Re: [Freedreno] [RFC PATCH v1 02/15] iommu/arm-smmu: Add split pagetable support for arm-smmu-v2

From: Rob Clark
Date: Fri Mar 01 2019 - 15:25:34 EST

Next message: Andrew Morton: "Re: [PATCH] log2: make is_power_of_2() integer constant expression when possible"
Previous message: Rob Clark: "Re: [RFC PATCH v1 15/15] drm/msm/a5xx: Support per-instance pagetables"
In reply to: Jordan Crouse: "[RFC PATCH v1 02/15] iommu/arm-smmu: Add split pagetable support for arm-smmu-v2"
Next in thread: Jordan Crouse: "[RFC PATCH v1 05/15] iommu/arm-smmu: Add auxiliary domain support for arm-smmuv2"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Fri, Mar 1, 2019 at 2:38 PM Jordan Crouse <jcrouse@xxxxxxxxxxxxxx> wrote:
>
> Add support for a split pagetable (TTBR0/TTBR1) scheme for
> arm-smmu-v2. If split pagetables are enabled, create a
> pagetable for TTBR1 and set up the sign extension bit so
> that all IOVAs with that bit set are mapped and translated
> from the TTBR1 pagetable.
>
> Signed-off-by: Jordan Crouse <jcrouse@xxxxxxxxxxxxxx>
> ---
>
> drivers/iommu/arm-smmu-regs.h | 18 +++++
> drivers/iommu/arm-smmu.c | 149 +++++++++++++++++++++++++++++++++++++----
> drivers/iommu/io-pgtable-arm.c | 3 +-
> 3 files changed, 154 insertions(+), 16 deletions(-)
>
> diff --git a/drivers/iommu/arm-smmu-regs.h b/drivers/iommu/arm-smmu-regs.h
> index a1226e4..56f9709 100644
> --- a/drivers/iommu/arm-smmu-regs.h
> +++ b/drivers/iommu/arm-smmu-regs.h
> @@ -193,7 +193,25 @@ enum arm_smmu_s2cr_privcfg {
> #define RESUME_RETRY (0 << 0)
> #define RESUME_TERMINATE (1 << 0)
>
> +#define TTBCR_EPD1 (1 << 23)
> +#define TTBCR_T1SZ_SHIFT 16
> +#define TTBCR_IRGN1_SHIFT 24
> +#define TTBCR_ORGN1_SHIFT 26
> +#define TTBCR_RGN_WBWA 1
> +#define TTBCR_SH1_SHIFT 28
> +#define TTBCR_SH_IS 3
> +
> +#define TTBCR_TG1_16K (1 << 30)
> +#define TTBCR_TG1_4K (2 << 30)
> +#define TTBCR_TG1_64K (3 << 30)
> +
> #define TTBCR2_SEP_SHIFT 15
> +#define TTBCR2_SEP_31 (0x0 << TTBCR2_SEP_SHIFT)
> +#define TTBCR2_SEP_35 (0x1 << TTBCR2_SEP_SHIFT)
> +#define TTBCR2_SEP_39 (0x2 << TTBCR2_SEP_SHIFT)
> +#define TTBCR2_SEP_41 (0x3 << TTBCR2_SEP_SHIFT)
> +#define TTBCR2_SEP_43 (0x4 << TTBCR2_SEP_SHIFT)
> +#define TTBCR2_SEP_47 (0x5 << TTBCR2_SEP_SHIFT)
> #define TTBCR2_SEP_UPSTREAM (0x7 << TTBCR2_SEP_SHIFT)
> #define TTBCR2_AS (1 << 4)
>
> diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
> index af18a7e..05eb126 100644
> --- a/drivers/iommu/arm-smmu.c
> +++ b/drivers/iommu/arm-smmu.c
> @@ -151,6 +151,7 @@ struct arm_smmu_cb {
> u32 tcr[2];
> u32 mair[2];
> struct arm_smmu_cfg *cfg;
> + u64 split_table_mask;
> };
>
> struct arm_smmu_master_cfg {
> @@ -208,6 +209,7 @@ struct arm_smmu_device {
> unsigned long va_size;
> unsigned long ipa_size;
> unsigned long pa_size;
> + unsigned long ubs_size;
> unsigned long pgsize_bitmap;
>
> u32 num_global_irqs;
> @@ -252,13 +254,14 @@ enum arm_smmu_domain_stage {
>
> struct arm_smmu_domain {
> struct arm_smmu_device *smmu;
> - struct io_pgtable_ops *pgtbl_ops;
> + struct io_pgtable_ops *pgtbl_ops[2];
> const struct iommu_gather_ops *tlb_ops;
> struct arm_smmu_cfg cfg;
> enum arm_smmu_domain_stage stage;
> bool non_strict;
> struct mutex init_mutex; /* Protects smmu pointer */
> spinlock_t cb_lock; /* Serialises ATS1* ops and TLB syncs */
> + u32 attributes;
> struct iommu_domain domain;
> };
>
> @@ -618,6 +621,69 @@ static irqreturn_t arm_smmu_global_fault(int irq, void *dev)
> return IRQ_HANDLED;
> }
>
> +static void arm_smmu_init_ttbr1(struct arm_smmu_domain *smmu_domain,
> + struct io_pgtable_cfg *pgtbl_cfg)
> +{
> + struct arm_smmu_device *smmu = smmu_domain->smmu;
> + struct arm_smmu_cfg *cfg = &smmu_domain->cfg;
> + struct arm_smmu_cb *cb = &smmu_domain->smmu->cbs[cfg->cbndx];
> + int pgsize = 1 << __ffs(pgtbl_cfg->pgsize_bitmap);
> +
> + /* Enable speculative walks through the TTBR1 */
> + cb->tcr[0] &= ~TTBCR_EPD1;
> +
> + cb->tcr[0] |= TTBCR_SH_IS << TTBCR_SH1_SHIFT;
> + cb->tcr[0] |= TTBCR_RGN_WBWA << TTBCR_IRGN1_SHIFT;
> + cb->tcr[0] |= TTBCR_RGN_WBWA << TTBCR_ORGN1_SHIFT;
> +
> + switch (pgsize) {
> + case SZ_4K:
> + cb->tcr[0] |= TTBCR_TG1_4K;
> + break;
> + case SZ_16K:
> + cb->tcr[0] |= TTBCR_TG1_16K;
> + break;
> + case SZ_64K:
> + cb->tcr[0] |= TTBCR_TG1_64K;
> + break;
> + }
> +
> + cb->tcr[0] |= (64ULL - smmu->va_size) << TTBCR_T1SZ_SHIFT;
> +
> + /* Clear the existing SEP configuration */
> + cb->tcr[1] &= ~TTBCR2_SEP_UPSTREAM;
> +
> + /* Set up the sign extend bit */
> + switch (smmu->va_size) {
> + case 32:
> + cb->tcr[1] |= TTBCR2_SEP_31;
> + cb->split_table_mask = (1ULL << 31);
> + break;
> + case 36:
> + cb->tcr[1] |= TTBCR2_SEP_35;
> + cb->split_table_mask = (1ULL << 35);
> + break;
> + case 40:
> + cb->tcr[1] |= TTBCR2_SEP_39;
> + cb->split_table_mask = (1ULL << 39);
> + break;
> + case 42:
> + cb->tcr[1] |= TTBCR2_SEP_41;
> + cb->split_table_mask = (1ULL << 41);
> + break;
> + case 44:
> + cb->tcr[1] |= TTBCR2_SEP_43;
> + cb->split_table_mask = (1ULL << 43);
> + break;
> + case 48:
> + cb->tcr[1] |= TTBCR2_SEP_UPSTREAM;
> + cb->split_table_mask = (1ULL << 48);
> + }
> +
> + cb->ttbr[1] = pgtbl_cfg->arm_lpae_s1_cfg.ttbr[0];
> + cb->ttbr[1] |= (u64)cfg->asid << TTBRn_ASID_SHIFT;
> +}
> +
> static void arm_smmu_init_context_bank(struct arm_smmu_domain *smmu_domain,
> struct io_pgtable_cfg *pgtbl_cfg)
> {
> @@ -650,8 +716,12 @@ static void arm_smmu_init_context_bank(struct arm_smmu_domain *smmu_domain,
> } else {
> cb->ttbr[0] = pgtbl_cfg->arm_lpae_s1_cfg.ttbr[0];
> cb->ttbr[0] |= (u64)cfg->asid << TTBRn_ASID_SHIFT;
> - cb->ttbr[1] = pgtbl_cfg->arm_lpae_s1_cfg.ttbr[1];
> - cb->ttbr[1] |= (u64)cfg->asid << TTBRn_ASID_SHIFT;
> +
> + /*
> + * Set TTBR1 to empty by default - it will get
> + * programmed later if it is enabled
> + */
> + cb->ttbr[1] = (u64)cfg->asid << TTBRn_ASID_SHIFT;
> }
> } else {
> cb->ttbr[0] = pgtbl_cfg->arm_lpae_s2_cfg.vttbr;
> @@ -760,11 +830,13 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain,
> {
> int irq, start, ret = 0;
> unsigned long ias, oas;
> - struct io_pgtable_ops *pgtbl_ops;
> + struct io_pgtable_ops *pgtbl_ops[2];
> struct io_pgtable_cfg pgtbl_cfg;
> enum io_pgtable_fmt fmt;
> struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> struct arm_smmu_cfg *cfg = &smmu_domain->cfg;
> + bool split_tables =
> + (smmu_domain->attributes & (1 << DOMAIN_ATTR_SPLIT_TABLES));

BIT(DOMAIN_ATTR_SPLIT_TABLES) ?

>
> mutex_lock(&smmu_domain->init_mutex);
> if (smmu_domain->smmu)
> @@ -794,8 +866,11 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain,
> *
> * Note that you can't actually request stage-2 mappings.
> */
> - if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S1))
> + if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S1)) {
> smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
> + /* FIXME: fail instead? */
> + split_tables = false;

yeah, I think we want to return an error somewhere if not supported.
I think we want to fall back to not using per-process pagetables if
this fails.

BR,
-R

> + }
> if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S2))
> smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
>
> @@ -812,8 +887,11 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain,
> if (IS_ENABLED(CONFIG_IOMMU_IO_PGTABLE_ARMV7S) &&
> !IS_ENABLED(CONFIG_64BIT) && !IS_ENABLED(CONFIG_ARM_LPAE) &&
> (smmu->features & ARM_SMMU_FEAT_FMT_AARCH32_S) &&
> - (smmu_domain->stage == ARM_SMMU_DOMAIN_S1))
> + (smmu_domain->stage == ARM_SMMU_DOMAIN_S1)) {
> + /* FIXME: fail instead? */
> + split_tables = false;
> cfg->fmt = ARM_SMMU_CTX_FMT_AARCH32_S;
> + }
> if ((IS_ENABLED(CONFIG_64BIT) || cfg->fmt == ARM_SMMU_CTX_FMT_NONE) &&
> (smmu->features & (ARM_SMMU_FEAT_FMT_AARCH64_64K |
> ARM_SMMU_FEAT_FMT_AARCH64_16K |
> @@ -903,8 +981,8 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain,
> pgtbl_cfg.quirks |= IO_PGTABLE_QUIRK_NON_STRICT;
>
> smmu_domain->smmu = smmu;
> - pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
> - if (!pgtbl_ops) {
> + pgtbl_ops[0] = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
> + if (!pgtbl_ops[0]) {
> ret = -ENOMEM;
> goto out_clear_smmu;
> }
> @@ -916,6 +994,22 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain,
>
> /* Initialise the context bank with our page table cfg */
> arm_smmu_init_context_bank(smmu_domain, &pgtbl_cfg);
> +
> + pgtbl_ops[1] = NULL;
> +
> + if (split_tables) {
> + /* FIXME: I think it is safe to reuse pgtbl_cfg here */
> + pgtbl_ops[1] = alloc_io_pgtable_ops(fmt, &pgtbl_cfg,
> + smmu_domain);
> + if (!pgtbl_ops[1]) {
> + free_io_pgtable_ops(pgtbl_ops[0]);
> + ret = -ENOMEM;
> + goto out_clear_smmu;
> + }
> +
> + arm_smmu_init_ttbr1(smmu_domain, &pgtbl_cfg);
> + }
> +
> arm_smmu_write_context_bank(smmu, cfg->cbndx);
>
> /*
> @@ -934,7 +1028,9 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain,
> mutex_unlock(&smmu_domain->init_mutex);
>
> /* Publish page table ops for map/unmap */
> - smmu_domain->pgtbl_ops = pgtbl_ops;
> + smmu_domain->pgtbl_ops[0] = pgtbl_ops[0];
> + smmu_domain->pgtbl_ops[1] = pgtbl_ops[1];
> +
> return 0;
>
> out_clear_smmu:
> @@ -970,7 +1066,9 @@ static void arm_smmu_destroy_domain_context(struct iommu_domain *domain)
> devm_free_irq(smmu->dev, irq, domain);
> }
>
> - free_io_pgtable_ops(smmu_domain->pgtbl_ops);
> + free_io_pgtable_ops(smmu_domain->pgtbl_ops[0]);
> + free_io_pgtable_ops(smmu_domain->pgtbl_ops[1]);
> +
> __arm_smmu_free_bitmap(smmu->context_map, cfg->cbndx);
>
> arm_smmu_rpm_put(smmu);
> @@ -1285,10 +1383,23 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
> return ret;
> }
>
> +static struct io_pgtable_ops *
> +arm_smmu_get_pgtbl_ops(struct iommu_domain *domain, unsigned long iova)
> +{
> + struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> + struct arm_smmu_cfg *cfg = &smmu_domain->cfg;
> + struct arm_smmu_cb *cb = &smmu_domain->smmu->cbs[cfg->cbndx];
> +
> + if (iova & cb->split_table_mask)
> + return smmu_domain->pgtbl_ops[1];
> +
> + return smmu_domain->pgtbl_ops[0];
> +}
> +
> static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova,
> phys_addr_t paddr, size_t size, int prot)
> {
> - struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
> + struct io_pgtable_ops *ops = arm_smmu_get_pgtbl_ops(domain, iova);
> struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu;
> int ret;
>
> @@ -1305,7 +1416,7 @@ 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)
> {
> - struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
> + struct io_pgtable_ops *ops = arm_smmu_get_pgtbl_ops(domain, iova);
> struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu;
> size_t ret;
>
> @@ -1349,7 +1460,7 @@ static phys_addr_t arm_smmu_iova_to_phys_hard(struct iommu_domain *domain,
> struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> struct arm_smmu_device *smmu = smmu_domain->smmu;
> struct arm_smmu_cfg *cfg = &smmu_domain->cfg;
> - struct io_pgtable_ops *ops= smmu_domain->pgtbl_ops;
> + struct io_pgtable_ops *ops = arm_smmu_get_pgtbl_ops(domain, iova);
> struct device *dev = smmu->dev;
> void __iomem *cb_base;
> u32 tmp;
> @@ -1397,7 +1508,7 @@ static phys_addr_t arm_smmu_iova_to_phys(struct iommu_domain *domain,
> dma_addr_t iova)
> {
> struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> - struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
> + struct io_pgtable_ops *ops = arm_smmu_get_pgtbl_ops(domain, iova);
>
> if (domain->type == IOMMU_DOMAIN_IDENTITY)
> return iova;
> @@ -1584,6 +1695,11 @@ static int arm_smmu_domain_get_attr(struct iommu_domain *domain,
> case DOMAIN_ATTR_NESTING:
> *(int *)data = (smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED);
> return 0;
> + case DOMAIN_ATTR_SPLIT_TABLES:
> + *((int *)data) =
> + !!(smmu_domain->attributes &
> + (1 << DOMAIN_ATTR_SPLIT_TABLES));
> + return 0;
> default:
> return -ENODEV;
> }
> @@ -1624,6 +1740,11 @@ static int arm_smmu_domain_set_attr(struct iommu_domain *domain,
> else
> smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
> break;
> + case DOMAIN_ATTR_SPLIT_TABLES:
> + if (*((int *)data))
> + smmu_domain->attributes |=
> + (1 << DOMAIN_ATTR_SPLIT_TABLES);
> + break;
> default:
> ret = -ENODEV;
> }
> diff --git a/drivers/iommu/io-pgtable-arm.c b/drivers/iommu/io-pgtable-arm.c
> index 237cacd..dc9fb2e 100644
> --- a/drivers/iommu/io-pgtable-arm.c
> +++ b/drivers/iommu/io-pgtable-arm.c
> @@ -475,8 +475,7 @@ static int arm_lpae_map(struct io_pgtable_ops *ops, unsigned long iova,
> if (!(iommu_prot & (IOMMU_READ | IOMMU_WRITE)))
> return 0;
>
> - if (WARN_ON(iova >= (1ULL << data->iop.cfg.ias) ||
> - paddr >= (1ULL << data->iop.cfg.oas)))
> + if (WARN_ON(paddr >= (1ULL << data->iop.cfg.oas)))
> return -ERANGE;
>
> prot = arm_lpae_prot_to_pte(data, iommu_prot);
> --
> 2.7.4
>
> _______________________________________________
> Freedreno mailing list
> Freedreno@xxxxxxxxxxxxxxxxxxxxx
> https://lists.freedesktop.org/mailman/listinfo/freedreno

Next message: Andrew Morton: "Re: [PATCH] log2: make is_power_of_2() integer constant expression when possible"
Previous message: Rob Clark: "Re: [RFC PATCH v1 15/15] drm/msm/a5xx: Support per-instance pagetables"
In reply to: Jordan Crouse: "[RFC PATCH v1 02/15] iommu/arm-smmu: Add split pagetable support for arm-smmu-v2"
Next in thread: Jordan Crouse: "[RFC PATCH v1 05/15] iommu/arm-smmu: Add auxiliary domain support for arm-smmuv2"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]