From patchwork Thu Nov 21 13:40:48 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hiroshi DOYU X-Patchwork-Id: 3218581 Return-Path: X-Original-To: patchwork-linux-arm@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 332A59F26C for ; Thu, 21 Nov 2013 14:07:21 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 495822074B for ; Thu, 21 Nov 2013 14:07:16 +0000 (UTC) Received: from casper.infradead.org (casper.infradead.org [85.118.1.10]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 3DC2420716 for ; Thu, 21 Nov 2013 14:07:11 +0000 (UTC) Received: from merlin.infradead.org ([2001:4978:20e::2]) by casper.infradead.org with esmtps (Exim 4.80.1 #2 (Red Hat Linux)) id 1VjUZ1-0004tW-Hx; Thu, 21 Nov 2013 13:45:01 +0000 Received: from localhost ([::1] helo=merlin.infradead.org) by merlin.infradead.org with esmtp (Exim 4.80.1 #2 (Red Hat Linux)) id 1VjUY4-0003Fk-0q; Thu, 21 Nov 2013 13:44:00 +0000 Received: from hqemgate16.nvidia.com ([216.228.121.65]) by merlin.infradead.org with esmtps (Exim 4.80.1 #2 (Red Hat Linux)) id 1VjUWM-00032J-33 for linux-arm-kernel@lists.infradead.org; Thu, 21 Nov 2013 13:42:26 +0000 Received: from hqnvupgp08.nvidia.com (Not Verified[216.228.121.13]) by hqemgate16.nvidia.com id ; Thu, 21 Nov 2013 05:41:39 -0800 Received: from hqemhub01.nvidia.com ([172.20.12.94]) by hqnvupgp08.nvidia.com (PGP Universal service); Thu, 21 Nov 2013 05:35:52 -0800 X-PGP-Universal: processed; by hqnvupgp08.nvidia.com on Thu, 21 Nov 2013 05:35:52 -0800 Received: from hqnvemgw02.nvidia.com (172.16.227.111) by hqemhub01.nvidia.com (172.20.150.30) with Microsoft SMTP Server id 8.3.327.1; Thu, 21 Nov 2013 05:41:54 -0800 Received: from sc-daphne.nvidia.com (Not Verified[172.20.232.60]) by hqnvemgw02.nvidia.com with MailMarshal (v7,1,2,5326) id ; Thu, 21 Nov 2013 05:41:54 -0800 Received: from oreo.Nvidia.com (dhcp-10-21-26-134.nvidia.com [10.21.26.134]) by sc-daphne.nvidia.com (8.13.8+Sun/8.8.8) with ESMTP id rALDf0sD014302; Thu, 21 Nov 2013 05:41:50 -0800 (PST) From: Hiroshi Doyu To: , , , , , , Subject: [PATCHv6 12/13] iommu/tegra: smmu: add SMMU to an global iommu list Date: Thu, 21 Nov 2013 15:40:48 +0200 Message-ID: <1385041249-7705-13-git-send-email-hdoyu@nvidia.com> X-Mailer: git-send-email 1.8.1.5 In-Reply-To: <1385041249-7705-1-git-send-email-hdoyu@nvidia.com> References: <1385041249-7705-1-git-send-email-hdoyu@nvidia.com> MIME-Version: 1.0 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20131121_084214_421739_852AF065 X-CRM114-Status: GOOD ( 15.05 ) X-Spam-Score: -2.4 (--) Cc: mark.rutland@arm.com, devicetree@vger.kernel.org, lorenzo.pieralisi@arm.com, linux-kernel@vger.kernel.org, iommu@lists.linux-foundation.org, galak@codeaurora.org, linux-tegra@vger.kernel.org, linux-arm-kernel@lists.infradead.org, Hiroshi Doyu X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+patchwork-linux-arm=patchwork.kernel.org@lists.infradead.org X-Spam-Status: No, score=-4.7 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 allows to inquire if SMMU is populated or not. Suggested by Thierry Reding and copied his example code. Signed-off-by: Hiroshi Doyu Cc: Thierry Reding --- v6: New for v6. --- drivers/iommu/tegra-smmu.c | 55 +++++++++++++++++++++++++--------------------- 1 file changed, 30 insertions(+), 25 deletions(-) diff --git a/drivers/iommu/tegra-smmu.c b/drivers/iommu/tegra-smmu.c index 1544f7c..590c9fe 100644 --- a/drivers/iommu/tegra-smmu.c +++ b/drivers/iommu/tegra-smmu.c @@ -225,6 +225,8 @@ struct smmu_debugfs_info { * Per SMMU device - IOMMU device */ struct smmu_device { + struct iommu iommu; + void __iomem *regbase; /* register offset base */ void __iomem **regs; /* register block start address array */ void __iomem **rege; /* register block end address array */ @@ -234,7 +236,6 @@ struct smmu_device { unsigned long page_count; /* total remappable size */ spinlock_t lock; char *name; - struct device *dev; struct rb_root clients; struct page *avp_vector_page; /* dummy page shared by all AS's */ @@ -371,7 +372,7 @@ static int register_smmu_client(struct smmu_device *smmu, return -EBUSY; } - client = devm_kzalloc(smmu->dev, sizeof(*client), GFP_KERNEL); + client = devm_kzalloc(smmu->iommu.dev, sizeof(*client), GFP_KERNEL); if (!client) return -ENOMEM; @@ -388,7 +389,7 @@ static int smmu_of_get_swgroups(struct device *dev, unsigned long *swgroups) of_property_for_each_phandle_with_args(dev->of_node, "iommus", "#iommu-cells", i, &args) { - if (args.np != smmu_handle->dev->of_node) + if (args.np != smmu_handle->iommu.dev->of_node) continue; BUG_ON(args.args_count != 2); @@ -527,7 +528,7 @@ static void free_ptbl(struct smmu_as *as, dma_addr_t iova) unsigned long *pdir = (unsigned long *)page_address(as->pdir_page); if (pdir[pdn] != _PDE_VACANT(pdn)) { - dev_dbg(as->smmu->dev, "pdn: %lx\n", pdn); + dev_dbg(as->smmu->iommu.dev, "pdn: %lx\n", pdn); ClearPageReserved(SMMU_EX_PTBL_PAGE(pdir[pdn])); __free_page(SMMU_EX_PTBL_PAGE(pdir[pdn])); @@ -542,7 +543,7 @@ static void free_pdir(struct smmu_as *as) { unsigned addr; int count; - struct device *dev = as->smmu->dev; + struct device *dev = as->smmu->iommu.dev; if (!as->pdir_page) return; @@ -585,11 +586,11 @@ static unsigned long *locate_pte(struct smmu_as *as, unsigned long addr = SMMU_PDN_TO_ADDR(pdn); /* Vacant - allocate a new page table */ - dev_dbg(as->smmu->dev, "New PTBL pdn: %lx\n", pdn); + dev_dbg(as->smmu->iommu.dev, "New PTBL pdn: %lx\n", pdn); *ptbl_page_p = alloc_page(GFP_ATOMIC); if (!*ptbl_page_p) { - dev_err(as->smmu->dev, + dev_err(as->smmu->iommu.dev, "failed to allocate smmu_device page table\n"); return NULL; } @@ -649,7 +650,7 @@ static int alloc_pdir(struct smmu_as *as) /* * do the allocation, then grab as->lock */ - cnt = devm_kzalloc(smmu->dev, + cnt = devm_kzalloc(smmu->iommu.dev, sizeof(cnt[0]) * SMMU_PDIR_COUNT, GFP_KERNEL); page = alloc_page(GFP_KERNEL | __GFP_DMA); @@ -663,7 +664,8 @@ static int alloc_pdir(struct smmu_as *as) } if (!page || !cnt) { - dev_err(smmu->dev, "failed to allocate at %s\n", __func__); + dev_err(smmu->iommu.dev, + "failed to allocate at %s\n", __func__); err = -ENOMEM; goto err_out; } @@ -693,7 +695,7 @@ static int alloc_pdir(struct smmu_as *as) err_out: spin_unlock_irqrestore(&as->lock, flags); - devm_kfree(smmu->dev, cnt); + devm_kfree(smmu->iommu.dev, cnt); if (page) __free_page(page); return err; @@ -748,7 +750,7 @@ static int smmu_iommu_map(struct iommu_domain *domain, unsigned long iova, unsigned long pfn = __phys_to_pfn(pa); unsigned long flags; - dev_info(as->smmu->dev, "[%d] %08lx:%pa\n", as->asid, iova, &pa); + dev_info(as->smmu->iommu.dev, "[%d] %08lx:%pa\n", as->asid, iova, &pa); if (!pfn_valid(pfn)) return -ENOMEM; @@ -765,7 +767,7 @@ static size_t smmu_iommu_unmap(struct iommu_domain *domain, unsigned long iova, struct smmu_as *as = domain->priv; unsigned long flags; - dev_dbg(as->smmu->dev, "[%d] %08lx\n", as->asid, iova); + dev_dbg(as->smmu->iommu.dev, "[%d] %08lx\n", as->asid, iova); spin_lock_irqsave(&as->lock, flags); __smmu_iommu_unmap(as, iova); @@ -788,7 +790,7 @@ static phys_addr_t smmu_iommu_iova_to_phys(struct iommu_domain *domain, pte = locate_pte(as, iova, true, &page, &count); pfn = *pte & SMMU_PFN_MASK; WARN_ON(!pfn_valid(pfn)); - dev_dbg(as->smmu->dev, + dev_dbg(as->smmu->iommu.dev, "iova:%08llx pfn:%08lx asid:%d\n", (unsigned long long)iova, pfn, as->asid); @@ -822,7 +824,7 @@ static int smmu_iommu_attach_dev(struct iommu_domain *domain, spin_lock(&as->client_lock); list_for_each_entry(c, &as->client, list) { if (c->dev == dev) { - dev_err(smmu->dev, + dev_err(smmu->iommu.dev, "%s is already attached\n", dev_name(c->dev)); err = -EINVAL; goto err_client; @@ -844,7 +846,7 @@ static int smmu_iommu_attach_dev(struct iommu_domain *domain, pr_info("Reserve \"page zero\" for AVP vectors using a common dummy\n"); } - dev_dbg(smmu->dev, "%s is attached\n", dev_name(dev)); + dev_dbg(smmu->iommu.dev, "%s is attached\n", dev_name(dev)); return 0; err_client: @@ -867,12 +869,12 @@ static void smmu_iommu_detach_dev(struct iommu_domain *domain, smmu_client_disable_swgroups(c); list_del(&c->list); c->as = NULL; - dev_dbg(smmu->dev, + dev_dbg(smmu->iommu.dev, "%s is detached\n", dev_name(c->dev)); goto out; } } - dev_err(smmu->dev, "Couldn't find %s\n", dev_name(dev)); + dev_err(smmu->iommu.dev, "Couldn't find %s\n", dev_name(dev)); out: spin_unlock(&as->client_lock); } @@ -899,7 +901,7 @@ static int smmu_iommu_domain_init(struct iommu_domain *domain) break; } if (i == smmu->num_as) - dev_err(smmu->dev, "no free AS\n"); + dev_err(smmu->iommu.dev, "no free AS\n"); return err; found: @@ -920,7 +922,7 @@ found: smmu->page_count * SMMU_PAGE_SIZE - 1; domain->geometry.force_aperture = true; - dev_dbg(smmu->dev, "smmu_as@%p\n", as); + dev_dbg(smmu->iommu.dev, "smmu_as@%p\n", as); return 0; } @@ -953,7 +955,7 @@ static void smmu_iommu_domain_destroy(struct iommu_domain *domain) spin_unlock_irqrestore(&as->lock, flags); domain->priv = NULL; - dev_dbg(smmu->dev, "smmu_as@%p\n", as); + dev_dbg(smmu->iommu.dev, "smmu_as@%p\n", as); } /* @@ -1085,7 +1087,7 @@ static ssize_t smmu_debugfs_stats_write(struct file *file, break; } - dev_dbg(smmu->dev, "%s() %08x, %08x @%08x\n", __func__, + dev_dbg(smmu->iommu.dev, "%s() %08x, %08x @%08x\n", __func__, val, smmu_read(smmu, offs), offs); return count; @@ -1107,7 +1109,7 @@ static int smmu_debugfs_stats_show(struct seq_file *s, void *v) val = smmu_read(smmu, offs); seq_printf(s, "%s:%08x ", stats[i], val); - dev_dbg(smmu->dev, "%s() %s %08x @%08x\n", __func__, + dev_dbg(smmu->iommu.dev, "%s() %s %08x @%08x\n", __func__, stats[i], val, offs); } seq_printf(s, "\n"); @@ -1145,7 +1147,7 @@ static void smmu_debugfs_create(struct smmu_device *smmu) if (!smmu->debugfs_info) return; - root = debugfs_create_dir(dev_name(smmu->dev), NULL); + root = debugfs_create_dir(dev_name(smmu->iommu.dev), NULL); if (!root) goto err_out; smmu->debugfs_root = root; @@ -1217,7 +1219,7 @@ static void tegra_smmu_create_default_map(struct smmu_device *smmu) smmu->map[i] = arm_iommu_create_mapping(&platform_bus_type, base, size, 0); if (IS_ERR(smmu->map[i])) - dev_err(smmu->dev, + dev_err(smmu->iommu.dev, "Couldn't create: asid=%d map=%p %pa-%pa\n", i, smmu->map[i], &base, &base + size - 1); } @@ -1285,7 +1287,7 @@ static int tegra_smmu_probe(struct platform_device *pdev) if (!smmu->ahb) return -ENODEV; - smmu->dev = dev; + smmu->iommu.dev = dev; smmu->num_as = asids; smmu->iovmm_base = base; smmu->page_count = size; @@ -1322,6 +1324,8 @@ static int tegra_smmu_probe(struct platform_device *pdev) smmu_handle = smmu; bus_set_iommu(&platform_bus_type, &smmu_iommu_ops); tegra_smmu_create_default_map(smmu); + + iommu_add(&smmu->iommu); return 0; } @@ -1337,6 +1341,7 @@ static int tegra_smmu_remove(struct platform_device *pdev) free_pdir(&smmu->as[i]); __free_page(smmu->avp_vector_page); smmu_handle = NULL; + iommu_del(&smmu->iommu); return 0; }