cacheline_group_end_aligned(cold); }; libeth_cacheline_set_assert(struct idpf_q_vector, 112, 24 + sizeof(struct napi_struct) + 2 * sizeof(struct dim), - 8 + sizeof(cpumask_var_t)); + 8); struct idpf_rx_queue_stats { u64_stats_t packets; @@ -934,7 +932,7 @@ static inline int idpf_q_vector_to_mem(const struct idpf_q_vector *q_vector) if (!q_vector) return NUMA_NO_NODE; - cpu = cpumask_first(q_vector->affinity_mask); + cpu = cpumask_first(&q_vector->napi.config->affinity_mask); return cpu < nr_cpu_ids ? cpu_to_mem(cpu) : NUMA_NO_NODE; }