From: Nicolin Chen <nicolinc@nvidia.com>
To: <will@kernel.org>, <robin.murphy@arm.com>, <jgg@nvidia.com>
Cc: <joro@8bytes.org>, <kevin.tian@intel.com>, <praan@google.com>,
<nathan@kernel.org>, <yi.l.liu@intel.com>, <peterz@infradead.org>,
<mshavit@google.com>, <jsnitsel@redhat.com>,
<smostafa@google.com>, <jeff.johnson@oss.qualcomm.com>,
<zhangzekun11@huawei.com>, <linux-arm-kernel@lists.infradead.org>,
<iommu@lists.linux.dev>, <linux-kernel@vger.kernel.org>,
<shameerali.kolothum.thodi@huawei.com>
Subject: [PATCH v2 11/11] iommu/arm-smmu-v3: Allow to share S2 nest_parent domain across vSMMUs
Date: Mon, 14 Apr 2025 21:57:46 -0700 [thread overview]
Message-ID: <e6a808167d61bc254aea397cbe75622e02e9d703.1744692494.git.nicolinc@nvidia.com> (raw)
In-Reply-To: <cover.1744692494.git.nicolinc@nvidia.com>
An S2 nest_parent domain used by one vSMMU can be shared with another vSMMU
so long as the underlying stage-2 page table is compatible by the physical
SMMU instance.
There is no direct information about the page table from the master device,
but a comparison can be done between the physical SMMU that the nest_parent
domain was allocated for and the physical SMMU that the device is behind.
Replace the smmu test in arm_vsmmu_alloc() with a compatibility test, which
goes through the physical SMMU parameters that were used to decide the page
table formats.
Signed-off-by: Nicolin Chen <nicolinc@nvidia.com>
---
.../arm/arm-smmu-v3/arm-smmu-v3-iommufd.c | 21 ++++++++++++++++++-
1 file changed, 20 insertions(+), 1 deletion(-)
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c
index 5d05f8a78215..f654e665739a 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c
@@ -501,6 +501,25 @@ static const struct iommufd_viommu_ops arm_vsmmu_ops = {
.cache_invalidate = arm_vsmmu_cache_invalidate,
};
+static bool arm_smmu_s2_parent_can_share(struct arm_smmu_domain *s2_parent,
+ struct arm_smmu_device *smmu)
+{
+ struct arm_smmu_device *s2_smmu = s2_parent->smmu;
+
+ if (s2_smmu == smmu)
+ return true;
+ if (s2_smmu->iommu.ops != smmu->iommu.ops)
+ return false;
+ if (s2_smmu->ias > smmu->ias || s2_smmu->oas > smmu->oas)
+ return false;
+ if (s2_smmu->pgsize_bitmap != smmu->pgsize_bitmap)
+ return false;
+ if ((s2_smmu->features & ARM_SMMU_FEAT_COHERENCY) !=
+ (smmu->features & ARM_SMMU_FEAT_COHERENCY))
+ return false;
+ return true;
+}
+
struct iommufd_viommu *arm_vsmmu_alloc(struct device *dev,
struct iommu_domain *parent,
struct iommufd_ctx *ictx,
@@ -520,7 +539,7 @@ struct iommufd_viommu *arm_vsmmu_alloc(struct device *dev,
if (!(smmu->features & ARM_SMMU_FEAT_NESTING))
return ERR_PTR(-EOPNOTSUPP);
- if (s2_parent->smmu != master->smmu)
+ if (!arm_smmu_s2_parent_can_share(s2_parent, master->smmu))
return ERR_PTR(-EINVAL);
/*
--
2.43.0
prev parent reply other threads:[~2025-04-15 4:58 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-15 4:57 [PATCH v2 00/11] iommu/arm-smmu-v3: Allocate vmid per vsmmu instead of s2_parent Nicolin Chen
2025-04-15 4:57 ` [PATCH v2 01/11] iommu/arm-smmu-v3: Pass in vmid to arm_smmu_make_s2_domain_ste() Nicolin Chen
2025-04-15 4:57 ` [PATCH v2 02/11] iommu/arm-smmu-v3: Pass in smmu/iommu_domain to __arm_smmu_tlb_inv_range() Nicolin Chen
2025-05-15 15:06 ` Will Deacon
2025-04-15 4:57 ` [PATCH v2 03/11] iommu/arm-smmu-v3: Share cmdq/cmd helpers with arm-smmu-v3-iommufd Nicolin Chen
2025-04-15 4:57 ` [PATCH v2 04/11] iommu/arm-smmu-v3: Add an inline arm_smmu_tlb_inv_vmid helper Nicolin Chen
2025-04-15 4:57 ` [PATCH v2 05/11] iommu/arm-smmu-v3: Rename arm_smmu_attach_prepare_vmaster Nicolin Chen
2025-04-15 4:57 ` [PATCH v2 06/11] iommu/arm-smmu-v3: Introduce arm_smmu_s2_parent_tlb_ invalidation helpers Nicolin Chen
2025-04-15 12:50 ` Jason Gunthorpe
2025-04-15 20:10 ` Nicolin Chen
2025-04-15 23:46 ` Jason Gunthorpe
2025-04-15 4:57 ` [PATCH v2 07/11] iommu/arm-smmu-v3: Introduce arm_vsmmu_atc_inv_domain() Nicolin Chen
2025-04-15 4:57 ` [PATCH v2 08/11] iommu/arm-smmu-v3: Use vSMMU helpers for S2 and ATC invalidations Nicolin Chen
2025-04-15 4:57 ` [PATCH v2 09/11] iommu/arm-smmu-v3: Clean up nested_ats_flush from master_domain Nicolin Chen
2025-04-15 4:57 ` [PATCH v2 10/11] iommu/arm-smmu-v3: Decouple vmid from S2 nest_parent domain Nicolin Chen
2025-04-15 4:57 ` Nicolin Chen [this message]
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=e6a808167d61bc254aea397cbe75622e02e9d703.1744692494.git.nicolinc@nvidia.com \
--to=nicolinc@nvidia.com \
--cc=iommu@lists.linux.dev \
--cc=jeff.johnson@oss.qualcomm.com \
--cc=jgg@nvidia.com \
--cc=joro@8bytes.org \
--cc=jsnitsel@redhat.com \
--cc=kevin.tian@intel.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mshavit@google.com \
--cc=nathan@kernel.org \
--cc=peterz@infradead.org \
--cc=praan@google.com \
--cc=robin.murphy@arm.com \
--cc=shameerali.kolothum.thodi@huawei.com \
--cc=smostafa@google.com \
--cc=will@kernel.org \
--cc=yi.l.liu@intel.com \
--cc=zhangzekun11@huawei.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
all inboxes | Powered by JetHome®