From patchwork Mon Jun 5 20:15:31 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Tom Zanussi X-Patchwork-Id: 103478 Return-Path: Delivered-To: ouuuleilei@gmail.com Received: by 2002:a59:994d:0:b0:3d9:f83d:47d9 with SMTP id k13csp2932740vqr; Mon, 5 Jun 2023 13:21:59 -0700 (PDT) X-Google-Smtp-Source: ACHHUZ4ZzVftDqXjd90mooBUsUhCtbRNYsHC2QLvjP7b0DkcXmlGrk95rAcQ0KQMiIZGIfI42Gv4 X-Received: by 2002:a05:620a:8084:b0:75b:23a0:d9d8 with SMTP id ef4-20020a05620a808400b0075b23a0d9d8mr804242qkb.46.1685996519721; Mon, 05 Jun 2023 13:21:59 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1685996519; cv=none; d=google.com; s=arc-20160816; b=gBITvAgTyhFElSuAKIKcMi4qmvDVZcbLyuK0aSJlWMMLqNAokTkKycsReeF3rhJvtW lNAjG1ehqgcEPwz2zt7ofkIYIsEmPIiidf1tqtDDlfn1SEMbUiEbfo9DivIPmd1JOc1r IRr9LLpqppRKUovcnhyEs93BN93fmCwArzT5HTDAAwwgHvJ3KCX0MijBlI7nyZbjIoWg Idxn7uTb1AHegrDgMbMI8hrmF4UAl1wFU9nIAkBcgnCOjZouDQsCjQj8KJQWnAxeaoPb VwXVniJWPozzcyaf/rCIncIql5dBiGkau8e/xohjZS5gCZB1fGtHDjHuWVwbDzZKG7CV 57iQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:content-transfer-encoding:mime-version :references:in-reply-to:message-id:date:subject:cc:to:from :dkim-signature; bh=WyuhP9EA8ysV5ghSTDxeb4n2sNpKjHxEMUR/+PfwmAQ=; b=nGs6IP1mjMGxgn71mvaJTR5Rck5HQtXX9vWCC375dYL7kK0tIb1QMelv3+WXlG76vq FuUeylnf/aesqEa3fPKYaIuaPE2KWfc7+HP1B57EjwaR8sOnfm5ONp7smHWhYJPqjr0R vNz+VFq1tSCaJvZoRLl0YjtySWq9IavWxIDoMGtVDsQ/1EIxdGXEf/VuFmR/WqQMKiPt QAAUDIlgIfgtL9eMsBSamFNNltdCZdqu7gn/Smp8dhn8Gj770t8Lk+Zg5QUgRbn9BsOv /dHiSVPoxMTQtPwdg9H+t7iMFtqhJoMOqlofmUQDlZW/no3dYIm4xuDhHXpi2QSEihsa HJBw== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@intel.com header.s=Intel header.b=L5zv9vo8; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=intel.com Received: from out1.vger.email (out1.vger.email. [2620:137:e000::1:20]) by mx.google.com with ESMTP id f20-20020a05620a15b400b0075cf4272692si354991qkk.586.2023.06.05.13.21.45; Mon, 05 Jun 2023 13:21:59 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20; Authentication-Results: mx.google.com; dkim=pass header.i=@intel.com header.s=Intel header.b=L5zv9vo8; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=intel.com Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233111AbjFEURM (ORCPT + 99 others); Mon, 5 Jun 2023 16:17:12 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:44274 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232470AbjFEUQg (ORCPT ); Mon, 5 Jun 2023 16:16:36 -0400 Received: from mga14.intel.com (mga14.intel.com [192.55.52.115]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7CDDFE43; Mon, 5 Jun 2023 13:16:16 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1685996176; x=1717532176; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=gwTTsreHzCk9i2/NeHjW08Gd30ZFhWIRVlEyw4/AHOs=; b=L5zv9vo8GwX6JpizpxQ9cnuqrLMmtsmraJ+JId+qcOfMhawLpZzXuXQX vwhyEkPpGwNK81SAXRfTapIk+LVkKsOBCrf0UlxGjzZBXdWBCoAEjg9EB kb8ODBKgWyy3Ic2PXBjlPfVCcVp6e4ZQdoF8zOsEqdshS0KOJbNl+CHx+ ENUa8kHpe/vsJombhQmzOnpA2ITuKsIPIy5bFYpvZMzJRBVCBjp4iUzCn YwEuGEjcrjZELpo0ncFEZOvRKLIgLHmYRZWq1xsS8puVfm7WtGWhFF/Vv iVVnT1GNtZRjaE71wdwFcbVPx9U3RGk/sAUBFCLb+Xsbgh5ju/AANiVBU Q==; X-IronPort-AV: E=McAfee;i="6600,9927,10732"; a="356480421" X-IronPort-AV: E=Sophos;i="6.00,218,1681196400"; d="scan'208";a="356480421" Received: from fmsmga007.fm.intel.com ([10.253.24.52]) by fmsmga103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 05 Jun 2023 13:16:00 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10732"; a="711934497" X-IronPort-AV: E=Sophos;i="6.00,218,1681196400"; d="scan'208";a="711934497" Received: from rkbrewer-mobl1.amr.corp.intel.com (HELO tzanussi-mobl1.intel.com) ([10.212.33.16]) by fmsmga007-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 05 Jun 2023 13:15:59 -0700 From: Tom Zanussi To: herbert@gondor.apana.org.au, davem@davemloft.net, fenghua.yu@intel.com, vkoul@kernel.org Cc: dave.jiang@intel.com, tony.luck@intel.com, wajdi.k.feghali@intel.com, james.guilford@intel.com, kanchana.p.sridhar@intel.com, vinodh.gopal@intel.com, giovanni.cabiddu@intel.com, linux-kernel@vger.kernel.org, linux-crypto@vger.kernel.org, dmaengine@vger.kernel.org Subject: [PATCH v6 10/15] crypto: iaa - Add per-cpu workqueue table with rebalancing Date: Mon, 5 Jun 2023 15:15:31 -0500 Message-Id: <20230605201536.738396-11-tom.zanussi@linux.intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230605201536.738396-1-tom.zanussi@linux.intel.com> References: <20230605201536.738396-1-tom.zanussi@linux.intel.com> MIME-Version: 1.0 X-Spam-Status: No, score=-4.3 required=5.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_EF,RCVD_IN_DNSWL_MED,SPF_HELO_NONE, SPF_NONE,T_SCC_BODY_TEXT_LINE autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org X-getmail-retrieved-from-mailbox: =?utf-8?q?INBOX?= X-GMAIL-THRID: =?utf-8?q?1767895486588856613?= X-GMAIL-MSGID: =?utf-8?q?1767895486588856613?= The iaa compression/decompression algorithms in later patches need a way to retrieve an appropriate IAA workqueue depending on how close the associated IAA device is to the current cpu. For this purpose, add a per-cpu array of workqueues such that an appropriate workqueue can be retrieved by simply accessing the per-cpu array. Whenever a new workqueue is bound to or unbound from the iaa_crypto driver, the available workqueues are 'rebalanced' such that work submitted from a particular CPU is given to the most appropriate workqueue available. There currently isn't any way for the user to tweak the way this is done internally - if necessary, knobs can be added later for that purpose. Current best practice is to configure and bind at least one workqueue for each IAA device, but as long as there is at least one workqueue configured and bound to any IAA device in the system, the iaa_crypto driver will work, albeit most likely not as efficiently. [ Based on work originally by George Powley, Jing Lin and Kyung Min Park ] Signed-off-by: Tom Zanussi --- drivers/crypto/intel/iaa/iaa_crypto.h | 7 + drivers/crypto/intel/iaa/iaa_crypto_main.c | 221 +++++++++++++++++++++ 2 files changed, 228 insertions(+) diff --git a/drivers/crypto/intel/iaa/iaa_crypto.h b/drivers/crypto/intel/iaa/iaa_crypto.h index 5d1fff7f4b8e..c25546fa87f7 100644 --- a/drivers/crypto/intel/iaa/iaa_crypto.h +++ b/drivers/crypto/intel/iaa/iaa_crypto.h @@ -27,4 +27,11 @@ struct iaa_device { struct list_head wqs; }; +struct wq_table_entry { + struct idxd_wq **wqs; + int max_wqs; + int n_wqs; + int cur_wq; +}; + #endif diff --git a/drivers/crypto/intel/iaa/iaa_crypto_main.c b/drivers/crypto/intel/iaa/iaa_crypto_main.c index ab1bd0b00d89..670e9c3c8f9a 100644 --- a/drivers/crypto/intel/iaa/iaa_crypto_main.c +++ b/drivers/crypto/intel/iaa/iaa_crypto_main.c @@ -22,6 +22,46 @@ /* number of iaa instances probed */ static unsigned int nr_iaa; +static unsigned int nr_cpus; +static unsigned int nr_nodes; +static unsigned int nr_cpus_per_node; + +/* Number of physical cpus sharing each iaa instance */ +static unsigned int cpus_per_iaa; + +/* Per-cpu lookup table for balanced wqs */ +static struct wq_table_entry __percpu *wq_table; + +static void wq_table_add(int cpu, struct idxd_wq *wq) +{ + struct wq_table_entry *entry = per_cpu_ptr(wq_table, cpu); + + if (WARN_ON(entry->n_wqs == entry->max_wqs)) + return; + + entry->wqs[entry->n_wqs++] = wq; + + pr_debug("%s: added iaa wq %d.%d to idx %d of cpu %d\n", __func__, + entry->wqs[entry->n_wqs - 1]->idxd->id, + entry->wqs[entry->n_wqs - 1]->id, entry->n_wqs - 1, cpu); +} + +static void wq_table_free_entry(int cpu) +{ + struct wq_table_entry *entry = per_cpu_ptr(wq_table, cpu); + + kfree(entry->wqs); + memset(entry, 0, sizeof(*entry)); +} + +static void wq_table_clear_entry(int cpu) +{ + struct wq_table_entry *entry = per_cpu_ptr(wq_table, cpu); + + entry->n_wqs = 0; + entry->cur_wq = 0; + memset(entry->wqs, 0, entry->max_wqs * sizeof(struct idxd_wq *)); +} static LIST_HEAD(iaa_devices); static DEFINE_MUTEX(iaa_devices_lock); @@ -141,6 +181,53 @@ static void del_iaa_wq(struct iaa_device *iaa_device, struct idxd_wq *wq) } } +static void clear_wq_table(void) +{ + int cpu; + + for (cpu = 0; cpu < nr_cpus; cpu++) + wq_table_clear_entry(cpu); + + pr_debug("cleared wq table\n"); +} + +static void free_wq_table(void) +{ + int cpu; + + for (cpu = 0; cpu < nr_cpus; cpu++) + wq_table_free_entry(cpu); + + free_percpu(wq_table); + + pr_debug("freed wq table\n"); +} + +static int alloc_wq_table(int max_wqs) +{ + struct wq_table_entry *entry; + int cpu; + + wq_table = alloc_percpu(struct wq_table_entry); + if (!wq_table) + return -ENOMEM; + + for (cpu = 0; cpu < nr_cpus; cpu++) { + entry = per_cpu_ptr(wq_table, cpu); + entry->wqs = kzalloc(GFP_KERNEL, max_wqs * sizeof(struct wq *)); + if (!entry->wqs) { + free_wq_table(); + return -ENOMEM; + } + + entry->max_wqs = max_wqs; + } + + pr_debug("initialized wq table\n"); + + return 0; +} + static int save_iaa_wq(struct idxd_wq *wq) { struct iaa_device *iaa_device, *found = NULL; @@ -195,6 +282,8 @@ static int save_iaa_wq(struct idxd_wq *wq) return -EINVAL; idxd_wq_get(wq); + + cpus_per_iaa = (nr_nodes * nr_cpus_per_node) / nr_iaa; out: return 0; } @@ -210,6 +299,116 @@ static void remove_iaa_wq(struct idxd_wq *wq) break; } } + + if (nr_iaa) + cpus_per_iaa = (nr_nodes * nr_cpus_per_node) / nr_iaa; + else + cpus_per_iaa = 0; +} + +static int wq_table_add_wqs(int iaa, int cpu) +{ + struct iaa_device *iaa_device, *found_device = NULL; + int ret = 0, cur_iaa = 0, n_wqs_added = 0; + struct idxd_device *idxd; + struct iaa_wq *iaa_wq; + struct pci_dev *pdev; + struct device *dev; + + list_for_each_entry(iaa_device, &iaa_devices, list) { + idxd = iaa_device->idxd; + pdev = idxd->pdev; + dev = &pdev->dev; + + if (cur_iaa != iaa) { + cur_iaa++; + continue; + } + + found_device = iaa_device; + dev_dbg(dev, "getting wq from iaa_device %d, cur_iaa %d\n", + found_device->idxd->id, cur_iaa); + break; + } + + if (!found_device) { + found_device = list_first_entry_or_null(&iaa_devices, + struct iaa_device, list); + if (!found_device) { + pr_debug("couldn't find any iaa devices with wqs!\n"); + ret = -EINVAL; + goto out; + } + cur_iaa = 0; + + idxd = found_device->idxd; + pdev = idxd->pdev; + dev = &pdev->dev; + dev_dbg(dev, "getting wq from only iaa_device %d, cur_iaa %d\n", + found_device->idxd->id, cur_iaa); + } + + list_for_each_entry(iaa_wq, &found_device->wqs, list) { + wq_table_add(cpu, iaa_wq->wq); + pr_debug("rebalance: added wq for cpu=%d: iaa wq %d.%d\n", + cpu, iaa_wq->wq->idxd->id, iaa_wq->wq->id); + n_wqs_added++; + }; + + if (!n_wqs_added) { + pr_debug("couldn't find any iaa wqs!\n"); + ret = -EINVAL; + goto out; + } +out: + return ret; +} + +/* + * Rebalance the wq table so that given a cpu, it's easy to find the + * closest IAA instance. The idea is to try to choose the most + * appropriate IAA instance for a caller and spread available + * workqueues around to clients. + */ +static void rebalance_wq_table(void) +{ + const struct cpumask *node_cpus; + int node, cpu, iaa = -1; + + if (nr_iaa == 0) + return; + + pr_debug("rebalance: nr_nodes=%d, nr_cpus %d, nr_iaa %d, cpus_per_iaa %d\n", + nr_nodes, nr_cpus, nr_iaa, cpus_per_iaa); + + clear_wq_table(); + + if (nr_iaa == 1) { + for (cpu = 0; cpu < nr_cpus; cpu++) { + if (WARN_ON(wq_table_add_wqs(0, cpu))) { + pr_debug("could not add any wqs for iaa 0 to cpu %d!\n", cpu); + return; + } + } + + return; + } + + for_each_online_node(node) { + node_cpus = cpumask_of_node(node); + + for (cpu = 0; cpu < nr_cpus_per_node; cpu++) { + int node_cpu = cpumask_nth(cpu, node_cpus); + + if ((cpu % cpus_per_iaa) == 0) + iaa++; + + if (WARN_ON(wq_table_add_wqs(iaa, node_cpu))) { + pr_debug("could not add any wqs for iaa %d to cpu %d!\n", iaa, cpu); + return; + } + } + } } static int iaa_crypto_probe(struct idxd_dev *idxd_dev) @@ -218,6 +417,7 @@ static int iaa_crypto_probe(struct idxd_dev *idxd_dev) struct idxd_device *idxd = wq->idxd; struct idxd_driver_data *data = idxd->data; struct device *dev = &idxd_dev->conf_dev; + bool first_wq = false; int ret = 0; if (idxd->state != IDXD_DEV_ENABLED) @@ -248,10 +448,19 @@ static int iaa_crypto_probe(struct idxd_dev *idxd_dev) mutex_lock(&iaa_devices_lock); + if (list_empty(&iaa_devices)) { + ret = alloc_wq_table(wq->idxd->max_wqs); + if (ret) + goto err_alloc; + first_wq = true; + } + ret = save_iaa_wq(wq); if (ret) goto err_save; + rebalance_wq_table(); + mutex_unlock(&iaa_devices_lock); out: mutex_unlock(&wq->wq_lock); @@ -259,6 +468,10 @@ static int iaa_crypto_probe(struct idxd_dev *idxd_dev) return ret; err_save: + if (first_wq) + free_wq_table(); +err_alloc: + mutex_unlock(&iaa_devices_lock); drv_disable_wq(wq); err: wq->type = IDXD_WQT_NONE; @@ -277,6 +490,10 @@ static void iaa_crypto_remove(struct idxd_dev *idxd_dev) remove_iaa_wq(wq); drv_disable_wq(wq); + rebalance_wq_table(); + + if (nr_iaa == 0) + free_wq_table(); mutex_unlock(&iaa_devices_lock); mutex_unlock(&wq->wq_lock); @@ -298,6 +515,10 @@ static int __init iaa_crypto_init_module(void) { int ret = 0; + nr_cpus = num_online_cpus(); + nr_nodes = num_online_nodes(); + nr_cpus_per_node = nr_cpus / nr_nodes; + ret = idxd_driver_register(&iaa_crypto_driver); if (ret) { pr_debug("IAA wq sub-driver registration failed\n");