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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 1333CC43458 for ; Tue, 30 Jun 2026 02:54:26 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 875E310E09E; Tue, 30 Jun 2026 02:54:26 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (1024-bit key; unprotected) header.d=amd.com header.i=@amd.com header.b="RWuzgRwi"; dkim-atps=neutral Received: from SN4PR2101CU001.outbound.protection.outlook.com (mail-southcentralusazon11012057.outbound.protection.outlook.com [40.93.195.57]) by gabe.freedesktop.org (Postfix) with ESMTPS id 3E0A410E09E for ; Tue, 30 Jun 2026 02:54:24 +0000 (UTC) ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=di5v+gC8Sf73GGssPPiB0BylIyQzKvNOPCNLkxKlolcmQ4D1ffxwEmllv5K9ORdCVaF7vidaYg8VPjAmj0CXCPhdGttuUxMRDQTXgtSNs/lBvEwIfB78mLsuPNLaZpB07e1YfvzdQQ72+lXue7pl2/lBEfi78Ferm/GMu2mbcPmVIrnM68KVhak8HrJ9tMJu3rEP5KM7s2MtBATRvWR07+B+G8WG2HkaFpdHlC0KysP881aPOZy/LsPhpjkV2QWT1QbzJr4UifrEAnSknQ1O24bs5af09iTGpTbcfEv2wXjX9iiozNt77XkvhSYI6DY2hhm6Cl4+e0C+5aBLrw5Ntg== 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=LpHWBqkgD3P8GwTNA/YxCNZ6SwTaSBnzYmSJMchSIfk=; b=I1iB44NsL5poZv/VDB5tw3J1h3RlpzIbqvL6piBISCYE8JzAJs7jFjY+B41p2Fbik962dwNZ99dITO7IWoqXkdYAfUxQbJycA/8j/XObXymiqdN+6J7WTA2GWHyG0PwHF7YZ5jPCxn9jjm06gxHvqpzBjzQeWE1NA61XEze2tHeRDqYm76sOB5gpaKD1/YUt/WkMkUCiYBsZnnPUTmKUJ5aFbvSoXNkBULCKZR0Hu745fl6PfLnetwCeYnn2cxE6P4f5s7NRmlwsdPqln6ZVYJm2kW3DpAKM1J6X/ErCUXl6lAyukz3GW8QOefDayfJFQG4oIs9aA6Mv7Zg/wo/0rA== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass (sender ip is 165.204.84.17) smtp.rcpttodomain=lists.freedesktop.org smtp.mailfrom=amd.com; dmarc=pass (p=quarantine sp=quarantine pct=100) action=none header.from=amd.com; dkim=none (message not signed); arc=none (0) 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=LpHWBqkgD3P8GwTNA/YxCNZ6SwTaSBnzYmSJMchSIfk=; b=RWuzgRwiyqTV6vCUa7eU+8ojcieGOcGy6mV+j5d/EjpIjTrdyxa4BG28ohrytKDpsTTlMl/4uiyyFNQJBWTNREn5h4jsR+PUV77ol1k4SztvvboVfwm289tynECqiD3Ij0Fdp1pdvN6e+ev/JC92+u3tzRV/5UAcx3VNsQG4xdI= Received: from CH0PR03CA0342.namprd03.prod.outlook.com (2603:10b6:610:11a::16) by SA1PR12MB7176.namprd12.prod.outlook.com (2603:10b6:806:2bd::19) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.159.19; Tue, 30 Jun 2026 02:54:17 +0000 Received: from CH2PEPF00000142.namprd02.prod.outlook.com (2603:10b6:610:11a:cafe::93) by CH0PR03CA0342.outlook.office365.com (2603:10b6:610:11a::16) with Microsoft SMTP Server (version=TLS1_3, cipher=TLS_AES_256_GCM_SHA384) id 15.21.159.19 via Frontend Transport; Tue, 30 Jun 2026 02:54:16 +0000 X-MS-Exchange-Authentication-Results: spf=pass (sender IP is 165.204.84.17) smtp.mailfrom=amd.com; dkim=none (message not signed) header.d=none;dmarc=pass action=none header.from=amd.com; Received-SPF: Pass (protection.outlook.com: domain of amd.com designates 165.204.84.17 as permitted sender) receiver=protection.outlook.com; client-ip=165.204.84.17; helo=satlexmb07.amd.com; pr=C Received: from satlexmb07.amd.com (165.204.84.17) by CH2PEPF00000142.mail.protection.outlook.com (10.167.244.75) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.21.181.6 via Frontend Transport; Tue, 30 Jun 2026 02:54:16 +0000 Received: from chonglidebug.amd.com (10.180.168.240) by satlexmb07.amd.com (10.181.42.216) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.2562.41; Mon, 29 Jun 2026 21:54:14 -0500 From: chong li To: CC: , , chong li Subject: [PATCH] [PATCH] drm/amdgpu: improve the amdgpu device init progress in sriov mode Date: Tue, 30 Jun 2026 10:53:20 +0800 Message-ID: <20260630025341.2101161-1-chongli2@amd.com> X-Mailer: git-send-email 2.48.1 MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8bit X-Originating-IP: [10.180.168.240] X-ClientProxiedBy: satlexmb07.amd.com (10.181.42.216) To satlexmb07.amd.com (10.181.42.216) X-EOPAttributedMessage: 0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: CH2PEPF00000142:EE_|SA1PR12MB7176:EE_ X-MS-Office365-Filtering-Correlation-Id: 3e253ecd-df21-41b1-40b4-08ded652e011 X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0; ARA:13230040|1800799024|23010399003|82310400026|376014|36860700016|18002099003|6133799003|11063799006|56012099006; X-Microsoft-Antispam-Message-Info: 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 X-Forefront-Antispam-Report: CIP:165.204.84.17; CTRY:US; LANG:en; SCL:1; SRV:; IPV:NLI; SFV:NSPM; H:satlexmb07.amd.com; PTR:InfoDomainNonexistent; CAT:NONE; SFS:(13230040)(1800799024)(23010399003)(82310400026)(376014)(36860700016)(18002099003)(6133799003)(11063799006)(56012099006); DIR:OUT; SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: 7LgnoCDEmAlFtcSrOukwiW+h45dMgFSHHyHZ+5IwjAyotz5p7FoN4xkGpvguxgq0p+pYSYlr/Kpidg2skF9K1A9+phpwH3MKMfKc1l6zK1GvsX4yxC82w3l6Bfu3Djzvvn+sUTLLneW7RUvsRx8so0582dzNSrVp4EvvF2ntJZoDJxRMUoSlacBIJqFaHd2eUDuWPu9i9g2TX4YN6vE7UvzISCGyufF261cPoFPye565/LArLc2pNZrMEraGxVUwf7kXmflWmMQeIHXIRPxTzjtDhA7jFw0cKpYWT2+UEEbHbwa1W/0T67jWP2hREQAk3tmRta556cO5thDs2DiJi/r3WTP6sVUFfRKOtM95jU4Ftken0OimLyuBazuYTPHHu3GXKgBYrricyxKOmqXAowWqYdfykh+88YIxk5fQcPufct3zINYALDVJeARTN56R X-OriginatorOrg: amd.com X-MS-Exchange-CrossTenant-OriginalArrivalTime: 30 Jun 2026 02:54:16.7544 (UTC) X-MS-Exchange-CrossTenant-Network-Message-Id: 3e253ecd-df21-41b1-40b4-08ded652e011 X-MS-Exchange-CrossTenant-Id: 3dd8961f-e488-4e60-8e11-a82d994e183d X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=3dd8961f-e488-4e60-8e11-a82d994e183d; Ip=[165.204.84.17]; Helo=[satlexmb07.amd.com] X-MS-Exchange-CrossTenant-AuthSource: CH2PEPF00000142.namprd02.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Anonymous X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem X-MS-Exchange-Transport-CrossTenantHeadersStamped: SA1PR12MB7176 X-BeenThere: amd-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Discussion list for AMD gfx List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: amd-gfx-bounces@lists.freedesktop.org Sender: "amd-gfx" Move the initialization of non-GPU resources out of the full GPU access region during AMDGPU device initialization Signed-off-by: chong li --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 10 +- drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 131 ++++++++++------- drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 138 +++++++++++++----- drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h | 2 + 4 files changed, 181 insertions(+), 100 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 5442a1fc1c37..247779cbf394 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -1953,10 +1953,6 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) amdgpu_device_enable_virtual_display(adev); if (amdgpu_sriov_vf(adev)) { - r = amdgpu_virt_request_full_gpu(adev, true); - if (r) - return r; - r = amdgpu_virt_init_critical_region(adev); if (r) return r; @@ -2118,6 +2114,12 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) if (!total) return -ENODEV; + if (amdgpu_sriov_vf(adev)) { + r = amdgpu_virt_request_full_gpu(adev, true); + if (r) + return r; + } + if (adev->gmc.xgmi.supported) amdgpu_xgmi_early_init(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index 029931f4e6ed..26916ebca452 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -26,6 +26,7 @@ #include "amdgpu.h" #include "amdgpu_discovery.h" +#include "amdgpu_virt.h" #include "soc15_hw_ip.h" #include "discovery.h" #include "amdgpu_ras.h" @@ -292,21 +293,54 @@ static int amdgpu_discovery_get_tmr_info(struct amdgpu_device *adev, u32 msg, tmr_offset_lo, tmr_offset_hi; int i, ret; - if (!amdgpu_sriov_vf(adev)) { - /* It can take up to two second for IFWI init to complete on some dGPUs, - * but generally it should be in the 60-100ms range. Normally this starts - * as soon as the device gets power so by the time the OS loads this has long - * completed. However, when a card is hotplugged via e.g., USB4, we need to - * wait for this to complete. Once the C2PMSG is updated, we can - * continue. - */ + if (amdgpu_sriov_vf(adev)) { + u32 ipd_size_kb = + adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_IPD_TABLE_ID].size_kb; + + *is_tmr_in_sysmem = false; + + if (ipd_size_kb) { + adev->discovery.offset = + adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_IPD_TABLE_ID].offset; + adev->discovery.size = ipd_size_kb << 10; + if (!adev->discovery.size) + return -EINVAL; + } else if (adev->virt.req_init_data_ver == GPU_CRIT_REGION_V1) { + vram_size = RREG32(mmRCC_CONFIG_MEMSIZE); + if (vram_size == U32_MAX) + return -ENXIO; + else if (!vram_size) + return -ENODEV; - for (i = 0; i < 2000; i++) { - msg = RREG32(mmMP0_SMN_C2PMSG_33); - if (msg & 0x80000000) - break; - msleep(1); + adev->discovery.size = DISCOVERY_TMR_SIZE; + adev->discovery.offset = (vram_size << 20) - DISCOVERY_TMR_OFFSET; + } else if (amdgpu_sriov_xgmi_connected_to_cpu(adev)) { + ret = amdgpu_acpi_get_tmr_info(adev, &tmr_offset, &tmr_size); + if (ret) + return ret; + + *is_tmr_in_sysmem = true; + adev->discovery.size = DISCOVERY_TMR_SIZE; + adev->discovery.offset = tmr_offset + tmr_size - DISCOVERY_TMR_OFFSET; + } else { + return -EINVAL; } + goto out; + } + + /* It can take up to two second for IFWI init to complete on some dGPUs, + * but generally it should be in the 60-100ms range. Normally this starts + * as soon as the device gets power so by the time the OS loads this has long + * completed. However, when a card is hotplugged via e.g., USB4, we need to + * wait for this to complete. Once the C2PMSG is updated, we can + * continue. + */ + + for (i = 0; i < 2000; i++) { + msg = RREG32(mmMP0_SMN_C2PMSG_33); + if (msg & 0x80000000) + break; + msleep(1); } vram_size = RREG32(mmRCC_CONFIG_MEMSIZE); @@ -322,35 +356,22 @@ static int amdgpu_discovery_get_tmr_info(struct amdgpu_device *adev, if (vram_size) adev->discovery.offset = (vram_size << 20) - DISCOVERY_TMR_OFFSET; - if (amdgpu_sriov_vf(adev)) { - if (adev->virt.is_dynamic_crit_regn_enabled) { - adev->discovery.offset = - adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_IPD_TABLE_ID].offset; - adev->discovery.size = - adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_IPD_TABLE_ID].size_kb << 10; - if (!adev->discovery.size) - return -EINVAL; - } else { - goto out; - } - } else { - tmr_size = RREG32(mmDRIVER_SCRATCH_2); - if (tmr_size) { - /* It's preferred to transition to PSP mailbox reg interface - * for both bare-metal and passthrough if available */ - adev->discovery.size = (u32)tmr_size; - tmr_offset_lo = RREG32(mmDRIVER_SCRATCH_0); - tmr_offset_hi = RREG32(mmDRIVER_SCRATCH_1); - adev->discovery.offset = ((u64)le32_to_cpu(tmr_offset_hi) << 32 | - le32_to_cpu(tmr_offset_lo)); - } else if (!vram_size) { - /* fall back to apci approach to query tmr offset if vram_size is 0 */ - ret = amdgpu_acpi_get_tmr_info(adev, &tmr_offset, &tmr_size); - if (ret) - return ret; - adev->discovery.size = DISCOVERY_TMR_SIZE; - adev->discovery.offset = tmr_offset + tmr_size - DISCOVERY_TMR_OFFSET; - } + tmr_size = RREG32(mmDRIVER_SCRATCH_2); + if (tmr_size) { + /* It's preferred to transition to PSP mailbox reg interface + * for both bare-metal and passthrough if available */ + adev->discovery.size = (u32)tmr_size; + tmr_offset_lo = RREG32(mmDRIVER_SCRATCH_0); + tmr_offset_hi = RREG32(mmDRIVER_SCRATCH_1); + adev->discovery.offset = ((u64)le32_to_cpu(tmr_offset_hi) << 32 | + le32_to_cpu(tmr_offset_lo)); + } else if (!vram_size) { + /* fall back to apci approach to query tmr offset if vram_size is 0 */ + ret = amdgpu_acpi_get_tmr_info(adev, &tmr_offset, &tmr_size); + if (ret) + return ret; + adev->discovery.size = DISCOVERY_TMR_SIZE; + adev->discovery.offset = tmr_offset + tmr_size - DISCOVERY_TMR_OFFSET; } out: adev->discovery.bin = kzalloc(adev->discovery.size, GFP_KERNEL); @@ -384,23 +405,21 @@ static int amdgpu_discovery_read_binary_from_mem(struct amdgpu_device *adev, uint8_t *binary, bool is_tmr_in_sysmem) { - int ret = 0; + if (is_tmr_in_sysmem) + return amdgpu_discovery_read_binary_from_sysmem(adev, binary); - if (!is_tmr_in_sysmem) { - if (amdgpu_sriov_vf(adev) && - amdgpu_sriov_xgmi_connected_to_cpu(adev)) { - ret = amdgpu_discovery_read_binary_from_sysmem(adev, binary); - } else { - amdgpu_device_vram_access(adev, adev->discovery.offset, - (uint32_t *)binary, - adev->discovery.size, false); - adev->discovery.reserve_tmr = true; - } - } else { - ret = amdgpu_discovery_read_binary_from_sysmem(adev, binary); + if (amdgpu_sriov_vf(adev)) { + adev->discovery.reserve_tmr = true; + return amdgpu_virt_read_vf_fb(adev, adev->discovery.offset, + binary, adev->discovery.size); } - return ret; + amdgpu_device_vram_access(adev, adev->discovery.offset, + (uint32_t *)binary, + adev->discovery.size, false); + adev->discovery.reserve_tmr = true; + + return 0; } static int amdgpu_discovery_read_binary_from_file(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index 7b0d42510161..9ade91ff9394 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -754,8 +754,7 @@ static int amdgpu_virt_read_exchange_data_from_mem(struct amdgpu_device *adev, u adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_DATAEXCHANGE_TABLE_ID].offset; uint32_t dataexchange_size = adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_DATAEXCHANGE_TABLE_ID].size_kb << 10; - uint64_t pos = 0; - + int ret; dev_info(adev->dev, "Got data exchange info from dynamic crit_region_table at offset 0x%x with size of 0x%x bytes.\n", dataexchange_offset, dataexchange_size); @@ -764,12 +763,10 @@ static int amdgpu_virt_read_exchange_data_from_mem(struct amdgpu_device *adev, u dev_err(adev->dev, "Data exchange data not aligned to 4 bytes\n"); return -EINVAL; } + ret = amdgpu_virt_read_vf_fb(adev, dataexchange_offset, pfvf_data, + dataexchange_size); - pos = (uint64_t)dataexchange_offset; - amdgpu_device_vram_access(adev, pos, pfvf_data, - dataexchange_size, false); - - return 0; + return ret; } void amdgpu_virt_fini_data_exchange(struct amdgpu_device *adev) @@ -802,7 +799,7 @@ void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev) schedule_delayed_work(&(adev->virt.vf2pf_work), msecs_to_jiffies(adev->virt.vf2pf_update_interval_ms)); } else if (adev->bios != NULL) { /* got through this logic in early init stage to get necessary flags, e.g. rlcg_acc related*/ - if (adev->virt.req_init_data_ver == GPU_CRIT_REGION_V2) { + if (adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_DATAEXCHANGE_TABLE_ID].size_kb) { pfvf_data = kzalloc(adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_DATAEXCHANGE_TABLE_ID].size_kb << 10, GFP_KERNEL); @@ -1014,38 +1011,89 @@ static uint8_t amdgpu_virt_crit_region_calc_checksum(uint8_t *buf_start, uint8_t return 0xffffffff - sum; } -int amdgpu_virt_init_critical_region(struct amdgpu_device *adev) +int amdgpu_virt_read_vf_fb(struct amdgpu_device *adev, u64 offset, + void *buf, size_t size) { - struct amd_sriov_msg_init_data_header *init_data_hdr = NULL; - u64 init_hdr_offset = adev->virt.init_data_header.offset; - u64 init_hdr_size = (u64)adev->virt.init_data_header.size_kb << 10; /* KB → bytes */ - u64 vram_size; + resource_size_t bar_start, bar_size, map_base; + void __iomem *vram; + size_t map_offset, map_size; u64 end; - int r = 0; - uint8_t checksum = 0; - /* Skip below init if critical region version != v2 */ - if (adev->virt.req_init_data_ver != GPU_CRIT_REGION_V2) - return 0; - - vram_size = RREG32(mmRCC_CONFIG_MEMSIZE); - if (!vram_size || vram_size == U32_MAX) + if (!buf || !size) return -EINVAL; - vram_size <<= 20; - if (check_add_overflow(init_hdr_offset, init_hdr_size, &end) || end > vram_size) { - dev_err(adev->dev, "init_data_header exceeds VRAM size, exiting\n"); + bar_size = pci_resource_len(adev->pdev, 0); + if (!bar_size) + return -ENODEV; + + if (check_add_overflow(offset, size, &end) || end > bar_size) { + dev_err(adev->dev, + "VF FB read exceeds BAR0 size: offset=0x%llx size=0x%zx BAR0=0x%llx\n", + offset, size, (u64)bar_size); return -EINVAL; } + bar_start = pci_resource_start(adev->pdev, 0); + map_offset = offset_in_page(offset); + map_base = bar_start + (offset & PAGE_MASK); + map_size = PAGE_ALIGN(map_offset + size); + + vram = ioremap_wc(map_base, map_size); + if (!vram) + return -ENOMEM; + + memcpy_fromio(buf, (u8 __iomem *)vram + map_offset, size); + iounmap(vram); + + return 0; +} + +static int amdgpu_virt_init_critical_region_v1(struct amdgpu_device *adev) +{ + memset(&adev->virt.crit_regn, 0, sizeof(adev->virt.crit_regn)); + memset(adev->virt.crit_regn_tbl, 0, sizeof(adev->virt.crit_regn_tbl)); + + adev->virt.crit_regn.offset = 0; + adev->virt.crit_regn.size_kb = AMD_SRIOV_MSG_INIT_DATA_TOT_SIZE_KB_V1; + + adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_VBIOS_IMG_TABLE_ID].offset = + AMD_SRIOV_MSG_VBIOS_OFFSET_V1; + adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_VBIOS_IMG_TABLE_ID].size_kb = + AMD_SRIOV_MSG_VBIOS_SIZE_KB_V1; + adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_RAS_TELEMETRY_TABLE_ID].offset = + AMD_SRIOV_MSG_RAS_TELEMETRY_OFFSET_KB_V1 << 10; + adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_RAS_TELEMETRY_TABLE_ID].size_kb = + AMD_SRIOV_MSG_RAS_TELEMETRY_SIZE_KB_V1; + adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_DATAEXCHANGE_TABLE_ID].offset = + AMD_SRIOV_MSG_DATAEXCHANGE_OFFSET_KB_V1 << 10; + adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_DATAEXCHANGE_TABLE_ID].size_kb = + AMD_SRIOV_MSG_DATAEXCHANGE_SIZE_KB_V1; + adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_BAD_PAGE_INFO_TABLE_ID].offset = + AMD_SRIOV_MSG_BAD_PAGE_OFFSET_KB_V1 << 10; + adev->virt.crit_regn_tbl[AMD_SRIOV_MSG_BAD_PAGE_INFO_TABLE_ID].size_kb = + AMD_SRIOV_MSG_BAD_PAGE_SIZE_KB_V1; + + adev->virt.is_dynamic_crit_regn_enabled = true; + + return 0; +} + +static int amdgpu_virt_init_critical_region_v2(struct amdgpu_device *adev) +{ + struct amd_sriov_msg_init_data_header *init_data_hdr = NULL; + u64 init_hdr_offset = adev->virt.init_data_header.offset; + int r = 0; + uint8_t checksum = 0; + /* Allocate for init_data_hdr */ init_data_hdr = kzalloc(sizeof(struct amd_sriov_msg_init_data_header), GFP_KERNEL); if (!init_data_hdr) return -ENOMEM; - amdgpu_device_vram_access(adev, (uint64_t)init_hdr_offset, (uint32_t *)init_data_hdr, - sizeof(struct amd_sriov_msg_init_data_header), false); - + r = amdgpu_virt_read_vf_fb(adev, init_hdr_offset, init_data_hdr, + sizeof(struct amd_sriov_msg_init_data_header)); + if (r) + goto out; /* Table validation */ if (strncmp(init_data_hdr->signature, AMDGPU_SRIOV_CRIT_DATA_SIGNATURE, @@ -1158,16 +1206,6 @@ int amdgpu_virt_init_critical_region(struct amdgpu_device *adev) goto out; } - /* reserved memory starts from crit region base offset with the size of 5MB */ - amdgpu_ttm_init_vram_resv(adev, AMDGPU_RESV_FW_VRAM_USAGE, - adev->virt.crit_regn.offset, - adev->virt.crit_regn.size_kb << 10, true); - dev_info(adev->dev, - "critical region v%d requested to reserve memory start at %08llx with %llu KB.\n", - init_data_hdr->version, - adev->mman.resv_region[AMDGPU_RESV_FW_VRAM_USAGE].offset, - adev->mman.resv_region[AMDGPU_RESV_FW_VRAM_USAGE].size >> 10); - adev->virt.is_dynamic_crit_regn_enabled = true; out: @@ -1177,6 +1215,16 @@ int amdgpu_virt_init_critical_region(struct amdgpu_device *adev) return r; } +int amdgpu_virt_init_critical_region(struct amdgpu_device *adev) +{ + if (adev->virt.req_init_data_ver == GPU_CRIT_REGION_V1) + return amdgpu_virt_init_critical_region_v1(adev); + else if (adev->virt.req_init_data_ver == GPU_CRIT_REGION_V2) + return amdgpu_virt_init_critical_region_v2(adev); + + return 0; +} + int amdgpu_virt_get_dynamic_data_info(struct amdgpu_device *adev, int data_id, uint8_t *binary, u32 *size) { @@ -1193,10 +1241,20 @@ int amdgpu_virt_get_dynamic_data_info(struct amdgpu_device *adev, /* Validate on input params */ if (!binary || !size || *size < (uint64_t)data_size) return -EINVAL; + /* + * Proceed to copy the dynamic content. During early VF init the normal + * VRAM aperture is not mapped yet, so read directly from the VF FB BAR. + */ + if (amdgpu_sriov_vf(adev)) { + int r; - /* Proceed to copy the dynamic content */ - amdgpu_device_vram_access(adev, - (uint64_t)data_offset, (uint32_t *)binary, data_size, false); + r = amdgpu_virt_read_vf_fb(adev, data_offset, binary, data_size); + if (r) + return r; + } else { + amdgpu_device_vram_access(adev, (uint64_t)data_offset, + (uint32_t *)binary, data_size, false); + } *size = (uint64_t)data_size; dev_dbg(adev->dev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index d8500c3e48a1..1ffd6065c6be 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -452,6 +452,8 @@ static inline bool is_virtual_machine(void) ((adev)->virt.is_xgmi_node_migrate_enabled && (adev)->gmc.xgmi.node_segment_size != 0) bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev); +int amdgpu_virt_read_vf_fb(struct amdgpu_device *adev, u64 offset, + void *buf, size_t size); void amdgpu_virt_init_setting(struct amdgpu_device *adev); int amdgpu_virt_request_full_gpu(struct amdgpu_device *adev, bool init); int amdgpu_virt_release_full_gpu(struct amdgpu_device *adev, bool init); -- 2.48.1