[PATCH v3 00/13] Add PASID support to SMMUv3 unmanaged domains

[PATCH v3 00/13] Add PASID support to SMMUv3 unmanaged domains

[Michael Shavit]

Hi all,

This patch series implements the set_dev_pasid operation for DMA
and UNMANAGED iommu domains.

The bulk of the series involves a refactor of stage 1 domains so that
they describe a single CD entry. On attach, stage 1 domains are inserted
into a CD table that is now owned by the arm_smmu_master struct. This is
a pre-requisite to the set_dev_pasid implementation but also results in
a conceptually cleaner arm_smmu_domain. Note that this does not preclude
from attaching domains that represent a CD table, such as for the
proposed iommufd NESTED domains.

The last few patches of the series make drive-by cleanups to the smmu
SVA implementation. A follow-up patch-series is planned so that the SVA
implementation can further take advantages of the changes introduced here. See
discussion on patch 14 of the v2 series.

This patch series is also available on gerrit with Jean's SMMU test
engine patches cherry-picked on top:
https://linux-review.googlesource.com/id/I0fcd9adc058d1c58a12d2599cc82fba73da7697a
This allowed testing of basic SVA functionality (e.g.: attaching, page
fault handling, and detaching).

Thanks,
Michael Shavit

Changelog
v3:
 * Dropped the bulk of the SVA refactoring to re-work as a follow-up
   series.
 * Reworded cover letter to omit dropped changes.
 * Rebased on 6.4 tip
v2:
https://lore.kernel.org/all/20230606120854.4170244-1-mshavit@google.com/
 * Reworded cover letter and commits based on v1 feedback.
 * Split and reworked `iommu/arm-smmu-v3: Move cdtable to arm_smmu_master`
 * Added SVA clean-up and refactor.
 * A few other small bug fixes and cosmetics.
v1:
https://lore.kernel.org/all/20230510205054.2667898-1-mshavit@google.com/


Michael Shavit (13):
  iommu/arm-smmu-v3: Move ctx_desc out of s1_cfg
  iommu/arm-smmu-v3: Add smmu_s1_cfg to smmu_master
  iommu/arm-smmu-v3: Refactor write_strtab_ent
  iommu/arm-smmu-v3: Refactor write_ctx_desc
  iommu/arm-smmu-v3: Use the master-owned s1_cfg
  iommu/arm-smmu-v3: Simplify arm_smmu_enable_ats
  iommu/arm-smmu-v3: Keep track of attached ssids
  iommu/arm-smmu-v3: Add helper for atc invalidation
  iommu/arm-smmu-v3: Implement set_dev_pasid
  iommu/arm-smmu-v3-sva: Remove bond refcount
  iommu/arm-smmu-v3-sva: Clean unused iommu_sva
  iommu/arm-smmu-v3-sva: Remove arm_smmu_bond
  iommu/arm-smmu-v3-sva: Add check when enabling sva

 .../iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c   | 156 +++---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c   | 447 ++++++++++++------
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h   |  49 +-
 3 files changed, 414 insertions(+), 238 deletions(-)


base-commit: b6dad5178ceaf23f369c3711062ce1f2afc33644
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 01/13] iommu/arm-smmu-v3: Move ctx_desc out of s1_cfg

[Michael Shavit]

s1_cfg describes the CD table that is inserted into an SMMU's STEs. It's
weird for s1_cfg to also own ctx_desc which describes a CD that is
inserted into that table. It is more appropriate for arm_smmu_domain to
own ctx_desc.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
 .../iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c   |  2 +-
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c   | 23 +++++++--------
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h   | 28 ++++++++++---------
 3 files changed, 28 insertions(+), 25 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
index a5a63b1c947eb..968559d625c40 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
@@ -62,7 +62,7 @@ arm_smmu_share_asid(struct mm_struct *mm, u16 asid)
 		return cd;
 	}
 
-	smmu_domain = container_of(cd, struct arm_smmu_domain, s1_cfg.cd);
+	smmu_domain = container_of(cd, struct arm_smmu_domain, cd);
 	smmu = smmu_domain->smmu;
 
 	ret = xa_alloc(&arm_smmu_asid_xa, &new_asid, cd,
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index 3fd83fb757227..beff04b897718 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -1863,7 +1863,7 @@ static void arm_smmu_tlb_inv_context(void *cookie)
 	 * careful, 007.
 	 */
 	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
-		arm_smmu_tlb_inv_asid(smmu, smmu_domain->s1_cfg.cd.asid);
+		arm_smmu_tlb_inv_asid(smmu, smmu_domain->cd.asid);
 	} else {
 		cmd.opcode	= CMDQ_OP_TLBI_S12_VMALL;
 		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
@@ -1946,7 +1946,7 @@ static void arm_smmu_tlb_inv_range_domain(unsigned long iova, size_t size,
 	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
 		cmd.opcode	= smmu_domain->smmu->features & ARM_SMMU_FEAT_E2H ?
 				  CMDQ_OP_TLBI_EL2_VA : CMDQ_OP_TLBI_NH_VA;
-		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
+		cmd.tlbi.asid	= smmu_domain->cd.asid;
 	} else {
 		cmd.opcode	= CMDQ_OP_TLBI_S2_IPA;
 		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
@@ -2077,7 +2077,7 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
 		mutex_lock(&arm_smmu_asid_lock);
 		if (cfg->cdcfg.cdtab)
 			arm_smmu_free_cd_tables(smmu_domain);
-		arm_smmu_free_asid(&cfg->cd);
+		arm_smmu_free_asid(&smmu_domain->cd);
 		mutex_unlock(&arm_smmu_asid_lock);
 	} else {
 		struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
@@ -2096,13 +2096,14 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
 	u32 asid;
 	struct arm_smmu_device *smmu = smmu_domain->smmu;
 	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
+	struct arm_smmu_ctx_desc *cd = &smmu_domain->cd;
 	typeof(&pgtbl_cfg->arm_lpae_s1_cfg.tcr) tcr = &pgtbl_cfg->arm_lpae_s1_cfg.tcr;
 
-	refcount_set(&cfg->cd.refs, 1);
+	refcount_set(&cd->refs, 1);
 
 	/* Prevent SVA from modifying the ASID until it is written to the CD */
 	mutex_lock(&arm_smmu_asid_lock);
-	ret = xa_alloc(&arm_smmu_asid_xa, &asid, &cfg->cd,
+	ret = xa_alloc(&arm_smmu_asid_xa, &asid, cd,
 		       XA_LIMIT(1, (1 << smmu->asid_bits) - 1), GFP_KERNEL);
 	if (ret)
 		goto out_unlock;
@@ -2115,23 +2116,23 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
 	if (ret)
 		goto out_free_asid;
 
-	cfg->cd.asid	= (u16)asid;
-	cfg->cd.ttbr	= pgtbl_cfg->arm_lpae_s1_cfg.ttbr;
-	cfg->cd.tcr	= FIELD_PREP(CTXDESC_CD_0_TCR_T0SZ, tcr->tsz) |
+	cd->asid	= (u16)asid;
+	cd->ttbr	= pgtbl_cfg->arm_lpae_s1_cfg.ttbr;
+	cd->tcr		= FIELD_PREP(CTXDESC_CD_0_TCR_T0SZ, tcr->tsz) |
 			  FIELD_PREP(CTXDESC_CD_0_TCR_TG0, tcr->tg) |
 			  FIELD_PREP(CTXDESC_CD_0_TCR_IRGN0, tcr->irgn) |
 			  FIELD_PREP(CTXDESC_CD_0_TCR_ORGN0, tcr->orgn) |
 			  FIELD_PREP(CTXDESC_CD_0_TCR_SH0, tcr->sh) |
 			  FIELD_PREP(CTXDESC_CD_0_TCR_IPS, tcr->ips) |
 			  CTXDESC_CD_0_TCR_EPD1 | CTXDESC_CD_0_AA64;
-	cfg->cd.mair	= pgtbl_cfg->arm_lpae_s1_cfg.mair;
+	cd->mair	= pgtbl_cfg->arm_lpae_s1_cfg.mair;
 
 	/*
 	 * Note that this will end up calling arm_smmu_sync_cd() before
 	 * the master has been added to the devices list for this domain.
 	 * This isn't an issue because the STE hasn't been installed yet.
 	 */
-	ret = arm_smmu_write_ctx_desc(smmu_domain, 0, &cfg->cd);
+	ret = arm_smmu_write_ctx_desc(smmu_domain, 0, cd);
 	if (ret)
 		goto out_free_cd_tables;
 
@@ -2141,7 +2142,7 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
 out_free_cd_tables:
 	arm_smmu_free_cd_tables(smmu_domain);
 out_free_asid:
-	arm_smmu_free_asid(&cfg->cd);
+	arm_smmu_free_asid(cd);
 out_unlock:
 	mutex_unlock(&arm_smmu_asid_lock);
 	return ret;
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
index b574c58a34876..68d519f21dbd8 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
@@ -593,7 +593,6 @@ struct arm_smmu_ctx_desc_cfg {
 
 struct arm_smmu_s1_cfg {
 	struct arm_smmu_ctx_desc_cfg	cdcfg;
-	struct arm_smmu_ctx_desc	cd;
 	u8				s1fmt;
 	u8				s1cdmax;
 };
@@ -707,25 +706,28 @@ enum arm_smmu_domain_stage {
 };
 
 struct arm_smmu_domain {
-	struct arm_smmu_device		*smmu;
-	struct mutex			init_mutex; /* Protects smmu pointer */
+	struct arm_smmu_device			*smmu;
+	struct mutex				init_mutex; /* Protects smmu pointer */
 
-	struct io_pgtable_ops		*pgtbl_ops;
-	bool				stall_enabled;
-	atomic_t			nr_ats_masters;
+	struct io_pgtable_ops			*pgtbl_ops;
+	bool					stall_enabled;
+	atomic_t				nr_ats_masters;
 
-	enum arm_smmu_domain_stage	stage;
+	enum arm_smmu_domain_stage		stage;
 	union {
-		struct arm_smmu_s1_cfg	s1_cfg;
-		struct arm_smmu_s2_cfg	s2_cfg;
+		struct {
+		struct arm_smmu_ctx_desc	cd;
+		struct arm_smmu_s1_cfg		s1_cfg;
+		};
+		struct arm_smmu_s2_cfg		s2_cfg;
 	};
 
-	struct iommu_domain		domain;
+	struct iommu_domain			domain;
 
-	struct list_head		devices;
-	spinlock_t			devices_lock;
+	struct list_head			devices;
+	spinlock_t				devices_lock;
 
-	struct list_head		mmu_notifiers;
+	struct list_head			mmu_notifiers;
 };
 
 static inline struct arm_smmu_domain *to_smmu_domain(struct iommu_domain *dom)
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 02/13] iommu/arm-smmu-v3: Add smmu_s1_cfg to smmu_master

[Michael Shavit]

Except for Nested domains, arm_smmu_master will own the STEs that are
inserted into the arm_smmu_device's STE table.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 28 +++++++++++++--------
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h |  1 +
 2 files changed, 18 insertions(+), 11 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index beff04b897718..023769f5ca79a 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -1126,15 +1126,16 @@ int arm_smmu_write_ctx_desc(struct arm_smmu_domain *smmu_domain, int ssid,
 	return 0;
 }
 
-static int arm_smmu_alloc_cd_tables(struct arm_smmu_domain *smmu_domain)
+static int arm_smmu_init_s1_cfg(struct arm_smmu_master *master,
+				struct arm_smmu_s1_cfg *cfg)
 {
 	int ret;
 	size_t l1size;
 	size_t max_contexts;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
+	struct arm_smmu_device *smmu = master->smmu;
 	struct arm_smmu_ctx_desc_cfg *cdcfg = &cfg->cdcfg;
 
+	cfg->s1cdmax = master->ssid_bits;
 	max_contexts = 1 << cfg->s1cdmax;
 
 	if (!(smmu->features & ARM_SMMU_FEAT_2_LVL_CDTAB) ||
@@ -1175,12 +1176,11 @@ static int arm_smmu_alloc_cd_tables(struct arm_smmu_domain *smmu_domain)
 	return ret;
 }
 
-static void arm_smmu_free_cd_tables(struct arm_smmu_domain *smmu_domain)
+static void arm_smmu_free_cd_tables(struct arm_smmu_device *smmu,
+				    struct arm_smmu_ctx_desc_cfg *cdcfg)
 {
 	int i;
 	size_t size, l1size;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-	struct arm_smmu_ctx_desc_cfg *cdcfg = &smmu_domain->s1_cfg.cdcfg;
 
 	if (cdcfg->l1_desc) {
 		size = CTXDESC_L2_ENTRIES * (CTXDESC_CD_DWORDS << 3);
@@ -2076,7 +2076,7 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
 		/* Prevent SVA from touching the CD while we're freeing it */
 		mutex_lock(&arm_smmu_asid_lock);
 		if (cfg->cdcfg.cdtab)
-			arm_smmu_free_cd_tables(smmu_domain);
+			arm_smmu_free_cd_tables(smmu, &cfg->cdcfg);
 		arm_smmu_free_asid(&smmu_domain->cd);
 		mutex_unlock(&arm_smmu_asid_lock);
 	} else {
@@ -2108,11 +2108,9 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
 	if (ret)
 		goto out_unlock;
 
-	cfg->s1cdmax = master->ssid_bits;
-
 	smmu_domain->stall_enabled = master->stall_enabled;
 
-	ret = arm_smmu_alloc_cd_tables(smmu_domain);
+	ret = arm_smmu_init_s1_cfg(master, cfg);
 	if (ret)
 		goto out_free_asid;
 
@@ -2140,7 +2138,7 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
 	return 0;
 
 out_free_cd_tables:
-	arm_smmu_free_cd_tables(smmu_domain);
+	arm_smmu_free_cd_tables(smmu, &cfg->cdcfg);
 out_free_asid:
 	arm_smmu_free_asid(cd);
 out_unlock:
@@ -2704,6 +2702,13 @@ static struct iommu_device *arm_smmu_probe_device(struct device *dev)
 	    smmu->features & ARM_SMMU_FEAT_STALL_FORCE)
 		master->stall_enabled = true;
 
+	ret = arm_smmu_init_s1_cfg(master, &master->owned_s1_cfg);
+	if (ret) {
+		arm_smmu_disable_pasid(master);
+		arm_smmu_remove_master(master);
+		goto err_free_master;
+	}
+
 	return &smmu->iommu;
 
 err_free_master:
@@ -2719,6 +2724,7 @@ static void arm_smmu_release_device(struct device *dev)
 	if (WARN_ON(arm_smmu_master_sva_enabled(master)))
 		iopf_queue_remove_device(master->smmu->evtq.iopf, dev);
 	arm_smmu_detach_dev(master);
+	arm_smmu_free_cd_tables(master->smmu, &master->owned_s1_cfg.cdcfg);
 	arm_smmu_disable_pasid(master);
 	arm_smmu_remove_master(master);
 	kfree(master);
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
index 68d519f21dbd8..053cc14c23969 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
@@ -688,6 +688,7 @@ struct arm_smmu_master {
 	struct arm_smmu_domain		*domain;
 	struct list_head		domain_head;
 	struct arm_smmu_stream		*streams;
+	struct arm_smmu_s1_cfg		owned_s1_cfg;
 	unsigned int			num_streams;
 	bool				ats_enabled;
 	bool				stall_enabled;
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 03/13] iommu/arm-smmu-v3: Refactor write_strtab_ent

[Michael Shavit]

Explicity keep track of the s1_cfg and s2_cfg that are attached to a
master in arm_smmu_master, regardless of whether they are owned by
arm_smmu_master, arm_smmu_domain or userspace.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 37 +++++++++------------
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h |  2 ++
 2 files changed, 17 insertions(+), 22 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index 023769f5ca79a..d79c6ef5d6ed4 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -1269,10 +1269,9 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
 	 */
 	u64 val = le64_to_cpu(dst[0]);
 	bool ste_live = false;
-	struct arm_smmu_device *smmu = NULL;
+	struct arm_smmu_device *smmu = master->smmu;
 	struct arm_smmu_s1_cfg *s1_cfg = NULL;
 	struct arm_smmu_s2_cfg *s2_cfg = NULL;
-	struct arm_smmu_domain *smmu_domain = NULL;
 	struct arm_smmu_cmdq_ent prefetch_cmd = {
 		.opcode		= CMDQ_OP_PREFETCH_CFG,
 		.prefetch	= {
@@ -1280,24 +1279,10 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
 		},
 	};
 
-	if (master) {
-		smmu_domain = master->domain;
-		smmu = master->smmu;
-	}
-
-	if (smmu_domain) {
-		switch (smmu_domain->stage) {
-		case ARM_SMMU_DOMAIN_S1:
-			s1_cfg = &smmu_domain->s1_cfg;
-			break;
-		case ARM_SMMU_DOMAIN_S2:
-		case ARM_SMMU_DOMAIN_NESTED:
-			s2_cfg = &smmu_domain->s2_cfg;
-			break;
-		default:
-			break;
-		}
-	}
+	if (master->s1_cfg)
+		s1_cfg = master->s1_cfg;
+	else if (master->s2_cfg)
+		s2_cfg = master->s2_cfg;
 
 	if (val & STRTAB_STE_0_V) {
 		switch (FIELD_GET(STRTAB_STE_0_CFG, val)) {
@@ -1319,8 +1304,8 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
 	val = STRTAB_STE_0_V;
 
 	/* Bypass/fault */
-	if (!smmu_domain || !(s1_cfg || s2_cfg)) {
-		if (!smmu_domain && disable_bypass)
+	if (!(s1_cfg || s2_cfg)) {
+		if (disable_bypass)
 			val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_ABORT);
 		else
 			val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_BYPASS);
@@ -2401,6 +2386,8 @@ static void arm_smmu_detach_dev(struct arm_smmu_master *master)
 
 	master->domain = NULL;
 	master->ats_enabled = false;
+	master->s1_cfg = NULL;
+	master->s2_cfg = NULL;
 	arm_smmu_install_ste_for_dev(master);
 }
 
@@ -2454,6 +2441,12 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 	}
 
 	master->domain = smmu_domain;
+	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
+		master->s1_cfg = &smmu_domain->s1_cfg;
+	} else if (smmu_domain->stage == ARM_SMMU_DOMAIN_S2 ||
+		   smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED) {
+		master->s2_cfg = &smmu_domain->s2_cfg;
+	}
 
 	/*
 	 * The SMMU does not support enabling ATS with bypass. When the STE is
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
index 053cc14c23969..3c614fbe2b8b9 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
@@ -689,6 +689,8 @@ struct arm_smmu_master {
 	struct list_head		domain_head;
 	struct arm_smmu_stream		*streams;
 	struct arm_smmu_s1_cfg		owned_s1_cfg;
+	struct arm_smmu_s1_cfg		*s1_cfg;
+	struct arm_smmu_s2_cfg		*s2_cfg;
 	unsigned int			num_streams;
 	bool				ats_enabled;
 	bool				stall_enabled;
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 04/13] iommu/arm-smmu-v3: Refactor write_ctx_desc

[Michael Shavit]

Update arm_smmu_write_ctx_desc and downstream functions to be agnostic
of the CD table's ownership. Note that in practice, it will only be
called in cases where the table is owned by arm_smmu_master. Whether
arm_smmu_write_ctx_desc will trigger a sync is also made part of the
API.

Note that this change isn't a nop refactor since SVA will call
arm_smmu_write_ctx_desc in a loop for every master the domain is
attached to despite the fact that they all share the same CD table. Note
that the next commit ceases this sharing of the s1_cfg which makes that
loop necessary.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
 .../iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c   | 36 ++++++++-
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c   | 80 +++++++++++--------
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h   |  6 +-
 3 files changed, 81 insertions(+), 41 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
index 968559d625c40..48fa8eb271a45 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
@@ -45,10 +45,12 @@ static struct arm_smmu_ctx_desc *
 arm_smmu_share_asid(struct mm_struct *mm, u16 asid)
 {
 	int ret;
+	unsigned long flags;
 	u32 new_asid;
 	struct arm_smmu_ctx_desc *cd;
 	struct arm_smmu_device *smmu;
 	struct arm_smmu_domain *smmu_domain;
+	struct arm_smmu_master *master;
 
 	cd = xa_load(&arm_smmu_asid_xa, asid);
 	if (!cd)
@@ -80,7 +82,11 @@ arm_smmu_share_asid(struct mm_struct *mm, u16 asid)
 	 * be some overlap between use of both ASIDs, until we invalidate the
 	 * TLB.
 	 */
-	arm_smmu_write_ctx_desc(smmu_domain, 0, cd);
+	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
+	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
+		arm_smmu_write_ctx_desc(smmu, master->s1_cfg, master, 0, cd);
+	}
+	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
 
 	/* Invalidate TLB entries previously associated with that context */
 	arm_smmu_tlb_inv_asid(smmu, asid);
@@ -211,6 +217,8 @@ static void arm_smmu_mm_release(struct mmu_notifier *mn, struct mm_struct *mm)
 {
 	struct arm_smmu_mmu_notifier *smmu_mn = mn_to_smmu(mn);
 	struct arm_smmu_domain *smmu_domain = smmu_mn->domain;
+	struct arm_smmu_master *master;
+	unsigned long flags;
 
 	mutex_lock(&sva_lock);
 	if (smmu_mn->cleared) {
@@ -222,7 +230,12 @@ static void arm_smmu_mm_release(struct mmu_notifier *mn, struct mm_struct *mm)
 	 * DMA may still be running. Keep the cd valid to avoid C_BAD_CD events,
 	 * but disable translation.
 	 */
-	arm_smmu_write_ctx_desc(smmu_domain, mm->pasid, &quiet_cd);
+	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
+	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
+		arm_smmu_write_ctx_desc(master->smmu, master->s1_cfg, master,
+					mm->pasid, &quiet_cd);
+	}
+	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
 
 	arm_smmu_tlb_inv_asid(smmu_domain->smmu, smmu_mn->cd->asid);
 	arm_smmu_atc_inv_domain(smmu_domain, mm->pasid, 0, 0);
@@ -248,8 +261,10 @@ arm_smmu_mmu_notifier_get(struct arm_smmu_domain *smmu_domain,
 			  struct mm_struct *mm)
 {
 	int ret;
+	unsigned long flags;
 	struct arm_smmu_ctx_desc *cd;
 	struct arm_smmu_mmu_notifier *smmu_mn;
+	struct arm_smmu_master *master;
 
 	list_for_each_entry(smmu_mn, &smmu_domain->mmu_notifiers, list) {
 		if (smmu_mn->mn.mm == mm) {
@@ -279,7 +294,12 @@ arm_smmu_mmu_notifier_get(struct arm_smmu_domain *smmu_domain,
 		goto err_free_cd;
 	}
 
-	ret = arm_smmu_write_ctx_desc(smmu_domain, mm->pasid, cd);
+	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
+	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
+		ret = arm_smmu_write_ctx_desc(master->smmu, master->s1_cfg,
+					      master, mm->pasid, cd);
+	}
+	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
 	if (ret)
 		goto err_put_notifier;
 
@@ -296,15 +316,23 @@ arm_smmu_mmu_notifier_get(struct arm_smmu_domain *smmu_domain,
 
 static void arm_smmu_mmu_notifier_put(struct arm_smmu_mmu_notifier *smmu_mn)
 {
+	unsigned long flags;
 	struct mm_struct *mm = smmu_mn->mn.mm;
 	struct arm_smmu_ctx_desc *cd = smmu_mn->cd;
+	struct arm_smmu_master *master;
 	struct arm_smmu_domain *smmu_domain = smmu_mn->domain;
 
 	if (!refcount_dec_and_test(&smmu_mn->refs))
 		return;
 
 	list_del(&smmu_mn->list);
-	arm_smmu_write_ctx_desc(smmu_domain, mm->pasid, NULL);
+
+	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
+	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
+		arm_smmu_write_ctx_desc(master->smmu, master->s1_cfg, master,
+					mm->pasid, NULL);
+	}
+	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
 
 	/*
 	 * If we went through clear(), we've already invalidated, and no
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index d79c6ef5d6ed4..b6f7cf60f8f3d 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -965,14 +965,13 @@ void arm_smmu_tlb_inv_asid(struct arm_smmu_device *smmu, u16 asid)
 	arm_smmu_cmdq_issue_cmd_with_sync(smmu, &cmd);
 }
 
-static void arm_smmu_sync_cd(struct arm_smmu_domain *smmu_domain,
+/* master may be null */
+static void arm_smmu_sync_cd(struct arm_smmu_master *master,
 			     int ssid, bool leaf)
 {
 	size_t i;
-	unsigned long flags;
-	struct arm_smmu_master *master;
 	struct arm_smmu_cmdq_batch cmds;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_device *smmu;
 	struct arm_smmu_cmdq_ent cmd = {
 		.opcode	= CMDQ_OP_CFGI_CD,
 		.cfgi	= {
@@ -981,16 +980,15 @@ static void arm_smmu_sync_cd(struct arm_smmu_domain *smmu_domain,
 		},
 	};
 
-	cmds.num = 0;
+	if (!master)
+		return;
 
-	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
-	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
-		for (i = 0; i < master->num_streams; i++) {
-			cmd.cfgi.sid = master->streams[i].id;
-			arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
-		}
+	smmu = master->smmu;
+	cmds.num = 0;
+	for (i = 0; i < master->num_streams; i++) {
+		cmd.cfgi.sid = master->streams[i].id;
+		arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
 	}
-	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
 
 	arm_smmu_cmdq_batch_submit(smmu, &cmds);
 }
@@ -1020,16 +1018,18 @@ static void arm_smmu_write_cd_l1_desc(__le64 *dst,
 	WRITE_ONCE(*dst, cpu_to_le64(val));
 }
 
-static __le64 *arm_smmu_get_cd_ptr(struct arm_smmu_domain *smmu_domain,
+/* master may be null */
+static __le64 *arm_smmu_get_cd_ptr(struct arm_smmu_device *smmu,
+				   struct arm_smmu_s1_cfg *s1_cfg,
+				   struct arm_smmu_master *master,
 				   u32 ssid)
 {
 	__le64 *l1ptr;
 	unsigned int idx;
 	struct arm_smmu_l1_ctx_desc *l1_desc;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-	struct arm_smmu_ctx_desc_cfg *cdcfg = &smmu_domain->s1_cfg.cdcfg;
+	struct arm_smmu_ctx_desc_cfg *cdcfg = &s1_cfg->cdcfg;
 
-	if (smmu_domain->s1_cfg.s1fmt == STRTAB_STE_0_S1FMT_LINEAR)
+	if (s1_cfg->s1fmt == STRTAB_STE_0_S1FMT_LINEAR)
 		return cdcfg->cdtab + ssid * CTXDESC_CD_DWORDS;
 
 	idx = ssid >> CTXDESC_SPLIT;
@@ -1041,13 +1041,21 @@ static __le64 *arm_smmu_get_cd_ptr(struct arm_smmu_domain *smmu_domain,
 		l1ptr = cdcfg->cdtab + idx * CTXDESC_L1_DESC_DWORDS;
 		arm_smmu_write_cd_l1_desc(l1ptr, l1_desc);
 		/* An invalid L1CD can be cached */
-		arm_smmu_sync_cd(smmu_domain, ssid, false);
+		arm_smmu_sync_cd(master, ssid, false);
 	}
 	idx = ssid & (CTXDESC_L2_ENTRIES - 1);
 	return l1_desc->l2ptr + idx * CTXDESC_CD_DWORDS;
 }
 
-int arm_smmu_write_ctx_desc(struct arm_smmu_domain *smmu_domain, int ssid,
+/*
+ * master must be provided if a CD sync is required but may be null otherwise
+ * (such as when the CD table isn't inserted into the STE yet, or is about to
+ * be detached.
+ */
+int arm_smmu_write_ctx_desc(struct arm_smmu_device *smmu,
+			    struct arm_smmu_s1_cfg *s1_cfg,
+			    struct arm_smmu_master *master,
+			    int ssid,
 			    struct arm_smmu_ctx_desc *cd)
 {
 	/*
@@ -1065,10 +1073,10 @@ int arm_smmu_write_ctx_desc(struct arm_smmu_domain *smmu_domain, int ssid,
 	bool cd_live;
 	__le64 *cdptr;
 
-	if (WARN_ON(ssid >= (1 << smmu_domain->s1_cfg.s1cdmax)))
+	if (WARN_ON(ssid >= (1 << s1_cfg->s1cdmax)))
 		return -E2BIG;
 
-	cdptr = arm_smmu_get_cd_ptr(smmu_domain, ssid);
+	cdptr = arm_smmu_get_cd_ptr(smmu, s1_cfg, master, ssid);
 	if (!cdptr)
 		return -ENOMEM;
 
@@ -1092,11 +1100,11 @@ int arm_smmu_write_ctx_desc(struct arm_smmu_domain *smmu_domain, int ssid,
 		cdptr[3] = cpu_to_le64(cd->mair);
 
 		/*
-		 * STE is live, and the SMMU might read dwords of this CD in any
-		 * order. Ensure that it observes valid values before reading
-		 * V=1.
+		 * STE may be live, and the SMMU might read dwords of this CD
+		 * in any order. Ensure that it observes valid values before
+		 * reading V=1.
 		 */
-		arm_smmu_sync_cd(smmu_domain, ssid, true);
+		arm_smmu_sync_cd(master, ssid, true);
 
 		val = cd->tcr |
 #ifdef __BIG_ENDIAN
@@ -1108,7 +1116,7 @@ int arm_smmu_write_ctx_desc(struct arm_smmu_domain *smmu_domain, int ssid,
 			FIELD_PREP(CTXDESC_CD_0_ASID, cd->asid) |
 			CTXDESC_CD_0_V;
 
-		if (smmu_domain->stall_enabled)
+		if (s1_cfg->stall_enabled)
 			val |= CTXDESC_CD_0_S;
 	}
 
@@ -1122,7 +1130,7 @@ int arm_smmu_write_ctx_desc(struct arm_smmu_domain *smmu_domain, int ssid,
 	 *   without first making the structure invalid.
 	 */
 	WRITE_ONCE(cdptr[0], cpu_to_le64(val));
-	arm_smmu_sync_cd(smmu_domain, ssid, true);
+	arm_smmu_sync_cd(master, ssid, true);
 	return 0;
 }
 
@@ -1136,6 +1144,7 @@ static int arm_smmu_init_s1_cfg(struct arm_smmu_master *master,
 	struct arm_smmu_ctx_desc_cfg *cdcfg = &cfg->cdcfg;
 
 	cfg->s1cdmax = master->ssid_bits;
+	cfg->stall_enabled = master->stall_enabled;
 	max_contexts = 1 << cfg->s1cdmax;
 
 	if (!(smmu->features & ARM_SMMU_FEAT_2_LVL_CDTAB) ||
@@ -2093,8 +2102,6 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
 	if (ret)
 		goto out_unlock;
 
-	smmu_domain->stall_enabled = master->stall_enabled;
-
 	ret = arm_smmu_init_s1_cfg(master, cfg);
 	if (ret)
 		goto out_free_asid;
@@ -2110,12 +2117,9 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
 			  CTXDESC_CD_0_TCR_EPD1 | CTXDESC_CD_0_AA64;
 	cd->mair	= pgtbl_cfg->arm_lpae_s1_cfg.mair;
 
-	/*
-	 * Note that this will end up calling arm_smmu_sync_cd() before
-	 * the master has been added to the devices list for this domain.
-	 * This isn't an issue because the STE hasn't been installed yet.
-	 */
-	ret = arm_smmu_write_ctx_desc(smmu_domain, 0, cd);
+	ret = arm_smmu_write_ctx_desc(smmu, cfg,
+				      NULL /*Not attached to a master yet */,
+				      0, cd);
 	if (ret)
 		goto out_free_cd_tables;
 
@@ -2386,6 +2390,11 @@ static void arm_smmu_detach_dev(struct arm_smmu_master *master)
 
 	master->domain = NULL;
 	master->ats_enabled = false;
+	if (master->s1_cfg)
+		arm_smmu_write_ctx_desc(
+			master->smmu, master->s1_cfg,
+			NULL /* Skip sync since we detach the CD table next*/,
+			0, NULL);
 	master->s1_cfg = NULL;
 	master->s2_cfg = NULL;
 	arm_smmu_install_ste_for_dev(master);
@@ -2435,7 +2444,8 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 		ret = -EINVAL;
 		goto out_unlock;
 	} else if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1 &&
-		   smmu_domain->stall_enabled != master->stall_enabled) {
+		   smmu_domain->s1_cfg.stall_enabled !=
+			   master->stall_enabled) {
 		ret = -EINVAL;
 		goto out_unlock;
 	}
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
index 3c614fbe2b8b9..00a493442d6f9 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
@@ -595,6 +595,7 @@ struct arm_smmu_s1_cfg {
 	struct arm_smmu_ctx_desc_cfg	cdcfg;
 	u8				s1fmt;
 	u8				s1cdmax;
+	bool				stall_enabled;
 };
 
 struct arm_smmu_s2_cfg {
@@ -713,7 +714,6 @@ struct arm_smmu_domain {
 	struct mutex				init_mutex; /* Protects smmu pointer */
 
 	struct io_pgtable_ops			*pgtbl_ops;
-	bool					stall_enabled;
 	atomic_t				nr_ats_masters;
 
 	enum arm_smmu_domain_stage		stage;
@@ -742,7 +742,9 @@ extern struct xarray arm_smmu_asid_xa;
 extern struct mutex arm_smmu_asid_lock;
 extern struct arm_smmu_ctx_desc quiet_cd;
 
-int arm_smmu_write_ctx_desc(struct arm_smmu_domain *smmu_domain, int ssid,
+int arm_smmu_write_ctx_desc(struct arm_smmu_device *smmu,
+			    struct arm_smmu_s1_cfg *s1_cfg,
+			    struct arm_smmu_master *smmu_master, int ssid,
 			    struct arm_smmu_ctx_desc *cd);
 void arm_smmu_tlb_inv_asid(struct arm_smmu_device *smmu, u16 asid);
 void arm_smmu_tlb_inv_range_asid(unsigned long iova, size_t size, int asid,
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 05/13] iommu/arm-smmu-v3: Use the master-owned s1_cfg

[Michael Shavit]

Insert CDs for STAGE_1 domains into a CD table owned by the
arm_smmu_master. Remove the CD table that was owned by arm_smmu_domain.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 43 ++++++---------------
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h |  3 --
 2 files changed, 12 insertions(+), 34 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index b6f7cf60f8f3d..08f440fe1da6d 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -2065,12 +2065,8 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
 
 	/* Free the CD and ASID, if we allocated them */
 	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
-		struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
-
 		/* Prevent SVA from touching the CD while we're freeing it */
 		mutex_lock(&arm_smmu_asid_lock);
-		if (cfg->cdcfg.cdtab)
-			arm_smmu_free_cd_tables(smmu, &cfg->cdcfg);
 		arm_smmu_free_asid(&smmu_domain->cd);
 		mutex_unlock(&arm_smmu_asid_lock);
 	} else {
@@ -2082,14 +2078,13 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
 	kfree(smmu_domain);
 }
 
-static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
+static int arm_smmu_domain_finalise_cd(struct arm_smmu_domain *smmu_domain,
 				       struct arm_smmu_master *master,
 				       struct io_pgtable_cfg *pgtbl_cfg)
 {
 	int ret;
 	u32 asid;
 	struct arm_smmu_device *smmu = smmu_domain->smmu;
-	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
 	struct arm_smmu_ctx_desc *cd = &smmu_domain->cd;
 	typeof(&pgtbl_cfg->arm_lpae_s1_cfg.tcr) tcr = &pgtbl_cfg->arm_lpae_s1_cfg.tcr;
 
@@ -2102,10 +2097,6 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
 	if (ret)
 		goto out_unlock;
 
-	ret = arm_smmu_init_s1_cfg(master, cfg);
-	if (ret)
-		goto out_free_asid;
-
 	cd->asid	= (u16)asid;
 	cd->ttbr	= pgtbl_cfg->arm_lpae_s1_cfg.ttbr;
 	cd->tcr		= FIELD_PREP(CTXDESC_CD_0_TCR_T0SZ, tcr->tsz) |
@@ -2117,19 +2108,9 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
 			  CTXDESC_CD_0_TCR_EPD1 | CTXDESC_CD_0_AA64;
 	cd->mair	= pgtbl_cfg->arm_lpae_s1_cfg.mair;
 
-	ret = arm_smmu_write_ctx_desc(smmu, cfg,
-				      NULL /*Not attached to a master yet */,
-				      0, cd);
-	if (ret)
-		goto out_free_cd_tables;
-
 	mutex_unlock(&arm_smmu_asid_lock);
 	return 0;
 
-out_free_cd_tables:
-	arm_smmu_free_cd_tables(smmu, &cfg->cdcfg);
-out_free_asid:
-	arm_smmu_free_asid(cd);
 out_unlock:
 	mutex_unlock(&arm_smmu_asid_lock);
 	return ret;
@@ -2192,7 +2173,7 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
 		ias = min_t(unsigned long, ias, VA_BITS);
 		oas = smmu->ias;
 		fmt = ARM_64_LPAE_S1;
-		finalise_stage_fn = arm_smmu_domain_finalise_s1;
+		finalise_stage_fn = arm_smmu_domain_finalise_cd;
 		break;
 	case ARM_SMMU_DOMAIN_NESTED:
 	case ARM_SMMU_DOMAIN_S2:
@@ -2439,20 +2420,20 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 	} else if (smmu_domain->smmu != smmu) {
 		ret = -EINVAL;
 		goto out_unlock;
-	} else if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1 &&
-		   master->ssid_bits != smmu_domain->s1_cfg.s1cdmax) {
-		ret = -EINVAL;
-		goto out_unlock;
-	} else if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1 &&
-		   smmu_domain->s1_cfg.stall_enabled !=
-			   master->stall_enabled) {
-		ret = -EINVAL;
-		goto out_unlock;
 	}
 
 	master->domain = smmu_domain;
 	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
-		master->s1_cfg = &smmu_domain->s1_cfg;
+		master->s1_cfg = &master->owned_s1_cfg;
+		ret = arm_smmu_write_ctx_desc(
+			smmu,
+			master->s1_cfg, NULL /*Not attached to a master yet */,
+			0, &smmu_domain->cd);
+		if (ret) {
+			master->s1_cfg = NULL;
+			master->domain = NULL;
+			goto out_unlock;
+		}
 	} else if (smmu_domain->stage == ARM_SMMU_DOMAIN_S2 ||
 		   smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED) {
 		master->s2_cfg = &smmu_domain->s2_cfg;
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
index 00a493442d6f9..dff0fa8345462 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
@@ -718,10 +718,7 @@ struct arm_smmu_domain {
 
 	enum arm_smmu_domain_stage		stage;
 	union {
-		struct {
 		struct arm_smmu_ctx_desc	cd;
-		struct arm_smmu_s1_cfg		s1_cfg;
-		};
 		struct arm_smmu_s2_cfg		s2_cfg;
 	};
 
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 06/13] iommu/arm-smmu-v3: Simplify arm_smmu_enable_ats

[Michael Shavit]

arm_smmu_enable_ats's call to inv_domain would trigger an invalidation
for all masters that a domain is attached to everytime it's attached to
another ATS-enabled master. It doesn't seem like those invalidations are
necessary, and it's easier to reason about arm_smmu_enable_ats if it
only issues invalidation commands for the current master.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
v1->v2: Fix commit message wrapping
---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index 08f440fe1da6d..dc7a59e87a2b4 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -2286,7 +2286,7 @@ static void arm_smmu_enable_ats(struct arm_smmu_master *master)
 	pdev = to_pci_dev(master->dev);
 
 	atomic_inc(&smmu_domain->nr_ats_masters);
-	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
+	arm_smmu_atc_inv_master(master);
 	if (pci_enable_ats(pdev, stu))
 		dev_err(master->dev, "Failed to enable ATS (STU %zu)\n", stu);
 }
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 07/13] iommu/arm-smmu-v3: Keep track of attached ssids

[Michael Shavit]

The arm-smmu-v3 driver keeps track of all masters that a domain is
attached to so that it can re-write their STEs when the domain's ASID is
upated by SVA. This tracking is also used to invalidate ATCs on all
masters that a domain is attached to.

This change introduces a new data structures to track all the CD entries
that a domain is attached to. This change is a pre-requisite to allow
domain attachment on non 0 SSIDs.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
v1->v2: Fix arm_smmu_atc_inv_cmd_set_ssid and other cosmetic changes
---
 .../iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c   | 53 +++++++----
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c   | 88 ++++++++++++-------
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h   | 18 ++--
 3 files changed, 105 insertions(+), 54 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
index 48fa8eb271a45..d07c08b53c5cf 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
@@ -51,6 +51,7 @@ arm_smmu_share_asid(struct mm_struct *mm, u16 asid)
 	struct arm_smmu_device *smmu;
 	struct arm_smmu_domain *smmu_domain;
 	struct arm_smmu_master *master;
+	struct arm_smmu_attached_domain *attached_domain;
 
 	cd = xa_load(&arm_smmu_asid_xa, asid);
 	if (!cd)
@@ -82,11 +83,14 @@ arm_smmu_share_asid(struct mm_struct *mm, u16 asid)
 	 * be some overlap between use of both ASIDs, until we invalidate the
 	 * TLB.
 	 */
-	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
-	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
-		arm_smmu_write_ctx_desc(smmu, master->s1_cfg, master, 0, cd);
+	spin_lock_irqsave(&smmu_domain->attached_domains_lock, flags);
+	list_for_each_entry(attached_domain, &smmu_domain->attached_domains,
+			    domain_head) {
+		master = attached_domain->master;
+		arm_smmu_write_ctx_desc(smmu, master->s1_cfg, master,
+					attached_domain->ssid, cd);
 	}
-	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
+	spin_unlock_irqrestore(&smmu_domain->attached_domains_lock, flags);
 
 	/* Invalidate TLB entries previously associated with that context */
 	arm_smmu_tlb_inv_asid(smmu, asid);
@@ -210,7 +214,7 @@ static void arm_smmu_mm_invalidate_range(struct mmu_notifier *mn,
 	if (!(smmu_domain->smmu->features & ARM_SMMU_FEAT_BTM))
 		arm_smmu_tlb_inv_range_asid(start, size, smmu_mn->cd->asid,
 					    PAGE_SIZE, false, smmu_domain);
-	arm_smmu_atc_inv_domain(smmu_domain, mm->pasid, start, size);
+	arm_smmu_atc_inv_domain_ssid(smmu_domain, mm->pasid, start, size);
 }
 
 static void arm_smmu_mm_release(struct mmu_notifier *mn, struct mm_struct *mm)
@@ -218,6 +222,7 @@ static void arm_smmu_mm_release(struct mmu_notifier *mn, struct mm_struct *mm)
 	struct arm_smmu_mmu_notifier *smmu_mn = mn_to_smmu(mn);
 	struct arm_smmu_domain *smmu_domain = smmu_mn->domain;
 	struct arm_smmu_master *master;
+	struct arm_smmu_attached_domain *attached_domain;
 	unsigned long flags;
 
 	mutex_lock(&sva_lock);
@@ -230,15 +235,21 @@ static void arm_smmu_mm_release(struct mmu_notifier *mn, struct mm_struct *mm)
 	 * DMA may still be running. Keep the cd valid to avoid C_BAD_CD events,
 	 * but disable translation.
 	 */
-	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
-	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
-		arm_smmu_write_ctx_desc(master->smmu, master->s1_cfg, master,
-					mm->pasid, &quiet_cd);
+	spin_lock_irqsave(&smmu_domain->attached_domains_lock, flags);
+	list_for_each_entry(attached_domain, &smmu_domain->attached_domains,
+			    domain_head) {
+		master = attached_domain->master;
+		/*
+		 * SVA domains piggyback on the attached_domain with SSID 0.
+		 */
+		if (attached_domain->ssid == 0)
+			arm_smmu_write_ctx_desc(master->smmu, master->s1_cfg,
+						master, mm->pasid, &quiet_cd);
 	}
-	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
+	spin_unlock_irqrestore(&smmu_domain->attached_domains_lock, flags);
 
 	arm_smmu_tlb_inv_asid(smmu_domain->smmu, smmu_mn->cd->asid);
-	arm_smmu_atc_inv_domain(smmu_domain, mm->pasid, 0, 0);
+	arm_smmu_atc_inv_domain_ssid(smmu_domain, mm->pasid, 0, 0);
 
 	smmu_mn->cleared = true;
 	mutex_unlock(&sva_lock);
@@ -265,6 +276,7 @@ arm_smmu_mmu_notifier_get(struct arm_smmu_domain *smmu_domain,
 	struct arm_smmu_ctx_desc *cd;
 	struct arm_smmu_mmu_notifier *smmu_mn;
 	struct arm_smmu_master *master;
+	struct arm_smmu_attached_domain *attached_domain;
 
 	list_for_each_entry(smmu_mn, &smmu_domain->mmu_notifiers, list) {
 		if (smmu_mn->mn.mm == mm) {
@@ -294,12 +306,14 @@ arm_smmu_mmu_notifier_get(struct arm_smmu_domain *smmu_domain,
 		goto err_free_cd;
 	}
 
-	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
-	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
+	spin_lock_irqsave(&smmu_domain->attached_domains_lock, flags);
+	list_for_each_entry(attached_domain, &smmu_domain->attached_domains,
+			    domain_head) {
+		master = attached_domain->master;
 		ret = arm_smmu_write_ctx_desc(master->smmu, master->s1_cfg,
 					      master, mm->pasid, cd);
 	}
-	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
+	spin_unlock_irqrestore(&smmu_domain->attached_domains_lock, flags);
 	if (ret)
 		goto err_put_notifier;
 
@@ -319,6 +333,7 @@ static void arm_smmu_mmu_notifier_put(struct arm_smmu_mmu_notifier *smmu_mn)
 	unsigned long flags;
 	struct mm_struct *mm = smmu_mn->mn.mm;
 	struct arm_smmu_ctx_desc *cd = smmu_mn->cd;
+	struct arm_smmu_attached_domain *attached_domain;
 	struct arm_smmu_master *master;
 	struct arm_smmu_domain *smmu_domain = smmu_mn->domain;
 
@@ -327,12 +342,14 @@ static void arm_smmu_mmu_notifier_put(struct arm_smmu_mmu_notifier *smmu_mn)
 
 	list_del(&smmu_mn->list);
 
-	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
-	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
+	spin_lock_irqsave(&smmu_domain->attached_domains_lock, flags);
+	list_for_each_entry(attached_domain, &smmu_domain->attached_domains,
+			    domain_head) {
+		master = attached_domain->master;
 		arm_smmu_write_ctx_desc(master->smmu, master->s1_cfg, master,
 					mm->pasid, NULL);
 	}
-	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
+	spin_unlock_irqrestore(&smmu_domain->attached_domains_lock, flags);
 
 	/*
 	 * If we went through clear(), we've already invalidated, and no
@@ -340,7 +357,7 @@ static void arm_smmu_mmu_notifier_put(struct arm_smmu_mmu_notifier *smmu_mn)
 	 */
 	if (!smmu_mn->cleared) {
 		arm_smmu_tlb_inv_asid(smmu_domain->smmu, cd->asid);
-		arm_smmu_atc_inv_domain(smmu_domain, mm->pasid, 0, 0);
+		arm_smmu_atc_inv_domain_ssid(smmu_domain, mm->pasid, 0, 0);
 	}
 
 	/* Frees smmu_mn */
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index dc7a59e87a2b4..70580ba7065dc 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -1711,7 +1711,14 @@ static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
 }
 
 static void
-arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size,
+arm_smmu_atc_inv_cmd_set_ssid(int ssid, struct arm_smmu_cmdq_ent *cmd)
+{
+	cmd->substream_valid = !!ssid;
+	cmd->atc.ssid = ssid;
+}
+
+static void
+arm_smmu_atc_inv_to_cmd(unsigned long iova, size_t size,
 			struct arm_smmu_cmdq_ent *cmd)
 {
 	size_t log2_span;
@@ -1736,8 +1743,8 @@ arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size,
 	 */
 	*cmd = (struct arm_smmu_cmdq_ent) {
 		.opcode			= CMDQ_OP_ATC_INV,
-		.substream_valid	= !!ssid,
-		.atc.ssid		= ssid,
+		.substream_valid	= false,
+		.atc.ssid		= 0,
 	};
 
 	if (!size) {
@@ -1783,8 +1790,7 @@ static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
 	struct arm_smmu_cmdq_ent cmd;
 	struct arm_smmu_cmdq_batch cmds;
 
-	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
-
+	arm_smmu_atc_inv_to_cmd(0, 0, &cmd);
 	cmds.num = 0;
 	for (i = 0; i < master->num_streams; i++) {
 		cmd.atc.sid = master->streams[i].id;
@@ -1794,13 +1800,19 @@ static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
 	return arm_smmu_cmdq_batch_submit(master->smmu, &cmds);
 }
 
-int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain, int ssid,
-			    unsigned long iova, size_t size)
+/*
+ * If ssid is non-zero, issue atc invalidations with the given ssid instead of
+ * the one the domain is attached to. This is used by SVA since it's pasid
+ * attachments aren't recorded in smmu_domain yet.
+ */
+int arm_smmu_atc_inv_domain_ssid(struct arm_smmu_domain *smmu_domain, int ssid,
+				 unsigned long iova, size_t size)
 {
 	int i;
 	unsigned long flags;
 	struct arm_smmu_cmdq_ent cmd;
 	struct arm_smmu_master *master;
+	struct arm_smmu_attached_domain *attached_domain;
 	struct arm_smmu_cmdq_batch cmds;
 
 	if (!(smmu_domain->smmu->features & ARM_SMMU_FEAT_ATS))
@@ -1823,25 +1835,37 @@ int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain, int ssid,
 	if (!atomic_read(&smmu_domain->nr_ats_masters))
 		return 0;
 
-	arm_smmu_atc_inv_to_cmd(ssid, iova, size, &cmd);
+	arm_smmu_atc_inv_to_cmd(iova, size, &cmd);
 
 	cmds.num = 0;
 
-	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
-	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
+	spin_lock_irqsave(&smmu_domain->attached_domains_lock, flags);
+	list_for_each_entry(attached_domain, &smmu_domain->attached_domains,
+			    domain_head) {
+		master = attached_domain->master;
 		if (!master->ats_enabled)
 			continue;
+		if (ssid != 0)
+			arm_smmu_atc_inv_cmd_set_ssid(ssid, &cmd);
+		else
+			arm_smmu_atc_inv_cmd_set_ssid(attached_domain->ssid, &cmd);
 
 		for (i = 0; i < master->num_streams; i++) {
 			cmd.atc.sid = master->streams[i].id;
 			arm_smmu_cmdq_batch_add(smmu_domain->smmu, &cmds, &cmd);
 		}
 	}
-	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
+	spin_unlock_irqrestore(&smmu_domain->attached_domains_lock, flags);
 
 	return arm_smmu_cmdq_batch_submit(smmu_domain->smmu, &cmds);
 }
 
+int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
+			    unsigned long iova, size_t size)
+{
+	return arm_smmu_atc_inv_domain_ssid(smmu_domain, 0, iova, size);
+}
+
 /* IO_PGTABLE API */
 static void arm_smmu_tlb_inv_context(void *cookie)
 {
@@ -1863,7 +1887,7 @@ static void arm_smmu_tlb_inv_context(void *cookie)
 		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
 		arm_smmu_cmdq_issue_cmd_with_sync(smmu, &cmd);
 	}
-	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
+	arm_smmu_atc_inv_domain(smmu_domain, 0, 0);
 }
 
 static void __arm_smmu_tlb_inv_range(struct arm_smmu_cmdq_ent *cmd,
@@ -1951,7 +1975,7 @@ static void arm_smmu_tlb_inv_range_domain(unsigned long iova, size_t size,
 	 * Unfortunately, this can't be leaf-only since we may have
 	 * zapped an entire table.
 	 */
-	arm_smmu_atc_inv_domain(smmu_domain, 0, iova, size);
+	arm_smmu_atc_inv_domain(smmu_domain, iova, size);
 }
 
 void arm_smmu_tlb_inv_range_asid(unsigned long iova, size_t size, int asid,
@@ -2031,8 +2055,8 @@ static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
 		return NULL;
 
 	mutex_init(&smmu_domain->init_mutex);
-	INIT_LIST_HEAD(&smmu_domain->devices);
-	spin_lock_init(&smmu_domain->devices_lock);
+	INIT_LIST_HEAD(&smmu_domain->attached_domains);
+	spin_lock_init(&smmu_domain->attached_domains_lock);
 	INIT_LIST_HEAD(&smmu_domain->mmu_notifiers);
 
 	return &smmu_domain->domain;
@@ -2270,12 +2294,12 @@ static bool arm_smmu_ats_supported(struct arm_smmu_master *master)
 	return dev_is_pci(dev) && pci_ats_supported(to_pci_dev(dev));
 }
 
-static void arm_smmu_enable_ats(struct arm_smmu_master *master)
+static void arm_smmu_enable_ats(struct arm_smmu_master *master,
+				struct arm_smmu_domain *smmu_domain)
 {
 	size_t stu;
 	struct pci_dev *pdev;
 	struct arm_smmu_device *smmu = master->smmu;
-	struct arm_smmu_domain *smmu_domain = master->domain;
 
 	/* Don't enable ATS at the endpoint if it's not enabled in the STE */
 	if (!master->ats_enabled)
@@ -2291,10 +2315,9 @@ static void arm_smmu_enable_ats(struct arm_smmu_master *master)
 		dev_err(master->dev, "Failed to enable ATS (STU %zu)\n", stu);
 }
 
-static void arm_smmu_disable_ats(struct arm_smmu_master *master)
+static void arm_smmu_disable_ats(struct arm_smmu_master *master,
+				 struct arm_smmu_domain *smmu_domain)
 {
-	struct arm_smmu_domain *smmu_domain = master->domain;
-
 	if (!master->ats_enabled)
 		return;
 
@@ -2358,18 +2381,17 @@ static void arm_smmu_disable_pasid(struct arm_smmu_master *master)
 static void arm_smmu_detach_dev(struct arm_smmu_master *master)
 {
 	unsigned long flags;
-	struct arm_smmu_domain *smmu_domain = master->domain;
+	struct arm_smmu_domain *smmu_domain = master->non_pasid_domain.domain;
 
 	if (!smmu_domain)
 		return;
 
-	arm_smmu_disable_ats(master);
+	arm_smmu_disable_ats(master, smmu_domain);
 
-	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
-	list_del(&master->domain_head);
-	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
+	spin_lock_irqsave(&smmu_domain->attached_domains_lock, flags);
+	list_del(&master->non_pasid_domain.domain_head);
+	spin_unlock_irqrestore(&smmu_domain->attached_domains_lock, flags);
 
-	master->domain = NULL;
 	master->ats_enabled = false;
 	if (master->s1_cfg)
 		arm_smmu_write_ctx_desc(
@@ -2378,6 +2400,7 @@ static void arm_smmu_detach_dev(struct arm_smmu_master *master)
 			0, NULL);
 	master->s1_cfg = NULL;
 	master->s2_cfg = NULL;
+	master->non_pasid_domain.domain = NULL;
 	arm_smmu_install_ste_for_dev(master);
 }
 
@@ -2422,7 +2445,6 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 		goto out_unlock;
 	}
 
-	master->domain = smmu_domain;
 	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
 		master->s1_cfg = &master->owned_s1_cfg;
 		ret = arm_smmu_write_ctx_desc(
@@ -2449,13 +2471,17 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 	if (smmu_domain->stage != ARM_SMMU_DOMAIN_BYPASS)
 		master->ats_enabled = arm_smmu_ats_supported(master);
 
+	master->non_pasid_domain.master = master;
+	master->non_pasid_domain.domain = smmu_domain;
+	master->non_pasid_domain.ssid = 0;
 	arm_smmu_install_ste_for_dev(master);
 
-	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
-	list_add(&master->domain_head, &smmu_domain->devices);
-	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
+	spin_lock_irqsave(&smmu_domain->attached_domains_lock, flags);
+	list_add(&master->non_pasid_domain.domain_head,
+		 &smmu_domain->attached_domains);
+	spin_unlock_irqrestore(&smmu_domain->attached_domains_lock, flags);
 
-	arm_smmu_enable_ats(master);
+	arm_smmu_enable_ats(master, smmu_domain);
 
 out_unlock:
 	mutex_unlock(&smmu_domain->init_mutex);
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
index dff0fa8345462..6929590530367 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
@@ -682,11 +682,19 @@ struct arm_smmu_stream {
 	struct rb_node			node;
 };
 
+/* List of {masters, ssid} that a domain is attached to */
+struct arm_smmu_attached_domain {
+	struct list_head	domain_head;
+	struct arm_smmu_domain  *domain;
+	struct arm_smmu_master  *master;
+	int			ssid;
+};
+
 /* SMMU private data for each master */
 struct arm_smmu_master {
 	struct arm_smmu_device		*smmu;
 	struct device			*dev;
-	struct arm_smmu_domain		*domain;
+	struct arm_smmu_attached_domain	non_pasid_domain;
 	struct list_head		domain_head;
 	struct arm_smmu_stream		*streams;
 	struct arm_smmu_s1_cfg		owned_s1_cfg;
@@ -724,8 +732,8 @@ struct arm_smmu_domain {
 
 	struct iommu_domain			domain;
 
-	struct list_head			devices;
-	spinlock_t				devices_lock;
+	struct list_head			attached_domains;
+	spinlock_t				attached_domains_lock;
 
 	struct list_head			mmu_notifiers;
 };
@@ -748,8 +756,8 @@ void arm_smmu_tlb_inv_range_asid(unsigned long iova, size_t size, int asid,
 				 size_t granule, bool leaf,
 				 struct arm_smmu_domain *smmu_domain);
 bool arm_smmu_free_asid(struct arm_smmu_ctx_desc *cd);
-int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain, int ssid,
-			    unsigned long iova, size_t size);
+int arm_smmu_atc_inv_domain_ssid(struct arm_smmu_domain *smmu_domain, int ssid,
+				 unsigned long iova, size_t size);
 
 #ifdef CONFIG_ARM_SMMU_V3_SVA
 bool arm_smmu_sva_supported(struct arm_smmu_device *smmu);
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 08/13] iommu/arm-smmu-v3: Add helper for atc invalidation

[Michael Shavit]

This will be used to invalidate ATC entries made on an SSID for a master
when detaching a domain with pasid.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
v1->v2: Make use of arm_smmu_atc_inv_cmd_set_ssid
---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index 70580ba7065dc..176013bb974b8 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -1784,13 +1784,15 @@ arm_smmu_atc_inv_to_cmd(unsigned long iova, size_t size,
 	cmd->atc.size	= log2_span;
 }
 
-static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
+static int arm_smmu_atc_inv_master_ssid(struct arm_smmu_master *master,
+					int ssid)
 {
 	int i;
 	struct arm_smmu_cmdq_ent cmd;
 	struct arm_smmu_cmdq_batch cmds;
 
 	arm_smmu_atc_inv_to_cmd(0, 0, &cmd);
+	arm_smmu_atc_inv_cmd_set_ssid(ssid, &cmd);
 	cmds.num = 0;
 	for (i = 0; i < master->num_streams; i++) {
 		cmd.atc.sid = master->streams[i].id;
@@ -1800,6 +1802,11 @@ static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
 	return arm_smmu_cmdq_batch_submit(master->smmu, &cmds);
 }
 
+static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
+{
+	return arm_smmu_atc_inv_master_ssid(master, 0);
+}
+
 /*
  * If ssid is non-zero, issue atc invalidations with the given ssid instead of
  * the one the domain is attached to. This is used by SVA since it's pasid
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 09/13] iommu/arm-smmu-v3: Implement set_dev_pasid

[Michael Shavit]

This change enables the use of the iommu_attach_dev_pasid API for
UNMANAGED domains. The primary use-case is to allow in-kernel users of
the iommu API to manage domains with PASID. This change also allows for
future support of pasid in the DMA api.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
v1->v2: Add missing atc invalidation when detaching with pasid
---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 168 +++++++++++++++++---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h |   1 +
 2 files changed, 149 insertions(+), 20 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index 176013bb974b8..a6fa56585c219 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -2173,6 +2173,10 @@ static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
 	return 0;
 }
 
+/*
+ * master may be null for domain types that are finalized before being attached
+ * to a master.
+ */
 static int arm_smmu_domain_finalise(struct iommu_domain *domain,
 				    struct arm_smmu_master *master)
 {
@@ -2369,6 +2373,11 @@ static int arm_smmu_enable_pasid(struct arm_smmu_master *master)
 	return 0;
 }
 
+static bool arm_smmu_master_has_pasid_domains(struct arm_smmu_master *master)
+{
+	return master->nr_attached_pasid_domains > 0;
+}
+
 static void arm_smmu_disable_pasid(struct arm_smmu_master *master)
 {
 	struct pci_dev *pdev;
@@ -2411,6 +2420,28 @@ static void arm_smmu_detach_dev(struct arm_smmu_master *master)
 	arm_smmu_install_ste_for_dev(master);
 }
 
+/*
+ * Once attached for the first time, a domain can no longer be attached to any
+ * master with a distinct upstream SMMU.
+ */
+static int arm_smmu_prepare_domain_for_smmu(struct arm_smmu_device *smmu,
+					    struct arm_smmu_domain *smmu_domain)
+{
+	int ret = 0;
+
+	mutex_lock(&smmu_domain->init_mutex);
+	if (!smmu_domain->smmu) {
+		smmu_domain->smmu = smmu;
+		ret = arm_smmu_domain_finalise(&smmu_domain->domain, NULL);
+		if (ret)
+			smmu_domain->smmu = NULL;
+	} else if (smmu_domain->smmu != smmu) {
+		ret = -EINVAL;
+	}
+	mutex_unlock(&smmu_domain->init_mutex);
+	return ret;
+}
+
 static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 {
 	int ret = 0;
@@ -2426,6 +2457,10 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 	master = dev_iommu_priv_get(dev);
 	smmu = master->smmu;
 
+	ret = arm_smmu_prepare_domain_for_smmu(smmu, smmu_domain);
+	if (ret)
+		return ret;
+
 	/*
 	 * Checking that SVA is disabled ensures that this device isn't bound to
 	 * any mm, and can be safely detached from its old domain. Bonds cannot
@@ -2436,22 +2471,18 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 		return -EBUSY;
 	}
 
-	arm_smmu_detach_dev(master);
-
-	mutex_lock(&smmu_domain->init_mutex);
-
-	if (!smmu_domain->smmu) {
-		smmu_domain->smmu = smmu;
-		ret = arm_smmu_domain_finalise(domain, master);
-		if (ret) {
-			smmu_domain->smmu = NULL;
-			goto out_unlock;
-		}
-	} else if (smmu_domain->smmu != smmu) {
-		ret = -EINVAL;
-		goto out_unlock;
+	/*
+	 * Attaching a bypass or stage 2 domain would break any domains attached
+	 * with pasid. Attaching an S1 domain should be feasible but requires
+	 * more complicated logic to handle.
+	 */
+	if (arm_smmu_master_has_pasid_domains(master)) {
+		dev_err(dev, "cannot attach - domain attached with pasid\n");
+		return -EBUSY;
 	}
 
+	arm_smmu_detach_dev(master);
+
 	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
 		master->s1_cfg = &master->owned_s1_cfg;
 		ret = arm_smmu_write_ctx_desc(
@@ -2460,8 +2491,7 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 			0, &smmu_domain->cd);
 		if (ret) {
 			master->s1_cfg = NULL;
-			master->domain = NULL;
-			goto out_unlock;
+			return ret;
 		}
 	} else if (smmu_domain->stage == ARM_SMMU_DOMAIN_S2 ||
 		   smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED) {
@@ -2490,11 +2520,75 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 
 	arm_smmu_enable_ats(master, smmu_domain);
 
-out_unlock:
-	mutex_unlock(&smmu_domain->init_mutex);
 	return ret;
 }
 
+static int arm_smmu_set_dev_pasid(struct iommu_domain *domain,
+				  struct device *dev, ioasid_t pasid)
+{
+	int ret = 0;
+	unsigned long flags;
+	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
+	struct arm_smmu_device *smmu;
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+	struct arm_smmu_attached_domain *attached_domain;
+	struct arm_smmu_master *master;
+
+	if (!fwspec)
+		return -ENOENT;
+
+	master = dev_iommu_priv_get(dev);
+	smmu = master->smmu;
+
+	ret = arm_smmu_prepare_domain_for_smmu(smmu, smmu_domain);
+	if (ret)
+		return ret;
+
+	if (pasid == 0) {
+		dev_err(dev, "pasid 0 is reserved for the device's primary domain\n");
+		return -ENODEV;
+	}
+
+	if (smmu_domain->stage != ARM_SMMU_DOMAIN_S1) {
+		dev_err(dev, "set_dev_pasid only supports stage 1 domains\n");
+		return -EINVAL;
+	}
+
+	if (!master->s1_cfg || master->s2_cfg)
+		return -EBUSY;
+
+	attached_domain = kzalloc(sizeof(*attached_domain), GFP_KERNEL);
+	if (!attached_domain)
+		return -ENOMEM;
+
+	attached_domain->master = master;
+	attached_domain->domain = smmu_domain;
+	attached_domain->ssid = pasid;
+
+	master->nr_attached_pasid_domains += 1;
+	/*
+	 * arm_smmu_share_asid may update the cd's asid value and write the
+	 * ctx_desc for every attached_domains in the list. There's a potential
+	 * race here regardless of whether we first write the ctx_desc or
+	 * first insert into the domain's list. Grabbing the asic_lock prevents
+	 * SVA from changing the cd's ASID while the cd is being attached.
+	 */
+	mutex_lock(&arm_smmu_asid_lock);
+	ret = arm_smmu_write_ctx_desc(master->smmu, master->s1_cfg, master,
+				      pasid, &smmu_domain->cd);
+	if (ret) {
+		mutex_unlock(&arm_smmu_asid_lock);
+		kfree(attached_domain);
+	}
+
+	spin_lock_irqsave(&smmu_domain->attached_domains_lock, flags);
+	list_add(&attached_domain->domain_head, &smmu_domain->attached_domains);
+	spin_unlock_irqrestore(&smmu_domain->attached_domains_lock, flags);
+	mutex_unlock(&arm_smmu_asid_lock);
+
+	return 0;
+}
+
 static int arm_smmu_map_pages(struct iommu_domain *domain, unsigned long iova,
 			      phys_addr_t paddr, size_t pgsize, size_t pgcount,
 			      int prot, gfp_t gfp, size_t *mapped)
@@ -2740,6 +2834,15 @@ static void arm_smmu_release_device(struct device *dev)
 
 	if (WARN_ON(arm_smmu_master_sva_enabled(master)))
 		iopf_queue_remove_device(master->smmu->evtq.iopf, dev);
+	if (WARN_ON(master->nr_attached_pasid_domains != 0)) {
+		/*
+		 * TODO: Do we need to handle this case?
+		 * This requires a mechanism to obtain all the pasid domains
+		 * that this master is attached to so that we can clean up the
+		 * domain's attached_domain list.
+		 */
+	}
+
 	arm_smmu_detach_dev(master);
 	arm_smmu_free_cd_tables(master->smmu, &master->owned_s1_cfg.cdcfg);
 	arm_smmu_disable_pasid(master);
@@ -2875,12 +2978,36 @@ static int arm_smmu_def_domain_type(struct device *dev)
 static void arm_smmu_remove_dev_pasid(struct device *dev, ioasid_t pasid)
 {
 	struct iommu_domain *domain;
+	struct arm_smmu_master *master = dev_iommu_priv_get(dev);
+	struct arm_smmu_domain *smmu_domain;
+	struct arm_smmu_attached_domain *attached_domain;
+	unsigned long flags;
 
-	domain = iommu_get_domain_for_dev_pasid(dev, pasid, IOMMU_DOMAIN_SVA);
+	if (!master || pasid == 0)
+		return;
+
+	domain = iommu_get_domain_for_dev_pasid(dev, pasid, 0);
 	if (WARN_ON(IS_ERR(domain)) || !domain)
 		return;
+	if (domain->type == IOMMU_DOMAIN_SVA)
+		return arm_smmu_sva_remove_dev_pasid(domain, dev, pasid);
 
-	arm_smmu_sva_remove_dev_pasid(domain, dev, pasid);
+	smmu_domain = to_smmu_domain(domain);
+	mutex_lock(&arm_smmu_asid_lock);
+	spin_lock_irqsave(&smmu_domain->attached_domains_lock, flags);
+	list_for_each_entry(attached_domain, &smmu_domain->attached_domains, domain_head) {
+		if (attached_domain->master != master ||
+		    attached_domain->ssid != pasid)
+			continue;
+		list_del(&attached_domain->domain_head);
+		break;
+	}
+	spin_unlock_irqrestore(&smmu_domain->attached_domains_lock, flags);
+	arm_smmu_write_ctx_desc(master->smmu, master->s1_cfg, master, pasid,
+				NULL);
+	arm_smmu_atc_inv_master_ssid(master, pasid);
+	master->nr_attached_pasid_domains -= 1;
+	mutex_unlock(&arm_smmu_asid_lock);
 }
 
 static struct iommu_ops arm_smmu_ops = {
@@ -2900,6 +3027,7 @@ static struct iommu_ops arm_smmu_ops = {
 	.owner			= THIS_MODULE,
 	.default_domain_ops = &(const struct iommu_domain_ops) {
 		.attach_dev		= arm_smmu_attach_dev,
+		.set_dev_pasid		= arm_smmu_set_dev_pasid,
 		.map_pages		= arm_smmu_map_pages,
 		.unmap_pages		= arm_smmu_unmap_pages,
 		.flush_iotlb_all	= arm_smmu_flush_iotlb_all,
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
index 6929590530367..48795a7287b69 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
@@ -707,6 +707,7 @@ struct arm_smmu_master {
 	bool				iopf_enabled;
 	struct list_head		bonds;
 	unsigned int			ssid_bits;
+	unsigned int			nr_attached_pasid_domains;
 };
 
 /* SMMU private data for an IOMMU domain */
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 10/13] iommu/arm-smmu-v3-sva: Remove bond refcount

[Michael Shavit]

The iommu-sva framework checks if a bond between a device and mm already
exists and handles refcounting at the iommu_domain level.
__arm_smmu_sva_bind is therefore only called once for a device/mm pair.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c | 12 +-----------
 1 file changed, 1 insertion(+), 11 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
index d07c08b53c5cf..20301d0a2c0b0 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
@@ -29,7 +29,6 @@ struct arm_smmu_bond {
 	struct mm_struct		*mm;
 	struct arm_smmu_mmu_notifier	*smmu_mn;
 	struct list_head		list;
-	refcount_t			refs;
 };
 
 #define sva_to_bond(handle) \
@@ -377,21 +376,12 @@ __arm_smmu_sva_bind(struct device *dev, struct mm_struct *mm)
 	if (!master || !master->sva_enabled)
 		return ERR_PTR(-ENODEV);
 
-	/* If bind() was already called for this {dev, mm} pair, reuse it. */
-	list_for_each_entry(bond, &master->bonds, list) {
-		if (bond->mm == mm) {
-			refcount_inc(&bond->refs);
-			return &bond->sva;
-		}
-	}
-
 	bond = kzalloc(sizeof(*bond), GFP_KERNEL);
 	if (!bond)
 		return ERR_PTR(-ENOMEM);
 
 	bond->mm = mm;
 	bond->sva.dev = dev;
-	refcount_set(&bond->refs, 1);
 
 	bond->smmu_mn = arm_smmu_mmu_notifier_get(smmu_domain, mm);
 	if (IS_ERR(bond->smmu_mn)) {
@@ -570,7 +560,7 @@ void arm_smmu_sva_remove_dev_pasid(struct iommu_domain *domain,
 		}
 	}
 
-	if (!WARN_ON(!bond) && refcount_dec_and_test(&bond->refs)) {
+	if (!WARN_ON(!bond)) {
 		list_del(&bond->list);
 		arm_smmu_mmu_notifier_put(bond->smmu_mn);
 		kfree(bond);
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 11/13] iommu/arm-smmu-v3-sva: Clean unused iommu_sva

[Michael Shavit]

The __arm_smmu_sva_bind function returned an unused iommu_sva handle
that can be removed.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
 .../iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c    | 18 ++++++------------
 1 file changed, 6 insertions(+), 12 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
index 20301d0a2c0b0..650c9c9ad52f1 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
@@ -25,7 +25,6 @@ struct arm_smmu_mmu_notifier {
 #define mn_to_smmu(mn) container_of(mn, struct arm_smmu_mmu_notifier, mn)
 
 struct arm_smmu_bond {
-	struct iommu_sva		sva;
 	struct mm_struct		*mm;
 	struct arm_smmu_mmu_notifier	*smmu_mn;
 	struct list_head		list;
@@ -364,8 +363,7 @@ static void arm_smmu_mmu_notifier_put(struct arm_smmu_mmu_notifier *smmu_mn)
 	arm_smmu_free_shared_cd(cd);
 }
 
-static struct iommu_sva *
-__arm_smmu_sva_bind(struct device *dev, struct mm_struct *mm)
+static int __arm_smmu_sva_bind(struct device *dev, struct mm_struct *mm)
 {
 	int ret;
 	struct arm_smmu_bond *bond;
@@ -374,14 +372,13 @@ __arm_smmu_sva_bind(struct device *dev, struct mm_struct *mm)
 	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
 
 	if (!master || !master->sva_enabled)
-		return ERR_PTR(-ENODEV);
+		return -ENODEV;
 
 	bond = kzalloc(sizeof(*bond), GFP_KERNEL);
 	if (!bond)
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 
 	bond->mm = mm;
-	bond->sva.dev = dev;
 
 	bond->smmu_mn = arm_smmu_mmu_notifier_get(smmu_domain, mm);
 	if (IS_ERR(bond->smmu_mn)) {
@@ -390,11 +387,11 @@ __arm_smmu_sva_bind(struct device *dev, struct mm_struct *mm)
 	}
 
 	list_add(&bond->list, &master->bonds);
-	return &bond->sva;
+	return 0;
 
 err_free_bond:
 	kfree(bond);
-	return ERR_PTR(ret);
+	return ret;
 }
 
 bool arm_smmu_sva_supported(struct arm_smmu_device *smmu)
@@ -572,13 +569,10 @@ static int arm_smmu_sva_set_dev_pasid(struct iommu_domain *domain,
 				      struct device *dev, ioasid_t id)
 {
 	int ret = 0;
-	struct iommu_sva *handle;
 	struct mm_struct *mm = domain->mm;
 
 	mutex_lock(&sva_lock);
-	handle = __arm_smmu_sva_bind(dev, mm);
-	if (IS_ERR(handle))
-		ret = PTR_ERR(handle);
+	ret = __arm_smmu_sva_bind(dev, mm);
 	mutex_unlock(&sva_lock);
 
 	return ret;
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 12/13] iommu/arm-smmu-v3-sva: Remove arm_smmu_bond

[Michael Shavit]

There's a 1:1 relationship between arm_smmu_bond and the iommu_domain
used in set_dev_pasid/remove_dev_pasid. arm_smmu_bond has become an
unnecessary complication. It's more natural to store any needed
information at the iommu_domain container level.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
 .../iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c   | 69 +++++++------------
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c   |  1 -
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h   |  2 +-
 3 files changed, 24 insertions(+), 48 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
index 650c9c9ad52f1..b615a85e6a54e 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
@@ -24,14 +24,13 @@ struct arm_smmu_mmu_notifier {
 
 #define mn_to_smmu(mn) container_of(mn, struct arm_smmu_mmu_notifier, mn)
 
-struct arm_smmu_bond {
-	struct mm_struct		*mm;
+struct arm_smmu_sva_domain {
+	struct iommu_domain		iommu_domain;
 	struct arm_smmu_mmu_notifier	*smmu_mn;
-	struct list_head		list;
 };
 
-#define sva_to_bond(handle) \
-	container_of(handle, struct arm_smmu_bond, sva)
+#define to_sva_domain(domain) \
+	container_of(domain, struct arm_smmu_sva_domain, iommu_domain)
 
 static DEFINE_MUTEX(sva_lock);
 
@@ -363,10 +362,10 @@ static void arm_smmu_mmu_notifier_put(struct arm_smmu_mmu_notifier *smmu_mn)
 	arm_smmu_free_shared_cd(cd);
 }
 
-static int __arm_smmu_sva_bind(struct device *dev, struct mm_struct *mm)
+static int __arm_smmu_sva_bind(struct device *dev,
+			       struct arm_smmu_sva_domain *sva_domain,
+			       struct mm_struct *mm)
 {
-	int ret;
-	struct arm_smmu_bond *bond;
 	struct arm_smmu_master *master = dev_iommu_priv_get(dev);
 	struct iommu_domain *domain = iommu_get_domain_for_dev(dev);
 	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
@@ -374,24 +373,14 @@ static int __arm_smmu_sva_bind(struct device *dev, struct mm_struct *mm)
 	if (!master || !master->sva_enabled)
 		return -ENODEV;
 
-	bond = kzalloc(sizeof(*bond), GFP_KERNEL);
-	if (!bond)
-		return -ENOMEM;
-
-	bond->mm = mm;
-
-	bond->smmu_mn = arm_smmu_mmu_notifier_get(smmu_domain, mm);
-	if (IS_ERR(bond->smmu_mn)) {
-		ret = PTR_ERR(bond->smmu_mn);
-		goto err_free_bond;
+	sva_domain->smmu_mn = arm_smmu_mmu_notifier_get(smmu_domain,
+							mm);
+	if (IS_ERR(sva_domain->smmu_mn)) {
+		sva_domain->smmu_mn = NULL;
+		return PTR_ERR(sva_domain->smmu_mn);
 	}
-
-	list_add(&bond->list, &master->bonds);
+	master->nr_attached_sva_domains += 1;
 	return 0;
-
-err_free_bond:
-	kfree(bond);
-	return ret;
 }
 
 bool arm_smmu_sva_supported(struct arm_smmu_device *smmu)
@@ -521,7 +510,7 @@ int arm_smmu_master_enable_sva(struct arm_smmu_master *master)
 int arm_smmu_master_disable_sva(struct arm_smmu_master *master)
 {
 	mutex_lock(&sva_lock);
-	if (!list_empty(&master->bonds)) {
+	if (master->nr_attached_sva_domains != 0) {
 		dev_err(master->dev, "cannot disable SVA, device is bound\n");
 		mutex_unlock(&sva_lock);
 		return -EBUSY;
@@ -545,23 +534,12 @@ void arm_smmu_sva_notifier_synchronize(void)
 void arm_smmu_sva_remove_dev_pasid(struct iommu_domain *domain,
 				   struct device *dev, ioasid_t id)
 {
-	struct mm_struct *mm = domain->mm;
-	struct arm_smmu_bond *bond = NULL, *t;
+	struct arm_smmu_sva_domain *sva_domain = to_sva_domain(domain);
 	struct arm_smmu_master *master = dev_iommu_priv_get(dev);
 
 	mutex_lock(&sva_lock);
-	list_for_each_entry(t, &master->bonds, list) {
-		if (t->mm == mm) {
-			bond = t;
-			break;
-		}
-	}
-
-	if (!WARN_ON(!bond)) {
-		list_del(&bond->list);
-		arm_smmu_mmu_notifier_put(bond->smmu_mn);
-		kfree(bond);
-	}
+	master->nr_attached_sva_domains -= 1;
+	arm_smmu_mmu_notifier_put(sva_domain->smmu_mn);
 	mutex_unlock(&sva_lock);
 }
 
@@ -572,7 +550,7 @@ static int arm_smmu_sva_set_dev_pasid(struct iommu_domain *domain,
 	struct mm_struct *mm = domain->mm;
 
 	mutex_lock(&sva_lock);
-	ret = __arm_smmu_sva_bind(dev, mm);
+	ret = __arm_smmu_sva_bind(dev, to_sva_domain(domain), mm);
 	mutex_unlock(&sva_lock);
 
 	return ret;
@@ -590,12 +568,11 @@ static const struct iommu_domain_ops arm_smmu_sva_domain_ops = {
 
 struct iommu_domain *arm_smmu_sva_domain_alloc(void)
 {
-	struct iommu_domain *domain;
+	struct arm_smmu_sva_domain *sva_domain;
 
-	domain = kzalloc(sizeof(*domain), GFP_KERNEL);
-	if (!domain)
+	sva_domain = kzalloc(sizeof(*sva_domain), GFP_KERNEL);
+	if (!sva_domain)
 		return NULL;
-	domain->ops = &arm_smmu_sva_domain_ops;
-
-	return domain;
+	sva_domain->iommu_domain.ops = &arm_smmu_sva_domain_ops;
+	return &sva_domain->iommu_domain;
 }
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index a6fa56585c219..b7f834dde85d1 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -2784,7 +2784,6 @@ static struct iommu_device *arm_smmu_probe_device(struct device *dev)
 
 	master->dev = dev;
 	master->smmu = smmu;
-	INIT_LIST_HEAD(&master->bonds);
 	dev_iommu_priv_set(dev, master);
 
 	ret = arm_smmu_insert_master(smmu, master);
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
index 48795a7287b69..3525d60668c23 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
@@ -705,7 +705,7 @@ struct arm_smmu_master {
 	bool				stall_enabled;
 	bool				sva_enabled;
 	bool				iopf_enabled;
-	struct list_head		bonds;
+	unsigned int			nr_attached_sva_domains;
 	unsigned int			ssid_bits;
 	unsigned int			nr_attached_pasid_domains;
 };
-- 
2.41.0.162.gfafddb0af9-goog

[PATCH v3 13/13] iommu/arm-smmu-v3-sva: Add check when enabling sva

[Michael Shavit]

SVA domains can only be attached when the master's STEs have a stage 1
domain.

Signed-off-by: Michael Shavit <mshavit@google.com>
---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
index b615a85e6a54e..e2a91f20f0906 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
@@ -499,9 +499,15 @@ int arm_smmu_master_enable_sva(struct arm_smmu_master *master)
 	int ret;
 
 	mutex_lock(&sva_lock);
+
+	if (!master->s1_cfg) {
+		ret = -EBUSY;
+		goto unlock;
+	}
 	ret = arm_smmu_master_sva_enable_iopf(master);
 	if (!ret)
 		master->sva_enabled = true;
+unlock:
 	mutex_unlock(&sva_lock);
 
 	return ret;
-- 
2.41.0.162.gfafddb0af9-goog

Re: [PATCH v3 07/13] iommu/arm-smmu-v3: Keep track of attached ssids

[kernel test robot]

Hi Michael,

kernel test robot noticed the following build warnings:

[auto build test WARNING on b6dad5178ceaf23f369c3711062ce1f2afc33644]

url:    https://github.com/intel-lab-lkp/linux/commits/Michael-Shavit/iommu-arm-smmu-v3-Move-ctx_desc-out-of-s1_cfg/20230614-234725
base:   b6dad5178ceaf23f369c3711062ce1f2afc33644
patch link:    https://lore.kernel.org/r/20230614154304.2860121-8-mshavit%40google.com
patch subject: [PATCH v3 07/13] iommu/arm-smmu-v3: Keep track of attached ssids
config: arm64-defconfig (https://download.01.org/0day-ci/archive/20230620/202306201817.Cn7Xxzds-lkp@intel.com/config)
compiler: aarch64-linux-gcc (GCC) 12.3.0
reproduce: (https://download.01.org/0day-ci/archive/20230620/202306201817.Cn7Xxzds-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202306201817.Cn7Xxzds-lkp@intel.com/

All warnings (new ones prefixed by >>):

>> drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c:1863:5: warning: no previous prototype for 'arm_smmu_atc_inv_domain' [-Wmissing-prototypes]
    1863 | int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
         |     ^~~~~~~~~~~~~~~~~~~~~~~
   drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c: In function 'arm_smmu_attach_dev':
   drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c:2456:31: error: 'struct arm_smmu_master' has no member named 'domain'
    2456 |                         master->domain = NULL;
         |                               ^~


vim +/arm_smmu_atc_inv_domain +1863 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c

  1862	
> 1863	int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
  1864				    unsigned long iova, size_t size)
  1865	{
  1866		return arm_smmu_atc_inv_domain_ssid(smmu_domain, 0, iova, size);
  1867	}
  1868	

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki