From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 43B73C433FE for ; Thu, 14 Oct 2021 08:59:16 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id CF273611C3 for ; Thu, 14 Oct 2021 08:59:15 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org CF273611C3 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=huawei.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=kvack.org Received: by kanga.kvack.org (Postfix) id 6E6B5940031; Thu, 14 Oct 2021 04:59:15 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 6968C94001C; Thu, 14 Oct 2021 04:59:15 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 585A5940031; Thu, 14 Oct 2021 04:59:15 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0108.hostedemail.com [216.40.44.108]) by kanga.kvack.org (Postfix) with ESMTP id 4943094001C for ; Thu, 14 Oct 2021 04:59:15 -0400 (EDT) Received: from smtpin28.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay04.hostedemail.com (Postfix) with ESMTP id 06F432DEBB for ; Thu, 14 Oct 2021 08:59:15 +0000 (UTC) X-FDA: 78694443870.28.FF76642 Received: from szxga03-in.huawei.com (szxga03-in.huawei.com [45.249.212.189]) by imf20.hostedemail.com (Postfix) with ESMTP id D2D99D0000A7 for ; Thu, 14 Oct 2021 08:59:12 +0000 (UTC) Received: from dggemv704-chm.china.huawei.com (unknown [172.30.72.56]) by szxga03-in.huawei.com (SkyGuard) with ESMTP id 4HVNbH27Hzz8tfS; Thu, 14 Oct 2021 16:58:03 +0800 (CST) Received: from dggpemm500002.china.huawei.com (7.185.36.229) by dggemv704-chm.china.huawei.com (10.3.19.47) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2308.8; Thu, 14 Oct 2021 16:59:11 +0800 Received: from [10.174.178.178] (10.174.178.178) by dggpemm500002.china.huawei.com (7.185.36.229) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2308.8; Thu, 14 Oct 2021 16:59:10 +0800 Message-ID: <8fc5e1ae-a356-6225-2e50-cf0e5ee26208@huawei.com> Date: Thu, 14 Oct 2021 16:59:10 +0800 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101 Thunderbird/91.0.3 Subject: Re: [PATCH] mm/vmalloc: fix numa spreading for large hash tables To: Shakeel Butt CC: Andrew Morton , Nicholas Piggin , Linux MM , LKML , Eric Dumazet , "Kefeng Wang" , References: <20210928121040.2547407-1-chenwandun@huawei.com> From: Chen Wandun In-Reply-To: Content-Type: text/plain; charset="UTF-8"; format=flowed X-Originating-IP: [10.174.178.178] X-ClientProxiedBy: dggems706-chm.china.huawei.com (10.3.19.183) To dggpemm500002.china.huawei.com (7.185.36.229) X-CFilter-Loop: Reflected X-Rspamd-Queue-Id: D2D99D0000A7 X-Stat-Signature: o8kgq4h83jnkt5r4eefez6aszrytysh4 Authentication-Results: imf20.hostedemail.com; dkim=none; dmarc=pass (policy=none) header.from=huawei.com; spf=pass (imf20.hostedemail.com: domain of chenwandun@huawei.com designates 45.249.212.189 as permitted sender) smtp.mailfrom=chenwandun@huawei.com X-Rspamd-Server: rspam02 X-HE-Tag: 1634201952-655267 Content-Transfer-Encoding: quoted-printable X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: =E5=9C=A8 2021/10/14 5:46, Shakeel Butt =E5=86=99=E9=81=93: > On Tue, Sep 28, 2021 at 5:03 AM Chen Wandun wro= te: >> >> Eric Dumazet reported a strange numa spreading info in [1], and found >> commit 121e6f3258fe ("mm/vmalloc: hugepage vmalloc mappings") introduc= ed >> this issue [2]. >> >> Dig into the difference before and after this patch, page allocation h= as >> some difference: >> >> before: >> alloc_large_system_hash >> __vmalloc >> __vmalloc_node(..., NUMA_NO_NODE, ...) >> __vmalloc_node_range >> __vmalloc_area_node >> alloc_page /* because NUMA_NO_NODE, so choose all= oc_page branch */ >> alloc_pages_current >> alloc_page_interleave /* can be proved by= print policy mode */ >> >> after: >> alloc_large_system_hash >> __vmalloc >> __vmalloc_node(..., NUMA_NO_NODE, ...) >> __vmalloc_node_range >> __vmalloc_area_node >> alloc_pages_node /* choose nid by nuam_mem_id() *= / >> __alloc_pages_node(nid, ....) >> >> So after commit 121e6f3258fe ("mm/vmalloc: hugepage vmalloc mappings")= , >> it will allocate memory in current node instead of interleaving alloca= te >> memory. >> >> [1] >> https://lore.kernel.org/linux-mm/CANn89iL6AAyWhfxdHO+jaT075iOa3XcYn9k6= JJc7JR2XYn6k_Q@mail.gmail.com/ >> >> [2] >> https://lore.kernel.org/linux-mm/CANn89iLofTR=3DAK-QOZY87RdUZENCZUT4O6= a0hvhu3_EwRMerOg@mail.gmail.com/ >> >> Fixes: 121e6f3258fe ("mm/vmalloc: hugepage vmalloc mappings") >> Reported-by: Eric Dumazet >> Signed-off-by: Chen Wandun >> --- >> mm/vmalloc.c | 33 ++++++++++++++++++++++++++------- >> 1 file changed, 26 insertions(+), 7 deletions(-) >> >> diff --git a/mm/vmalloc.c b/mm/vmalloc.c >> index f884706c5280..48e717626e94 100644 >> --- a/mm/vmalloc.c >> +++ b/mm/vmalloc.c >> @@ -2823,6 +2823,8 @@ vm_area_alloc_pages(gfp_t gfp, int nid, >> unsigned int order, unsigned int nr_pages, struct pag= e **pages) >> { >> unsigned int nr_allocated =3D 0; >> + struct page *page; >> + int i; >> >> /* >> * For order-0 pages we make use of bulk allocator, if >> @@ -2833,6 +2835,7 @@ vm_area_alloc_pages(gfp_t gfp, int nid, >> if (!order) { >=20 > Can you please replace the above with if (!order && nid !=3D NUMA_NO_NO= DE)? >=20 >> while (nr_allocated < nr_pages) { >> unsigned int nr, nr_pages_request; >> + page =3D NULL; >> >> /* >> * A maximum allowed request is hard-coded an= d is 100 >> @@ -2842,9 +2845,23 @@ vm_area_alloc_pages(gfp_t gfp, int nid, >> */ >> nr_pages_request =3D min(100U, nr_pages - nr_= allocated); >> >=20 > Undo the following change in this if block. Yes, It seem like more simpler as you suggested, But it still have=20 performance regression, I plan to change the following to consider both mempolcy and alloc_pages_bulk. Thanks, Wandun >=20 >> - nr =3D alloc_pages_bulk_array_node(gfp, nid, >> - nr_pages_request, pages + nr_allocated= ); >> - >> + if (nid =3D=3D NUMA_NO_NODE) { >> + for (i =3D 0; i < nr_pages_request; i+= +) { >> + page =3D alloc_page(gfp); >> + if (page) >> + pages[nr_allocated + i= ] =3D page; >> + else { >> + nr =3D i; >> + break; >> + } >> + } >> + if (i >=3D nr_pages_request) >> + nr =3D nr_pages_request; >> + } else { >> + nr =3D alloc_pages_bulk_array_node(gfp= , nid, >> + nr_pages_reque= st, >> + pages + nr_all= ocated); >> + } >> nr_allocated +=3D nr; >> cond_resched(); >> >> @@ -2863,11 +2880,13 @@ vm_area_alloc_pages(gfp_t gfp, int nid, >=20 > Put the following line under "else if (order)" >=20 >> gfp |=3D __GFP_COMP; >> >> /* High-order pages or fallback path if "bulk" fails. */ >> - while (nr_allocated < nr_pages) { >=20 > Keep the following declarations inside the while loop. >=20 >> - struct page *page; >> - int i; >> >> - page =3D alloc_pages_node(nid, gfp, order); >> + page =3D NULL; >> + while (nr_allocated < nr_pages) { >> + if (nid =3D=3D NUMA_NO_NODE) >> + page =3D alloc_pages(gfp, order); >> + else >> + page =3D alloc_pages_node(nid, gfp, order); >> if (unlikely(!page)) >> break; >> >> -- >> 2.25.1 >> > .=20 >=20