From patchwork Fri Mar 14 22:21:12 2025 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Zi Yan X-Patchwork-Id: 14017550 Received: from NAM10-DM6-obe.outbound.protection.outlook.com (mail-dm6nam10on2083.outbound.protection.outlook.com [40.107.93.83]) (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 00EEB1C8632; Fri, 14 Mar 2025 22:21:29 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=fail smtp.client-ip=40.107.93.83 ARC-Seal: i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1741990891; cv=fail; b=PWsiMelJTrWEzjV09XnXpCZtOunqwZ26nMYHTn5bdTp2NAXEVJEB2lYqeBuWKtf8Ft5tQ/kOszV22isWFmHQQK05ZBQ30xDJgudo1Eai89XcAzPiZya4ug18b/66Lrd3nQNmdUDGgxwjRanmOX/UVErdsAyOrmvqyIFshJ17SQM= ARC-Message-Signature: i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1741990891; c=relaxed/simple; bh=eSrkF9WnEy9WvdbV3jHXPc5gfAFFvEJjCG3z8X/4BvM=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: Content-Type:MIME-Version; b=ZWtzRoTgM9Sxl6FWfm+ZpWV2bsYMmqCWXznipTJ0z4etO6Yo6VdeDkUDmP+q3fJ6+ZMpH4ZJ7kxa4QyT+aEYb+AumrPWknih5nRU4mqn5gQk+MOG2+ojb9QGgxhC052FIcIFrTS0qbPughM88lgRzcCZiVe/gXRC9Qzv+ieR834= 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=usCmkQ0G; arc=fail smtp.client-ip=40.107.93.83 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="usCmkQ0G" ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=BWOxhsGo2/O/XzwSNFl9yNKy0iLvWuui69w50b72CE8ftVvF5xg8rUYfwHdwOmATdE4DbY2pmj1dn9vAVktaSEis80x7n7SOViWI22V8U4JwSZ2GNiAN7A5b5pfS6+rMHL4v5heJliFYhMUuvxef1gD4Q7k2q238YohsnEEdSoPa+8RRIPj7K0r2AuH2EF3DLt8lbvqhjb/hsWCG+F2uG122+eruolv4NEZAtuyopS424VaYAsCtc5oSOQgtKzqwyalBz8VZewRjjTS2s1+30VY8+UBp37WSpMYsItKkk5q6BIcDOYe4McGDw5mo7ColyLjftoJqEsXVEf+X8n3a8A== 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=hYriJ3ifTBG6aSAfwgreb84HuxgdP+89iJ6qwOfd5lA=; b=mFYof6wum1LTIV/PDcJX80ja4hYsJ2cmOSuschGcTBlDolsU1dTdZavG3mxSoYBr44GRdSBxA4fDeGohGhcXiymJwv1YId0QZP3JOg4eB+I3k/G+mtzJoMOjFYCzqLenUKSnIxljKvC4ten943t1l8/+eE/HhNV6ILtrdRuZ/gS+d3ZPssvNeLFcm6EuxUcNbpPn5y4BgSXYgPr2eiQKuPn0bI9x/StikDd5yn/VqEbshn4+kd3VP6Rh7eTudfmmU5JPtdxhikGThbd8moAUvBvBbSEts+CHvgbr2/MV9XLYKFW1tuz6Lf1QC3MJ4zOtOE4OWlWFaWdXquA6igdrBg== 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=hYriJ3ifTBG6aSAfwgreb84HuxgdP+89iJ6qwOfd5lA=; b=usCmkQ0GzZ3G5PTP8X/AZxLbd93shEYt56z2vbrz91vOV9L3zIeB00BBJ5ns1TRwxJYfZv2t0/02ab2I5o9ZRJSlV1HBPi8piu+iW4uQOcDDjAvcN2WBaACtb02wJxjvzkmNKOjwe240w2WR8cteh/Hvt+NqtdLTVK/VBMN9SJgqM2EP5Og7VklTjwAw8PWYX32TlCeSu/bME2mtbY40g0Da3FZRu+jnSxlhy5PUSyjDo2Z76BHmhJSFlACFj82O7TaMg1jbjx2f86zraJL43Ypge4f0lU07PkHrXdvsc73snjA51owOO4Sm30I7uiAfhPRMRhqxbynd1uZw9vbujQ== Authentication-Results: dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=nvidia.com; Received: from DS7PR12MB9473.namprd12.prod.outlook.com (2603:10b6:8:252::5) by PH7PR12MB9076.namprd12.prod.outlook.com (2603:10b6:510:2f6::21) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.8534.28; Fri, 14 Mar 2025 22:21:25 +0000 Received: from DS7PR12MB9473.namprd12.prod.outlook.com ([fe80::5189:ecec:d84a:133a]) by DS7PR12MB9473.namprd12.prod.outlook.com ([fe80::5189:ecec:d84a:133a%5]) with mapi id 15.20.8534.027; Fri, 14 Mar 2025 22:21:25 +0000 From: Zi Yan To: Andrew Morton , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org Cc: Baolin Wang , Matthew Wilcox , Hugh Dickins , Kairui Song , Miaohe Lin , SeongJae Park , linux-kernel@vger.kernel.org, Zi Yan , David Hildenbrand , John Hubbard , Kefeng Wang , "Kirill A. Shuemov" , Ryan Roberts , Yang Shi , Yu Zhao Subject: [PATCH RESEND v3 1/2] mm/filemap: use xas_try_split() in __filemap_add_folio() Date: Fri, 14 Mar 2025 18:21:12 -0400 Message-ID: <20250314222113.711703-2-ziy@nvidia.com> X-Mailer: git-send-email 2.47.2 In-Reply-To: <20250314222113.711703-1-ziy@nvidia.com> References: <20250314222113.711703-1-ziy@nvidia.com> X-ClientProxiedBy: BLAP220CA0028.NAMP220.PROD.OUTLOOK.COM (2603:10b6:208:32c::33) To DS7PR12MB9473.namprd12.prod.outlook.com (2603:10b6:8:252::5) Precedence: bulk X-Mailing-List: linux-fsdevel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: DS7PR12MB9473:EE_|PH7PR12MB9076:EE_ X-MS-Office365-Filtering-Correlation-Id: d13e0440-fadb-40e2-b10a-08dd63468eee X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230040|366016|7416014|376014|1800799024; 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)(366016)(7416014)(376014)(1800799024);DIR:OUT;SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: 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 X-OriginatorOrg: Nvidia.com X-MS-Exchange-CrossTenant-Network-Message-Id: d13e0440-fadb-40e2-b10a-08dd63468eee X-MS-Exchange-CrossTenant-AuthSource: DS7PR12MB9473.namprd12.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 14 Mar 2025 22:21:25.4798 (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: TxJGLKvGSsl8ZNmDVpJjd6ery1a7bHQGUoVWiyvwQ+sucwN55Nv6rSX0K5pQwceX X-MS-Exchange-Transport-CrossTenantHeadersStamped: PH7PR12MB9076 During __filemap_add_folio(), a shadow entry is covering n slots and a folio covers m slots with m < n is to be added. Instead of splitting all n slots, only the m slots covered by the folio need to be split and the remaining n-m shadow entries can be retained with orders ranging from m to n-1. This method only requires (n/XA_CHUNK_SHIFT) - (m/XA_CHUNK_SHIFT) new xa_nodes instead of (n % XA_CHUNK_SHIFT) * ((n/XA_CHUNK_SHIFT) - (m/XA_CHUNK_SHIFT)) new xa_nodes, compared to the original xas_split_alloc() + xas_split() one. For example, to insert an order-0 folio when an order-9 shadow entry is present (assuming XA_CHUNK_SHIFT is 6), 1 xa_node is needed instead of 8. xas_try_split_min_order() is introduced to reduce the number of calls to xas_try_split() during split. Cc: Baolin Wang Cc: Hugh Dickins Cc: Kairui Song Cc: Miaohe Lin Cc: Mattew Wilcox Cc: David Hildenbrand Cc: John Hubbard Cc: Kefeng Wang Cc: Kirill A. Shuemov Cc: Ryan Roberts Cc: Yang Shi Cc: Yu Zhao Signed-off-by: Zi Yan --- include/linux/xarray.h | 7 +++++++ lib/xarray.c | 25 +++++++++++++++++++++++ mm/filemap.c | 45 +++++++++++++++++------------------------- 3 files changed, 50 insertions(+), 27 deletions(-) diff --git a/include/linux/xarray.h b/include/linux/xarray.h index 4010195201c9..78eede109b1a 100644 --- a/include/linux/xarray.h +++ b/include/linux/xarray.h @@ -1556,6 +1556,7 @@ int xas_get_order(struct xa_state *xas); void xas_split(struct xa_state *, void *entry, unsigned int order); void xas_split_alloc(struct xa_state *, void *entry, unsigned int order, gfp_t); void xas_try_split(struct xa_state *xas, void *entry, unsigned int order); +unsigned int xas_try_split_min_order(unsigned int order); #else static inline int xa_get_order(struct xarray *xa, unsigned long index) { @@ -1582,6 +1583,12 @@ static inline void xas_try_split(struct xa_state *xas, void *entry, unsigned int order) { } + +static inline unsigned int xas_try_split_min_order(unsigned int order) +{ + return 0; +} + #endif /** diff --git a/lib/xarray.c b/lib/xarray.c index 3bae48558e21..9644b18af18d 100644 --- a/lib/xarray.c +++ b/lib/xarray.c @@ -1134,6 +1134,28 @@ void xas_split(struct xa_state *xas, void *entry, unsigned int order) } EXPORT_SYMBOL_GPL(xas_split); +/** + * xas_try_split_min_order() - Minimal split order xas_try_split() can accept + * @order: Current entry order. + * + * xas_try_split() can split a multi-index entry to smaller than @order - 1 if + * no new xa_node is needed. This function provides the minimal order + * xas_try_split() supports. + * + * Return: the minimal order xas_try_split() supports + * + * Context: Any context. + * + */ +unsigned int xas_try_split_min_order(unsigned int order) +{ + if (order % XA_CHUNK_SHIFT == 0) + return order == 0 ? 0 : order - 1; + + return order - (order % XA_CHUNK_SHIFT); +} +EXPORT_SYMBOL_GPL(xas_try_split_min_order); + /** * xas_try_split() - Try to split a multi-index entry. * @xas: XArray operation state. @@ -1145,6 +1167,9 @@ EXPORT_SYMBOL_GPL(xas_split); * needed, the function will use GFP_NOWAIT to get one if xas->xa_alloc is * NULL. If more new xa_node are needed, the function gives EINVAL error. * + * NOTE: use xas_try_split_min_order() to get next split order instead of + * @order - 1 if you want to minmize xas_try_split() calls. + * * Context: Any context. The caller should hold the xa_lock. */ void xas_try_split(struct xa_state *xas, void *entry, unsigned int order) diff --git a/mm/filemap.c b/mm/filemap.c index 152993a86de3..cc69f174f76b 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -857,11 +857,10 @@ EXPORT_SYMBOL_GPL(replace_page_cache_folio); noinline int __filemap_add_folio(struct address_space *mapping, struct folio *folio, pgoff_t index, gfp_t gfp, void **shadowp) { - XA_STATE(xas, &mapping->i_pages, index); - void *alloced_shadow = NULL; - int alloced_order = 0; + XA_STATE_ORDER(xas, &mapping->i_pages, index, folio_order(folio)); bool huge; long nr; + unsigned int forder = folio_order(folio); VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio); VM_BUG_ON_FOLIO(folio_test_swapbacked(folio), folio); @@ -870,7 +869,6 @@ noinline int __filemap_add_folio(struct address_space *mapping, mapping_set_update(&xas, mapping); VM_BUG_ON_FOLIO(index & (folio_nr_pages(folio) - 1), folio); - xas_set_order(&xas, index, folio_order(folio)); huge = folio_test_hugetlb(folio); nr = folio_nr_pages(folio); @@ -880,7 +878,7 @@ noinline int __filemap_add_folio(struct address_space *mapping, folio->index = xas.xa_index; for (;;) { - int order = -1, split_order = 0; + int order = -1; void *entry, *old = NULL; xas_lock_irq(&xas); @@ -898,21 +896,25 @@ noinline int __filemap_add_folio(struct address_space *mapping, order = xas_get_order(&xas); } - /* entry may have changed before we re-acquire the lock */ - if (alloced_order && (old != alloced_shadow || order != alloced_order)) { - xas_destroy(&xas); - alloced_order = 0; - } - if (old) { - if (order > 0 && order > folio_order(folio)) { + if (order > 0 && order > forder) { + unsigned int split_order = max(forder, + xas_try_split_min_order(order)); + /* How to handle large swap entries? */ BUG_ON(shmem_mapping(mapping)); - if (!alloced_order) { - split_order = order; - goto unlock; + + while (order > forder) { + xas_set_order(&xas, index, split_order); + xas_try_split(&xas, old, order); + if (xas_error(&xas)) + goto unlock; + order = split_order; + split_order = + max(xas_try_split_min_order( + split_order), + forder); } - xas_split(&xas, old, order); xas_reset(&xas); } if (shadowp) @@ -936,17 +938,6 @@ noinline int __filemap_add_folio(struct address_space *mapping, unlock: xas_unlock_irq(&xas); - /* split needed, alloc here and retry. */ - if (split_order) { - xas_split_alloc(&xas, old, split_order, gfp); - if (xas_error(&xas)) - goto error; - alloced_shadow = old; - alloced_order = split_order; - xas_reset(&xas); - continue; - } - if (!xas_nomem(&xas, gfp)) break; } From patchwork Fri Mar 14 22:21:13 2025 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Zi Yan X-Patchwork-Id: 14017551 Received: from NAM10-DM6-obe.outbound.protection.outlook.com (mail-dm6nam10on2083.outbound.protection.outlook.com [40.107.93.83]) (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 F219B1FF1AF; Fri, 14 Mar 2025 22:21:31 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=fail smtp.client-ip=40.107.93.83 ARC-Seal: i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1741990894; cv=fail; b=GRKnaeuFWPjnOetZpMU+NyFaPIH29Us+tFP10BWALBmx8855u56j8e0EPlZergDOWOEsdGwBIR7W5hLctL0WViHN5h+KBqPCF7WVd81igeR8oMiaDON3wL1RoVOwxypJafl6FTc+WQXCVpnHPlxvhn2Rf7hBbbhOG+whLFqWUC8= ARC-Message-Signature: i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1741990894; c=relaxed/simple; bh=8HxKjQiXzFOjObuQ350/K7vUpTPUEWL4tXaXpCUvna8=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: Content-Type:MIME-Version; b=dX3VDPA+XWjlg5Mx5jd9lOJf1ZuMtriZwb0MSAMql6EFEeHXcBkFICaH3pHz1Z5pLjLA7AG3xKuJyoesKrR8zW4/Zwysii+2WWvQN3J6rrVK46wUac/+fZ23xvP/FiUeXIdgZIVE179ObhqXNEx0hFbmZ44uzSRKH+/onwobGx0= 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=fEIyhnK0; arc=fail smtp.client-ip=40.107.93.83 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="fEIyhnK0" ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=HWRcXeuqnt55bxBKk7MSiWLlVloYATE9G529gxAV+U8OmdbGIgVKxjhfslFWAh5SqeZFVia/MPvESwMQQVoqRguaj3iZ81Ubbh+JOmXHc9Y+AchRi7X+K1qfnIYzeCMNG5bsheqxKawb2Ej83gKk92uc1ypvvw3t9oLkyRLl482E83UZLQAFlXIM3eDU0hSoBHbx5afbbV5qf75D6RtHaKcGW5JWm04R6asAbnb2FV3vuiUMFiNQO1S7Wp84fFszBb4iRaUVPcXWRE9+KW16FXz+NlU6UjBt9iPYbGc/wQYDE+/iDOG5GsJcoFFgqFTV1K9x1vz6utHomks25zd54Q== 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=m3vCypCdoWVadCZlANYmq2YAis7NDT4qJ+tzvN0ESSc=; b=apkspoFUWCACgfUvZSWNwdxUfw8on1farTBRvf2zs5gcho8C4by1P3/rrLEtQRPgh7cRCSRrSS1DAttnYY+AdmF5DXAB6rfE9EboqcDYsiHLgQp5yEPZ83w7z6xn/rOfALp4NQlYfOeVzwQWEa4Ac2kSv08SgB4JptCVCwla6rwwG6Tt/EurGY2zy3Gd7m6tcYrussd8Irz9ogNQmicm3Aml5Sh4S096oQKww4ooJHWnwOctNgyDY2I4/WSa5ygB6b+1dZpgmNTkgifWpxHTwT/j7SO4YEq7GBUjSfckY0pqCcjNuFdzexharUxOJ8jI6v/Ey3cZ+ajBxzgc1gqCtQ== 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=m3vCypCdoWVadCZlANYmq2YAis7NDT4qJ+tzvN0ESSc=; b=fEIyhnK0c4DLYL/liscW/qI4a8q2fqaM0+XPG8g2E7pnaM1O6Q/7eFH96zFSagUsq5wNT+Np2i3hEqi8nWM2IZRMXkQeQp+79ZPHrZYY2CdBBJ1vm1MBzK9An5XnvhSsG7fJVU5M2z3MR10MNGkAQiuqrPhOhnxn1IhIJGh9NmqoPkW0tZ8zL94KRHayzdc53c2v4yFLuNxF/9Uf14IOjxIfPQ7Fjk/VoNbPl2rxT9ZV9JgwUM1qjvP9t88OCPdMPqcsSB7kT7yHhsfO1gJcCE6SM40hYgFbh6WV0mpossBFW5gPtW4I7NVKm9uq80Uj1EkXijumjYo06omHSJzgdw== Authentication-Results: dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=nvidia.com; Received: from DS7PR12MB9473.namprd12.prod.outlook.com (2603:10b6:8:252::5) by PH7PR12MB9076.namprd12.prod.outlook.com (2603:10b6:510:2f6::21) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.8534.28; Fri, 14 Mar 2025 22:21:26 +0000 Received: from DS7PR12MB9473.namprd12.prod.outlook.com ([fe80::5189:ecec:d84a:133a]) by DS7PR12MB9473.namprd12.prod.outlook.com ([fe80::5189:ecec:d84a:133a%5]) with mapi id 15.20.8534.027; Fri, 14 Mar 2025 22:21:26 +0000 From: Zi Yan To: Andrew Morton , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org Cc: Baolin Wang , Matthew Wilcox , Hugh Dickins , Kairui Song , Miaohe Lin , SeongJae Park , linux-kernel@vger.kernel.org, Zi Yan , David Hildenbrand , John Hubbard , Kefeng Wang , "Kirill A. Shuemov" , Ryan Roberts , Yang Shi , Yu Zhao Subject: [PATCH RESEND v3 2/2] mm/shmem: use xas_try_split() in shmem_split_large_entry() Date: Fri, 14 Mar 2025 18:21:13 -0400 Message-ID: <20250314222113.711703-3-ziy@nvidia.com> X-Mailer: git-send-email 2.47.2 In-Reply-To: <20250314222113.711703-1-ziy@nvidia.com> References: <20250314222113.711703-1-ziy@nvidia.com> X-ClientProxiedBy: BLAP220CA0016.NAMP220.PROD.OUTLOOK.COM (2603:10b6:208:32c::21) To DS7PR12MB9473.namprd12.prod.outlook.com (2603:10b6:8:252::5) Precedence: bulk X-Mailing-List: linux-fsdevel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: DS7PR12MB9473:EE_|PH7PR12MB9076:EE_ X-MS-Office365-Filtering-Correlation-Id: 1d2b7342-8ed7-4ce9-c9d5-08dd63468fb2 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230040|366016|7416014|376014|1800799024; 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)(366016)(7416014)(376014)(1800799024);DIR:OUT;SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: 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 X-OriginatorOrg: Nvidia.com X-MS-Exchange-CrossTenant-Network-Message-Id: 1d2b7342-8ed7-4ce9-c9d5-08dd63468fb2 X-MS-Exchange-CrossTenant-AuthSource: DS7PR12MB9473.namprd12.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 14 Mar 2025 22:21:26.7507 (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: 9sCE0RSIKd565HoNOhvb7OIj+wJXP4agNmQ8YzJm7RTMvTc2Hx8ukhyOPNHuMBfI X-MS-Exchange-Transport-CrossTenantHeadersStamped: PH7PR12MB9076 During shmem_split_large_entry(), large swap entries are covering n slots and an order-0 folio needs to be inserted. Instead of splitting all n slots, only the 1 slot covered by the folio need to be split and the remaining n-1 shadow entries can be retained with orders ranging from 0 to n-1. This method only requires (n/XA_CHUNK_SHIFT) new xa_nodes instead of (n % XA_CHUNK_SHIFT) * (n/XA_CHUNK_SHIFT) new xa_nodes, compared to the original xas_split_alloc() + xas_split() one. For example, to split an order-9 large swap entry (assuming XA_CHUNK_SHIFT is 6), 1 xa_node is needed instead of 8. xas_try_split_min_order() is used to reduce the number of calls to xas_try_split() during split. Cc: Baolin Wang Cc: Hugh Dickins Cc: Kairui Song Cc: Mattew Wilcox Cc: Miaohe Lin Cc: David Hildenbrand Cc: John Hubbard Cc: Kefeng Wang Cc: Kirill A. Shuemov Cc: Ryan Roberts Cc: Yang Shi Cc: Yu Zhao Reviewed-by: Baolin Wang Tested-by: Baolin Wang Signed-off-by: Zi Yan --- mm/shmem.c | 59 ++++++++++++++++++++++++++---------------------------- 1 file changed, 28 insertions(+), 31 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index 5586903950b3..7b738d8d6581 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2153,15 +2153,16 @@ static int shmem_split_large_entry(struct inode *inode, pgoff_t index, { struct address_space *mapping = inode->i_mapping; XA_STATE_ORDER(xas, &mapping->i_pages, index, 0); - void *alloced_shadow = NULL; - int alloced_order = 0, i; + int split_order = 0, entry_order; + int i; /* Convert user data gfp flags to xarray node gfp flags */ gfp &= GFP_RECLAIM_MASK; for (;;) { - int order = -1, split_order = 0; void *old = NULL; + int cur_order; + pgoff_t swap_index; xas_lock_irq(&xas); old = xas_load(&xas); @@ -2170,60 +2171,56 @@ static int shmem_split_large_entry(struct inode *inode, pgoff_t index, goto unlock; } - order = xas_get_order(&xas); + entry_order = xas_get_order(&xas); - /* Swap entry may have changed before we re-acquire the lock */ - if (alloced_order && - (old != alloced_shadow || order != alloced_order)) { - xas_destroy(&xas); - alloced_order = 0; - } + if (!entry_order) + goto unlock; /* Try to split large swap entry in pagecache */ - if (order > 0) { - if (!alloced_order) { - split_order = order; + cur_order = entry_order; + swap_index = round_down(index, 1 << entry_order); + + split_order = xas_try_split_min_order(cur_order); + + while (cur_order > 0) { + pgoff_t aligned_index = + round_down(index, 1 << cur_order); + pgoff_t swap_offset = aligned_index - swap_index; + + xas_set_order(&xas, index, split_order); + xas_try_split(&xas, old, cur_order); + if (xas_error(&xas)) goto unlock; - } - xas_split(&xas, old, order); /* * Re-set the swap entry after splitting, and the swap * offset of the original large entry must be continuous. */ - for (i = 0; i < 1 << order; i++) { - pgoff_t aligned_index = round_down(index, 1 << order); + for (i = 0; i < 1 << cur_order; + i += (1 << split_order)) { swp_entry_t tmp; - tmp = swp_entry(swp_type(swap), swp_offset(swap) + i); + tmp = swp_entry(swp_type(swap), + swp_offset(swap) + swap_offset + + i); __xa_store(&mapping->i_pages, aligned_index + i, swp_to_radix_entry(tmp), 0); } + cur_order = split_order; + split_order = xas_try_split_min_order(split_order); } unlock: xas_unlock_irq(&xas); - /* split needed, alloc here and retry. */ - if (split_order) { - xas_split_alloc(&xas, old, split_order, gfp); - if (xas_error(&xas)) - goto error; - alloced_shadow = old; - alloced_order = split_order; - xas_reset(&xas); - continue; - } - if (!xas_nomem(&xas, gfp)) break; } -error: if (xas_error(&xas)) return xas_error(&xas); - return alloced_order; + return entry_order; } /*