From patchwork Tue Dec 10 09:39:14 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Li RongQing X-Patchwork-Id: 1206993 X-Patchwork-Delegate: davem@davemloft.net Return-Path: X-Original-To: patchwork-incoming-netdev@ozlabs.org Delivered-To: patchwork-incoming-netdev@ozlabs.org Authentication-Results: ozlabs.org; spf=none (no SPF record) smtp.mailfrom=vger.kernel.org (client-ip=209.132.180.67; helo=vger.kernel.org; envelope-from=netdev-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=none (p=none dis=none) header.from=baidu.com Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 47XFPz5KvCz9sR0 for ; Tue, 10 Dec 2019 20:39:19 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727162AbfLJJjS (ORCPT ); Tue, 10 Dec 2019 04:39:18 -0500 Received: from mx21.baidu.com ([220.181.3.85]:33750 "EHLO baidu.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1726915AbfLJJjR (ORCPT ); Tue, 10 Dec 2019 04:39:17 -0500 Received: from BJHW-Mail-Ex16.internal.baidu.com (unknown [10.127.64.39]) by Forcepoint Email with ESMTPS id A3039B19AABF1; Tue, 10 Dec 2019 17:39:13 +0800 (CST) Received: from BJHW-Mail-Ex13.internal.baidu.com (10.127.64.36) by BJHW-Mail-Ex16.internal.baidu.com (10.127.64.39) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA256_P256) id 15.1.1713.5; Tue, 10 Dec 2019 17:39:14 +0800 Received: from BJHW-Mail-Ex13.internal.baidu.com ([100.100.100.36]) by BJHW-Mail-Ex13.internal.baidu.com ([100.100.100.36]) with mapi id 15.01.1713.004; Tue, 10 Dec 2019 17:39:14 +0800 From: "Li,Rongqing" To: Yunsheng Lin , Saeed Mahameed , "brouer@redhat.com" CC: "ilias.apalodimas@linaro.org" , "jonathan.lemon@gmail.com" , "netdev@vger.kernel.org" Subject: =?utf-8?b?562U5aSNOiBbUEFUQ0hdW3YyXSBwYWdlX3Bvb2w6IGhhbmRsZSBw?= =?utf-8?q?age_recycle_for_NUMA=5FNO=5FNODE_condition?= Thread-Topic: [PATCH][v2] page_pool: handle page recycle for NUMA_NO_NODE condition Thread-Index: AQHVrBgioGDhH/MP9UuNcNvu7zNf3aeuC1gAgAMqtQCAAL3wgIAAIMoAgAEG1JA= Date: Tue, 10 Dec 2019 09:39:14 +0000 Message-ID: References: <1575624767-3343-1-git-send-email-lirongqing@baidu.com> <9fecbff3518d311ec7c3aee9ae0315a73682a4af.camel@mellanox.com> <20191209131416.238d4ae4@carbon> <816bc34a7d25881f35e0c3e21dc2283ffeffb093.camel@mellanox.com> In-Reply-To: Accept-Language: zh-CN, en-US Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [172.22.198.19] x-baidu-bdmsfe-datecheck: 1_BJHW-Mail-Ex16_2019-12-10 17:39:14:464 x-baidu-bdmsfe-viruscheck: BJHW-Mail-Ex16_GRAY_Inside_WithoutAtta_2019-12-10 17:39:14:448 MIME-Version: 1.0 Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org > > static int mvneta_create_page_pool(struct mvneta_port *pp, > struct mvneta_rx_queue *rxq, int size) { > struct bpf_prog *xdp_prog = READ_ONCE(pp->xdp_prog); > struct page_pool_params pp_params = { > .order = 0, > .flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV, > .pool_size = size, > .nid = cpu_to_node(0), This kind of device should only be installed to vendor's platform which did not support numa But as you say , Saeed advice maybe cause that recycle always fail, if nid is configured like upper, and different from running NAPI node id And maybe we can catch this case by the below -Li > .dev = pp->dev->dev.parent, > .dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE, > .offset = pp->rx_offset_correction, > .max_len = MVNETA_MAX_RX_BUF_SIZE, > }; > > the pool->p.nid is not NUMA_NO_NODE, then the node of page allocated for rx > may not be numa_mem_id() when running in the NAPI polling, because > pool->p.nid is not the same as the node of cpu running in the NAPI polling. > > Does the page pool support recycling for above case? > > Or we "fix' the above case by setting pool->p.nid to > NUMA_NO_NODE/dev_to_node(), or by calling pool_update_nid() in NAPI > polling as mlx5 does? > > > > > > Thanks, > > Saeed. > > > > > > > > > > > > > > diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 3c8b51ccd1c1..973235c09487 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -328,6 +328,11 @@ static bool pool_page_reusable(struct page_pool *pool, struct page *page) void __page_pool_put_page(struct page_pool *pool, struct page *page, unsigned int dma_sync_size, bool allow_direct) { + allow_direct = allow_direct && in_serving_softirq(); + + if (allow_direct) + WARN_ON_ONCE((pool->p.nid != NUMA_NO_NODE) && + (pool->p.nid != numa_mem_id())); /* This allocator is optimized for the XDP mode that uses * one-frame-per-page, but have fallbacks that act like the * regular page allocator APIs. @@ -342,7 +347,7 @@ void __page_pool_put_page(struct page_pool *pool, struct page *page, page_pool_dma_sync_for_device(pool, page, dma_sync_size); - if (allow_direct && in_serving_softirq()) + if (allow_direct) if (__page_pool_recycle_direct(page, pool)) return;