From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1) server-digest SHA256) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 71C7D606A692 for ; Fri, 20 Nov 2026 21:28:44 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key, unprotected) header.d=amd.com header.i=@amd.com header.a=rsa-sha256 header.s=selector1 header.b=t/XrZv1T Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id C5E2B112D3A; Fri, 20 Nov 2026 21:28:38 +0000 (UTC) Received: from BN1PR04CU002.outbound.protection.outlook.com (mail-eastus2azon11010000.outbound.protection.outlook.com [52.101.56.0]) by gabe.freedesktop.org (Postfix) with ESMTPS id 833B210E348 for ; Fri, 20 Nov 2026 21:28:32 +0000 (UTC) ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=EJ7hMkU1EES5ua/SlyxKWoHVMCCrW5Y1Wr/0fBqiZzwEs+QcJ9gUGPHuNpe19hQxSUuhzdpLh8hprvw8tn1tp5KBQieSZqpYW4GNDIhizwBoAkyV+5unr6g6+uYDHs8oNSc50DFcNZ1kBcI6oL31i1fCO9+QpyrY1xua9MHyxO1ylE8RC1tl7d8LJVB3vPZ9KFXHKGU2DVCMsvZ7gXf2p3g4WNlmg3Zu3tIg8N+kzuZBstgt/GsZYtumteOjlllarVaVni4PWuyJy1IsHlN4PxBYyqaHkCMyl+uQwWRCaPQjQpiiO8/KYcmRNZtrbnRTC1Lv2JCAZU0AyoyQ3hGHkQ== 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=nIVg+KmdOULtyvX9w3Gpe54fRoI87SNPOeN+si2A+Ek=; b=M2K4LTZxumdH7RNjDlrddoZuggddKqlaUN70jQy4Zx0WYD+wdKNxjQbcVpHEZBep9xcLpzQfW7dI7Adzzs4l3Qs5741VtdPneYmtvWiy5mxLXpb6rX5JGnDn5DdjujLIlBcJH6U/9ExPsQJ2GIZJAN349feZj+pcXkZPzFVBzZ74uUCctDnL7wIZnUmqkq760W812CA7/KPWz7MKu8nK1F7KYrHSovm2jSL0G1v/vBnsH+RLcy0EJuElVO9C39AB76v0ftG3weHgQPgc7m82oaFXYAPfaIiiPOLDVBUu/NKjXnXoQvubaW7fIzpJIBEJ5tGckm2Z4VFrziaICD5DRg== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=amd.com; dmarc=pass action=none header.from=amd.com; dkim=pass header.d=amd.com; arc=none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=amd.com; s=selector1; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=nIVg+KmdOULtyvX9w3Gpe54fRoI87SNPOeN+si2A+Ek=; b=t/XrZv1TiXZpNoEt4D0k/KJg/eDss19p3qFvDbwCFeHxXjUN/5JaYS+lGQJYQB+HYgcx8XCGaSWhniPI4wJ4qDVIdE/Ql7hpX4MFIFHUTRLdcE4qujaQYIXB4nEhh7GBwHT7dYqpu+BCwyiRyK1fTo8aO/Mck9rWHxRWy9jro5A= Authentication-Results: mx.microsoft.com 1; dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=amd.com; Received: from LV8SPRMB0063.namprd12.prod.outlook.com (2603:10b6:408:242::10) by MN0PR12MB5954.namprd12.prod.outlook.com (2603:10b6:208:37d::15) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.639.12; Fri, 20 Nov 2026 21:28:27 +0000 Received: from LV8SPRMB0063.namprd12.prod.outlook.com ([fe80::6cb:fff:7dc4:2fc0]) by LV8SPRMB0063.namprd12.prod.outlook.com ([fe80::6cb:fff:7dc4:2fc0%4]) with mapi id 15.21.0615.013; Fri, 20 Nov 2026 21:28:27 +0000 Message-ID: <9dacfc65-7554-466e-bb71-63b853386278@amd.com> Date: Fri, 20 Nov 2026 16:28:24 -0500 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH 1/4] drm/amdgpu: limit userptr sg segments to half of sg->length To: William Palacek , amd-gfx@lists.freedesktop.org Cc: Harish.Kasiviswanathan@amd.com, christian.koenig@amd.com, philip.yang@amd.com References: <5a6eb18599b44e0457abdc9d54a00dbb0060f10c.1795014593.git.William.Palacek@amd.com> Content-Language: en-US From: Felix Kuehling Organization: AMD Inc. In-Reply-To: <5a6eb18599b44e0457abdc9d54a00dbb0060f10c.1795014593.git.William.Palacek@amd.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-ClientProxiedBy: YQBPR0101CA0212.CANPRD01.PROD.OUTLOOK.COM (2603:10b6:c01:67::18) To LV8SPRMB0063.namprd12.prod.outlook.com (2603:10b6:408:242::10) X-Archived-By: MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: LV8SPRMB0063:EE_|MN0PR12MB5954:EE_ X-MS-Office365-Filtering-Correlation-Id: 84ad4965-04ac-48c4-47d8-08df474d36bf X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0; ARA:13230040|376014|23010399003|366016|1800799024|10067099003|261022105027099003|4143699003|56012099006|11063799006|261009223027099003|261021121427099003|261021123027099003|261102123227099003|261026140227099003|261028144727099003|261026121527099003|18002099003|22082099003|6133799003; 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:LV8SPRMB0063.namprd12.prod.outlook.com; PTR:; CAT:NONE; SFS:(13230040)(376014)(23010399003)(366016)(1800799024)(10067099003)(261022105027099003)(4143699003)(56012099006)(11063799006)(261009223027099003)(261021121427099003)(261021123027099003)(261102123227099003)(261026140227099003)(261028144727099003)(261026121527099003)(18002099003)(22082099003)(6133799003); DIR:OUT; SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?cXduZTh1YzluSU5VQVYrejk0ZEt0NmFNU2E1YnU4VE55VXpPck83eVVlZC9h?= =?utf-8?B?cWY0Wm9VdU9qak8xMGV3MWduZnFkZTlkS1FMNkxVbENTWEEwdEc2U2EzQXRk?= =?utf-8?B?NDVRbU1kdEVuaUpPemJxMnVGcDFTdEFMcTM5eTBUSnlXVWVwczR2SkxvNVNh?= =?utf-8?B?Q1N1dHFrRFNyZFFNRU5oSWF3M0pHS0tJKy9mTEFXYzRhcEZIWWlDaHU5Nm1J?= =?utf-8?B?MzB5MytUeE5mcFFSejhPZVJ2dXV1TmEyM0ViaDhlamVzUHNGL3B0K2hVSkRR?= =?utf-8?B?SlVlZ0lrSU9oZk9RY24yNmpLNUdVMTI5MmhWb0FwbXRMeG96N1hDRU43UHZ5?= =?utf-8?B?YXR0TWI3dlpLVnFmMitVOGs4UW5WajRnY25wMFl6cHdUeXQvNFFtQ0ZlSERi?= =?utf-8?B?LzJZZ2xqbUx5YUFMRmxQZld0eUk0ZWxPVVY5d3JXalg2d1JTL0pCMGRNRmgy?= =?utf-8?B?Qyt5b0ZuWVhVeVc0OUtnNms5WVE3U0h0WXZRRzMybG9mUy9JUVhhbHlmVzNy?= =?utf-8?B?RjIvcWJMS3lGQ3pjNHpnZW8zdGFXQ1dMUkl5MnFEcUREVjFQd09RQlN6SnNX?= =?utf-8?B?SThya0tJN2wrUHpQcWdjSGJFcTZ4KzUyalJ6MG45L1Z3bzhaL3ZIZm1sQmVE?= =?utf-8?B?d3Rkci9tTzI2cFhRMjZwd01KQ2RrWkZpNFBrbGthVDhuMzQ1M1BLT2hJZG1L?= =?utf-8?B?a2wza1puT2ROWW5jb3pKaTEyblBZWUZYdUs4emFRMnFzYnQwMFpISnBiOXBo?= =?utf-8?B?QUsyamhYTzdNZkRlaVNJcG9oK1I4aGg4V3BFeFZVYk1iQWlwbVM0aFRFNzhR?= =?utf-8?B?Y3NJa0s2TStWWXg0S2lhdURuKzdaVXE1SDlWVFpuL1FHQmp1Q1RVeDN6Sys2?= =?utf-8?B?L3J2NEpUSDd4MDJOL1hoM040aFl1blNHSlNXYXhhYjBlQ2tDRE1NMU9BYWs1?= =?utf-8?B?V0hXQXJzVXZZN2NYWDdLcGdZeDVBa2JoclRoN25zV1luSHQvaHUwbHI4eU1C?= =?utf-8?B?UEdBMU9CUjN3L3dnR3U1ZVZ2NFhrVUNCaHFXbEhJQ3h6WTJKeWxPdUwrMFNh?= =?utf-8?B?WXlDa1lUOWVpV2RuWVpNWC9WY2draDliYkFrbHdjRTFvME10b3lDSlAxWXYw?= =?utf-8?B?Vk9XbkJ1MnY2a1B5SGJwQjd5bUNRd3NOc2FPQXAyWGtsNkhrN0Y4Z1lPNVh0?= =?utf-8?B?R2tsUlorNGhNUEpuVHVSVFQxSWo1dHFkSWlKTHFjQnprRDZaTGVjUXBMNU9P?= =?utf-8?B?U0dhemZQS2hFd2s2aFRaeVZXdDc4SGY5dExBaHd4Nk5wRkErdThjcHRkcFVh?= =?utf-8?B?YVFoSTJOeFZjcUFLaUFFM2dxQk1hYm9tQmd6TE8wemtFTGE5ejJpdFg1a3U4?= =?utf-8?B?blIrdVBtWnRoZ3krM0dNcCtPSFBOcjFLOWcxZU1RUTdPSmxORXFtcHlmSlF3?= =?utf-8?B?VVdjVk16SjQ1U0FLaTBXS2l0MEUxTjRJR2J1SHkrOVVmaS9jd1M2Rk5JVmsz?= =?utf-8?B?ZXViZ3FESEVWVUo1cGxyK2loZVZ5Yk0zTVhCM3lPTW5TVzVOcnhsVGlpZDdK?= =?utf-8?B?MllvZERucElhUUZUb1BiQ1M3dS9VU2VFejBiSDUrT2pIYTRodVFqMHZORExo?= =?utf-8?B?WjQ0MzFNYjdpWnYycE5IRW51bWJWWWhBWFNSb3RkRVpLcVJwU3FvU1ZKeGtt?= =?utf-8?B?T0Zpd1RRVUcwT1Vra1JSNXBHbi9RUEphN2lWcDlPVEFMSHkrb2M2NTJnUFIz?= =?utf-8?B?UWFMd01CcXMvY3hhdkNzZmtOZGxvT1cxMzhBNzJ4YzA2NGRyUlhhRzZ5NGlI?= =?utf-8?B?QkZuMHJhaURMTWFmNTNybXNrT2JmNWlwV2wwYThBcXQ4d3JXMkkxaXZmajJI?= =?utf-8?B?MmRqTjRDRGhCNlMzWm9ZUGxQN2d2cFoxbXczdEk0SmI1MHFZeW9nWmcybDRp?= =?utf-8?B?TUV2Yk5DVTdxQ2R4ZkVxRjc5cTQyRmNoajhGNFBjMmQxYjhyaGhaR1NLc0FG?= =?utf-8?B?OXVzVm9RT3paaVdvbFBXV29pbCtWd3g1QzlzNnB3UjFIcTg4bmJKNkFCb2Rh?= =?utf-8?B?WnlkTXF1aG1LRTZQNXlIenIwNFBSNisvNk5PUDR0YktMekxIelk3QzdaU3F4?= =?utf-8?B?d1o3WDZKVE5PbnhiL3JhSnRmRXlheC8reG9vY3BsZzBFWkliWmlaWjB0aDYz?= =?utf-8?B?d2hiUUFQWGN6WHpQWEVnRGtXMU5pNEhpQlVDaWR5L0hGbDF4QlMyQzJ2WHE0?= =?utf-8?B?NjE1Y0RqRkgzTDRubjNGc1dpLzBCYm51NHU4L3Q3Mk5JM3hWT1ZwYWlyeWVW?= =?utf-8?B?enRub2U3dDR0Uk1JbkRsL3ZNa0txelF0K29GYzd6TzJibTBmNm1ldz09?= X-OriginatorOrg: amd.com X-MS-Exchange-CrossTenant-Network-Message-Id: 84ad4965-04ac-48c4-47d8-08df474d36bf X-MS-Exchange-CrossTenant-AuthSource: LV8SPRMB0063.namprd12.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 20 Nov 2026 21:28:27.2675 (UTC) X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted X-MS-Exchange-CrossTenant-Id: 3dd8961f-e488-4e60-8e11-a82d994e183d X-MS-Exchange-CrossTenant-MailboxType: HOSTED X-MS-Exchange-CrossTenant-UserPrincipalName: QOYuWumhtr1IucfFSQaRItIvk4IRLF1Iz943rDhN3qLD8kZFhHaF1eOBwRsDUkVFBEd/9IvMJqdh53J21UbCFQ== X-MS-Exchange-Transport-CrossTenantHeadersStamped: MN0PR12MB5954 X-BeenThere: amd-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Discussion list for AMD gfx List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "amd-gfx" X-Spam-Status: No, score=-17.3 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_CR_TRAILER,INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE, SPF_PASS,USER_AGENT_SANE_1 autolearn=ham autolearn_force=no version=4.0.1 X-Spam-Checker-Version: SpamAssassin 4.0.1 (2024-03-26) on aws-us-west-2-korg-listserv-alma10-1 On 2026-11-18 10:10, William Palacek wrote: > Userptr sg tables are built with sg_alloc_table_from_pages(), which > merges physically contiguous pages into segments of up to > 0xfffff000 bytes. iommu_dma_map_sg() pads a segment so that the next > one does not cross the DMA segment boundary, and adds the padding to > the 32-bit sg->length. With two large contiguous segments the padded > length reaches 4 GiB and wraps to 0. > > On kernels without commit 1c8f6308e557 ("iommu/dma: Catch scatterlist > length overflows") the wrapped length makes iommu_map_sg() map less I recommend you don't mention upstream bugs that are already fixed, since this commit is meant for upstream. Your patch seems to be applicable regardless. > than the allocated IOVA range. The IOVA is then freed without > unmapping it, and a later map at that address fails with -EADDRINUSE. > On kernels with that commit, iommu_dma_map_sg() returns -EOVERFLOW > instead. In both cases a large userptr allocation fails to pin. On > MI300X with the default DMA-FQ IOMMU domain, the first 32 GiB > hipHostMalloc after boot built 16 segments of up to 3.7 GiB, mapped > only 15.7 GiB and failed. > > The padding is shorter than the next segment, so a padded segment > still fits sg->length when no segment exceeds half of it. This statement doesn't make sense. sg->length is per scatterlist entry. Do you mean "no unpadded segment exceeds half of the 4GB maximum size"? The commit headline is equally confusing and should probably also be reworded. > Limit both > userptr sg tables to that, as xe does in commit 1b1d3710380d > ("drm/xe: Apply upper limit to sg element size"). The XE commit also change a call to dma_set_max_seg_size. You're leaving the call in amdgpu_ttm_init unchanged with UINT_MAX. Is that an oversight? Regards,   Felix > > Signed-off-by: William Palacek > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 8 ++++---- > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 7 ++++--- > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 7 +++++++ > 3 files changed, 15 insertions(+), 7 deletions(-) > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c > index f060b206c255..565ecd7e012a 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c > @@ -589,10 +589,10 @@ kfd_mem_dmamap_userptr(struct kgd_mem *mem, > return -ENOMEM; > > /* Same sequence as in amdgpu_ttm_tt_pin_userptr */ > - ret = sg_alloc_table_from_pages(ttm->sg, src_ttm->pages, > - ttm->num_pages, 0, > - (u64)ttm->num_pages << PAGE_SHIFT, > - GFP_KERNEL); > + ret = sg_alloc_table_from_pages_segment(ttm->sg, src_ttm->pages, > + ttm->num_pages, 0, > + (u64)ttm->num_pages << PAGE_SHIFT, > + AMDGPU_SG_MAX_SEGMENT, GFP_KERNEL); > if (unlikely(ret)) > goto free_sg; > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > index 8dfa92967b63..b0672f1747ee 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > @@ -837,9 +837,10 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_device *bdev, > int r; > > /* Allocate an SG array and squash pages into it */ > - r = sg_alloc_table_from_pages(ttm->sg, ttm->pages, ttm->num_pages, 0, > - (u64)ttm->num_pages << PAGE_SHIFT, > - GFP_KERNEL); > + r = sg_alloc_table_from_pages_segment(ttm->sg, ttm->pages, > + ttm->num_pages, 0, > + (u64)ttm->num_pages << PAGE_SHIFT, > + AMDGPU_SG_MAX_SEGMENT, GFP_KERNEL); > if (r) > goto release_sg; > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > index 9856a2fc2d08..da7d7621aad1 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > @@ -44,6 +44,13 @@ > > #define AMDGPU_GTT_MAX_TRANSFER_SIZE (1ULL << 22) > > +/* > + * iommu_dma_map_sg() pads a segment so that the next one does not cross the > + * DMA segment boundary. The padding is shorter than the next segment, so the > + * padded length still fits sg->length if no segment exceeds half of it. > + */ > +#define AMDGPU_SG_MAX_SEGMENT round_down(UINT_MAX / 2, PAGE_SIZE) > + > extern const struct attribute_group amdgpu_vram_mgr_attr_group; > extern const struct attribute_group amdgpu_gtt_mgr_attr_group; >