From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from BL0PR03CU003.outbound.protection.outlook.com (mail-eastusazon11012007.outbound.protection.outlook.com [52.101.53.7]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 97E62370ADF for ; Fri, 9 Oct 2026 19:09:25 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=fail smtp.client-ip=52.101.53.7 ARC-Seal:i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1791572968; cv=fail; b=ulYrc7v9X9MaEJFDWlQynpPt451QtMzUkUxsEn7K3ROXQx/2/4IDYdWH9JxjjuGBojfurDtDKzIvBB+9j5/dCErA/AsIo0G7V/wHDxiVexvu550OJ9fwnZf/GVNphrmu4UxlQzGKzI7MyNH8k57vhy/P/XIE08+KdREEAh80Zn8= ARC-Message-Signature:i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1791572968; c=relaxed/simple; bh=s6gptLRL+xkb/An/+vARSJ7RLk1GcHudbDXH3Wx+DNQ=; h=Content-Type:Date:Message-Id:From:Subject:Cc:To:References: In-Reply-To:MIME-Version; b=Fyd+kncm7gd97/en1ti/uXKlDQdGy+MOP2fHwDIjC8ppKiwCphODKOo+WcfCGwVMaQWLQj8J2wmVbK8sJO8+MAPd8jA9gGmQfEuHISPVk9p9zcnFiDLeJWMjJU3gya5VxbmLR11TOGQfBGKi6GqSvQOje1KRwquflHzNqFbVI4o= ARC-Authentication-Results:i=2; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=nvidia.com; spf=fail smtp.mailfrom=nvidia.com; dkim=pass (2048-bit key) header.d=Nvidia.com header.i=@Nvidia.com header.b=CuQwWlkF; arc=fail smtp.client-ip=52.101.53.7 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=nvidia.com Authentication-Results: smtp.subspace.kernel.org; spf=fail smtp.mailfrom=nvidia.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=Nvidia.com header.i=@Nvidia.com header.b="CuQwWlkF" ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=xy5fuby9g4v7zipgQWqJxO6JTrIVJZuXBPcsJyL5tGie+5qHDmsUJsqmliRiX7GuMWh8TvTZNK3+cnzaxyEPDGlUHS4bGciu8XRfukvD5n1YfRmXDEc24Kd0dDGll3vMhWckGsd+fMLEphwh0c5NClPIcDjmEatKfl1e1xGKVBg4ThtetmDYQEWALdUWNj0Q8AHxhLBlkshs31Gg91ldm3BWmlrQa5DgMI+LbX5oakgtcxKk7P4Xd+M11qd7IjKUJBy/x/Wnk4cc8ocyXyM2f/QdJDkMENRVpncR6GCYlEfBg7aGugvduemI3RCR69KVfzFlj9o+BFRuVgWig2CpOQ== 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=UuO4z7CXxh0LjmaJRr8ob5oXJWieqzVbtQFbcZrsQrg=; b=CEltnd67dGFFmmABgbklmNf/eBbLomD4W7vu3kFXRgLmiUROyZuhzsLhcNUWk1QRhDN3eA39KzXFPxyI2DYEr0IK6u0XhEbwnuFSLTVbXQEGwo1AdmYsZoN16aVz2bcfXiEEdkwdFphsWHdro/wk6jv19vug9rMOiL8NOVlRI+iFSp+1yeWJ5ph0wbHDLKPnXFZnuxwvTc2JyC2F1xIWgXmYxqUe7NYFJMHdCPa9MuUZ5m9TklWtcyCOefjvuZo3w+c0mkqMIz1Pj7zNbLrP0BmipJR/cBJqYEqVTSVcjcringeijK9IW6bV8DIl4cwVGLtGl9jNf0U00wOZm4f0XA== 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=UuO4z7CXxh0LjmaJRr8ob5oXJWieqzVbtQFbcZrsQrg=; b=CuQwWlkF7erXx4/+wiPYgKiLWPov+/xXVz+fA8m7C0buqusH0KUJsKP0g7K1q8jSTtaFzt4InSnfF5FDGAnKeHdhg0vddPo/iBara3SLohgJ2hYmHHrieXAKc9Qlh9o9gVT+NVEyE2Txla23+YRuVbcLHFxMwuQ7rXtXikApDZtUYIfozhde1qofu6OIodsdZ4hEv1kZICm3Wrsm+AyPtYxoAXKqaLCAa4Jz76x4SEknu6OdqDis+364hROpRZTyXj+kSVSEdi4nBQmZ0uJ62ZkHtEeWmD4OGY6Pyk/uTkt2tbUjdW0DEUHpwICOACDhGt3vsVp9EwYn/pY1qc9o7Q== Authentication-Results: mx.microsoft.com 1; dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=nvidia.com; Received: from IA0PR12MB8374.namprd12.prod.outlook.com (2603:10b6:208:40e::7) by MW6PR12MB8760.namprd12.prod.outlook.com (2603:10b6:303:23a::6) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.496.17; Fri, 9 Oct 2026 19:09:11 +0000 Received: from IA0PR12MB8374.namprd12.prod.outlook.com ([fe80::d85f:4c87:ae84:3f16]) by IA0PR12MB8374.namprd12.prod.outlook.com ([fe80::d85f:4c87:ae84:3f16%6]) with mapi id 15.21.0472.016; Fri, 9 Oct 2026 19:09:10 +0000 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=UTF-8 Date: Fri, 09 Oct 2026 15:09:09 -0400 Message-Id: From: "Zi Yan" Subject: Re: [PATCH v8 15/30] mm: swap in PMD swap entries as whole THPs during swapoff Cc: , "Baoquan He" , , , , , , , , , , , "Nico Pache" , "Liam R.Howlett" , , "Vlastimil Babka" , , , , , , , , To: "Usama Arif" , "Andrew Morton" , , , , , X-Mailer: aerc 0.22.0 References: <20261002095503.3585565-1-usama.arif@linux.dev> <20261002095503.3585565-16-usama.arif@linux.dev> In-Reply-To: <20261002095503.3585565-16-usama.arif@linux.dev> X-ClientProxiedBy: BN9PR03CA0297.namprd03.prod.outlook.com (2603:10b6:408:f5::32) To IA0PR12MB8374.namprd12.prod.outlook.com (2603:10b6:208:40e::7) Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: IA0PR12MB8374:EE_|MW6PR12MB8760:EE_ X-MS-Office365-Filtering-Correlation-Id: 282fbff5-acdc-429b-bdad-08df2638ccb5 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230040|1800799024|366016|376014|7416014|23010399003|22082099003|18002099003|56012099006|11063799006|4143699003|10067099003; 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:IA0PR12MB8374.namprd12.prod.outlook.com;PTR:;CAT:NONE;SFS:(13230040)(1800799024)(366016)(376014)(7416014)(23010399003)(22082099003)(18002099003)(56012099006)(11063799006)(4143699003)(10067099003);DIR:OUT;SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?bjVBRlpKL25FSC9XR3FKMFd4WC82cHE5ZjBjUEYzZGFxbzR5NklnWTVCN2dJ?= =?utf-8?B?N2lJNDZISTFNdFJIczl4UTR5ejNLY3ZBQXNIT3ErZWNINjRoOWxIY2NmUTgw?= =?utf-8?B?WnhuczZXV1J4d203SmpFL2t2d0U4QlFQMXgxYkprK0JDSlU1THRlcy9ENXZK?= =?utf-8?B?M3ppaVA0QmZtWGk5aVV0Z2tCbHJhb0p6dXJwVklWallKb24vVEs2MDZTVFk1?= =?utf-8?B?anJXTlI5MjB4L253TitjOUhBbHlKYnJDc3FrZVZlYzJXT3B6TVJiY1BLYmFL?= =?utf-8?B?Wm9FOTNmY1g2NDY1NlQ0Rzdqek01SDNPRVN0Z3ArbXZiYk51VGFtazZ1RUR6?= =?utf-8?B?N3gybFd6N0pWT2dnRitvek9yZy8zaVZON0xrRDIzYUY3N1ZUL1lidFE1cFdw?= =?utf-8?B?aWRsNGdyUHVhM3NLTkt4VEl3cGkyYmxuSUhraW9OcmlsbkZQVTI1QjNlWUpj?= =?utf-8?B?MVdLOFZ3RTJ1NUwrZ2QzbE9kczhOa1RCSVFHT3NXbWZKYjBnRE90VWM1S0FU?= =?utf-8?B?bDVCWjVrUVRnb3BUZEVhMTFWbU0vcFBtNkphM2htaTVNUkZSNUF0aTNTamxy?= =?utf-8?B?bHV1VFNlczFlZzBvcDEwYk1ub1dYRjRZdW9idmVYMHRsUTVyMDlSdU8valFr?= =?utf-8?B?YkF4NWNIU1g1NUNlVWxxNDZZSmkwUnRhUFRINFc3YmJQdHNPcGRwY3VhdnEv?= =?utf-8?B?QjA0OCsvYk8wWXd3MHdsWklVRXZSL21GZkxzejJicGJTNFZZaW8xV2dBTjdC?= =?utf-8?B?SXZmYjJ3dGUxbkZFb1lONGU0Q0ZrMjd3MGUrbkoyakovUE1LcXRGWmJBdUNP?= =?utf-8?B?b2JsUjRtZ2NsUmxNeHMrSmp1K2svdnRxUVllRnA5RVpIZDdOZ1cvaDNZLzJi?= =?utf-8?B?ZzVrR3VZbytxaVNWNTh4MnhabjhEbGVlMTJYRnREamNyMWMxQ041OVZoWDlw?= =?utf-8?B?amFrcHVIQ3dZNCtXVkRCdkowV29FKzZiRDhRSlFMZDhnemhSK1NSN0hROFk4?= =?utf-8?B?N29ma1A1dmpLTCtqWEtpL2RaTGNLYVlQc3liRkNjZ2Y5YUowQnIvYTIwaVBC?= =?utf-8?B?WnBwSjRSelY3bW5xUGxWMlFTZVB4Q1hsNURhNGVwVStMTDhhT1BISzhiTlZp?= =?utf-8?B?SzlKOW9sZEZUbmcrWERiK3FJSytTeDYyOVUvU21OL3lCYW11T3NaYlpFY2Rk?= =?utf-8?B?ZG5aWU45VElvdFFXdlFhYk5lYkZuc1luK01lM0R3dUk2eEZseUhXTUN2amhi?= =?utf-8?B?bDIvZTRRT0ZnTnpTUkFBNlQ4bEV1bGxUWExtYkZSWW50RnROT3hkWnIzU0pT?= =?utf-8?B?Nkt4R1hrUWU2K21DZXJFY2Z1UzZDem1FQmZWcTVCR3k0dXIzRmMrZjI4d0E1?= =?utf-8?B?WXk2TFRTTG1Kb1BkQ25QenJsY3F4RS91WDNxaEJOUjJjTnFWOWZROEpLeHdk?= =?utf-8?B?dldyOFZ5eHVHaUUxbytqc2w0OVU2dURoWXlaWHZNWlFHZ3laeEh1eWp4ajkz?= =?utf-8?B?NjlMRG9SZGw1UWgvUmtXSU5jN0lKWWxjaGgvUGdvWFlpUnphbjU4V2gyMWZn?= =?utf-8?B?Ym9NVmYyK3JSci9wMEVHNG5LVmk3NXZwM2lMWm1zNkVHYUQvUjdzclZzVEhD?= =?utf-8?B?dDdzMHRxRk1wS2hlZmJwTFM4d3MzQ096NmZQTFR6ZjVzWVJUN3QvNCtSclBO?= =?utf-8?B?cDl4MmdSMHJ5SWtRTk14TWEycjhjdXhJVmNPUU96WHhMUFNrTllncGptY1V2?= =?utf-8?B?ZndPRjJyeVNEdU9GY3FiTGtReHFoM2hCWGNrS05zaUtPNDYrUldIQTJUbjBu?= =?utf-8?B?OGlDQ3lkbVVSbEZDbWFjY01MUjVGa2hvZTMwcnhtdjYyanlERFExaHhwQy9p?= =?utf-8?B?OUt4R29tdVBrdjNSUHNHeXFHS3hjYjVMazlSUTNmRFNZR1AvZy9kNWZHTU1v?= =?utf-8?B?YWNTb0t3VmlZUXJ4STFObDhYbzZuYnpJQUdISTByY3VuNWREZjAyTFlFdm0w?= =?utf-8?B?ZmRtZDZrcjlsRjZPNW5iWXgyT2Y3aG1YVFdNZVl4OEZacVZoektqQlR0MEJR?= =?utf-8?B?R3d0YStMMnMza0xMUXpzb0JMcnFGK3FTQnNLV0RCSTdrZW05MHYwQkFueUFJ?= =?utf-8?B?Ly9Pd1Evd1dzVTR5eUNzbEtlMmlKaElEbllTbWY3c2pYelNFRVViU2xWV1o1?= =?utf-8?B?eFh6Q1BsQVBpRnRCTFNvd0hhZS95L0FXeld5NlFjc2FEaXBiWkF0aUcvSGxW?= =?utf-8?B?SXdDeXFWSFhhMno1NHBBTm9FdDFucUlsQU1INDlROVFKa1Z3WExjcHBPNXNm?= =?utf-8?Q?5bQSoeIWTHDhTpUYIn?= X-OriginatorOrg: Nvidia.com X-MS-Exchange-CrossTenant-Network-Message-Id: 282fbff5-acdc-429b-bdad-08df2638ccb5 X-MS-Exchange-CrossTenant-AuthSource: IA0PR12MB8374.namprd12.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 09 Oct 2026 19:09:10.6390 (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: 4tJTOgt83C1+Lu3f8IMAXPe0+5+hRfHIOZUGVFE7skv47HIbzJa9UEvTcmUUfHzj X-MS-Exchange-Transport-CrossTenantHeadersStamped: MW6PR12MB8760 On Fri Oct 2, 2026 at 5:52 AM EDT, Usama Arif wrote: > swapoff walks every mm and faults every slot of the device back in. > unuse_pte_range() only understands PTEs, so a PMD swap entry would never = be > found and swapoff would never complete. > > A PMD swap entry is a compact encoding for HPAGE_PMD_NR slots, not a > promise that the swap cache holds one folio for them. Add > swap_pmd_cache_lookup() to classify the covered range as empty, one > PMD-sized folio, or already split, and unuse_pmd() to map the first two > cases back in as one THP, preserving soft-dirty, exclusive and UFFD state= . > > Everything else falls back to PTEs: a split cache, per-page zswap state, = a > failed PMD-order allocation or read, or a poisoned subpage. Check > PageHWPoison on every subpage rather than the folio-level flag, which > memory_failure() only sets after taking the folio lock. > > All the fallback reasons are observed without the PMD lock and possibly > after sleeping, so they share one exit that re-checks the PMD is still th= e > entry we were called for before splitting it. That exit also drops a foli= o > that is not uptodate, or that has never been mapped, from the swap cache: > the PTE path cannot re-read the first, and would add a single-page rmap t= o > the second. > > Signed-off-by: Usama Arif > --- > mm/internal.h | 16 ++++ > mm/swap.h | 17 +++++ > mm/swap_state.c | 44 +++++++++++ > mm/swapfile.c | 189 ++++++++++++++++++++++++++++++++++++++++++++++++ > 4 files changed, 266 insertions(+) > > diff --git a/mm/internal.h b/mm/internal.h > index 05179c4b2090e..ec7f007bc2c0d 100644 > --- a/mm/internal.h > +++ b/mm/internal.h > @@ -24,6 +24,22 @@ > =20 > struct folio_batch; > =20 > +/* A high level description like below might be better. Check every page in a folio for HWPoison flag. Return true if there is any. > + * Unlike folio_contain_hwpoisoned_page(), this does not rely on the fol= io-level > + * PG_has_hwpoisoned, which memory_failure() only sets after taking the = folio > + * lock and so can lag a tail-page poison. > + */ > +static inline bool folio_has_hwpoisoned_subpage(const struct folio *foli= o) s/subpage/page, IIRC, we are moving away from subpage. Maybe folio_has_any_hwpoisoned_page()? > +{ > + long nr =3D folio_nr_pages(folio); > + long i; > + > + for (i =3D 0; i < nr; i++) > + if (PageHWPoison(folio_page(folio, i))) > + return true; > + return false; > +} > + --=20 Best Regards, Yan, Zi