From patchwork Sat Nov 8 18:37:32 2014 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Oded Gabbay X-Patchwork-Id: 5258711 Return-Path: X-Original-To: patchwork-dri-devel@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork1.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.19.201]) by patchwork1.web.kernel.org (Postfix) with ESMTP id E943C9F39B for ; Sat, 8 Nov 2014 18:42:36 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id D206920148 for ; Sat, 8 Nov 2014 18:42:35 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) by mail.kernel.org (Postfix) with ESMTP id BCE8020138 for ; Sat, 8 Nov 2014 18:42:34 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 3CDBF6E29C; Sat, 8 Nov 2014 10:42:34 -0800 (PST) X-Original-To: dri-devel@lists.freedesktop.org Delivered-To: dri-devel@lists.freedesktop.org Received: from na01-bl2-obe.outbound.protection.outlook.com (mail-bl2on0139.outbound.protection.outlook.com [65.55.169.139]) by gabe.freedesktop.org (Postfix) with ESMTP id 798246E29C for ; Sat, 8 Nov 2014 10:42:32 -0800 (PST) Received: from BN1PR02CA0038.namprd02.prod.outlook.com (10.141.56.38) by CO1PR02MB206.namprd02.prod.outlook.com (10.242.165.144) with Microsoft SMTP Server (TLS) id 15.1.6.9; Sat, 8 Nov 2014 18:42:30 +0000 Received: from BL2FFO11FD026.protection.gbl (2a01:111:f400:7c09::140) by BN1PR02CA0038.outlook.office365.com (2a01:111:e400:2a::38) with Microsoft SMTP Server (TLS) id 15.1.16.15 via Frontend Transport; Sat, 8 Nov 2014 18:42:29 +0000 Received: from atltwp02.amd.com (165.204.84.222) by BL2FFO11FD026.mail.protection.outlook.com (10.173.161.105) with Microsoft SMTP Server id 15.1.6.13 via Frontend Transport; Sat, 8 Nov 2014 18:42:29 +0000 X-WSS-ID: 0NEQHAO-08-GDR-02 X-M-MSG: Received: from satlvexedge01.amd.com (satlvexedge01.amd.com [10.177.96.28]) (using TLSv1 with cipher AES128-SHA (128/128 bits)) (No client certificate requested) by atltwp02.amd.com (Axway MailGate 5.3.1) with ESMTPS id 21E2CD16028; Sat, 8 Nov 2014 12:42:24 -0600 (CST) Received: from SATLEXDAG02.amd.com (10.181.40.5) by satlvexedge01.amd.com (10.177.96.28) with Microsoft SMTP Server (TLS) id 14.3.195.1; Sat, 8 Nov 2014 12:42:38 -0600 Received: from odedg-kaveri.amd.com (10.180.168.240) by SATLEXDAG02.amd.com (10.181.40.5) with Microsoft SMTP Server (TLS) id 14.3.195.1; Sat, 8 Nov 2014 13:42:27 -0500 From: Oded Gabbay To: David Airlie , Alex Deucher , Jerome Glisse Subject: [PATCH v5 11/24] amdkfd: Add binding/unbinding calls to amd_iommu driver Date: Sat, 8 Nov 2014 20:37:32 +0200 Message-ID: <1415471865-1893-12-git-send-email-oded.gabbay@amd.com> X-Mailer: git-send-email 2.1.0 In-Reply-To: <1415471865-1893-1-git-send-email-oded.gabbay@amd.com> References: <1415471865-1893-1-git-send-email-oded.gabbay@amd.com> MIME-Version: 1.0 X-Originating-IP: [10.180.168.240] X-EOPAttributedMessage: 0 X-Forefront-Antispam-Report: CIP:165.204.84.222; CTRY:US; IPV:NLI; EFV:NLI; SFV:NSPM; SFS:(10019020)(6009001)(428002)(189002)(199003)(33646002)(53416004)(50466002)(106466001)(4396001)(50226001)(99396003)(95666004)(62966003)(120916001)(229853001)(105586002)(107046002)(102836001)(68736004)(77096003)(77156002)(64706001)(87286001)(48376002)(87936001)(88136002)(89996001)(20776003)(44976005)(47776003)(104166001)(46102003)(36756003)(86362001)(31966008)(50986999)(97736003)(84676001)(19580395003)(92566001)(101416001)(93916002)(21056001)(92726001)(76176999)(19580405001); DIR:OUT; SFP:1102; SCL:1; SRVR:CO1PR02MB206; H:atltwp02.amd.com; FPR:; MLV:sfv; PTR:InfoDomainNonexistent; MX:1; A:1; LANG:en; X-Microsoft-Antispam: UriScan:; X-Microsoft-Antispam: BCL:0;PCL:0;RULEID:;SRVR:CO1PR02MB206; X-Exchange-Antispam-Report-Test: UriScan:; X-Exchange-Antispam-Report-CFA: BCL:0;PCL:0;RULEID:;SRVR:CO1PR02MB206; X-Forefront-PRVS: 0389EDA07F Received-SPF: None (protection.outlook.com: amd.com does not designate permitted sender hosts) Authentication-Results: spf=none (sender IP is 165.204.84.222) smtp.mailfrom=Oded.Gabbay@amd.com; X-Exchange-Antispam-Report-CFA: BCL:0;PCL:0;RULEID:;SRVR:CO1PR02MB206; X-OriginatorOrg: amd4.onmicrosoft.com Cc: Jay Cornwall , Andrew Lewycky , linux-kernel@vger.kernel.org, Evgeny Pinchuk , dri-devel@lists.freedesktop.org X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" X-Spam-Status: No, score=-4.8 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_MED, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP This patch adds the functions to bind and unbind pasid from a device through the amd_iommu driver. The unbind function is called when the mm_struct of the process is released. The bind function is not called here because it is called only in the IOCTLs which are not yet implemented at this stage of the patchset. Signed-off-by: Oded Gabbay --- drivers/gpu/drm/amd/amdkfd/kfd_device.c | 90 +++++++++++++++++++++++++++++++- drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 1 + drivers/gpu/drm/amd/amdkfd/kfd_process.c | 12 +++++ 3 files changed, 102 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index 71a03f7..4b1cced 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -100,6 +100,65 @@ struct kfd_dev *kgd2kfd_probe(struct kgd_dev *kgd, struct pci_dev *pdev) return kfd; } +static bool device_iommu_pasid_init(struct kfd_dev *kfd) +{ + const u32 required_iommu_flags = AMD_IOMMU_DEVICE_FLAG_ATS_SUP | + AMD_IOMMU_DEVICE_FLAG_PRI_SUP | + AMD_IOMMU_DEVICE_FLAG_PASID_SUP; + + struct amd_iommu_device_info iommu_info; + unsigned int pasid_limit; + int err; + + err = amd_iommu_device_info(kfd->pdev, &iommu_info); + if (err < 0) { + dev_err(kfd_device, + "error getting iommu info. is the iommu enabled?\n"); + return false; + } + + if ((iommu_info.flags & required_iommu_flags) != required_iommu_flags) { + dev_err(kfd_device, "error required iommu flags ats(%i), pri(%i), pasid(%i)\n", + (iommu_info.flags & AMD_IOMMU_DEVICE_FLAG_ATS_SUP) != 0, + (iommu_info.flags & AMD_IOMMU_DEVICE_FLAG_PRI_SUP) != 0, + (iommu_info.flags & AMD_IOMMU_DEVICE_FLAG_PASID_SUP) != 0); + return false; + } + + pasid_limit = min_t(unsigned int, + (unsigned int)1 << kfd->device_info->max_pasid_bits, + iommu_info.max_pasids); + /* + * last pasid is used for kernel queues doorbells + * in the future the last pasid might be used for a kernel thread. + */ + pasid_limit = min_t(unsigned int, + pasid_limit, + kfd->doorbell_process_limit - 1); + + err = amd_iommu_init_device(kfd->pdev, pasid_limit); + if (err < 0) { + dev_err(kfd_device, "error initializing iommu device\n"); + return false; + } + + if (!kfd_set_pasid_limit(pasid_limit)) { + dev_err(kfd_device, "error setting pasid limit\n"); + amd_iommu_free_device(kfd->pdev); + return false; + } + + return true; +} + +static void iommu_pasid_shutdown_callback(struct pci_dev *pdev, int pasid) +{ + struct kfd_dev *dev = kfd_device_by_pci_dev(pdev); + + if (dev) + kfd_unbind_process_from_device(dev, pasid); +} + bool kgd2kfd_device_init(struct kfd_dev *kfd, const struct kgd2kfd_shared_resources *gpu_resources) { @@ -131,6 +190,14 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, goto kfd_topology_add_device_error; } + if (!device_iommu_pasid_init(kfd)) { + dev_err(kfd_device, + "Error initializing iommuv2 for device (%x:%x)\n", + kfd->pdev->vendor, kfd->pdev->device); + goto device_iommu_pasid_error; + } + amd_iommu_set_invalidate_ctx_cb(kfd->pdev, + iommu_pasid_shutdown_callback); kfd->init_complete = true; dev_info(kfd_device, "added device (%x:%x)\n", kfd->pdev->vendor, @@ -138,6 +205,8 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, goto out; +device_iommu_pasid_error: + kfd_topology_remove_device(kfd); kfd_topology_add_device_error: kfd2kgd->fini_sa_manager(kfd->kgd); dev_err(kfd_device, @@ -149,7 +218,10 @@ out: void kgd2kfd_device_exit(struct kfd_dev *kfd) { - kfd_topology_remove_device(kfd); + if (kfd->init_complete) { + amd_iommu_free_device(kfd->pdev); + kfd_topology_remove_device(kfd); + } kfree(kfd); } @@ -157,12 +229,28 @@ void kgd2kfd_device_exit(struct kfd_dev *kfd) void kgd2kfd_suspend(struct kfd_dev *kfd) { BUG_ON(kfd == NULL); + + if (kfd->init_complete) + amd_iommu_free_device(kfd->pdev); } int kgd2kfd_resume(struct kfd_dev *kfd) { + unsigned int pasid_limit; + int err; + BUG_ON(kfd == NULL); + pasid_limit = kfd_get_pasid_limit(); + + if (kfd->init_complete) { + err = amd_iommu_init_device(kfd->pdev, pasid_limit); + if (err < 0) + return -ENXIO; + amd_iommu_set_invalidate_ctx_cb(kfd->pdev, + iommu_pasid_shutdown_callback); + } + return 0; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index 77d15db..431a09b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -194,6 +194,7 @@ void kfd_process_destroy_wq(void); struct kfd_process *kfd_create_process(const struct task_struct *); struct kfd_process *kfd_get_process(const struct task_struct *); +void kfd_unbind_process_from_device(struct kfd_dev *dev, unsigned int pasid); struct kfd_process_device *kfd_get_process_device_data(struct kfd_dev *dev, struct kfd_process *p, int create_pdd); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 5596f69..5084794 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -24,6 +24,7 @@ #include #include #include +#include #include struct mm_struct; @@ -163,6 +164,7 @@ static void kfd_process_wq_release(struct work_struct *work) list_for_each_entry_safe(pdd, temp, &p->per_device_data, per_device_list) { + amd_iommu_unbind_pasid(pdd->dev->pdev, p->pasid); list_del(&pdd->per_device_list); kfree(pdd); @@ -316,6 +318,7 @@ struct kfd_process_device *kfd_bind_process_to_device(struct kfd_dev *dev, struct kfd_process *p) { struct kfd_process_device *pdd = kfd_get_process_device_data(dev, p, 1); + int err; if (pdd == NULL) return ERR_PTR(-ENOMEM); @@ -323,6 +326,15 @@ struct kfd_process_device *kfd_bind_process_to_device(struct kfd_dev *dev, if (pdd->bound) return pdd; + err = amd_iommu_bind_pasid(dev->pdev, p->pasid, p->lead_thread); + if (err < 0) + return ERR_PTR(err); + + if (err < 0) { + amd_iommu_unbind_pasid(dev->pdev, p->pasid); + return ERR_PTR(err); + } + pdd->bound = true; return pdd;