#define IDR0_ST_LVL_SHIFT 27
#define IDR0_ST_LVL_MASK 0x3
#define IDR0_ST_LVL_2LVL (1 << IDR0_ST_LVL_SHIFT)
-#define IDR0_STALL_MODEL (3 << 24)
+#define IDR0_STALL_MODEL_SHIFT 24
+#define IDR0_STALL_MODEL_MASK 0x3
+#define IDR0_STALL_MODEL_STALL (0 << IDR0_STALL_MODEL_SHIFT)
+#define IDR0_STALL_MODEL_FORCE (2 << IDR0_STALL_MODEL_SHIFT)
#define IDR0_TTENDIAN_SHIFT 21
#define IDR0_TTENDIAN_MASK 0x3
#define IDR0_TTENDIAN_LE (2 << IDR0_TTENDIAN_SHIFT)
#define STRTAB_STE_1_STRW_EL2 2UL
#define STRTAB_STE_1_STRW_SHIFT 30
+#define STRTAB_STE_1_SHCFG_INCOMING 1UL
+#define STRTAB_STE_1_SHCFG_SHIFT 44
+
#define STRTAB_STE_2_S2VMID_SHIFT 0
#define STRTAB_STE_2_S2VMID_MASK 0xffffUL
#define STRTAB_STE_2_VTCR_SHIFT 32
#define PRIQ_0_SID_MASK 0xffffffffUL
#define PRIQ_0_SSID_SHIFT 32
#define PRIQ_0_SSID_MASK 0xfffffUL
-#define PRIQ_0_OF (1UL << 57)
#define PRIQ_0_PERM_PRIV (1UL << 58)
#define PRIQ_0_PERM_EXEC (1UL << 59)
#define PRIQ_0_PERM_READ (1UL << 60)
};
dev_err(smmu->dev, "CMDQ error (cons 0x%08x): %s\n", cons,
- cerror_str[idx]);
+ idx < ARRAY_SIZE(cerror_str) ? cerror_str[idx] : "Unknown");
switch (idx) {
- case CMDQ_ERR_CERROR_ILL_IDX:
- break;
case CMDQ_ERR_CERROR_ABT_IDX:
dev_err(smmu->dev, "retrying command fetch\n");
case CMDQ_ERR_CERROR_NONE_IDX:
return;
+ case CMDQ_ERR_CERROR_ILL_IDX:
+ /* Fallthrough */
+ default:
+ break;
}
/*
val |= disable_bypass ? STRTAB_STE_0_CFG_ABORT
: STRTAB_STE_0_CFG_BYPASS;
dst[0] = cpu_to_le64(val);
+ dst[1] = cpu_to_le64(STRTAB_STE_1_SHCFG_INCOMING
+ << STRTAB_STE_1_SHCFG_SHIFT);
dst[2] = 0; /* Nuke the VMID */
if (ste_live)
arm_smmu_sync_ste_for_sid(smmu, sid);
STRTAB_STE_1_S1C_CACHE_WBRA
<< STRTAB_STE_1_S1COR_SHIFT |
STRTAB_STE_1_S1C_SH_ISH << STRTAB_STE_1_S1CSH_SHIFT |
- STRTAB_STE_1_S1STALLD |
#ifdef CONFIG_PCI_ATS
STRTAB_STE_1_EATS_TRANS << STRTAB_STE_1_EATS_SHIFT |
#endif
STRTAB_STE_1_STRW_NSEL1 << STRTAB_STE_1_STRW_SHIFT);
+ if (smmu->features & ARM_SMMU_FEAT_STALLS)
+ dst[1] |= cpu_to_le64(STRTAB_STE_1_S1STALLD);
+
val |= (ste->s1_cfg->cdptr_dma & STRTAB_STE_0_S1CTXPTR_MASK
<< STRTAB_STE_0_S1CTXPTR_SHIFT) |
STRTAB_STE_0_CFG_S1_TRANS;
strtab = &cfg->strtab[(sid >> STRTAB_SPLIT) * STRTAB_L1_DESC_DWORDS];
desc->span = STRTAB_SPLIT + 1;
- desc->l2ptr = dma_zalloc_coherent(smmu->dev, size, &desc->l2ptr_dma,
- GFP_KERNEL);
+ desc->l2ptr = dmam_alloc_coherent(smmu->dev, size, &desc->l2ptr_dma,
+ GFP_KERNEL | __GFP_ZERO);
if (!desc->l2ptr) {
dev_err(smmu->dev,
"failed to allocate l2 stream table for SID %u\n",
static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
{
- u32 gerror, gerrorn;
+ u32 gerror, gerrorn, active;
struct arm_smmu_device *smmu = dev;
gerror = readl_relaxed(smmu->base + ARM_SMMU_GERROR);
gerrorn = readl_relaxed(smmu->base + ARM_SMMU_GERRORN);
- gerror ^= gerrorn;
- if (!(gerror & GERROR_ERR_MASK))
+ active = gerror ^ gerrorn;
+ if (!(active & GERROR_ERR_MASK))
return IRQ_NONE; /* No errors pending */
dev_warn(smmu->dev,
"unexpected global error reported (0x%08x), this could be serious\n",
- gerror);
+ active);
- if (gerror & GERROR_SFM_ERR) {
+ if (active & GERROR_SFM_ERR) {
dev_err(smmu->dev, "device has entered Service Failure Mode!\n");
arm_smmu_device_disable(smmu);
}
- if (gerror & GERROR_MSI_GERROR_ABT_ERR)
+ if (active & GERROR_MSI_GERROR_ABT_ERR)
dev_warn(smmu->dev, "GERROR MSI write aborted\n");
- if (gerror & GERROR_MSI_PRIQ_ABT_ERR) {
+ if (active & GERROR_MSI_PRIQ_ABT_ERR) {
dev_warn(smmu->dev, "PRIQ MSI write aborted\n");
arm_smmu_priq_handler(irq, smmu->dev);
}
- if (gerror & GERROR_MSI_EVTQ_ABT_ERR) {
+ if (active & GERROR_MSI_EVTQ_ABT_ERR) {
dev_warn(smmu->dev, "EVTQ MSI write aborted\n");
arm_smmu_evtq_handler(irq, smmu->dev);
}
- if (gerror & GERROR_MSI_CMDQ_ABT_ERR) {
+ if (active & GERROR_MSI_CMDQ_ABT_ERR) {
dev_warn(smmu->dev, "CMDQ MSI write aborted\n");
arm_smmu_cmdq_sync_handler(irq, smmu->dev);
}
- if (gerror & GERROR_PRIQ_ABT_ERR)
+ if (active & GERROR_PRIQ_ABT_ERR)
dev_err(smmu->dev, "PRIQ write aborted -- events may have been lost\n");
- if (gerror & GERROR_EVTQ_ABT_ERR)
+ if (active & GERROR_EVTQ_ABT_ERR)
dev_err(smmu->dev, "EVTQ write aborted -- events may have been lost\n");
- if (gerror & GERROR_CMDQ_ERR)
+ if (active & GERROR_CMDQ_ERR)
arm_smmu_cmdq_skip_err(smmu);
writel(gerror, smmu->base + ARM_SMMU_GERRORN);
}
static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
- bool leaf, void *cookie)
+ size_t granule, bool leaf, void *cookie)
{
struct arm_smmu_domain *smmu_domain = cookie;
struct arm_smmu_device *smmu = smmu_domain->smmu;
cmd.tlbi.vmid = smmu_domain->s2_cfg.vmid;
}
- arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+ do {
+ arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+ cmd.tlbi.addr += granule;
+ } while (size -= granule);
}
static struct iommu_gather_ops arm_smmu_gather_ops = {
struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
if (cfg->cdptr) {
- dma_free_coherent(smmu_domain->smmu->dev,
- CTXDESC_CD_DWORDS << 3,
- cfg->cdptr,
- cfg->cdptr_dma);
+ dmam_free_coherent(smmu_domain->smmu->dev,
+ CTXDESC_CD_DWORDS << 3,
+ cfg->cdptr,
+ cfg->cdptr_dma);
arm_smmu_bitmap_free(smmu->asid_map, cfg->cd.asid);
}
if (IS_ERR_VALUE(asid))
return asid;
- cfg->cdptr = dma_zalloc_coherent(smmu->dev, CTXDESC_CD_DWORDS << 3,
- &cfg->cdptr_dma, GFP_KERNEL);
+ cfg->cdptr = dmam_alloc_coherent(smmu->dev, CTXDESC_CD_DWORDS << 3,
+ &cfg->cdptr_dma,
+ GFP_KERNEL | __GFP_ZERO);
if (!cfg->cdptr) {
dev_warn(smmu->dev, "failed to allocate context descriptor\n");
ret = -ENOMEM;
smmu = arm_smmu_get_for_pci_dev(pdev);
if (!smmu) {
ret = -ENOENT;
- goto out_put_group;
+ goto out_remove_dev;
}
smmu_group = kzalloc(sizeof(*smmu_group), GFP_KERNEL);
if (!smmu_group) {
ret = -ENOMEM;
- goto out_put_group;
+ goto out_remove_dev;
}
smmu_group->ste.valid = true;
for (i = 0; i < smmu_group->num_sids; ++i) {
/* If we already know about this SID, then we're done */
if (smmu_group->sids[i] == sid)
- return 0;
+ goto out_put_group;
}
/* Check the SID is in range of the SMMU and our stream table */
if (!arm_smmu_sid_in_range(smmu, sid)) {
ret = -ERANGE;
- goto out_put_group;
+ goto out_remove_dev;
}
/* Ensure l2 strtab is initialised */
if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
ret = arm_smmu_init_l2_strtab(smmu, sid);
if (ret)
- goto out_put_group;
+ goto out_remove_dev;
}
/* Resize the SID array for the group */
if (!sids) {
smmu_group->num_sids--;
ret = -ENOMEM;
- goto out_put_group;
+ goto out_remove_dev;
}
/* Add the new SID */
sids[smmu_group->num_sids - 1] = sid;
smmu_group->sids = sids;
- return 0;
out_put_group:
+ iommu_group_put(group);
+ return 0;
+
+out_remove_dev:
+ iommu_group_remove_device(dev);
iommu_group_put(group);
return ret;
}
{
size_t qsz = ((1 << q->max_n_shift) * dwords) << 3;
- q->base = dma_alloc_coherent(smmu->dev, qsz, &q->base_dma, GFP_KERNEL);
+ q->base = dmam_alloc_coherent(smmu->dev, qsz, &q->base_dma, GFP_KERNEL);
if (!q->base) {
dev_err(smmu->dev, "failed to allocate queue (0x%zx bytes)\n",
qsz);
return 0;
}
-static void arm_smmu_free_one_queue(struct arm_smmu_device *smmu,
- struct arm_smmu_queue *q)
-{
- size_t qsz = ((1 << q->max_n_shift) * q->ent_dwords) << 3;
-
- dma_free_coherent(smmu->dev, qsz, q->base, q->base_dma);
-}
-
-static void arm_smmu_free_queues(struct arm_smmu_device *smmu)
-{
- arm_smmu_free_one_queue(smmu, &smmu->cmdq.q);
- arm_smmu_free_one_queue(smmu, &smmu->evtq.q);
-
- if (smmu->features & ARM_SMMU_FEAT_PRI)
- arm_smmu_free_one_queue(smmu, &smmu->priq.q);
-}
-
static int arm_smmu_init_queues(struct arm_smmu_device *smmu)
{
int ret;
ret = arm_smmu_init_one_queue(smmu, &smmu->cmdq.q, ARM_SMMU_CMDQ_PROD,
ARM_SMMU_CMDQ_CONS, CMDQ_ENT_DWORDS);
if (ret)
- goto out;
+ return ret;
/* evtq */
ret = arm_smmu_init_one_queue(smmu, &smmu->evtq.q, ARM_SMMU_EVTQ_PROD,
ARM_SMMU_EVTQ_CONS, EVTQ_ENT_DWORDS);
if (ret)
- goto out_free_cmdq;
+ return ret;
/* priq */
if (!(smmu->features & ARM_SMMU_FEAT_PRI))
return 0;
- ret = arm_smmu_init_one_queue(smmu, &smmu->priq.q, ARM_SMMU_PRIQ_PROD,
- ARM_SMMU_PRIQ_CONS, PRIQ_ENT_DWORDS);
- if (ret)
- goto out_free_evtq;
-
- return 0;
-
-out_free_evtq:
- arm_smmu_free_one_queue(smmu, &smmu->evtq.q);
-out_free_cmdq:
- arm_smmu_free_one_queue(smmu, &smmu->cmdq.q);
-out:
- return ret;
-}
-
-static void arm_smmu_free_l2_strtab(struct arm_smmu_device *smmu)
-{
- int i;
- size_t size;
- struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
-
- size = 1 << (STRTAB_SPLIT + ilog2(STRTAB_STE_DWORDS) + 3);
- for (i = 0; i < cfg->num_l1_ents; ++i) {
- struct arm_smmu_strtab_l1_desc *desc = &cfg->l1_desc[i];
-
- if (!desc->l2ptr)
- continue;
-
- dma_free_coherent(smmu->dev, size, desc->l2ptr,
- desc->l2ptr_dma);
- }
+ return arm_smmu_init_one_queue(smmu, &smmu->priq.q, ARM_SMMU_PRIQ_PROD,
+ ARM_SMMU_PRIQ_CONS, PRIQ_ENT_DWORDS);
}
static int arm_smmu_init_l1_strtab(struct arm_smmu_device *smmu)
void *strtab;
u64 reg;
u32 size, l1size;
- int ret;
struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
/*
size, smmu->sid_bits);
l1size = cfg->num_l1_ents * (STRTAB_L1_DESC_DWORDS << 3);
- strtab = dma_zalloc_coherent(smmu->dev, l1size, &cfg->strtab_dma,
- GFP_KERNEL);
+ strtab = dmam_alloc_coherent(smmu->dev, l1size, &cfg->strtab_dma,
+ GFP_KERNEL | __GFP_ZERO);
if (!strtab) {
dev_err(smmu->dev,
"failed to allocate l1 stream table (%u bytes)\n",
<< STRTAB_BASE_CFG_SPLIT_SHIFT;
cfg->strtab_base_cfg = reg;
- ret = arm_smmu_init_l1_strtab(smmu);
- if (ret)
- dma_free_coherent(smmu->dev,
- l1size,
- strtab,
- cfg->strtab_dma);
- return ret;
+ return arm_smmu_init_l1_strtab(smmu);
}
static int arm_smmu_init_strtab_linear(struct arm_smmu_device *smmu)
struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
size = (1 << smmu->sid_bits) * (STRTAB_STE_DWORDS << 3);
- strtab = dma_zalloc_coherent(smmu->dev, size, &cfg->strtab_dma,
- GFP_KERNEL);
+ strtab = dmam_alloc_coherent(smmu->dev, size, &cfg->strtab_dma,
+ GFP_KERNEL | __GFP_ZERO);
if (!strtab) {
dev_err(smmu->dev,
"failed to allocate linear stream table (%u bytes)\n",
return 0;
}
-static void arm_smmu_free_strtab(struct arm_smmu_device *smmu)
-{
- struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
- u32 size = cfg->num_l1_ents;
-
- if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
- arm_smmu_free_l2_strtab(smmu);
- size *= STRTAB_L1_DESC_DWORDS << 3;
- } else {
- size *= STRTAB_STE_DWORDS * 3;
- }
-
- dma_free_coherent(smmu->dev, size, cfg->strtab, cfg->strtab_dma);
-}
-
static int arm_smmu_init_structures(struct arm_smmu_device *smmu)
{
int ret;
if (ret)
return ret;
- ret = arm_smmu_init_strtab(smmu);
- if (ret)
- goto out_free_queues;
-
- return 0;
-
-out_free_queues:
- arm_smmu_free_queues(smmu);
- return ret;
-}
-
-static void arm_smmu_free_structures(struct arm_smmu_device *smmu)
-{
- arm_smmu_free_strtab(smmu);
- arm_smmu_free_queues(smmu);
+ return arm_smmu_init_strtab(smmu);
}
static int arm_smmu_write_reg_sync(struct arm_smmu_device *smmu, u32 val,
dev_warn(smmu->dev, "IDR0.COHACC overridden by dma-coherent property (%s)\n",
coherent ? "true" : "false");
- if (reg & IDR0_STALL_MODEL)
+ switch (reg & IDR0_STALL_MODEL_MASK << IDR0_STALL_MODEL_SHIFT) {
+ case IDR0_STALL_MODEL_STALL:
+ /* Fallthrough */
+ case IDR0_STALL_MODEL_FORCE:
smmu->features |= ARM_SMMU_FEAT_STALLS;
+ }
if (reg & IDR0_S1P)
smmu->features |= ARM_SMMU_FEAT_TRANS_S1;
platform_set_drvdata(pdev, smmu);
/* Reset the device */
- ret = arm_smmu_device_reset(smmu);
- if (ret)
- goto out_free_structures;
-
- return 0;
-
-out_free_structures:
- arm_smmu_free_structures(smmu);
- return ret;
+ return arm_smmu_device_reset(smmu);
}
static int arm_smmu_device_remove(struct platform_device *pdev)
struct arm_smmu_device *smmu = platform_get_drvdata(pdev);
arm_smmu_device_disable(smmu);
- arm_smmu_free_structures(smmu);
return 0;
}
#define io_pgtable_to_data(x) \
container_of((x), struct arm_lpae_io_pgtable, iop)
-#define io_pgtable_ops_to_pgtable(x) \
- container_of((x), struct io_pgtable, ops)
-
#define io_pgtable_ops_to_data(x) \
io_pgtable_to_data(io_pgtable_ops_to_pgtable(x))
((((d)->levels - ((l) - ARM_LPAE_START_LVL(d) + 1)) \
* (d)->bits_per_level) + (d)->pg_shift)
+#define ARM_LPAE_GRANULE(d) (1UL << (d)->pg_shift)
+
#define ARM_LPAE_PAGES_PER_PGD(d) \
- DIV_ROUND_UP((d)->pgd_size, 1UL << (d)->pg_shift)
+ DIV_ROUND_UP((d)->pgd_size, ARM_LPAE_GRANULE(d))
/*
* Calculate the index at level l used to map virtual address a using the
/* IOPTE accessors */
#define iopte_deref(pte,d) \
(__va((pte) & ((1ULL << ARM_LPAE_MAX_ADDR_BITS) - 1) \
- & ~((1ULL << (d)->pg_shift) - 1)))
+ & ~(ARM_LPAE_GRANULE(d) - 1ULL)))
#define iopte_type(pte,l) \
(((pte) >> ARM_LPAE_PTE_TYPE_SHIFT) & ARM_LPAE_PTE_TYPE_MASK)
/* Grab a pointer to the next level */
pte = *ptep;
if (!pte) {
- cptep = __arm_lpae_alloc_pages(1UL << data->pg_shift,
+ cptep = __arm_lpae_alloc_pages(ARM_LPAE_GRANULE(data),
GFP_ATOMIC, cfg);
if (!cptep)
return -ENOMEM;
arm_lpae_iopte *start, *end;
unsigned long table_size;
- /* Only leaf entries at the last level */
- if (lvl == ARM_LPAE_MAX_LEVELS - 1)
- return;
-
if (lvl == ARM_LPAE_START_LVL(data))
table_size = data->pgd_size;
else
- table_size = 1UL << data->pg_shift;
+ table_size = ARM_LPAE_GRANULE(data);
start = ptep;
- end = (void *)ptep + table_size;
+
+ /* Only leaf entries at the last level */
+ if (lvl == ARM_LPAE_MAX_LEVELS - 1)
+ end = ptep;
+ else
+ end = (void *)ptep + table_size;
while (ptep != end) {
arm_lpae_iopte pte = *ptep++;
__arm_lpae_set_pte(ptep, table, cfg);
iova &= ~(blk_size - 1);
- cfg->tlb->tlb_add_flush(iova, blk_size, true, data->iop.cookie);
+ cfg->tlb->tlb_add_flush(iova, blk_size, blk_size, true, data->iop.cookie);
return size;
}
void *cookie = data->iop.cookie;
size_t blk_size = ARM_LPAE_BLOCK_SIZE(lvl, data);
+ /* Something went horribly wrong and we ran out of page table */
+ if (WARN_ON(lvl == ARM_LPAE_MAX_LEVELS))
+ return 0;
+
ptep += ARM_LPAE_LVL_IDX(iova, lvl, data);
pte = *ptep;
-
- /* Something went horribly wrong and we ran out of page table */
- if (WARN_ON(!pte || (lvl == ARM_LPAE_MAX_LEVELS)))
+ if (WARN_ON(!pte))
return 0;
/* If the size matches this level, we're in the right place */
if (!iopte_leaf(pte, lvl)) {
/* Also flush any partial walks */
- tlb->tlb_add_flush(iova, size, false, cookie);
+ tlb->tlb_add_flush(iova, size, ARM_LPAE_GRANULE(data),
+ false, cookie);
tlb->tlb_sync(cookie);
ptep = iopte_deref(pte, data);
__arm_lpae_free_pgtable(data, lvl + 1, ptep);
} else {
- tlb->tlb_add_flush(iova, size, true, cookie);
+ tlb->tlb_add_flush(iova, size, size, true, cookie);
}
return size;
return 0;
found_translation:
- iova &= ((1 << data->pg_shift) - 1);
+ iova &= (ARM_LPAE_GRANULE(data) - 1);
return ((phys_addr_t)iopte_to_pfn(pte,data) << data->pg_shift) | iova;
}
(ARM_LPAE_TCR_RGN_WBWA << ARM_LPAE_TCR_IRGN0_SHIFT) |
(ARM_LPAE_TCR_RGN_WBWA << ARM_LPAE_TCR_ORGN0_SHIFT);
- switch (1 << data->pg_shift) {
+ switch (ARM_LPAE_GRANULE(data)) {
case SZ_4K:
reg |= ARM_LPAE_TCR_TG0_4K;
break;
sl = ARM_LPAE_START_LVL(data);
- switch (1 << data->pg_shift) {
+ switch (ARM_LPAE_GRANULE(data)) {
case SZ_4K:
reg |= ARM_LPAE_TCR_TG0_4K;
sl++; /* SL0 format is different for 4K granule size */
WARN_ON(cookie != cfg_cookie);
}
-static void dummy_tlb_add_flush(unsigned long iova, size_t size, bool leaf,
- void *cookie)
+static void dummy_tlb_add_flush(unsigned long iova, size_t size,
+ size_t granule, bool leaf, void *cookie)
{
WARN_ON(cookie != cfg_cookie);
WARN_ON(!(size & cfg_cookie->pgsize_bitmap));