From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from BYAPR05CU005.outbound.protection.outlook.com (mail-westusazon11010001.outbound.protection.outlook.com [52.101.85.1]) (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 4AD153D994 for ; Wed, 30 Sep 2026 00:20:46 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=fail smtp.client-ip=52.101.85.1 ARC-Seal:i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1790727648; cv=fail; b=lxsEV4r9RNYht62DrSt7BdHsoZXOvtE3AlCZ7PqM/guZjURtCgypwhWPgjkPujQ7V7J819ic3nIgcDgZU6Y0teZLc2Y36PZeu3oZImCW/AMZhJ9d/6qDG+lD4k3zo13uW6yvjIJJYngcV9ZFLyycW5thcHSyp6Put7O/7uzEIVU= ARC-Message-Signature:i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1790727648; c=relaxed/simple; bh=dAGOIc5sOh39Xz9FAmWoDzsYYwTFMQq0UnObnr/eRLU=; h=Message-ID:Date:Subject:To:Cc:References:From:In-Reply-To: Content-Type:MIME-Version; b=N3c0OVSeforjE9y4J6qHp1xadTIjhLEnztPFziCvh0ep4V22BIpbFAdCDUWsqvWXSyjxOT2sbPu3QvTdliTRP8jp89GVTb5r8a6bsToGbL88uDzqANpvv41mT4U5x+fZfqzFMwi62BM8FnIYV5RBX78LStslTdGl195O397idJQ= ARC-Authentication-Results:i=2; smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=amd.com; spf=fail smtp.mailfrom=amd.com; dkim=pass (1024-bit key) header.d=amd.com header.i=@amd.com header.b=xF46vTuS; arc=fail smtp.client-ip=52.101.85.1 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=amd.com Authentication-Results: smtp.subspace.kernel.org; spf=fail smtp.mailfrom=amd.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=amd.com header.i=@amd.com header.b="xF46vTuS" ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=XXpjgQyVqBYPWR9pCfcCFeND+uAMhlnEcQsCSj7gZshElm9PyPbKxyrFOxO56WYdpd9zXmUbTUWZEjQtNXYh1RuWtYEIjMnMkRuSd21lY9OnKY4riBBnbJNlWmekfKnDnSqmlAjL49qkD5JURTYpSsTEB1dTQKOCWf4fKo9y45TvDJe/5cNEfDim3hgOrImlVroMANJyKjwjqrdo+pJUiEHUw4n27gyUMYwMRUaFX+qtgKf5xjetRItjV46FyY7yTnBEt4UjOlej7GxCuBh8NhDQ/YsTyXUxt/aJvSEmi3vZPZB70ZKaFMEd0/Yt2gHvsBu+jQAg6Svc2j+gmqnf7Q== 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=2E6SZgQWf29lNgUdSSlX9bD58CnfIvNlhB8EpnifMyo=; b=lrsMtu6lRDsG2MfWYjZiieVPCoit9nWWE6EMcUC+aXSU2mgP519T6Dop2LxK2mBFmdibYIiOlaK1/dReV6XSpb/lw+Iars/w9xaLFODcLJ86DxUMbsmqn3tjfmM285YVlahtzRyidZCnVEQZrykdZuTbo1Vb5gC6JeKZlAPyuxmwsK8OoAmxsI2bRtuaheJKkh+gqTLqWnkfJqXjzeAr+vNFbtEFbGG6moykSGP89sQUi4cd4OhKMEYDSAue46o6wHzpk1egvAaQYXLgRWDVSk9+dwGX6CKoYNLBqs+oiMbIAelWNxPh3V6jf4LDuRcI8jdSC0Ef/1Jph2rB31QF0Q== 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=2E6SZgQWf29lNgUdSSlX9bD58CnfIvNlhB8EpnifMyo=; b=xF46vTuSZafv6bPsvtbVwK+WEsb7QN9jX9iuf6Ex9uruVOH/VN2CCthmoadJ0TRJQtOoGgo5EYIJYa1sxNtx9m4r0zFlRVMaEooa66k6+w2chLObtkQOl9Lzo8oWFBzywuZ7CL2w6JLFrRPu4JFz7/W9MxizaWJXnf1qRsU0i/I= Authentication-Results: mx.microsoft.com 1; dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=amd.com; Received: from IA1PR12MB9062.namprd12.prod.outlook.com (2603:10b6:208:3aa::14) by SA1PR12MB9247.namprd12.prod.outlook.com (2603:10b6:806:3af::18) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.451.25; Wed, 30 Sep 2026 00:20:42 +0000 Received: from IA1PR12MB9062.namprd12.prod.outlook.com ([fe80::28c6:3cbd:6e97:5257]) by IA1PR12MB9062.namprd12.prod.outlook.com ([fe80::28c6:3cbd:6e97:5257%6]) with mapi id 15.21.0451.022; Wed, 30 Sep 2026 00:20:40 +0000 Message-ID: <906d1899-26af-4829-8a4b-9d451ff13917@amd.com> Date: Tue, 29 Sep 2026 17:20:37 -0700 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH V0 14/21] accel/amdxdna: Fix fence timeline name and context allocation To: Lizhi Hou , quic_jhugo@quicinc.com, karol.wachowski@linux.intel.com, max.zhen@amd.com, ogabbay@kernel.org, dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org Cc: sonal.santan@amd.com, mario.limonciello@amd.com References: <20260926013448.3840921-1-yidong.zhang@amd.com> <20260926013448.3840921-15-yidong.zhang@amd.com> Content-Language: en-US From: "Zhang, Yidong (David)" In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-ClientProxiedBy: SJ0PR03CA0049.namprd03.prod.outlook.com (2603:10b6:a03:33e::24) To IA1PR12MB9062.namprd12.prod.outlook.com (2603:10b6:208:3aa::14) 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: IA1PR12MB9062:EE_|SA1PR12MB9247:EE_ X-MS-Office365-Filtering-Correlation-Id: 86ac9979-f982-4c0c-7e4b-08df1e88a873 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230040|1800799024|23010399003|366016|376014|10067099003|11063799006|56012099006|3023799007|4143699003|22082099003|18002099003; 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:IA1PR12MB9062.namprd12.prod.outlook.com;PTR:;CAT:NONE;SFS:(13230040)(1800799024)(23010399003)(366016)(376014)(10067099003)(11063799006)(56012099006)(3023799007)(4143699003)(22082099003)(18002099003);DIR:OUT;SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?UzkwcCtjY2NiQlp1d0huRkpoQVZaQ0kwT3h4RGc3cExnMXQ0bDIySGlWdytl?= =?utf-8?B?allpelFZMHV1aGVJRHZVcEt5VnpISitFb2k1R0dIaVJkWkJDcWFxNktHblQ5?= =?utf-8?B?ZVducENrLzZ4aUtmVFZCa1c4Q1E0VW9xdTBQU0tiS3oxaDF6ZlFCc2pIOWhT?= =?utf-8?B?OGVja1dYbU15MW13azFINWhZWTB3OEExdlpvbmwrcFYrc3RrcjBOSmNMV3ZD?= =?utf-8?B?NG5mbi9zcndqUjhxZyt0UnlJUVBVc2ptSG9RWEhVNUI1dFdNWjhuck9YYWZs?= =?utf-8?B?RjIrUWE4OVc3ZHhScHN3QnBBeHpJZi9HQk1HSklNNkxvbWJMdHFBSGlzL2hP?= =?utf-8?B?WlhWalVFSHg0RGx0R3dYTmo5ZElLMnRUOVEyZFBVNjJvdEZwQitmcDcyNTJH?= =?utf-8?B?L2lWVjNxdEFHY3RWYnRReElKa05JSzhBSndSMkZYTGo3a1pzU2Z6eWZ5Vkpm?= =?utf-8?B?RVVuSWY4VXkrUVFKbkdNdy9PL0xHNUtlVDN2bDAxcHkwRzBKWlE1WlB2dEt6?= =?utf-8?B?azYrdWZSdGJlSk5ydFVIR2ptVjc0SmYwZk55WlQxNThDTXc2VUFJMWRmc2F3?= =?utf-8?B?RlN4bGhtN3NmNllYWG9mbGNONGJaRWU0Rmo5QWFNeU1LYUxBOVNlNWpFNFNw?= =?utf-8?B?R2xHS1kwWlh1dHdWYlpWSThLTUtvQ1V5WVg0V3BHVGsrdTZoVm8zSXZTazBE?= =?utf-8?B?M2hGYmR2VGtNMm0yN3FjNHVHbkZjM1RsTHR6aEZzWkNTYnpFekpjK1dlUDJV?= =?utf-8?B?VXBCTzJwcnRGZldjQ3pwQzFHMzU2YmpnTmxiTVBmWEJ6UjhWNC9CaWQ1ZmZq?= =?utf-8?B?eDNESGJnNjNQUmNPaE4wZVhkWkh6VXByMkFVVmRCVGVreFMvQ1QwcFFZdDEv?= =?utf-8?B?dTRzK21nc25SQmxWZmowZHE0dHZDVVRPSG5mWlZhMmVlNU4xSkVhdHljRkx2?= =?utf-8?B?UHVrcnZVNTYzbldGdW1CeGhibW9YVWI2MjJ2TDNPY0dtOVFhaS81TWtDOER3?= =?utf-8?B?ZkluSFduWEZ0NGp4YlNtMVNiWWRDdVY1Uk1VbHRPckJ6YXozY2EySTRGbnNO?= =?utf-8?B?dUZQaHUxd1J1UXJxc2psUlBqMVQ1Nk1hUUZjV3FTc3FTSkwvbzNHWDJXdlZJ?= =?utf-8?B?aURsSXJ6Ui96c0MySnlqWDdYQnU3L2pUam5aRy9KTzhGMHR6YlRVbXdXOE9p?= =?utf-8?B?aGl2NWkxamNZa09ZYlRWY0gxU08xMy9JVWFqUnkvdzQ1cjVhUEZlOVBvOVBv?= =?utf-8?B?MDF1SUhiZ0xYWmsxMTNvWVkzRGxKd1htSjNFM1ZKR2ZYZVV5ZUxjT3ZxOWRX?= =?utf-8?B?VVloY3QxN1dIdWozNjlXOUxXV2RiQzBuVUdpNkN4a3hzblZYT05PSWdTWDBi?= =?utf-8?B?Z3dQMjR4Z0Vac3FSTEhQQlRacnZkeVpma05JcnF6Y1J6bE8zV0Vua0VqdEd5?= =?utf-8?B?S1B0L3BsRjhsU3ptWDFzakZGVVNNdndJUThIby83c25xRHg2WmFXZ2RyRHAx?= =?utf-8?B?SHU2OXVBUENXZnplY2hXTUs4eFBiU1B1dlhoU09GcGZNc045dGZPRFNqZFBT?= =?utf-8?B?Wm94R1FGMncxd2o2WjlkcFEwZWZDSTdzZ0R5VzYvblJDV0xQK3ZEYVJtMU1B?= =?utf-8?B?NWlhUWNLZmdWc2FZRUxudmFHUW50WjF3RXUrMGVCcmxoOGc2Q0trbDN4emdJ?= =?utf-8?B?NFRMM1ZpOUFhSGtQTUNac1ZsZ1E5Vk9PSzlzR1ZEVDRWd1NDSmxjNXZsZTAx?= =?utf-8?B?cTl3cDRpbUJYYWl5ZS9DQkhIYUlxRVlYa1U0OTBLTzk2UVo2TVZEdEJURnNF?= =?utf-8?B?dDBNb3hBWlArOHlFbWNRdDFVZitrOUkxNEEyWEl3U3B3MGQrU3hQR0UrYnZs?= =?utf-8?B?WTUyTnoyZkl6MmJOV2lsVzZJTjgreWYrQmUyUjMraGpkZHhQcmNrNklFb0M0?= =?utf-8?B?WnkvZGUxUDVsT2NFZUJ4YzhlNTlDSFZtclUvOVdGYXJrMDlxUlFMZG1mUHFZ?= =?utf-8?B?QVZ3TkVIVFp4TmVoS0RwTFVTdWZFbVRqRTg2eDBLS0c3TkNXWWQxd3YwemJZ?= =?utf-8?B?MnFrMmI1bHU0aHhDQ0dMUnFZZ0JPYTkwNXo2Z3pOdUxhK09MVGpOc3BqUUlC?= =?utf-8?B?RnpEZ2pyV0JnQzcvS1ZDWUt6WnowTDhmMll2OE9ONjdQdG1GN1VTcDQ2dW8w?= =?utf-8?B?OTExK1Z4MDBYVnZJRFE2TVVjdWREdWU0d3liT0wxVHp5ekdidW93Z1h0VVhv?= =?utf-8?B?all4RldDVW1ISlVFbjBxczBzZy9yM3B0L2s0ZCtUeFM3SFpHY21md3RDOG11?= =?utf-8?B?OW1KcTBoeGN4eTVYVWFUOWNGTHB6a20vUVdNWjlDek9JRmVpSHJYdz09?= X-OriginatorOrg: amd.com X-MS-Exchange-CrossTenant-Network-Message-Id: 86ac9979-f982-4c0c-7e4b-08df1e88a873 X-MS-Exchange-CrossTenant-AuthSource: IA1PR12MB9062.namprd12.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 30 Sep 2026 00:20:40.4590 (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: iXFP3z+wzYxgrGCuqc1fFeTK0gYE2ZMoTS3f1XaE4xv+ZUF9+S15/JmMFlcngKmrMqIy/Vp/ZCzt5RD++BZT4A== X-MS-Exchange-Transport-CrossTenantHeadersStamped: SA1PR12MB9247 On 9/28/2026 2:06 PM, Lizhi Hou wrote: > > On 9/25/26 18:34, David Zhang wrote: >> This is part of the fix to align BO reservation locking and fence >> management with aie2. >> >> Fences published into BO reservation objects via dma_resv_add_fence() >> can outlive the hardware context (e.g. when a BO is exported as a >> dma-buf and imported by another process). Using hwctx->name for the >> fence >> timeline name risks a use-after-free once the hwctx is destroyed. >> Switch timeline name to dev_name() which is backed by the device that >> outlives any individual context. >> >> Additionally, allocate a unique fence context via >> dma_fence_context_alloc(1) for each job fence so that >> dma_resv_add_fence() >> does not evict a prior in-flight job's fence from a shared BO's >> reservation object when multiple jobs touch the same BO. Also guard >> hwctx_fini call in amdxdna_hwctx_destroy_rcu() against NULL ops. > > Is this aie4 kernel submission specific? aie2 does not publish this > fence. > > If it does not fix any existing issue, please describe it clearly. I have addressed your comments with a newer PATCH V1. This change is merged with KMQ patches. It is AIE4 specific. > >> >> The corresponding AIE4 command submission BO locking and fence >> attachment logic is implemented in a subsequent patch ("accel/amdxdna: >> Implement AIE4 command packet building and submission"). >> >> Co-developed-by: Max Zhen >> Signed-off-by: Max Zhen >> Signed-off-by: David Zhang >> --- >>   drivers/accel/amdxdna/amdxdna_ctx.c | 29 ++++++++++++++++++++++------- >>   1 file changed, 22 insertions(+), 7 deletions(-) >> >> diff --git a/drivers/accel/amdxdna/amdxdna_ctx.c >> b/drivers/accel/amdxdna/amdxdna_ctx.c >> index 888e857ec558..6ca7774150d5 100644 >> --- a/drivers/accel/amdxdna/amdxdna_ctx.c >> +++ b/drivers/accel/amdxdna/amdxdna_ctx.c >> @@ -25,7 +25,7 @@ >>   struct amdxdna_fence { >>       struct dma_fence    base; >>       spinlock_t        lock; /* for base */ >> -    struct amdxdna_hwctx    *hwctx; >> +    struct device        *dev; >>   }; >>     static const char *amdxdna_fence_get_driver_name(struct dma_fence >> *fence) >> @@ -39,7 +39,14 @@ static const char >> *amdxdna_fence_get_timeline_name(struct dma_fence *fence) >>         xdna_fence = container_of(fence, struct amdxdna_fence, base); >>   -    return xdna_fence->hwctx->name; >> +    /* >> +     * Use device name rather than hwctx name: the fence is >> published into >> +     * BO reservation objects via dma_resv_add_fence() and can >> outlive the >> +     * hwctx (e.g. when a BO is exported as a dma-buf and imported by >> +     * another process). The device outlives any individual context, so >> +     * dev_name() is safe to call at any point during the fence's >> lifetime. >> +     */ >> +    return dev_name(xdna_fence->dev); >>   } >>     static const struct dma_fence_ops fence_ops = { >> @@ -55,9 +62,17 @@ static struct dma_fence >> *amdxdna_fence_create(struct amdxdna_hwctx *hwctx) >>       if (!fence) >>           return NULL; >>   -    fence->hwctx = hwctx; >> +    fence->dev = hwctx->client->xdna->ddev.dev; >>       spin_lock_init(&fence->lock); >> -    dma_fence_init(&fence->base, &fence_ops, &fence->lock, >> hwctx->id, 0); >> +    /* >> +     * Part of the fix to align BO reservation locking and fence >> +     * management with AIE2: each job fence needs a unique context so >> +     * dma_resv_add_fence() does not evict a prior job's fence from a >> +     * shared BO's reservation object when two in-flight jobs touch >> +     * the same BO. The corresponding AIE4 command submission locking >> +     * and fence attachment is implemented in aie4_cmd_submit(). >> +     */ >> +    dma_fence_init(&fence->base, &fence_ops, &fence->lock, >> dma_fence_context_alloc(1), 0); >>       return &fence->base; >>   } >>   @@ -81,13 +96,13 @@ static void >> amdxdna_hwctx_release_expanded_heap(struct amdxdna_hwctx *hwctx) >>   static void amdxdna_hwctx_destroy_rcu(struct amdxdna_hwctx *hwctx, >>                         struct srcu_struct *ss) >>   { >> -    struct amdxdna_client *client = hwctx->client; >> -    struct amdxdna_dev *xdna = client->xdna; >> +    struct amdxdna_dev *xdna = hwctx->client->xdna; >>         synchronize_srcu(ss); >>         /* At this point, user is not able to submit new commands */ >> -    xdna->dev_info->ops->hwctx_fini(hwctx); >> +    if (xdna->dev_info->ops->hwctx_fini) >> +        xdna->dev_info->ops->hwctx_fini(hwctx); > > This seems unrelated. Please remove from the patch. Removed. David > > Lizhi > >> amdxdna_hwctx_release_expanded_heap(hwctx); >>       kfree(hwctx->name);