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 kanga.kvack.org (kanga.kvack.org [205.233.56.17]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 666A2F3D33A for ; Thu, 5 Mar 2026 17:33:07 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 9354D6B0092; Thu, 5 Mar 2026 12:33:06 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 8E3216B00A3; Thu, 5 Mar 2026 12:33:06 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 7E6696B00A6; Thu, 5 Mar 2026 12:33:06 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0013.hostedemail.com [216.40.44.13]) by kanga.kvack.org (Postfix) with ESMTP id 678786B0092 for ; Thu, 5 Mar 2026 12:33:06 -0500 (EST) Received: from smtpin09.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay06.hostedemail.com (Postfix) with ESMTP id 0D0411B6FBB for ; Thu, 5 Mar 2026 17:33:06 +0000 (UTC) X-FDA: 84512705172.09.B181842 Received: from CH1PR05CU001.outbound.protection.outlook.com (mail-northcentralusazon11010005.outbound.protection.outlook.com [52.101.193.5]) by imf07.hostedemail.com (Postfix) with ESMTP id 12C7040018 for ; Thu, 5 Mar 2026 17:33:02 +0000 (UTC) Authentication-Results: imf07.hostedemail.com; dkim=pass header.d=Nvidia.com header.s=selector2 header.b=EFtcKv0s; spf=pass (imf07.hostedemail.com: domain of ziy@nvidia.com designates 52.101.193.5 as permitted sender) smtp.mailfrom=ziy@nvidia.com; dmarc=pass (policy=reject) header.from=nvidia.com; arc=pass ("microsoft.com:s=arcselector10001:i=1") ARC-Message-Signature: i=2; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1772731983; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references:dkim-signature; bh=Yp/wRZgfQB4QDhqrCSoGnF2BkptrbPVP4GBP3vhAVoU=; b=XzahSHSFzFknxBeS7Y/zuJNA3vmvdoUglfPHSVGQijWaVUkL28qKaPwfTNW/VxPYUNpnCO oZGOcxV6EYlxUd/R+1xI383k0gWaWlrH6BljOjDx9huUrNYMJuHzCgTGVZCptsLaeb7j/h 5kHHNeqYQeAqZSrqH3Dm18QXyF+TzRA= ARC-Authentication-Results: i=2; imf07.hostedemail.com; dkim=pass header.d=Nvidia.com header.s=selector2 header.b=EFtcKv0s; spf=pass (imf07.hostedemail.com: domain of ziy@nvidia.com designates 52.101.193.5 as permitted sender) smtp.mailfrom=ziy@nvidia.com; dmarc=pass (policy=reject) header.from=nvidia.com; arc=pass ("microsoft.com:s=arcselector10001:i=1") ARC-Seal: i=2; s=arc-20220608; d=hostedemail.com; t=1772731983; a=rsa-sha256; cv=pass; b=Z+2um7Smzn2pJt8MprZdODFfFu95p8dZvw5MDx6o7YG8MXUgpkkwe147435YpBk74CHPvP klOqcz8AJVCgBOig60l4Av5oNYUo4SSWw3RPUKNOhu1bf6XmImjN/Q1bv+bdMsMVChYdhd I/pizPxKMYJi1Dam0C4t3mvk+E8tnDs= ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=m3u+sGk8xoM2XEVtYWdtPREEkb5yCp/Qo2V5pmvzyfiUAlXfrmcldbDw93cw+LEHYFd3wZZwm26HDkfDEzwS57j1H0QU6K6DnZZsuX8TQxiXRvixQ5AIyrhtayT+z/fJnAPfgqfOv6b6LriZ2/8Q7pmkWUQJ2efAcpWYYi4bpdn7jLywn/GNYr551H5kF59M+QG/y0IvD8SxQ0bEkqiPL46uohmMpUIXV9Ljfc3voCTidGDkacvyBy9XvsIX/cju7Pp3zG8pKgsqpB1dhLEK7SJ8LuBmua1DTy4OYuAkfVwvzPf6IU9lSIXNDUBmbLclDGcdYduWo9+gZTCh1FzRZQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector10001; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=Yp/wRZgfQB4QDhqrCSoGnF2BkptrbPVP4GBP3vhAVoU=; b=qXhyGMa7l4weEDKosfNBZ9tvm6E/8aPbiGANToZMSjmGEnbdxNEWZMDXkr8XI1UP6gZwu/ZTRHzlhCORYfT1FCs+t7uc1/QbvP4gogKj/gdqM/eNLIuT9dugcyl9vczQcWsW7ZYHwJSOyvfgyTEr2MADTOB1rrDqIh/yyNQdYjFTYGF8q2SMrHZi6Qy4iF2B0VkuGsWA+CO6LKj4rczGCvyu1FDXGDcnDCiuOaue/EcjhyceUc24Nw7PZyOjj7HWqfzLpJlSiOpdvIwawg2eQWCLdRTrugB0RAeqj3SyZANzqSIpg8s7gVXS9MbVZHWzWgkTqIRDrdSek9F50IaxQA== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=nvidia.com; dmarc=pass action=none header.from=nvidia.com; dkim=pass header.d=nvidia.com; arc=none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=Nvidia.com; s=selector2; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=Yp/wRZgfQB4QDhqrCSoGnF2BkptrbPVP4GBP3vhAVoU=; b=EFtcKv0swp8+V7dVsAUfHi6+EhnI0VRh2aAqJSOVhUNPA7lVWT7w64ixHw3AKh7P1ufWUGPq1061I7byRUcKJpRNhFdvre40jOH4rOXGE1Flws6OUx/zn5p7ZxM/KmzBCZeM7pFWg6OdvDaEkkMu9YrJ2YWyWIqAnrQ1VAKSY5wxLOB1QmclM+iCvPxl8XpEavo+40sQhnCrylSNn5YXwxL0UzHee3UASfR8ZBJIgAplE9Co57pOkIDTIg/FjcwmYuqg0x0x5RKEg8WTbJ3v1ikF5SWxHCkzcC1Ed6ZXIWf6tWLHPTR9b79Dak5Y50+EnwQyS4i58MPcdpx8OCXlLA== Received: from DS7PR12MB9473.namprd12.prod.outlook.com (2603:10b6:8:252::5) by DS0PR12MB7511.namprd12.prod.outlook.com (2603:10b6:8:139::8) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.9678.18; Thu, 5 Mar 2026 17:32:56 +0000 Received: from DS7PR12MB9473.namprd12.prod.outlook.com ([fe80::f01d:73d2:2dda:c7b2]) by DS7PR12MB9473.namprd12.prod.outlook.com ([fe80::f01d:73d2:2dda:c7b2%4]) with mapi id 15.20.9654.020; Thu, 5 Mar 2026 17:32:56 +0000 From: Zi Yan To: Usama Arif Cc: =?utf-8?q?Mika_Penttil=C3=A4?= , Balbir Singh , Kiryl Shutsemau , matthew.brost@intel.com, npache@redhat.com, david@kernel.org, Usama Arif , Andrew Morton , linux-mm@kvack.org, joshua.hahnjy@gmail.com, hannes@cmpxchg.org, rakie.kim@sk.com, byungchul@sk.com, gourry@gourry.net, ying.huang@linux.alibaba.com, apopple@nvidia.com, riel@surriel.com, shakeel.butt@linux.dev, linux-kernel@vger.kernel.org, kernel-team@meta.com Subject: Re: [PATCH] mm/migrate_device: fix folio refcount leak on folio_split_unmapped failure Date: Thu, 05 Mar 2026 12:32:51 -0500 X-Mailer: MailMate (2.0r6290) Message-ID: <1EAE2E58-7A71-4B59-B1EF-3A3C753DDC1E@nvidia.com> In-Reply-To: <7996d5c5-24db-4ef2-b88a-1b9d33f9e976@linux.dev> References: <20260304120132.3973445-1-usamaarif642@gmail.com> <5e59c077-9f06-4e45-86e1-ca696e6105b4@nvidia.com> <622eb392-8c04-473d-b42a-ecdc489799c4@linux.dev> <942f2df4-6fb5-415f-b7d4-87a83315890b@redhat.com> <80683d6d-ea38-4326-af5e-e4c173bb1930@redhat.com> <332c9e16-46c3-4e1c-898e-2cb0a87ba1fc@linux.dev> <7996d5c5-24db-4ef2-b88a-1b9d33f9e976@linux.dev> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable X-ClientProxiedBy: BY5PR04CA0001.namprd04.prod.outlook.com (2603:10b6:a03:1d0::11) To DS7PR12MB9473.namprd12.prod.outlook.com (2603:10b6:8:252::5) MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: DS7PR12MB9473:EE_|DS0PR12MB7511:EE_ X-MS-Office365-Filtering-Correlation-Id: a54a0f2d-0a29-4970-7ee5-08de7add3d02 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230040|376014|7416014|1800799024|366016|7053199007|13003099007; X-Microsoft-Antispam-Message-Info: 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 X-Forefront-Antispam-Report: CIP:255.255.255.255;CTRY:;LANG:en;SCL:1;SRV:;IPV:NLI;SFV:NSPM;H:DS7PR12MB9473.namprd12.prod.outlook.com;PTR:;CAT:NONE;SFS:(13230040)(376014)(7416014)(1800799024)(366016)(7053199007)(13003099007);DIR:OUT;SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?VGszamd4NzRxS0ZMdlhpMUZhS1dCdDZLMGtoUCtVbW0yYmt0VXlqYXF2a2pI?= =?utf-8?B?SFFYSmNQc1ZNeEVWdGRQRkd3WSsvRjFmNVBERVVIYy9VdWdmRWpYR1JCSUpa?= =?utf-8?B?Wlh1Z0pMc2tZLzcyQzBMNEdVYWRMKzAxd2RlRnB5bStNaTAwb0l1MjNlaDFY?= =?utf-8?B?Z1UzK0tmK3F6MGpyMEVHbkhmd0Q4MGFtcVpUYU1YOHdPV1FEanR4anp5cXEz?= =?utf-8?B?S256NlpWRTVpUVFIZkV2WThVSU5PU2ZKa1NpV1ljRE1BaE00cTIwT045dEF1?= =?utf-8?B?U3d3QitkQks3WWs1N0RVd2owcmI2d3B3dG9kd1JzSFk2OG9GdjllZ0xpRVZy?= =?utf-8?B?cmZuSkpBV0Q2MmVXejJZa0laT0l0Yy9pS3R2QTE2ejBaSVp5RWtsRUhSTmtr?= =?utf-8?B?czFhQXRlTnUzVXlYdXFtZW90QklYUE5NNDdTc0xvSnRjdkliSDE5UVpPbjgv?= =?utf-8?B?YkVlM0FZalo2T2dyQWJ6Y2d6UXFwZTg4TEI0RHJ3aHB3blBkY0tUclo2RExU?= =?utf-8?B?S1NCT0FScXV2V1E0TE92SUZLOFZndXgzRVdraWhEY09hRDRBa2RJKzlsRmFn?= =?utf-8?B?UzlqQ2VaKzhjUzM2aGhPRUoyckhNZ0doYXVhTFQ3NkMvcHo3V1JvU0w2cjZY?= =?utf-8?B?MGJRN2VWd1I4d1dkUnFCSTZ6SE5PSEJkN3dTZnl5T1YrYzZMNEhzYTVVUENk?= =?utf-8?B?SFlhZkovNEh5LzJBNHQ4dEJienpGSXFzU1Y3U0RuTXpiLzJ4QmM2STFPajl1?= =?utf-8?B?cW04aytiMzhHeGErRnJFVXdpdUQ1dzVOZGU2cldET3Q3RStDaUdwZm5xMnFq?= =?utf-8?B?S2laTjhJZlg4WEhKTTd0bzJOS0ZpM3JYMzAreWxaUlI2S3h6YTRDcVJCZU9L?= =?utf-8?B?MzFLWGdhaTVIK0JvQW5BZ0FRTUVER29WcEFOVHlZeDRoZWVlL3FFZWxqbWhG?= =?utf-8?B?dnEwU2Rjcm9vTEY4aDY4Tnp4Yit4QVdzMlM2WU1IdE11a2lYTm5VUG5oa1kr?= =?utf-8?B?dlZMYVQyb3prazliSGE1SWlicnEyd2ZHMUVYMTJMeEptMG1OSW9YdTkzanFG?= =?utf-8?B?UjNmOE9QNWlzd05EejRweURaM1BpWDUrL0pkd2FwemY4ZXNJc0o4c1FPR1ht?= =?utf-8?B?YXQ1ZHhObytwTit2V0xMYWhMeWFwK1pRRVEvT0I3cGIyYlEvQTdVallEWFgv?= =?utf-8?B?QzhHTGxkWTVUSDQ4bk9ldGJJQUxaelpVZzU1NlR6aXFON3BpZzFQWngwV0xz?= =?utf-8?B?WFFNUzRGTGsrUFRZaExTU1ljbWd4WE1IRHM2dEJXcS9NTHVPT2tsc2JuRUd1?= =?utf-8?B?QlFGbVBWQkVUaWZzT1RHN2RrME9rRzNCcXNlcGRBZjZidFBuSExpb1J4S2Zx?= =?utf-8?B?L21hVWhzNnI0TUxBeGVTOXJzaCsxNkozeHB2VERFNTN0aGZNWmVlY2JTbk53?= =?utf-8?B?cWtSeDhVeFdTQ25sUElVSnRXUHhCRkhUaGkzQnM2TU1lY05ScTdKQXdDelNa?= =?utf-8?B?alFKbnNTSTFKWHFERUIzaXBQVVJVS1l4eDVIZjF3MU1Bby9oSnQ4WUU5M3c1?= =?utf-8?B?dUFWRWZwcG9YYWhFSEVZMDUvZE9BUXFUUzlZc0R0dFoxN29XMk9TM2ZFU29y?= =?utf-8?B?cFFKZjZyL1dpQStLNGtTajJjVGdBWFA3TDIrYks4N3JtcjF4dWtkRytiRW5j?= =?utf-8?B?cm9rMHJDcUxqa0d1SnJubmpKdnJKazVOQklXYnRGSm9YeDE4eDU1RStRWElz?= =?utf-8?B?NUZqWjBGajdJcUtKVU8xTFdsNGtaZWljcDZRQm4vc01JZEd4a0dmaWxHM2RX?= =?utf-8?B?WnlkUTB0ZS9GRWJLSEZrRmFUOXdPNW92eDEvZFA3bHBKL1ZkeTIvbWJLZHl6?= =?utf-8?B?eGR2YVcvZVA5amVqM0l1ZnZLQWxoTm5nelk1UzRJNUR3R2hLWEdXcE16dHRY?= =?utf-8?B?Z3dqbmVCUFJYVEljWktubXhKYk9BWjJ6T2IvS0dxYlNpc3NsZEdVS3A4VmlL?= =?utf-8?B?YWx4TXI5L2IxanhUSzVGY0twUDU1SUNyUGVvNlRYakVoNEJFM2pCc0VLUld6?= =?utf-8?B?T0phM1Q4Tm9CTE9TTTVBS0txQmRHZFUzdUdmdVVRRjZUSWdKK040dm43WGRM?= =?utf-8?B?aVlTTmdIMS9jUW90U01IRHlxLzRNbTBLV1M2eWJKS3p1dGI0b0NQQXNqbTY4?= =?utf-8?B?c3R3M1RGS1E4a0lCRHlzQk85MDN2KzRFV25XVC9ld2thUVBSMFlUWnhJd1Vh?= =?utf-8?B?cSt0NFBWM05CMXkycVBnVVBUeTRtYnRhOUU5cTNodEpLQUFrNGtzQklRVkFS?= =?utf-8?B?d1pSS1g1bGQ2TnJRNWhFQVZKVHN6VW1vTVJ4LzMvUFYrUGE5NnZlUT09?= X-OriginatorOrg: Nvidia.com X-MS-Exchange-CrossTenant-Network-Message-Id: a54a0f2d-0a29-4970-7ee5-08de7add3d02 X-MS-Exchange-CrossTenant-AuthSource: DS7PR12MB9473.namprd12.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 05 Mar 2026 17:32:56.4342 (UTC) X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted X-MS-Exchange-CrossTenant-Id: 43083d15-7273-40c1-b7db-39efd9ccc17a X-MS-Exchange-CrossTenant-MailboxType: HOSTED X-MS-Exchange-CrossTenant-UserPrincipalName: x9u5rf3jpT0a2nJb1yyfW3B96jN4wK1V7UjOJ+yNj9ILuOoK7QPdHis9F+eLIrOY X-MS-Exchange-Transport-CrossTenantHeadersStamped: DS0PR12MB7511 X-Rspamd-Server: rspam10 X-Rspamd-Queue-Id: 12C7040018 X-Stat-Signature: tspj4smg5i1fcirxtk1wi6hqnnq8mak7 X-Rspam-User: X-HE-Tag: 1772731982-507730 X-HE-Meta: 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 Aw/+gPQc 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 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: List-Subscribe: List-Unsubscribe: On 5 Mar 2026, at 12:00, Usama Arif wrote: > On 05/03/2026 16:39, Zi Yan wrote: >> On 5 Mar 2026, at 11:36, Usama Arif wrote: >> >>> On 05/03/2026 12:09, Mika Penttil=C3=A4 wrote: >>>> On 3/5/26 13:44, Usama Arif wrote: >>>> >>>>> >>>>> On 05/03/2026 06:09, Mika Penttil=C3=A4 wrote: >>>>>> Hi! >>>>>> >>>>>> On 3/5/26 01:28, Usama Arif wrote: >>>>>> >>>>>>> On 04/03/2026 22:09, Balbir Singh wrote: >>>>>>>> On 3/5/26 08:54, Zi Yan wrote: >>>>>>>>> On 4 Mar 2026, at 16:48, Balbir Singh wrote: >>>>>>>>> >>>>>>>>>> On 3/5/26 02:17, Zi Yan wrote: >>>>>>>>>>> On 4 Mar 2026, at 7:01, Usama Arif wrote: >>>>>>>>>>> >>>>>>>>>>>> From: Usama Arif >>>>>>>>>>>> >>>>>>>>>>>> migrate_vma_split_unmapped_folio() takes an extra reference vi= a >>>>>>>>>>>> folio_get() before calling folio_split_unmapped(). On success= , the >>>>>>>>>>>> split consumes this reference: __folio_freeze_and_split_unmapp= ed() >>>>>>>>>>>> expects the +1 in its folio_ref_freeze() check, and distribute= s it >>>>>>>>>>>> across the resulting sub-folios via folio_ref_unfreeze(...+1),= which >>>>>>>>>>>> are later balanced by folio_put() calls in __migrate_device_fi= nalize(). >>>>>>>>>>>> >>>>>>>>>>>> If folio_split_unmapped() fails (e.g., unexpected pinning retu= rns >>>>>>>>>>>> -EAGAIN), the function returns without calling folio_put(). T= he extra >>>>>>>>>>>> reference is never released. >>>>>>>>>>>> >>>>>>>>>>>> Add the missing folio_put() on the error path. >>>>>>>>>>>> >>>>>>>>>>>> Fixes: 4265d67e405a4 ("mm/migrate_device: add THP splitting du= ring migration") >>>>>>>>>>>> Closes: https://lore.kernel.org/all/CAA1CXcDyqPPwf_-W7B+PFQtL8= HdoJGCEqVsVxq7DhOUB=3DL4PQA@mail.gmail.com/ >>>>>>>>>>>> Reported-by: Nico Pache >>>>>>>>>>>> Signed-off-by: Usama Arif >>>>>>>>>>>> --- >>>>>>>>>>>> mm/migrate_device.c | 4 +++- >>>>>>>>>>>> 1 file changed, 3 insertions(+), 1 deletion(-) >>>>>>>>>>>> >>>>>>>>>>>> diff --git a/mm/migrate_device.c b/mm/migrate_device.c >>>>>>>>>>>> index 0a8b31939640f..351ecd9065d13 100644 >>>>>>>>>>>> --- a/mm/migrate_device.c >>>>>>>>>>>> +++ b/mm/migrate_device.c >>>>>>>>>>>> @@ -917,8 +917,10 @@ static int migrate_vma_split_unmapped_fol= io(struct migrate_vma *migrate, >>>>>>>>>>>> folio_get(folio); >>>>>>>>>>>> split_huge_pmd_address(migrate->vma, addr, true); >>>>>>>>>>>> ret =3D folio_split_unmapped(folio, 0); >>>>>>>>>>>> - if (ret) >>>>>>>>>>>> + if (ret) { >>>>>>>>>>>> + folio_put(folio); >>>>>>>>>>>> return ret; >>>>>>>>>>>> + } >>>>>>>>>>>> migrate->src[idx] &=3D ~MIGRATE_PFN_COMPOUND; >>>>>>>>>>>> flags =3D migrate->src[idx] & ((1UL << MIGRATE_PFN_SHIFT) - = 1); >>>>>>>>>>>> pfn =3D migrate->src[idx] >> MIGRATE_PFN_SHIFT; >>>>>>>>>>>> --=20 >>>>>>>>>>>> 2.47.3 >>>>>>>>>>> Add Balbir, who wrote the code, to comment on this. >>>>>>>>>>> >>>>>>>>>> Thanks Zi! >>>>>>>>>> >>>>>>>>>> Just wondering if there is a reproducer for the issue and how th= e fix was tested? >>>>>>>>>> I expect migrate_vma_finalize() to be called for folios, even wh= en split failed and >>>>>>>>>> drop the lock. >>>>>>>>> Does migrate_vma_finalize() do folio_put() for failed-to-split fo= lios? >>>>>>>>> If so, how does it distinguish between split folios and failed-to= -split folios? >>>>>>>>> By comparing source and destination folio orders? >>>>>>>>> >>>>>>>> We reset the MIGRATE_PFN_MIGRATE flag for failing to migrate pfns.= We do a folio_put >>>>>>>> on the src in finalize, if it is split then on all the split folio= s as well. >>>>>>>> >>>>>>>>> What we see from migrate_vma_split_unmapped_folio() is that >>>>>>>>> it adds a refcount for all input folios, but only drops a refcoun= t >>>>>>>>> for the split folio. Isn=E2=80=99t it cause failed-to-split folio= s to have >>>>>>>>> additional refcount? >>>>>>>>> >>>>>>> Hello! >>>>>>> >>>>>>> Thanks for reviewing everyone. So its very difficult to create a re= producer I think >>>>>>> the extra reference would need to appear after migrate_device_unmap= () but before >>>>>>> folio_split_unmapped() in migrate_vma_pages()? That's hard to trigg= er reliably from >>>>>>> userspace. >>>>>>> >>>>>>> The fix came about when Nico indicated there might be an issue if s= plit_huge_pmd_address >>>>>>> fails in my patch [1]. >>>>>>> >>>>>>> Below is my understanding of how refcounting is working over here s= tep by step. I >>>>>>> might very well be wrong on this, and the refcounting is a bit all = over the place >>>>>>> and I might miss a reference change somewhere so would really appre= ciate if someone >>>>>>> can confirm this! >>>>>>> >>>>>>> >>>>>>> 1. migrate_vma_collect_huge_pmd(): >>>>>>> a) folio_get(folio) -> +1 (collect reference) >>>>>>> 2. migrate_device_unmap(): >>>>>>> a) folio_isolate_lru() -> +1 (isolation reference) >>>>>>> b) folio_put() -> -1 (drops the collect reference) >>>>>>> >>>>>>> >>>>>>> Without this patch fix: >>>>>>> >>>>>>> 3. migrate_vma_split_unmapped_folio(): >>>>>>> a) folio_get(folio) -> +1 (split reference) >>>>>>> b) folio_split_unmapped() -> fails >>>>>>> c) Returns error =E2=80=94 without folio_put() which is the fix >>>>>>> 4. Caller in migrate_vma_pages(): clears MIGRATE_PFN_MIGRATE | MIGR= ATE_PFN_COMPOUND >>>>>>> 5. __migrate_device_finalize(): sees !(src_pfns[i] & MIGRATE_PFN_MI= GRATE), restores the folio: >>>>>>> a) remove_migration_ptes(src, src) =E2=80=94 re-establishes user = PTEs >>>>>>> b) folio_unlock(src) >>>>>>> c) folio_put(src) -> -1 (drops the isolation reference) >>>>>>> >>>>>>> The split reference in 3.a is never released and the folio has a pe= rmanently elevated refcount. >>>>>>> Unless I missed a folio_put somewhere for the refcount increase in = folio_isolate_lru() (2.b)? >>>>>>> >>>>>>> Please let me know if this makes sense! >>>>>>> >>>>>>> [1] https://lore.kernel.org/all/CAA1CXcDyqPPwf_-W7B+PFQtL8HdoJGCEqV= sVxq7DhOUB=3DL4PQA@mail.gmail.com/ >>>>>>> >>>>>>>> Thanks! Yes, the patch makes sense >>>>>>>> >>>>>>>> Acked-by: Balbir Singh >>>>>>>> >>>>>>>> Balbir >>>>>> I remember stumbling on this while ago also. The folio_get() in migr= ate_vma_split_unmapped_folio() >>>>>> is balanced with put_page() in __split_huge_pmd_locked() (freeze =3D= true), can't fail for device pages. >>>>>> Folios at this point are unmapped but have 1 refcount from "collecti= ng". >>>>>> After folio_split_unmapped() the refcount(s) is still 1. >>>>>> >>>>>> So it seems the code is good as is? A comment though would be good f= or the extra folio_get.. >>>>>> >>>>> hmm I dont think the put_page() in __split_huge_pmd_locked() is there= to balance the folio_get() in >>>>> migrate_vma_split_unmapped_folio(). There are other points where spli= t_huge_pmd_locked() is called >>>>> with freeze =3D true [1] and they don't get a reference before callin= g split_huge_pmd. >>>>> >>>>> I think the folio_put() in __split_huge_pmd_locked() freeze =3D true = case is there as migration >>>>> entries are being installed? >>>>> >>>>> [1] https://elixir.bootlin.com/linux/v6.19.3/source/mm/rmap.c#L2334 >>>>> >>>>> >>>> Yes normally you want to drop the reference when installing migration = entries but in this context >>>> you have already done the collecting for the THP folio and you want to= balance with the folio_get() >>>> the put_page() to keep the refs unchanged. Is that right Balbir? >>>> >>>> --Mika >>>> >>> >>> Hi Mika, >>> >>> You are right, This patch is wrong. I tried the below diff to force fol= io_split_unmapped to return >>> -EAGAIN. I ran tools/testing/selftests/mm/hmm-tests -r hmm.hmm_device_p= rivate.migrate_anon_huge_err >>> to trigger the path for folio_split_unmapped. >>> >>> diff --git a/mm/huge_memory.c b/mm/huge_memory.c >>> index 8e2746ea74adf..6df33b4990a13 100644 >>> --- a/mm/huge_memory.c >>> +++ b/mm/huge_memory.c >>> @@ -4140,6 +4140,8 @@ int folio_split_unmapped(struct folio *folio, uns= igned int new_order) >>> if (folio_expected_ref_count(folio) !=3D folio_ref_count(folio)= - 1) >>> return -EAGAIN; >>> >>> + return -EAGAIN; >>> + >>> local_irq_disable(); >>> ret =3D __folio_freeze_and_split_unmapped(folio, new_order, &fo= lio->page, NULL, >>> NULL, false, NULL, SPLI= T_TYPE_UNIFORM, >>> >>> >>> >>> I inserted a lot of traces to keep track of refcounts [1]. Without this= patch, I get >>> .... >>> hmm-tests-129 [000] ..... 1.476233: __migrate_device_pag= es: SPLIT_UNMAPPED: folio=3Dffc536e2c4100000 refcount=3D0 AFTER error NO fo= lio_put >>> hmm-tests-129 [000] ..... 1.476234: __migrate_device_pag= es: PAGES: split FAILED folio=3Dffc536e2c4100000 refcount=3D0 >>> hmm-tests-129 [000] ..... 1.476236: __migrate_device_fin= alize: FINALIZE[0]: src=3Dffc536e2c4100000 dst=3Dffc536e2c4100000 src=3D=3D= dst=3D1 refcount_src=3D1 mapcount_src=3D0 order_src=3D0 migrate=3D0 BEFORE = remove_migration_ptes >>> hmm-tests-129 [000] ..... 1.476237: __migrate_device_fin= alize: FINALIZE[0]: src=3Dffc536e2c4100000 refcount=3D1 mapcount=3D0 AFTER = remove_migration_ptes >>> hmm-tests-129 [000] ..... 1.476237: __migrate_device_fin= alize: FINALIZE[0]: src=3Dffc536e2c4100000 refcount=3D0 AFTER folio_put(src= ) >>> >>> i.e. refcount =3D 512, which is correct as split_huge_pmd_address was s= uccessful. Full output is >>> at [2]. >>> >>> With this patch, I get: >>> >>> BUG: Bad rss-counter state mm:00000000cfe88d5e type:MM_FILEPAGES val:-5= 11 Comm:bash Pid:63 >>> BUG: Bad rss-counter state mm:00000000cfe88d5e type:MM_ANONPAGES val:51= 1 Comm:bash Pid:63 >>> ... >>> hmm-tests-129 [000] ..... 1.468315: __migrate_device_pag= es: SPLIT_UNMAPPED: folio=3Dffed210c840f0000 refcount=3D1 AFTER error folio= _put FIX PRESENT >>> hmm-tests-129 [000] ..... 1.468315: __migrate_device_pag= es: PAGES: split FAILED folio=3Dffed210c840f0000 refcount=3D1 >>> hmm-tests-129 [000] ..... 1.468318: __migrate_device_fin= alize: FINALIZE[0]: src=3Dffed210c840f0000 dst=3Dffed210c840f0000 src=3D=3D= dst=3D1 refcount_src=3D1 mapcount_src=3D0 order_src=3D9 migrate=3D0 BEFORE = remove_migration_ptes >>> hmm-tests-129 [000] ..... 1.468357: __migrate_device_fin= alize: FINALIZE[0]: src=3Dffed210c840f0000 refcount=3D513 mapcount=3D512 AF= TER remove_migration_ptes >>> hmm-tests-129 [000] ..... 1.468357: __migrate_device_fin= alize: FINALIZE[0]: src=3Dffed210c840f0000 refcount=3D512 AFTER folio_put(s= rc) >>> >>> refcount=3D0 means the folio would be freed which is not correct. The f= ull output is at [3]. >>> >>> Thank you for clearing this up! >> >> Thank you for doing the investigation. Can you send a patch to add a com= ment >> in migrate_vma_split_unmapped_folio() about this to avoid the confusion >> in the future? >> > > Yeah this was really confusing. > > Does something like below look good? > > diff --git a/mm/migrate_device.c b/mm/migrate_device.c > index 78c7acf024615..a302f9d3ce921 100644 > --- a/mm/migrate_device.c > +++ b/mm/migrate_device.c > @@ -910,6 +910,11 @@ static int migrate_vma_split_unmapped_folio(struct m= igrate_vma *migrate, > > folio_get(folio); > split_huge_pmd_address(migrate->vma, addr, true); > + /* > + * split_huge_pmd_address consumes the folio_get reference above. > + * Therefore no folio_put is needed on the folio_split_unmapped > + * error path. > + */ > ret =3D folio_split_unmapped(folio, 0); > if (ret) > return ret; I do not think there is a need to explain why there is no folio_put() below. How about below? 1. it makes sure the folio has the right ref count, 2. it explains folio_get() is for split_huge_pmd_address() instead of folio_split_unmapped(). diff --git a/mm/migrate_device.c b/mm/migrate_device.c index 0a8b31939640f..0b31b878210ba 100644 --- a/mm/migrate_device.c +++ b/mm/migrate_device.c @@ -914,8 +914,14 @@ static int migrate_vma_split_unmapped_folio(struct mig= rate_vma *migrate, unsigned long flags; int ret =3D 0; + VM_WARN_ON_ONCE(folio_ref_count(folio) =3D=3D 1); + /* + * take a reference, since split_huge_pmd_address() with freeze =3D true + * drops a reference at the end. + */ folio_get(folio); split_huge_pmd_address(migrate->vma, addr, true); +=09 ret =3D folio_split_unmapped(folio, 0); if (ret) return ret; > >>> >>> >>> [1] https://gist.github.com/uarif1/65e1e816af7aa0ae38dd6ec64d62a993 >>> [2] https://gist.github.com/uarif1/79ea9500667daa4e2ef09cb5d308f041 >>> [3] https://gist.github.com/uarif1/8a35a6c65ba8b3a1c1dfe72dc30e821d >> >> >> Best Regards, >> Yan, Zi Best Regards, Yan, Zi