Hi Robin,

On Tue, Sep 25, 2018 at 6:01 PM Robin Murphy [off-list ref] wrote:

On 10/09/18 07:25, Vivek Gautam wrote:

quoted

Qcom's implementation of arm,mmu-500 require to serialize all
TLB invalidations for context banks.

What does "serailize all TLB invalidations" actually mean, because it's
not entirely clear from context, and furthermore this patch appears to
behave subtly differently to the downstream code so I'm really
struggling to figure out whether it's actually doing what it's intended
to do.

Adding Pratik Patel from downstream team.

Thanks for taking a look at this.
We want to space out the TLB invalidation and then the workaround
to toggle wait-safe logic would let the safe checks in HW work and
only allow invalidation to occur when device is expected to not
run into underruns.

quoted

In case the TLB invalidation requests don't go through the first
time, there's a way to disable/enable the wait for safe logic.
Disabling this logic expadites the TLBIs.

Different bootloaders with their access control policies allow this
register access differntly. With one, we should be able to directly
make qcom-scm call to do io read/write, while with other we should
use the specific SCM command to send request to do the complete
register configuration.
A separate device tree flag for arm-smmu will allow to identify
which firmware configuration of the two mentioned above we use.

Signed-off-by: Vivek Gautam <redacted>
---
  drivers/iommu/arm-smmu-regs.h |   2 +
  drivers/iommu/arm-smmu.c      | 133 +++++++++++++++++++++++++++++++++++++++++-
  2 files changed, 133 insertions(+), 2 deletions(-)

diff --git a/drivers/iommu/arm-smmu-regs.h b/drivers/iommu/arm-smmu-regs.h
index a1226e4ab5f8..71662cae9806 100644
--- a/drivers/iommu/arm-smmu-regs.h
+++ b/drivers/iommu/arm-smmu-regs.h

@@ -177,6 +177,8 @@ enum arm_smmu_s2cr_privcfg {
  #define ARM_SMMU_CB_ATS1PR          0x800
  #define ARM_SMMU_CB_ATSR            0x8f0

+#define ARM_SMMU_GID_QCOM_CUSTOM_CFG 0x300
+
  #define SCTLR_S1_ASIDPNE            (1 << 12)
  #define SCTLR_CFCFG                 (1 << 7)
  #define SCTLR_CFIE                  (1 << 6)

diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
index 411e5ac57c64..de9c4a5bf686 100644
--- a/drivers/iommu/arm-smmu.c
+++ b/drivers/iommu/arm-smmu.c

@@ -49,6 +49,7 @@
  #include <linux/pci.h>
  #include <linux/platform_device.h>
  #include <linux/pm_runtime.h>
+#include <linux/qcom_scm.h>
  #include <linux/slab.h>
  #include <linux/spinlock.h>

@@ -181,7 +182,8 @@ struct arm_smmu_device {
  #define ARM_SMMU_FEAT_EXIDS         (1 << 12)
      u32                             features;

-#define ARM_SMMU_OPT_SECURE_CFG_ACCESS (1 << 0)
+#define ARM_SMMU_OPT_SECURE_CFG_ACCESS        (1 << 0)
+#define ARM_SMMU_OPT_QCOM_FW_IMPL_ERRATA (1 << 1)
      u32                             options;
      enum arm_smmu_arch_version      version;
      enum arm_smmu_implementation    model;

@@ -266,6 +268,7 @@ static bool using_legacy_binding, using_generic_binding;

  static struct arm_smmu_option_prop arm_smmu_options[] = {
      { ARM_SMMU_OPT_SECURE_CFG_ACCESS, "calxeda,smmu-secure-config-access" },
+     { ARM_SMMU_OPT_QCOM_FW_IMPL_ERRATA, "qcom,smmu-500-fw-impl-errata" },
      { 0, NULL},
  };

@@ -531,12 +534,134 @@ static void arm_smmu_tlb_inv_vmid_nosync(unsigned long iova, size_t size,
      writel_relaxed(smmu_domain->cfg.vmid, base + ARM_SMMU_GR0_TLBIVMID);
  }

+#define CUSTOM_CFG_MDP_SAFE_ENABLE           BIT(15)
+#define CUSTOM_CFG_IFE1_SAFE_ENABLE          BIT(14)
+#define CUSTOM_CFG_IFE0_SAFE_ENABLE          BIT(13)
+
+static int __qsmmu500_wait_safe_toggle(struct arm_smmu_device *smmu, int en)
+{
+     int ret;
+     u32 val, gid_phys_base;
+     phys_addr_t reg;
+     struct vm_struct *vm;
+
+     /* We want physical address of SMMU, so the vm_area */
+     vm = find_vm_area(smmu->base);
+
+     /*
+      * GID (implementation defined address space) is located at
+      * SMMU_BASE + (2 ? PAGESIZE).
+      */
+     gid_phys_base = vm->phys_addr + (2 << (smmu)->pgshift);
+     reg = gid_phys_base + ARM_SMMU_GID_QCOM_CUSTOM_CFG;
+
+     ret = qcom_scm_io_readl_atomic(reg, &val);
+     if (ret)
+             return ret;
+
+     if (en)
+             val |= CUSTOM_CFG_MDP_SAFE_ENABLE |
+                    CUSTOM_CFG_IFE0_SAFE_ENABLE |
+                    CUSTOM_CFG_IFE1_SAFE_ENABLE;
+     else
+             val &= ~(CUSTOM_CFG_MDP_SAFE_ENABLE |
+                      CUSTOM_CFG_IFE0_SAFE_ENABLE |
+                      CUSTOM_CFG_IFE1_SAFE_ENABLE);
+
+     ret = qcom_scm_io_writel_atomic(reg, val);
+
+     return ret;
+}
+
+static int qsmmu500_wait_safe_toggle(struct arm_smmu_device *smmu,
+                                  int en, bool is_fw_impl)
+{
+     if (is_fw_impl)
+             return qcom_scm_qsmmu500_wait_safe_toggle(en);
+     else
+             return __qsmmu500_wait_safe_toggle(smmu, en);
+}
+
+static void qcom_errata_tlb_sync(struct arm_smmu_domain *smmu_domain)
+{
+     struct arm_smmu_device *smmu = smmu_domain->smmu;
+     void __iomem *base = ARM_SMMU_CB(smmu, smmu_domain->cfg.cbndx);
+     bool is_fw_impl;
+     u32 val;
+
+     writel_relaxed(0, base + ARM_SMMU_CB_TLBSYNC);
+
+     if (!readl_poll_timeout_atomic(base + ARM_SMMU_CB_TLBSTATUS, val,
+                                    !(val & sTLBGSTATUS_GSACTIVE), 0, 100))
+             return;
+
+     is_fw_impl = smmu->options & ARM_SMMU_OPT_QCOM_FW_IMPL_ERRATA ?
+                     true : false;
+
+     /* SCM call here to disable the wait-for-safe logic. */

I take it this is a global state, so it can't just be turned off for the
relevant contexts and left that way?

Yes this is a global state disable/enable. But can we disable it for
the required context altogether forever, I will have to find it out.

quoted

+     if (WARN(qsmmu500_wait_safe_toggle(smmu, false, is_fw_impl),
+              "Failed to disable wait-safe logic, bad hw state\n"))
+             return;
+
+     if (!readl_poll_timeout_atomic(base + ARM_SMMU_CB_TLBSTATUS, val,
+                                    !(val & sTLBGSTATUS_GSACTIVE), 0, 10000))
+             return;
+
+     /* SCM call here to re-enable the wait-for-safe logic. */
+     WARN(qsmmu500_wait_safe_toggle(smmu, true, is_fw_impl),
+          "Failed to re-enable wait-safe logic, bad hw state\n");
+
+     dev_err_ratelimited(smmu->dev,
+                         "TLB sync timed out -- SMMU in bad state\n");
+}
+
+static void qcom_errata_tlb_sync_context(void *cookie)
+{
+     struct arm_smmu_domain *smmu_domain = cookie;
+     unsigned long flags;
+
+     spin_lock_irqsave(&smmu_domain->cb_lock, flags);
+     qcom_errata_tlb_sync(smmu_domain);
+     spin_unlock_irqrestore(&smmu_domain->cb_lock, flags);
+}
+
+static void qcom_errata_tlb_inv_context_s1(void *cookie)
+{
+     struct arm_smmu_domain *smmu_domain = cookie;
+     struct arm_smmu_cfg *cfg = &smmu_domain->cfg;
+     void __iomem *base = ARM_SMMU_CB(smmu_domain->smmu, cfg->cbndx);
+     unsigned long flags;
+
+     spin_lock_irqsave(&smmu_domain->cb_lock, flags);
+     writel_relaxed(cfg->asid, base + ARM_SMMU_CB_S1_TLBIASID);
+     qcom_errata_tlb_sync(cookie);
+     spin_unlock_irqrestore(&smmu_domain->cb_lock, flags);
+}
+
+static void qcom_errata_tlb_inv_range_nosync(unsigned long iova, size_t size,
+                                          size_t granule, bool leaf,
+                                          void *cookie)
+{
+     struct arm_smmu_domain *smmu_domain = cookie;
+     unsigned long flags;
+
+     spin_lock_irqsave(&smmu_domain->cb_lock, flags);
+     arm_smmu_tlb_inv_range_nosync(iova, size, granule, leaf, cookie);
+     spin_unlock_irqrestore(&smmu_domain->cb_lock, flags);

I don't get what this locking is trying to achieve - the only thing it
protects is one or more writes to TLBIVA{L}, which *must* surely be
"serialised" by the interconnect anyway?

Okay, right. This callback is just a write to TLBIVA{L} registers, so this
should be good without any locking.
A subsequent .tlb_sync has the locking anyways.

The downstream code doesn't appear to implement .tlb_add_flush at all,
so something smells off.

Yes, the downstream pg-table driver never uses tlb_add_flush and
tlb_sync to do TLBIs.
Rather it relies on tlb_flush_all to flush the entire TLB after the
unmap is finished.
Moreover, in downstream a global remote spinlock is used to protect
the invalidation
requests owing to other entities besides kernel handling the TLB operations.

Assuming we don't want to do tlb_flush_all, we still want to serialize
the invalidation
requests, and then toggle the wait-safe logic too.
We would also not want the remote spinlock as we are invalidating based on ASIDs
in the kernel.

Regards
Vivek

Robin.

quoted

+}
+
  static const struct iommu_gather_ops arm_smmu_s1_tlb_ops = {
      .tlb_flush_all  = arm_smmu_tlb_inv_context_s1,
      .tlb_add_flush  = arm_smmu_tlb_inv_range_nosync,
      .tlb_sync       = arm_smmu_tlb_sync_context,
  };

+static const struct iommu_gather_ops qcom_errata_s1_tlb_ops = {
+     .tlb_flush_all  = qcom_errata_tlb_inv_context_s1,
+     .tlb_add_flush  = qcom_errata_tlb_inv_range_nosync,
+     .tlb_sync       = qcom_errata_tlb_sync_context,
+};
+
  static const struct iommu_gather_ops arm_smmu_s2_tlb_ops_v2 = {
      .tlb_flush_all  = arm_smmu_tlb_inv_context_s2,
      .tlb_add_flush  = arm_smmu_tlb_inv_range_nosync,

@@ -824,7 +949,11 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain,
                      ias = min(ias, 32UL);
                      oas = min(oas, 32UL);
              }
-             smmu_domain->tlb_ops = &arm_smmu_s1_tlb_ops;
+             if (of_device_is_compatible(smmu->dev->of_node,
+                                         "qcom,sdm845-smmu-500"))
+                     smmu_domain->tlb_ops = &qcom_errata_s1_tlb_ops;
+             else
+                     smmu_domain->tlb_ops = &arm_smmu_s1_tlb_ops;
              break;
      case ARM_SMMU_DOMAIN_NESTED:
              /*

_______________________________________________
iommu mailing list
iommu at lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation