dmaengine.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jacob Pan <jacob.jun.pan@linux.intel.com>
To: LKML <linux-kernel@vger.kernel.org>,
	iommu@lists.linux.dev, Jason Gunthorpe <jgg@nvidia.com>,
	"Lu Baolu" <baolu.lu@linux.intel.com>,
	Joerg Roedel <joro@8bytes.org>,
	dmaengine@vger.kernel.org, vkoul@kernel.org
Cc: "Robin Murphy" <robin.murphy@arm.com>,
	"Will Deacon" <will@kernel.org>,
	David Woodhouse <dwmw2@infradead.org>,
	Raj Ashok <ashok.raj@intel.com>,
	"Tian, Kevin" <kevin.tian@intel.com>, Yi Liu <yi.l.liu@intel.com>,
	"Yu, Fenghua" <fenghua.yu@intel.com>,
	Dave Jiang <dave.jiang@intel.com>,
	Tony Luck <tony.luck@intel.com>,
	"Zanussi, Tom" <tom.zanussi@intel.com>,
	Jacob Pan <jacob.jun.pan@linux.intel.com>
Subject: [PATCH 1/4] iommu/vt-d: Implement set device pasid op for default domain
Date: Wed,  1 Mar 2023 16:59:56 -0800	[thread overview]
Message-ID: <20230302005959.2695267-2-jacob.jun.pan@linux.intel.com> (raw)
In-Reply-To: <20230302005959.2695267-1-jacob.jun.pan@linux.intel.com>

On VT-d platforms, legacy DMA requests without PASID use device’s
default domain, where RID_PASID is always attached. Device drivers
can then use the DMA API for all in-kernel DMA on the RID.

Ideally, devices capable of using ENQCMDS can also transparently use the
default domain, consequently DMA API. However, VT-d architecture
dictates that the PASID used by ENQCMDS must be different from the
RID_PASID value.

To provide support for transparent use of DMA API with non-RID_PASID
value, this patch implements the set_dev_pasid() function for the
default domain. The idea is that device drivers wishing to use ENQCMDS
to submit work on buffers mapped by DMA API will call
iommu_attach_device_pasid() beforehand.

Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
---
 drivers/iommu/intel/iommu.c | 32 ++++++++++++++++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
index 10f657828d3a..a0cb3bc851ac 100644
--- a/drivers/iommu/intel/iommu.c
+++ b/drivers/iommu/intel/iommu.c
@@ -4665,6 +4665,10 @@ static void intel_iommu_remove_dev_pasid(struct device *dev, ioasid_t pasid)
 		case IOMMU_DOMAIN_SVA:
 			intel_svm_remove_dev_pasid(dev, pasid);
 			break;
+		case IOMMU_DOMAIN_DMA:
+		case IOMMU_DOMAIN_DMA_FQ:
+		case IOMMU_DOMAIN_IDENTITY:
+			break;
 		default:
 			/* should never reach here */
 			WARN_ON(1);
@@ -4675,6 +4679,33 @@ static void intel_iommu_remove_dev_pasid(struct device *dev, ioasid_t pasid)
 	intel_pasid_tear_down_entry(iommu, dev, pasid, false);
 }
 
+static int intel_iommu_set_dev_pasid(struct iommu_domain *domain,
+				   struct device *dev, ioasid_t pasid)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct dmar_domain *dmar_domain = to_dmar_domain(domain);
+	struct intel_iommu *iommu = info->iommu;
+	int ret = 0;
+
+	if (!sm_supported(iommu) || !info)
+		return -ENODEV;
+
+	if (WARN_ON(pasid == PASID_RID2PASID))
+		return -EINVAL;
+
+	if (hw_pass_through && domain_type_is_si(dmar_domain))
+		ret = intel_pasid_setup_pass_through(iommu, dmar_domain,
+						     dev, pasid);
+	else if (dmar_domain->use_first_level)
+		ret = domain_setup_first_level(iommu, dmar_domain,
+					       dev, pasid);
+	else
+		ret = intel_pasid_setup_second_level(iommu, dmar_domain,
+						     dev, pasid);
+
+	return ret;
+}
+
 const struct iommu_ops intel_iommu_ops = {
 	.capable		= intel_iommu_capable,
 	.domain_alloc		= intel_iommu_domain_alloc,
@@ -4702,6 +4733,7 @@ const struct iommu_ops intel_iommu_ops = {
 		.iova_to_phys		= intel_iommu_iova_to_phys,
 		.free			= intel_iommu_domain_free,
 		.enforce_cache_coherency = intel_iommu_enforce_cache_coherency,
+		.set_dev_pasid		= intel_iommu_set_dev_pasid,
 	}
 };
 
-- 
2.25.1


  reply	other threads:[~2023-03-02  0:56 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-02  0:59 [PATCH 0/4] Re-enable IDXD kernel workqueue under DMA API Jacob Pan
2023-03-02  0:59 ` Jacob Pan [this message]
2023-03-02  9:37   ` [PATCH 1/4] iommu/vt-d: Implement set device pasid op for default domain Tian, Kevin
2023-03-02 19:25     ` Jacob Pan
2023-03-02 14:06   ` Baolu Lu
2023-03-03  2:36     ` Tian, Kevin
2023-03-03  2:48       ` Baolu Lu
2023-03-03  3:02         ` Tian, Kevin
2023-03-03  4:38           ` Baolu Lu
2023-03-03  5:35             ` Tian, Kevin
2023-03-06 19:04               ` Jacob Pan
2023-03-06 19:02                 ` Jason Gunthorpe
2023-03-06 23:45                   ` Jacob Pan
2023-03-07  0:45                     ` Jacob Pan
2023-03-03  5:38     ` Tian, Kevin
2023-03-03 16:35       ` Jacob Pan
2023-03-05  3:05         ` Baolu Lu
2023-03-06  8:18           ` Tian, Kevin
2023-03-06 18:43             ` Jacob Pan
2023-03-06 18:29           ` Jacob Pan
2023-03-06 12:57   ` Jason Gunthorpe
2023-03-06 17:36     ` Jacob Pan
2023-03-06 17:41       ` Jason Gunthorpe
2023-03-07  2:15         ` Baolu Lu
2023-03-02  0:59 ` [PATCH 2/4] iommu/vt-d: Use non-privileged mode for all PASIDs Jacob Pan
2023-03-02 14:11   ` Baolu Lu
2023-03-03 21:40     ` Jacob Pan
2023-03-02  0:59 ` [PATCH 3/4] iommu/sva: Support reservation of global PASIDs Jacob Pan
2023-03-02  3:06   ` kernel test robot
2023-03-02  3:19   ` kernel test robot
2023-03-02  9:43   ` Tian, Kevin
2023-03-03 21:47     ` Jacob Pan
2023-03-06 13:01       ` Jason Gunthorpe
2023-03-06 17:44         ` Jacob Pan
2023-03-06 17:43           ` Jason Gunthorpe
2023-03-06 17:57             ` Jacob Pan
2023-03-06 18:19               ` Jason Gunthorpe
2023-03-06 18:48                 ` Luck, Tony
2023-03-06 19:05                   ` Jason Gunthorpe
2023-03-09 17:06                     ` Jacob Pan
2023-03-16  7:25                       ` Tian, Kevin
2023-03-20 17:22                         ` Jason Gunthorpe
2023-03-02  0:59 ` [PATCH 4/4] dmaengine/idxd: Re-enable kernel workqueue under DMA API Jacob Pan
2023-03-02  1:03   ` Dave Jiang
2023-03-02  9:47   ` Tian, Kevin
2023-03-02 12:57     ` Jason Gunthorpe
2023-03-03 21:49       ` Jacob Pan
2023-03-03 22:12     ` Jacob Pan
2023-03-03  1:19   ` Baolu Lu
2023-03-03 21:52     ` Jacob Pan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230302005959.2695267-2-jacob.jun.pan@linux.intel.com \
    --to=jacob.jun.pan@linux.intel.com \
    --cc=ashok.raj@intel.com \
    --cc=baolu.lu@linux.intel.com \
    --cc=dave.jiang@intel.com \
    --cc=dmaengine@vger.kernel.org \
    --cc=dwmw2@infradead.org \
    --cc=fenghua.yu@intel.com \
    --cc=iommu@lists.linux.dev \
    --cc=jgg@nvidia.com \
    --cc=joro@8bytes.org \
    --cc=kevin.tian@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=robin.murphy@arm.com \
    --cc=tom.zanussi@intel.com \
    --cc=tony.luck@intel.com \
    --cc=vkoul@kernel.org \
    --cc=will@kernel.org \
    --cc=yi.l.liu@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).