Commit 1554240f authored by Will Deacon's avatar Will Deacon
Browse files

Merge branches 'for-joerg/arm-smmu/smmu-v2' and 'for-joerg/arm-smmu/smmu-v3'... 

Merge branches 'for-joerg/arm-smmu/smmu-v2' and 'for-joerg/arm-smmu/smmu-v3' into for-joerg/arm-smmu/updates

* for-joerg/arm-smmu/smmu-v2:
  Refactoring to allow for implementation-specific hooks in 'arm-smmu-impl.c'

* for-joerg/arm-smmu/smmu-v3:
  Support for deferred TLB invalidation and batching of commands
  Rework ATC invalidation for ATS-enabled PCIe masters
parents d720e641 a91bcc2b

drivers/gpu/drm/panfrost/panfrost_mmu.c

+16 −8
Original line number Diff line number Diff line
@@ -222,7 +222,7 @@ void panfrost_mmu_unmap(struct panfrost_gem_object *bo) 
		size_t unmapped_page;

		size_t pgsize = get_pgsize(iova, len - unmapped_len);



		unmapped_page = ops->unmap(ops, iova, pgsize);

		unmapped_page = ops->unmap(ops, iova, pgsize, NULL);

		if (!unmapped_page)

			break;
@@ -247,20 +247,28 @@ static void mmu_tlb_inv_context_s1(void *cookie) 
	mmu_hw_do_operation(pfdev, 0, 0, ~0UL, AS_COMMAND_FLUSH_MEM);

}



static void mmu_tlb_inv_range_nosync(unsigned long iova, size_t size,

				     size_t granule, bool leaf, void *cookie)

{}



static void mmu_tlb_sync_context(void *cookie)

{

	//struct panfrost_device *pfdev = cookie;

	// TODO: Wait 1000 GPU cycles for HW_ISSUE_6367/T60X

}



static const struct iommu_gather_ops mmu_tlb_ops = {

static void mmu_tlb_flush_walk(unsigned long iova, size_t size, size_t granule,

			       void *cookie)

{

	mmu_tlb_sync_context(cookie);

}



static void mmu_tlb_flush_leaf(unsigned long iova, size_t size, size_t granule,

			       void *cookie)

{

	mmu_tlb_sync_context(cookie);

}



static const struct iommu_flush_ops mmu_tlb_ops = {

	.tlb_flush_all	= mmu_tlb_inv_context_s1,

	.tlb_add_flush	= mmu_tlb_inv_range_nosync,

	.tlb_sync	= mmu_tlb_sync_context,

	.tlb_flush_walk = mmu_tlb_flush_walk,

	.tlb_flush_leaf = mmu_tlb_flush_leaf,

};



static const char *access_type_name(struct panfrost_device *pfdev,

drivers/iommu/amd_iommu.c

+6 −5
Original line number Diff line number Diff line
@@ -3055,7 +3055,8 @@ static int amd_iommu_map(struct iommu_domain *dom, unsigned long iova, 
}



static size_t amd_iommu_unmap(struct iommu_domain *dom, unsigned long iova,

			   size_t page_size)

			      size_t page_size,

			      struct iommu_iotlb_gather *gather)

{

	struct protection_domain *domain = to_pdomain(dom);

	size_t unmap_size;
@@ -3196,9 +3197,10 @@ static void amd_iommu_flush_iotlb_all(struct iommu_domain *domain) 
	domain_flush_complete(dom);

}



static void amd_iommu_iotlb_range_add(struct iommu_domain *domain,

				      unsigned long iova, size_t size)

static void amd_iommu_iotlb_sync(struct iommu_domain *domain,

				 struct iommu_iotlb_gather *gather)

{

	amd_iommu_flush_iotlb_all(domain);

}



const struct iommu_ops amd_iommu_ops = {
@@ -3219,8 +3221,7 @@ const struct iommu_ops amd_iommu_ops = { 
	.is_attach_deferred = amd_iommu_is_attach_deferred,

	.pgsize_bitmap	= AMD_IOMMU_PGSIZES,

	.flush_iotlb_all = amd_iommu_flush_iotlb_all,

	.iotlb_range_add = amd_iommu_iotlb_range_add,

	.iotlb_sync = amd_iommu_flush_iotlb_all,

	.iotlb_sync = amd_iommu_iotlb_sync,

};



/*****************************************************************************

drivers/iommu/arm-smmu-v3.c

+736 −237

File changed.

Preview size limit exceeded, changes collapsed.

drivers/iommu/arm-smmu.c

+70 −24
Original line number Diff line number Diff line
@@ -366,7 +366,7 @@ static void arm_smmu_tlb_inv_range_s2(unsigned long iova, size_t size, 
 * On MMU-401 at least, the cost of firing off multiple TLBIVMIDs appears

 * almost negligible, but the benefit of getting the first one in as far ahead

 * of the sync as possible is significant, hence we don't just make this a

 * no-op and set .tlb_sync to arm_smmu_inv_context_s2() as you might think.

 * no-op and set .tlb_sync to arm_smmu_tlb_inv_context_s2() as you might think.

 */

static void arm_smmu_tlb_inv_vmid_nosync(unsigned long iova, size_t size,

					 size_t granule, bool leaf, void *cookie)
@@ -380,21 +380,66 @@ static void arm_smmu_tlb_inv_vmid_nosync(unsigned long iova, size_t size, 
	arm_smmu_gr0_write(smmu, ARM_SMMU_GR0_TLBIVMID, smmu_domain->cfg.vmid);

}



static const struct iommu_gather_ops arm_smmu_s1_tlb_ops = {

static void arm_smmu_tlb_inv_walk(unsigned long iova, size_t size,

				  size_t granule, void *cookie)

{

	struct arm_smmu_domain *smmu_domain = cookie;

	const struct arm_smmu_flush_ops *ops = smmu_domain->flush_ops;



	ops->tlb_inv_range(iova, size, granule, false, cookie);

	ops->tlb_sync(cookie);

}



static void arm_smmu_tlb_inv_leaf(unsigned long iova, size_t size,

				  size_t granule, void *cookie)

{

	struct arm_smmu_domain *smmu_domain = cookie;

	const struct arm_smmu_flush_ops *ops = smmu_domain->flush_ops;



	ops->tlb_inv_range(iova, size, granule, true, cookie);

	ops->tlb_sync(cookie);

}



static void arm_smmu_tlb_add_page(struct iommu_iotlb_gather *gather,

				  unsigned long iova, size_t granule,

				  void *cookie)

{

	struct arm_smmu_domain *smmu_domain = cookie;

	const struct arm_smmu_flush_ops *ops = smmu_domain->flush_ops;



	ops->tlb_inv_range(iova, granule, granule, true, cookie);

}



static const struct arm_smmu_flush_ops arm_smmu_s1_tlb_ops = {

	.tlb = {

		.tlb_flush_all	= arm_smmu_tlb_inv_context_s1,

	.tlb_add_flush	= arm_smmu_tlb_inv_range_s1,

		.tlb_flush_walk	= arm_smmu_tlb_inv_walk,

		.tlb_flush_leaf	= arm_smmu_tlb_inv_leaf,

		.tlb_add_page	= arm_smmu_tlb_add_page,

	},

	.tlb_inv_range		= arm_smmu_tlb_inv_range_s1,

	.tlb_sync		= arm_smmu_tlb_sync_context,

};



static const struct iommu_gather_ops arm_smmu_s2_tlb_ops_v2 = {

static const struct arm_smmu_flush_ops arm_smmu_s2_tlb_ops_v2 = {

	.tlb = {

		.tlb_flush_all	= arm_smmu_tlb_inv_context_s2,

	.tlb_add_flush	= arm_smmu_tlb_inv_range_s2,

		.tlb_flush_walk	= arm_smmu_tlb_inv_walk,

		.tlb_flush_leaf	= arm_smmu_tlb_inv_leaf,

		.tlb_add_page	= arm_smmu_tlb_add_page,

	},

	.tlb_inv_range		= arm_smmu_tlb_inv_range_s2,

	.tlb_sync		= arm_smmu_tlb_sync_context,

};



static const struct iommu_gather_ops arm_smmu_s2_tlb_ops_v1 = {

static const struct arm_smmu_flush_ops arm_smmu_s2_tlb_ops_v1 = {

	.tlb = {

		.tlb_flush_all	= arm_smmu_tlb_inv_context_s2,

	.tlb_add_flush	= arm_smmu_tlb_inv_vmid_nosync,

		.tlb_flush_walk	= arm_smmu_tlb_inv_walk,

		.tlb_flush_leaf	= arm_smmu_tlb_inv_leaf,

		.tlb_add_page	= arm_smmu_tlb_add_page,

	},

	.tlb_inv_range		= arm_smmu_tlb_inv_vmid_nosync,

	.tlb_sync		= arm_smmu_tlb_sync_vmid,

};
@@ -667,7 +712,7 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain, 
			ias = min(ias, 32UL);

			oas = min(oas, 32UL);

		}

		smmu_domain->tlb_ops = &arm_smmu_s1_tlb_ops;

		smmu_domain->flush_ops = &arm_smmu_s1_tlb_ops;

		break;

	case ARM_SMMU_DOMAIN_NESTED:

		/*
@@ -687,9 +732,9 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain, 
			oas = min(oas, 40UL);

		}

		if (smmu->version == ARM_SMMU_V2)

			smmu_domain->tlb_ops = &arm_smmu_s2_tlb_ops_v2;

			smmu_domain->flush_ops = &arm_smmu_s2_tlb_ops_v2;

		else

			smmu_domain->tlb_ops = &arm_smmu_s2_tlb_ops_v1;

			smmu_domain->flush_ops = &arm_smmu_s2_tlb_ops_v1;

		break;

	default:

		ret = -EINVAL;
@@ -725,7 +770,7 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain, 
		.ias		= ias,

		.oas		= oas,

		.coherent_walk	= smmu->features & ARM_SMMU_FEAT_COHERENT_WALK,

		.tlb		= smmu_domain->tlb_ops,

		.tlb		= &smmu_domain->flush_ops->tlb,

		.iommu_dev	= smmu->dev,

	};
@@ -1131,7 +1176,7 @@ static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova, 
}



static size_t arm_smmu_unmap(struct iommu_domain *domain, unsigned long iova,

			     size_t size)

			     size_t size, struct iommu_iotlb_gather *gather)

{

	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;

	struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu;
@@ -1141,7 +1186,7 @@ static size_t arm_smmu_unmap(struct iommu_domain *domain, unsigned long iova, 
		return 0;



	arm_smmu_rpm_get(smmu);

	ret = ops->unmap(ops, iova, size);

	ret = ops->unmap(ops, iova, size, gather);

	arm_smmu_rpm_put(smmu);



	return ret;
@@ -1152,21 +1197,22 @@ static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain) 
	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);

	struct arm_smmu_device *smmu = smmu_domain->smmu;



	if (smmu_domain->tlb_ops) {

	if (smmu_domain->flush_ops) {

		arm_smmu_rpm_get(smmu);

		smmu_domain->tlb_ops->tlb_flush_all(smmu_domain);

		smmu_domain->flush_ops->tlb.tlb_flush_all(smmu_domain);

		arm_smmu_rpm_put(smmu);

	}

}



static void arm_smmu_iotlb_sync(struct iommu_domain *domain)

static void arm_smmu_iotlb_sync(struct iommu_domain *domain,

				struct iommu_iotlb_gather *gather)

{

	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);

	struct arm_smmu_device *smmu = smmu_domain->smmu;



	if (smmu_domain->tlb_ops) {

	if (smmu_domain->flush_ops) {

		arm_smmu_rpm_get(smmu);

		smmu_domain->tlb_ops->tlb_sync(smmu_domain);

		smmu_domain->flush_ops->tlb_sync(smmu_domain);

		arm_smmu_rpm_put(smmu);

	}

}

drivers/iommu/arm-smmu.h

+8 −1
Original line number Diff line number Diff line
@@ -304,10 +304,17 @@ enum arm_smmu_domain_stage { 
	ARM_SMMU_DOMAIN_BYPASS,

};



struct arm_smmu_flush_ops {

	struct iommu_flush_ops		tlb;

	void (*tlb_inv_range)(unsigned long iova, size_t size, size_t granule,

			      bool leaf, void *cookie);

	void (*tlb_sync)(void *cookie);

};



struct arm_smmu_domain {

	struct arm_smmu_device		*smmu;

	struct io_pgtable_ops		*pgtbl_ops;

	const struct iommu_gather_ops	*tlb_ops;

	const struct arm_smmu_flush_ops	*flush_ops;

	struct arm_smmu_cfg		cfg;

	enum arm_smmu_domain_stage	stage;

	bool				non_strict;

CRA Git | Maintained and supported by SUSTech CRA and CCSE