From patchwork Fri Jan 25 09:53:43 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ming Lei X-Patchwork-Id: 10780879 X-Patchwork-Delegate: bhelgaas@google.com Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id AD7B013B5 for ; Fri, 25 Jan 2019 09:54:19 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 9C7B62CC85 for ; Fri, 25 Jan 2019 09:54:19 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 8F7EE2E6BE; Fri, 25 Jan 2019 09:54:19 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.9 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI, RCVD_IN_DNSWL_HI autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 8B8E62CC85 for ; Fri, 25 Jan 2019 09:54:18 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729112AbfAYJyM (ORCPT ); Fri, 25 Jan 2019 04:54:12 -0500 Received: from mx1.redhat.com ([209.132.183.28]:47404 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727975AbfAYJyM (ORCPT ); Fri, 25 Jan 2019 04:54:12 -0500 Received: from smtp.corp.redhat.com (int-mx05.intmail.prod.int.phx2.redhat.com [10.5.11.15]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id BCB4413268F; Fri, 25 Jan 2019 09:54:11 +0000 (UTC) Received: from localhost (ovpn-8-28.pek2.redhat.com [10.72.8.28]) by smtp.corp.redhat.com (Postfix) with ESMTP id 299EA5B6B0; Fri, 25 Jan 2019 09:54:06 +0000 (UTC) From: Ming Lei To: Christoph Hellwig , Bjorn Helgaas , Thomas Gleixner Cc: Jens Axboe , linux-block@vger.kernel.org, Sagi Grimberg , linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org, Ming Lei Subject: [PATCH 1/5] genirq/affinity: move allocation of 'node_to_cpumask' to irq_build_affinity_masks Date: Fri, 25 Jan 2019 17:53:43 +0800 Message-Id: <20190125095347.17950-2-ming.lei@redhat.com> In-Reply-To: <20190125095347.17950-1-ming.lei@redhat.com> References: <20190125095347.17950-1-ming.lei@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.15 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.38]); Fri, 25 Jan 2019 09:54:11 +0000 (UTC) Sender: linux-pci-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-pci@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP 'node_to_cpumask' is just one temparay variable for irq_build_affinity_masks(), so move it into irq_build_affinity_masks(). No functioanl change. Signed-off-by: Ming Lei Reviewed-by: Bjorn Helgaas --- kernel/irq/affinity.c | 27 +++++++++++++-------------- 1 file changed, 13 insertions(+), 14 deletions(-) diff --git a/kernel/irq/affinity.c b/kernel/irq/affinity.c index 45b68b4ea48b..118b66d64a53 100644 --- a/kernel/irq/affinity.c +++ b/kernel/irq/affinity.c @@ -175,18 +175,22 @@ static int __irq_build_affinity_masks(const struct irq_affinity *affd, */ static int irq_build_affinity_masks(const struct irq_affinity *affd, int startvec, int numvecs, int firstvec, - cpumask_var_t *node_to_cpumask, struct irq_affinity_desc *masks) { int curvec = startvec, nr_present, nr_others; int ret = -ENOMEM; cpumask_var_t nmsk, npresmsk; + cpumask_var_t *node_to_cpumask; if (!zalloc_cpumask_var(&nmsk, GFP_KERNEL)) return ret; if (!zalloc_cpumask_var(&npresmsk, GFP_KERNEL)) - goto fail; + goto fail_nmsk; + + node_to_cpumask = alloc_node_to_cpumask(); + if (!node_to_cpumask) + goto fail_npresmsk; ret = 0; /* Stabilize the cpumasks */ @@ -217,9 +221,12 @@ static int irq_build_affinity_masks(const struct irq_affinity *affd, if (nr_present < numvecs) WARN_ON(nr_present + nr_others < numvecs); + free_node_to_cpumask(node_to_cpumask); + + fail_npresmsk: free_cpumask_var(npresmsk); - fail: + fail_nmsk: free_cpumask_var(nmsk); return ret; } @@ -236,7 +243,6 @@ irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) { int affvecs = nvecs - affd->pre_vectors - affd->post_vectors; int curvec, usedvecs; - cpumask_var_t *node_to_cpumask; struct irq_affinity_desc *masks = NULL; int i, nr_sets; @@ -247,13 +253,9 @@ irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) if (nvecs == affd->pre_vectors + affd->post_vectors) return NULL; - node_to_cpumask = alloc_node_to_cpumask(); - if (!node_to_cpumask) - return NULL; - masks = kcalloc(nvecs, sizeof(*masks), GFP_KERNEL); if (!masks) - goto outnodemsk; + return NULL; /* Fill out vectors at the beginning that don't need affinity */ for (curvec = 0; curvec < affd->pre_vectors; curvec++) @@ -271,11 +273,10 @@ irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) int ret; ret = irq_build_affinity_masks(affd, curvec, this_vecs, - curvec, node_to_cpumask, masks); + curvec, masks); if (ret) { kfree(masks); - masks = NULL; - goto outnodemsk; + return NULL; } curvec += this_vecs; usedvecs += this_vecs; @@ -293,8 +294,6 @@ irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) for (i = affd->pre_vectors; i < nvecs - affd->post_vectors; i++) masks[i].is_managed = 1; -outnodemsk: - free_node_to_cpumask(node_to_cpumask); return masks; } From patchwork Fri Jan 25 09:53:44 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ming Lei X-Patchwork-Id: 10780883 X-Patchwork-Delegate: bhelgaas@google.com Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id DCCA613BF for ; Fri, 25 Jan 2019 09:54:22 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id CAAB82CC85 for ; Fri, 25 Jan 2019 09:54:22 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id BE8BB2E6B8; Fri, 25 Jan 2019 09:54:22 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.9 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI, RCVD_IN_DNSWL_HI autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 69DAA2E6B8 for ; Fri, 25 Jan 2019 09:54:22 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729145AbfAYJyV (ORCPT ); Fri, 25 Jan 2019 04:54:21 -0500 Received: from mx1.redhat.com ([209.132.183.28]:44418 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726888AbfAYJyV (ORCPT ); Fri, 25 Jan 2019 04:54:21 -0500 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.phx2.redhat.com [10.5.11.16]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id BCC61C057E22; Fri, 25 Jan 2019 09:54:20 +0000 (UTC) Received: from localhost (ovpn-8-28.pek2.redhat.com [10.72.8.28]) by smtp.corp.redhat.com (Postfix) with ESMTP id E5F654A4; Fri, 25 Jan 2019 09:54:16 +0000 (UTC) From: Ming Lei To: Christoph Hellwig , Bjorn Helgaas , Thomas Gleixner Cc: Jens Axboe , linux-block@vger.kernel.org, Sagi Grimberg , linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org, Ming Lei Subject: [PATCH 2/5] genirq/affinity: allow driver to setup managed IRQ's affinity Date: Fri, 25 Jan 2019 17:53:44 +0800 Message-Id: <20190125095347.17950-3-ming.lei@redhat.com> In-Reply-To: <20190125095347.17950-1-ming.lei@redhat.com> References: <20190125095347.17950-1-ming.lei@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.16 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.31]); Fri, 25 Jan 2019 09:54:20 +0000 (UTC) Sender: linux-pci-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-pci@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP This patch introduces callback of .setup_affinity into 'struct irq_affinity', so that: 1) allow drivers to customize the affinity for managed IRQ, for example, now NVMe has special requirement for read queues & poll queues 2) 6da4b3ab9a6e9 ("genirq/affinity: Add support for allocating interrupt sets") makes pci_alloc_irq_vectors_affinity() a bit difficult to use for allocating interrupt sets: 'max_vecs' is required to same with 'min_vecs'. With this patch, driver can implement their own .setup_affinity to customize the affinity, then the above thing can be solved easily. Signed-off-by: Ming Lei --- include/linux/interrupt.h | 26 +++++++++++++++++--------- kernel/irq/affinity.c | 6 ++++++ 2 files changed, 23 insertions(+), 9 deletions(-) diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index c672f34235e7..f6cea778cf50 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -242,30 +242,38 @@ struct irq_affinity_notify { }; /** + * struct irq_affinity_desc - Interrupt affinity descriptor + * @mask: cpumask to hold the affinity assignment + */ +struct irq_affinity_desc { + struct cpumask mask; + unsigned int is_managed : 1; +}; + +/** * struct irq_affinity - Description for automatic irq affinity assignements * @pre_vectors: Don't apply affinity to @pre_vectors at beginning of * the MSI(-X) vector space * @post_vectors: Don't apply affinity to @post_vectors at end of * the MSI(-X) vector space + * @setup_affinity: Use driver's method to setup irq vectors affinity, + * and driver has to handle pre_vectors & post_vectors + * correctly, set 'is_managed' flag correct too + * @priv: Private data of @setup_affinity * @nr_sets: Length of passed in *sets array * @sets: Number of affinitized sets */ struct irq_affinity { int pre_vectors; int post_vectors; + int (*setup_affinity)(const struct irq_affinity *, + struct irq_affinity_desc *, + unsigned int); + void *priv; int nr_sets; int *sets; }; -/** - * struct irq_affinity_desc - Interrupt affinity descriptor - * @mask: cpumask to hold the affinity assignment - */ -struct irq_affinity_desc { - struct cpumask mask; - unsigned int is_managed : 1; -}; - #if defined(CONFIG_SMP) extern cpumask_var_t irq_default_affinity; diff --git a/kernel/irq/affinity.c b/kernel/irq/affinity.c index 118b66d64a53..7b77cbdf739c 100644 --- a/kernel/irq/affinity.c +++ b/kernel/irq/affinity.c @@ -257,6 +257,12 @@ irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) if (!masks) return NULL; + if (affd->setup_affinity) { + if (affd->setup_affinity(affd, masks, nvecs)) + return NULL; + return masks; + } + /* Fill out vectors at the beginning that don't need affinity */ for (curvec = 0; curvec < affd->pre_vectors; curvec++) cpumask_copy(&masks[curvec].mask, irq_default_affinity); From patchwork Fri Jan 25 09:53:45 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ming Lei X-Patchwork-Id: 10780887 X-Patchwork-Delegate: bhelgaas@google.com Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 7F02113BF for ; Fri, 25 Jan 2019 09:54:32 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 6F9302CC85 for ; Fri, 25 Jan 2019 09:54:32 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 63CB62E6F7; Fri, 25 Jan 2019 09:54:32 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.9 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI, RCVD_IN_DNSWL_HI autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 0FC562E6B8 for ; Fri, 25 Jan 2019 09:54:32 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729210AbfAYJyb (ORCPT ); Fri, 25 Jan 2019 04:54:31 -0500 Received: from mx1.redhat.com ([209.132.183.28]:51504 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726888AbfAYJyb (ORCPT ); Fri, 25 Jan 2019 04:54:31 -0500 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.phx2.redhat.com [10.5.11.14]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id A13F6A7FDE; Fri, 25 Jan 2019 09:54:30 +0000 (UTC) Received: from localhost (ovpn-8-28.pek2.redhat.com [10.72.8.28]) by smtp.corp.redhat.com (Postfix) with ESMTP id 9350F1834E; Fri, 25 Jan 2019 09:54:24 +0000 (UTC) From: Ming Lei To: Christoph Hellwig , Bjorn Helgaas , Thomas Gleixner Cc: Jens Axboe , linux-block@vger.kernel.org, Sagi Grimberg , linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org, Ming Lei Subject: [PATCH 3/5] genirq/affinity: introduce irq_build_affinity() Date: Fri, 25 Jan 2019 17:53:45 +0800 Message-Id: <20190125095347.17950-4-ming.lei@redhat.com> In-Reply-To: <20190125095347.17950-1-ming.lei@redhat.com> References: <20190125095347.17950-1-ming.lei@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.14 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.28]); Fri, 25 Jan 2019 09:54:30 +0000 (UTC) Sender: linux-pci-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-pci@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Drivers may use this API to build customized irq affinity, one example is NVMe, which needs to build multiple irq sets, on each of which all CPUs are spread. Signed-off-by: Ming Lei --- include/linux/interrupt.h | 12 ++++++++++++ kernel/irq/affinity.c | 27 +++++++++++++++++++++++++++ 2 files changed, 39 insertions(+) diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index f6cea778cf50..b820b07f3b55 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -323,6 +323,10 @@ irq_set_affinity_notifier(unsigned int irq, struct irq_affinity_notify *notify); struct irq_affinity_desc * irq_create_affinity_masks(int nvec, const struct irq_affinity *affd); +int irq_build_affinity(const struct irq_affinity *affd, int startvec, + int numvecs, int firstvec, + struct irq_affinity_desc *masks, int nmasks); + int irq_calc_affinity_vectors(int minvec, int maxvec, const struct irq_affinity *affd); #else /* CONFIG_SMP */ @@ -368,6 +372,14 @@ irq_calc_affinity_vectors(int minvec, int maxvec, const struct irq_affinity *aff return maxvec; } +static inline int +irq_build_affinity(const struct irq_affinity *affd, int startvec, + int numvecs, int firstvec, + struct irq_affinity_desc *masks, int nmasks) +{ + return 0; +} + #endif /* CONFIG_SMP */ /* diff --git a/kernel/irq/affinity.c b/kernel/irq/affinity.c index 7b77cbdf739c..524fdcda9f85 100644 --- a/kernel/irq/affinity.c +++ b/kernel/irq/affinity.c @@ -232,6 +232,33 @@ static int irq_build_affinity_masks(const struct irq_affinity *affd, } /** + * irq_build_affinity - build affinity masks for multiqueue spreading + * @affd: Description of the affinity requirements + * @startvec: The start vector for building affinity masks + * @numvec: The number of vectors is needed for building affinity + * @firstvec: It is the IRQ vector which we jump to for continue spread + * after the last vector(@startvec + @numvec - 1) is built. + * @masks: The mask array for storing the affinity masks + * @nmasks: The total number of @masks + * + * Both @startvec and @firstvec are relative to the 1st irq vectorc + * allocated to the device. + * + * Returns 0 if affinty masks is built successfully. + */ +int irq_build_affinity(const struct irq_affinity *affd, int startvec, + int numvecs, int firstvec, + struct irq_affinity_desc *masks, int nmasks) +{ + if (startvec >= nmasks || firstvec >= nmasks || numvecs > nmasks) + return -EINVAL; + + return irq_build_affinity_masks(affd, startvec, numvecs, firstvec, + masks); +} +EXPORT_SYMBOL_GPL(irq_build_affinity); + +/** * irq_create_affinity_masks - Create affinity masks for multiqueue spreading * @nvecs: The total number of vectors * @affd: Description of the affinity requirements From patchwork Fri Jan 25 09:53:46 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ming Lei X-Patchwork-Id: 10780891 X-Patchwork-Delegate: bhelgaas@google.com Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 3E56C13BF for ; Fri, 25 Jan 2019 09:54:45 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 2DE8D2CC85 for ; Fri, 25 Jan 2019 09:54:45 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 213102E6F7; Fri, 25 Jan 2019 09:54:45 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.9 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI, RCVD_IN_DNSWL_HI autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id ACA0A2E6B8 for ; Fri, 25 Jan 2019 09:54:44 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729044AbfAYJyn (ORCPT ); Fri, 25 Jan 2019 04:54:43 -0500 Received: from mx1.redhat.com ([209.132.183.28]:40812 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728459AbfAYJyn (ORCPT ); Fri, 25 Jan 2019 04:54:43 -0500 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.13]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 515E2C05FFE0; Fri, 25 Jan 2019 09:54:43 +0000 (UTC) Received: from localhost (ovpn-8-28.pek2.redhat.com [10.72.8.28]) by smtp.corp.redhat.com (Postfix) with ESMTP id 3A26F608C3; Fri, 25 Jan 2019 09:54:33 +0000 (UTC) From: Ming Lei To: Christoph Hellwig , Bjorn Helgaas , Thomas Gleixner Cc: Jens Axboe , linux-block@vger.kernel.org, Sagi Grimberg , linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org, Ming Lei Subject: [PATCH 4/5] nvme-pci: simplify nvme_setup_irqs() via .setup_affinity callback Date: Fri, 25 Jan 2019 17:53:46 +0800 Message-Id: <20190125095347.17950-5-ming.lei@redhat.com> In-Reply-To: <20190125095347.17950-1-ming.lei@redhat.com> References: <20190125095347.17950-1-ming.lei@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.13 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.32]); Fri, 25 Jan 2019 09:54:43 +0000 (UTC) Sender: linux-pci-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-pci@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Use the callback of .setup_affinity() to re-caculate number of queues, and build irqs affinity with help of irq_build_affinity(). Then nvme_setup_irqs() gets simplified a lot. Signed-off-by: Ming Lei --- drivers/nvme/host/pci.c | 97 ++++++++++++++++++++++++------------------------- 1 file changed, 48 insertions(+), 49 deletions(-) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index 9bc585415d9b..24496de0a29b 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -2078,17 +2078,58 @@ static void nvme_calc_io_queues(struct nvme_dev *dev, unsigned int irq_queues) } } +static int nvme_setup_affinity(const struct irq_affinity *affd, + struct irq_affinity_desc *masks, + unsigned int nmasks) +{ + struct nvme_dev *dev = affd->priv; + int affvecs = nmasks - affd->pre_vectors - affd->post_vectors; + int curvec, usedvecs; + int i; + + nvme_calc_io_queues(dev, nmasks); + + /* Fill out vectors at the beginning that don't need affinity */ + for (curvec = 0; curvec < affd->pre_vectors; curvec++) + cpumask_copy(&masks[curvec].mask, cpu_possible_mask); + + for (i = 0, usedvecs = 0; i < HCTX_TYPE_POLL; i++) { + int this_vecs = dev->io_queues[i]; + int ret; + + if (!this_vecs) + break; + + ret = irq_build_affinity(affd, curvec, this_vecs, curvec, + masks, nmasks); + if (ret) + return ret; + + curvec += this_vecs; + usedvecs += this_vecs; + } + + /* Fill out vectors at the end that don't need affinity */ + curvec = affd->pre_vectors + min(usedvecs, affvecs); + for (; curvec < nmasks; curvec++) + cpumask_copy(&masks[curvec].mask, cpu_possible_mask); + + /* Mark the managed interrupts */ + for (i = affd->pre_vectors; i < nmasks - affd->post_vectors; i++) + masks[i].is_managed = 1; + + return 0; +} + static int nvme_setup_irqs(struct nvme_dev *dev, unsigned int nr_io_queues) { struct pci_dev *pdev = to_pci_dev(dev->dev); - int irq_sets[2]; struct irq_affinity affd = { .pre_vectors = 1, - .nr_sets = ARRAY_SIZE(irq_sets), - .sets = irq_sets, + .setup_affinity = nvme_setup_affinity, + .priv = dev, }; - int result = 0; - unsigned int irq_queues, this_p_queues; + int result, irq_queues, this_p_queues; /* * Poll queues don't need interrupts, but we need at least one IO @@ -2103,50 +2144,8 @@ static int nvme_setup_irqs(struct nvme_dev *dev, unsigned int nr_io_queues) } dev->io_queues[HCTX_TYPE_POLL] = this_p_queues; - /* - * For irq sets, we have to ask for minvec == maxvec. This passes - * any reduction back to us, so we can adjust our queue counts and - * IRQ vector needs. - */ - do { - nvme_calc_io_queues(dev, irq_queues); - irq_sets[0] = dev->io_queues[HCTX_TYPE_DEFAULT]; - irq_sets[1] = dev->io_queues[HCTX_TYPE_READ]; - if (!irq_sets[1]) - affd.nr_sets = 1; - - /* - * If we got a failure and we're down to asking for just - * 1 + 1 queues, just ask for a single vector. We'll share - * that between the single IO queue and the admin queue. - * Otherwise, we assign one independent vector to admin queue. - */ - if (irq_queues > 1) - irq_queues = irq_sets[0] + irq_sets[1] + 1; - - result = pci_alloc_irq_vectors_affinity(pdev, irq_queues, - irq_queues, - PCI_IRQ_ALL_TYPES | PCI_IRQ_AFFINITY, &affd); - - /* - * Need to reduce our vec counts. If we get ENOSPC, the - * platform should support mulitple vecs, we just need - * to decrease our ask. If we get EINVAL, the platform - * likely does not. Back down to ask for just one vector. - */ - if (result == -ENOSPC) { - irq_queues--; - if (!irq_queues) - return result; - continue; - } else if (result == -EINVAL) { - irq_queues = 1; - continue; - } else if (result <= 0) - return -EIO; - break; - } while (1); - + result = pci_alloc_irq_vectors_affinity(pdev, 1, irq_queues, + PCI_IRQ_ALL_TYPES | PCI_IRQ_AFFINITY, &affd); return result; } From patchwork Fri Jan 25 09:53:47 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ming Lei X-Patchwork-Id: 10780895 X-Patchwork-Delegate: bhelgaas@google.com Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id EF74413B5 for ; Fri, 25 Jan 2019 09:54:53 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id DEE8B2CC85 for ; Fri, 25 Jan 2019 09:54:53 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id D2F162E6F7; Fri, 25 Jan 2019 09:54:53 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.9 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI, RCVD_IN_DNSWL_HI autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 703E92E6B8 for ; Fri, 25 Jan 2019 09:54:53 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727975AbfAYJyw (ORCPT ); Fri, 25 Jan 2019 04:54:52 -0500 Received: from mx1.redhat.com ([209.132.183.28]:40930 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726761AbfAYJyw (ORCPT ); Fri, 25 Jan 2019 04:54:52 -0500 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id A2A85C0C4265; Fri, 25 Jan 2019 09:54:51 +0000 (UTC) Received: from localhost (ovpn-8-28.pek2.redhat.com [10.72.8.28]) by smtp.corp.redhat.com (Postfix) with ESMTP id 5B75C6B476; Fri, 25 Jan 2019 09:54:45 +0000 (UTC) From: Ming Lei To: Christoph Hellwig , Bjorn Helgaas , Thomas Gleixner Cc: Jens Axboe , linux-block@vger.kernel.org, Sagi Grimberg , linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org, Ming Lei Subject: [PATCH 5/5] genirq/affinity: remove support for allocating interrupt sets Date: Fri, 25 Jan 2019 17:53:47 +0800 Message-Id: <20190125095347.17950-6-ming.lei@redhat.com> In-Reply-To: <20190125095347.17950-1-ming.lei@redhat.com> References: <20190125095347.17950-1-ming.lei@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.32]); Fri, 25 Jan 2019 09:54:51 +0000 (UTC) Sender: linux-pci-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-pci@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Now allocating interrupt sets can be done via .setup_affinity() easily, so remove the support for allocating interrupt sets. With this change, we don't need the limit of 'minvec == maxvec' any more in pci_alloc_irq_vectors_affinity(). Meantime irq_create_affinity_masks() gets simplified a lot. Signed-off-by: Ming Lei Acked-by: Bjorn Helgaas # pci/msi.c parts --- drivers/pci/msi.c | 14 ------------- include/linux/interrupt.h | 4 ---- kernel/irq/affinity.c | 52 +++++++++++------------------------------------ 3 files changed, 12 insertions(+), 58 deletions(-) diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 4c0b47867258..331483de1294 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -1035,13 +1035,6 @@ static int __pci_enable_msi_range(struct pci_dev *dev, int minvec, int maxvec, if (maxvec < minvec) return -ERANGE; - /* - * If the caller is passing in sets, we can't support a range of - * vectors. The caller needs to handle that. - */ - if (affd && affd->nr_sets && minvec != maxvec) - return -EINVAL; - if (WARN_ON_ONCE(dev->msi_enabled)) return -EINVAL; @@ -1093,13 +1086,6 @@ static int __pci_enable_msix_range(struct pci_dev *dev, if (maxvec < minvec) return -ERANGE; - /* - * If the caller is passing in sets, we can't support a range of - * supported vectors. The caller needs to handle that. - */ - if (affd && affd->nr_sets && minvec != maxvec) - return -EINVAL; - if (WARN_ON_ONCE(dev->msix_enabled)) return -EINVAL; diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index b820b07f3b55..a035e165f405 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -260,8 +260,6 @@ struct irq_affinity_desc { * and driver has to handle pre_vectors & post_vectors * correctly, set 'is_managed' flag correct too * @priv: Private data of @setup_affinity - * @nr_sets: Length of passed in *sets array - * @sets: Number of affinitized sets */ struct irq_affinity { int pre_vectors; @@ -270,8 +268,6 @@ struct irq_affinity { struct irq_affinity_desc *, unsigned int); void *priv; - int nr_sets; - int *sets; }; #if defined(CONFIG_SMP) diff --git a/kernel/irq/affinity.c b/kernel/irq/affinity.c index 524fdcda9f85..e8fea65325d9 100644 --- a/kernel/irq/affinity.c +++ b/kernel/irq/affinity.c @@ -269,9 +269,9 @@ struct irq_affinity_desc * irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) { int affvecs = nvecs - affd->pre_vectors - affd->post_vectors; - int curvec, usedvecs; + int curvec; struct irq_affinity_desc *masks = NULL; - int i, nr_sets; + int i; /* * If there aren't any vectors left after applying the pre/post @@ -293,34 +293,14 @@ irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) /* Fill out vectors at the beginning that don't need affinity */ for (curvec = 0; curvec < affd->pre_vectors; curvec++) cpumask_copy(&masks[curvec].mask, irq_default_affinity); - /* - * Spread on present CPUs starting from affd->pre_vectors. If we - * have multiple sets, build each sets affinity mask separately. - */ - nr_sets = affd->nr_sets; - if (!nr_sets) - nr_sets = 1; - - for (i = 0, usedvecs = 0; i < nr_sets; i++) { - int this_vecs = affd->sets ? affd->sets[i] : affvecs; - int ret; - - ret = irq_build_affinity_masks(affd, curvec, this_vecs, - curvec, masks); - if (ret) { - kfree(masks); - return NULL; - } - curvec += this_vecs; - usedvecs += this_vecs; + + if (irq_build_affinity_masks(affd, curvec, affvecs, curvec, masks)) { + kfree(masks); + return NULL; } /* Fill out vectors at the end that don't need affinity */ - if (usedvecs >= affvecs) - curvec = affd->pre_vectors + affvecs; - else - curvec = affd->pre_vectors + usedvecs; - for (; curvec < nvecs; curvec++) + for (curvec = affd->pre_vectors + affvecs; curvec < nvecs; curvec++) cpumask_copy(&masks[curvec].mask, irq_default_affinity); /* Mark the managed interrupts */ @@ -340,21 +320,13 @@ int irq_calc_affinity_vectors(int minvec, int maxvec, const struct irq_affinity { int resv = affd->pre_vectors + affd->post_vectors; int vecs = maxvec - resv; - int set_vecs; + int ret; if (resv > minvec) return 0; - if (affd->nr_sets) { - int i; - - for (i = 0, set_vecs = 0; i < affd->nr_sets; i++) - set_vecs += affd->sets[i]; - } else { - get_online_cpus(); - set_vecs = cpumask_weight(cpu_possible_mask); - put_online_cpus(); - } - - return resv + min(set_vecs, vecs); + get_online_cpus(); + ret = min_t(int, cpumask_weight(cpu_possible_mask), vecs) + resv; + put_online_cpus(); + return ret; }