[v3,03/13] hw/riscv: add RISC-V IOMMU base emulation

From: Tomasz Jeznach <tjeznach@rivosinc.com>

The RISC-V IOMMU specification is now ratified as-per the RISC-V
international process. The latest frozen specifcation can be found
at:

https://github.com/riscv-non-isa/riscv-iommu/releases/download/v1.0/riscv-iommu.pdf

Add the foundation of the device emulation for RISC-V IOMMU, which
includes an IOMMU that has no capabilities but MSI interrupt support and
fault queue interfaces. We'll add add more features incrementally in the
next patches.

Co-developed-by: Sebastien Boeuf <seb@rivosinc.com>
Signed-off-by: Sebastien Boeuf <seb@rivosinc.com>
Signed-off-by: Tomasz Jeznach <tjeznach@rivosinc.com>
Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
---
 hw/riscv/Kconfig         |    4 +
 hw/riscv/meson.build     |    1 +
 hw/riscv/riscv-iommu.c   | 1602 ++++++++++++++++++++++++++++++++++++++
 hw/riscv/riscv-iommu.h   |  141 ++++
 hw/riscv/trace-events    |   11 +
 hw/riscv/trace.h         |    1 +
 include/hw/riscv/iommu.h |   36 +
 meson.build              |    1 +
 8 files changed, 1797 insertions(+)
 create mode 100644 hw/riscv/riscv-iommu.c
 create mode 100644 hw/riscv/riscv-iommu.h
 create mode 100644 hw/riscv/trace-events
 create mode 100644 hw/riscv/trace.h
 create mode 100644 include/hw/riscv/iommu.h

On 5/24/2024 1:39 AM, Daniel Henrique Barboza wrote:
> From: Tomasz Jeznach <tjeznach@rivosinc.com>
> 
> The RISC-V IOMMU specification is now ratified as-per the RISC-V
> international process. The latest frozen specifcation can be found
> at:
> 
> https://github.com/riscv-non-isa/riscv-iommu/releases/download/v1.0/riscv-iommu.pdf
> 
> Add the foundation of the device emulation for RISC-V IOMMU, which
> includes an IOMMU that has no capabilities but MSI interrupt support and
> fault queue interfaces. We'll add add more features incrementally in the
						      ^^^  ^^^
repeated 'add'

> next patches.
> 
> Co-developed-by: Sebastien Boeuf <seb@rivosinc.com>
> Signed-off-by: Sebastien Boeuf <seb@rivosinc.com>
> Signed-off-by: Tomasz Jeznach <tjeznach@rivosinc.com>
> Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
> ---
>   hw/riscv/Kconfig         |    4 +
>   hw/riscv/meson.build     |    1 +
>   hw/riscv/riscv-iommu.c   | 1602 ++++++++++++++++++++++++++++++++++++++
>   hw/riscv/riscv-iommu.h   |  141 ++++
>   hw/riscv/trace-events    |   11 +
>   hw/riscv/trace.h         |    1 +
>   include/hw/riscv/iommu.h |   36 +
>   meson.build              |    1 +
>   8 files changed, 1797 insertions(+)
>   create mode 100644 hw/riscv/riscv-iommu.c
>   create mode 100644 hw/riscv/riscv-iommu.h
>   create mode 100644 hw/riscv/trace-events
>   create mode 100644 hw/riscv/trace.h
>   create mode 100644 include/hw/riscv/iommu.h
> 
> diff --git a/hw/riscv/Kconfig b/hw/riscv/Kconfig
> index a2030e3a6f..f69d6e3c8e 100644
> --- a/hw/riscv/Kconfig
> +++ b/hw/riscv/Kconfig
> @@ -1,3 +1,6 @@
> +config RISCV_IOMMU
> +    bool
> +
>   config RISCV_NUMA
>       bool
>   
> @@ -47,6 +50,7 @@ config RISCV_VIRT
>       select SERIAL
>       select RISCV_ACLINT
>       select RISCV_APLIC
> +    select RISCV_IOMMU
>       select RISCV_IMSIC
>       select SIFIVE_PLIC
>       select SIFIVE_TEST
> diff --git a/hw/riscv/meson.build b/hw/riscv/meson.build
> index f872674093..cbc99c6e8e 100644
> --- a/hw/riscv/meson.build
> +++ b/hw/riscv/meson.build
> @@ -10,5 +10,6 @@ riscv_ss.add(when: 'CONFIG_SIFIVE_U', if_true: files('sifive_u.c'))
>   riscv_ss.add(when: 'CONFIG_SPIKE', if_true: files('spike.c'))
>   riscv_ss.add(when: 'CONFIG_MICROCHIP_PFSOC', if_true: files('microchip_pfsoc.c'))
>   riscv_ss.add(when: 'CONFIG_ACPI', if_true: files('virt-acpi-build.c'))
> +riscv_ss.add(when: 'CONFIG_RISCV_IOMMU', if_true: files('riscv-iommu.c'))
>   
>   hw_arch += {'riscv': riscv_ss}
> diff --git a/hw/riscv/riscv-iommu.c b/hw/riscv/riscv-iommu.c
> new file mode 100644
> index 0000000000..39b4ff1405
> --- /dev/null
> +++ b/hw/riscv/riscv-iommu.c
> @@ -0,0 +1,1602 @@
> +/*
> + * QEMU emulation of an RISC-V IOMMU
> + *
> + * Copyright (C) 2021-2023, Rivos Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#include "qemu/osdep.h"
> +#include "qom/object.h"
> +#include "hw/pci/pci_bus.h"
> +#include "hw/pci/pci_device.h"
> +#include "hw/qdev-properties.h"
> +#include "hw/riscv/riscv_hart.h"
> +#include "migration/vmstate.h"
> +#include "qapi/error.h"
> +#include "qemu/timer.h"
> +
> +#include "cpu_bits.h"
> +#include "riscv-iommu.h"
> +#include "riscv-iommu-bits.h"
> +#include "trace.h"
> +
> +#define LIMIT_CACHE_CTX               (1U << 7)
> +#define LIMIT_CACHE_IOT               (1U << 20)
> +
> +/* Physical page number coversions */
> +#define PPN_PHYS(ppn)                 ((ppn) << TARGET_PAGE_BITS)
> +#define PPN_DOWN(phy)                 ((phy) >> TARGET_PAGE_BITS)
> +
> +typedef struct RISCVIOMMUContext RISCVIOMMUContext;
> +typedef struct RISCVIOMMUEntry RISCVIOMMUEntry;
> +
> +/* Device assigned I/O address space */
> +struct RISCVIOMMUSpace {
> +    IOMMUMemoryRegion iova_mr;  /* IOVA memory region for attached device */
> +    AddressSpace iova_as;       /* IOVA address space for attached device */
> +    RISCVIOMMUState *iommu;     /* Managing IOMMU device state */
> +    uint32_t devid;             /* Requester identifier, AKA device_id */
> +    bool notifier;              /* IOMMU unmap notifier enabled */
> +    QLIST_ENTRY(RISCVIOMMUSpace) list;
> +};
> +
> +/* Device translation context state. */
> +struct RISCVIOMMUContext {
> +    uint64_t devid:24;          /* Requester Id, AKA device_id */
> +    uint64_t pasid:20;          /* Process Address Space ID */
> +    uint64_t __rfu:20;          /* reserved */
> +    uint64_t tc;                /* Translation Control */
> +    uint64_t ta;                /* Translation Attributes */
> +    uint64_t msi_addr_mask;     /* MSI filtering - address mask */
> +    uint64_t msi_addr_pattern;  /* MSI filtering - address pattern */
> +    uint64_t msiptp;            /* MSI redirection page table pointer */
> +};

Can we alias use (devid + pasid + __rfu) by union? so that can easily compare 
the key, esp. I assume functions like __ctx_equal() are on hot path.

And, pasid, is the term in PCI context. I suggest use more general name in spec: 
process_id.

e.g. below (just compiled, not tested)

diff --git a/hw/riscv/riscv-iommu.c b/hw/riscv/riscv-iommu.c
index 1b34d226f9..74011c7f1f 100644
--- a/hw/riscv/riscv-iommu.c
+++ b/hw/riscv/riscv-iommu.c
@@ -53,9 +53,12 @@ struct RISCVIOMMUSpace {

  /* Device translation context state. */
  struct RISCVIOMMUContext {
-    uint64_t devid:24;          /* Requester Id, AKA device_id */
-    uint64_t pasid:20;          /* Process Address Space ID */
-    uint64_t __rfu:20;          /* reserved */
+    union {
+        uint64_t devid:24,          /* Requester Id, AKA device_id */
+                 pasid:20,          /* Process Address Space ID */
+                 __rfu:20;          /* reserved */
+        uint64_t key;
+    };
      uint64_t tc;                /* Translation Control */
      uint64_t ta;                /* Translation Attributes */
      uint64_t satp;              /* S-Stage address translation and protection */
@@ -943,14 +946,14 @@ static gboolean __ctx_equal(gconstpointer v1, 
gconstpointer v2)
  {
      RISCVIOMMUContext *c1 = (RISCVIOMMUContext *) v1;
      RISCVIOMMUContext *c2 = (RISCVIOMMUContext *) v2;
-    return c1->devid == c2->devid && c1->pasid == c2->pasid;
+    return c1->key == c2->key;
  }

  static guint __ctx_hash(gconstpointer v)
  {
      RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) v;
      /* Generate simple hash of (pasid, devid), assuming 24-bit wide devid */
-    return (guint)(ctx->devid) + ((guint)(ctx->pasid) << 24);
+    return (guint)ctx->key;
  }

  static void __ctx_inval_devid_pasid(gpointer key, gpointer value, gpointer data)
@@ -958,8 +961,7 @@ static void __ctx_inval_devid_pasid(gpointer key, gpointer 
value, gpointer data)
      RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
      RISCVIOMMUContext *arg = (RISCVIOMMUContext *) data;
      if (ctx->tc & RISCV_IOMMU_DC_TC_V &&
-        ctx->devid == arg->devid &&
-        ctx->pasid == arg->pasid) {
+        ctx->key == arg->key) {
          ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
      }
  }
@@ -989,6 +991,7 @@ static void riscv_iommu_ctx_inval(RISCVIOMMUState *s, GHFunc 
func,
      RISCVIOMMUContext key = {
          .devid = devid,
          .pasid = pasid,
+        .__rfu = 0,
      };
      ctx_cache = g_hash_table_ref(s->ctx_cache);
      g_hash_table_foreach(ctx_cache, func, &key);
@@ -1004,6 +1007,7 @@ static RISCVIOMMUContext *riscv_iommu_ctx(RISCVIOMMUState *s,
      RISCVIOMMUContext key = {
          .devid = devid,
          .pasid = pasid,
+        .__rfu = 0,
      };

      ctx_cache = g_hash_table_ref(s->ctx_cache);


> +
> +/* IOMMU index for transactions without PASID specified. */
> +#define RISCV_IOMMU_NOPASID 0
> +
> +static void riscv_iommu_notify(RISCVIOMMUState *s, int vec)
> +{
> +    const uint32_t fctl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FCTL);
> +    uint32_t ipsr, ivec;
> +
> +    if (fctl & RISCV_IOMMU_FCTL_WSI || !s->notify) {
> +        return;
> +    }
> +
> +    ipsr = riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_IPSR, (1 << vec), 0);
> +    ivec = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_IVEC);
> +
> +    if (!(ipsr & (1 << vec))) {
> +        s->notify(s, (ivec >> (vec * 4)) & 0x0F);
> +    }
> +}
> +
> +static void riscv_iommu_fault(RISCVIOMMUState *s,
> +                              struct riscv_iommu_fq_record *ev)
> +{
> +    uint32_t ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
> +    uint32_t head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQH) & s->fq_mask;
> +    uint32_t tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQT) & s->fq_mask;
> +    uint32_t next = (tail + 1) & s->fq_mask;
> +    uint32_t devid = get_field(ev->hdr, RISCV_IOMMU_FQ_HDR_DID);
> +
> +    trace_riscv_iommu_flt(s->parent_obj.id, PCI_BUS_NUM(devid), PCI_SLOT(devid),
> +                          PCI_FUNC(devid), ev->hdr, ev->iotval);
> +
> +    if (!(ctrl & RISCV_IOMMU_FQCSR_FQON) ||
> +        !!(ctrl & (RISCV_IOMMU_FQCSR_FQOF | RISCV_IOMMU_FQCSR_FQMF))) {
> +        return;
> +    }
> +
> +    if (head == next) {
> +        riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR,
> +                              RISCV_IOMMU_FQCSR_FQOF, 0);
> +    } else {
> +        dma_addr_t addr = s->fq_addr + tail * sizeof(*ev);
> +        if (dma_memory_write(s->target_as, addr, ev, sizeof(*ev),
> +                             MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR,
> +                                  RISCV_IOMMU_FQCSR_FQMF, 0);
> +        } else {
> +            riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_FQT, next);
> +        }
> +    }
> +
> +    if (ctrl & RISCV_IOMMU_FQCSR_FIE) {
> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_FQ);
> +    }
> +}
> +
> +static void riscv_iommu_pri(RISCVIOMMUState *s,
> +    struct riscv_iommu_pq_record *pr)
> +{
> +    uint32_t ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
> +    uint32_t head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQH) & s->pq_mask;
> +    uint32_t tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQT) & s->pq_mask;
> +    uint32_t next = (tail + 1) & s->pq_mask;
> +    uint32_t devid = get_field(pr->hdr, RISCV_IOMMU_PREQ_HDR_DID);
> +
> +    trace_riscv_iommu_pri(s->parent_obj.id, PCI_BUS_NUM(devid), PCI_SLOT(devid),
> +                          PCI_FUNC(devid), pr->payload);
> +
> +    if (!(ctrl & RISCV_IOMMU_PQCSR_PQON) ||
> +        !!(ctrl & (RISCV_IOMMU_PQCSR_PQOF | RISCV_IOMMU_PQCSR_PQMF))) {
> +        return;
> +    }
> +
> +    if (head == next) {
> +        riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR,
> +                              RISCV_IOMMU_PQCSR_PQOF, 0);
> +    } else {
> +        dma_addr_t addr = s->pq_addr + tail * sizeof(*pr);
> +        if (dma_memory_write(s->target_as, addr, pr, sizeof(*pr),
> +                             MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR,
> +                                  RISCV_IOMMU_PQCSR_PQMF, 0);
> +        } else {
> +            riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_PQT, next);
> +        }
> +    }
> +
> +    if (ctrl & RISCV_IOMMU_PQCSR_PIE) {
> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_PQ);
> +    }
> +}
> +
> +/* Portable implementation of pext_u64, bit-mask extraction. */
> +static uint64_t _pext_u64(uint64_t val, uint64_t ext)
> +{
> +    uint64_t ret = 0;
> +    uint64_t rot = 1;
> +
> +    while (ext) {
> +        if (ext & 1) {
> +            if (val & 1) {
> +                ret |= rot;
> +            }
> +            rot <<= 1;
> +        }
> +        val >>= 1;
> +        ext >>= 1;
> +    }
> +
> +    return ret;
> +}
> +
> +/* Check if GPA matches MSI/MRIF pattern. */
> +static bool riscv_iommu_msi_check(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
> +    dma_addr_t gpa)
> +{
> +    if (get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_MODE) !=
> +        RISCV_IOMMU_DC_MSIPTP_MODE_FLAT) {
> +        return false; /* Invalid MSI/MRIF mode */
> +    }
> +
> +    if ((PPN_DOWN(gpa) ^ ctx->msi_addr_pattern) & ~ctx->msi_addr_mask) {
> +        return false; /* GPA not in MSI range defined by AIA IMSIC rules. */
> +    }
> +
> +    return true;
> +}
> +
> +/* RISCV IOMMU Address Translation Lookup - Page Table Walk */
> +static int riscv_iommu_spa_fetch(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
> +    IOMMUTLBEntry *iotlb)
> +{
> +    /* Early check for MSI address match when IOVA == GPA */
> +    if (iotlb->perm & IOMMU_WO &&
> +        riscv_iommu_msi_check(s, ctx, iotlb->iova)) {
> +        iotlb->target_as = &s->trap_as;
> +        iotlb->translated_addr = iotlb->iova;
> +        iotlb->addr_mask = ~TARGET_PAGE_MASK;
> +        return 0;
> +    }
> +
> +    /* Exit early for pass-through mode. */
> +    iotlb->translated_addr = iotlb->iova;
> +    iotlb->addr_mask = ~TARGET_PAGE_MASK;
> +    /* Allow R/W in pass-through mode */
> +    iotlb->perm = IOMMU_RW;
> +    return 0;
> +}
> +
> +static void riscv_iommu_report_fault(RISCVIOMMUState *s,
> +                                     RISCVIOMMUContext *ctx,
> +                                     uint32_t fault_type, uint32_t cause,
> +                                     bool pv,
> +                                     uint64_t iotval, uint64_t iotval2)
> +{
> +    struct riscv_iommu_fq_record ev = { 0 };
> +
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_DTF) {
> +        switch (cause) {
> +        case RISCV_IOMMU_FQ_CAUSE_DMA_DISABLED:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_INVALID:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_CORRUPTED:
> +        case RISCV_IOMMU_FQ_CAUSE_INTERNAL_DP_ERROR:
> +        case RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT:
> +            break;
> +        default:
> +            /* DTF prevents reporting a fault for this given cause */
> +            return;
> +        }
> +    }
> +
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_CAUSE, cause);
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_TTYPE, fault_type);
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_DID, ctx->devid);
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_PV, true);
> +
> +    if (pv) {
> +        ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_PID, ctx->pasid);
> +    }
> +
> +    ev.iotval = iotval;
> +    ev.iotval2 = iotval2;
> +
> +    riscv_iommu_fault(s, &ev);
> +}
> +
> +/* Redirect MSI write for given GPA. */
> +static MemTxResult riscv_iommu_msi_write(RISCVIOMMUState *s,
> +    RISCVIOMMUContext *ctx, uint64_t gpa, uint64_t data,
> +    unsigned size, MemTxAttrs attrs)
> +{
> +    MemTxResult res;
> +    dma_addr_t addr;
> +    uint64_t intn;
> +    uint32_t n190;
> +    uint64_t pte[2];
> +    int fault_type = RISCV_IOMMU_FQ_TTYPE_UADDR_WR;
> +    int cause;
> +
> +    if (!riscv_iommu_msi_check(s, ctx, gpa)) {
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    /* Interrupt File Number */
> +    intn = _pext_u64(PPN_DOWN(gpa), ctx->msi_addr_mask);
> +    if (intn >= 256) {
> +        /* Interrupt file number out of range */
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    /* fetch MSI PTE */
> +    addr = PPN_PHYS(get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_PPN));
> +    addr = addr | (intn * sizeof(pte));
> +    res = dma_memory_read(s->target_as, addr, &pte, sizeof(pte),
> +            MEMTXATTRS_UNSPECIFIED);
> +    if (res != MEMTX_OK) {
> +        if (res == MEMTX_DECODE_ERROR) {
> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_PT_CORRUPTED;
> +        } else {
> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        }
> +        goto err;
> +    }
> +
> +    le64_to_cpus(&pte[0]);
> +    le64_to_cpus(&pte[1]);
> +
> +    if (!(pte[0] & RISCV_IOMMU_MSI_PTE_V) || (pte[0] & RISCV_IOMMU_MSI_PTE_C)) {
> +        /*
> +         * The spec mentions that: "If msipte.C == 1, then further
> +         * processing to interpret the PTE is implementation
> +         * defined.". We'll abort with cause = 262 for this
> +         * case too.
> +         */
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_INVALID;
> +        goto err;
> +    }
> +
> +    switch (get_field(pte[0], RISCV_IOMMU_MSI_PTE_M)) {
> +    case RISCV_IOMMU_MSI_PTE_M_BASIC:
> +        /* MSI Pass-through mode */
> +        addr = PPN_PHYS(get_field(pte[0], RISCV_IOMMU_MSI_PTE_PPN));
> +        addr = addr | (gpa & TARGET_PAGE_MASK);
> +
> +        trace_riscv_iommu_msi(s->parent_obj.id, PCI_BUS_NUM(ctx->devid),
> +                              PCI_SLOT(ctx->devid), PCI_FUNC(ctx->devid),
> +                              gpa, addr);
> +
> +        res = dma_memory_write(s->target_as, addr, &data, size, attrs);
> +        if (res != MEMTX_OK) {
> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
> +            goto err;
> +        }
> +
> +        return MEMTX_OK;
> +    case RISCV_IOMMU_MSI_PTE_M_MRIF:
> +        /* MRIF mode, continue. */
> +        break;
> +    default:
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_MISCONFIGURED;
> +        goto err;
> +    }
> +
> +    /*
> +     * Report an error for interrupt identities exceeding the maximum allowed
> +     * for an IMSIC interrupt file (2047) or destination address is not 32-bit
> +     * aligned. See IOMMU Specification, Chapter 2.3. MSI page tables.
> +     */
> +    if ((data > 2047) || (gpa & 3)) {
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_MISCONFIGURED;
> +        goto err;
> +    }
> +
> +    /* MSI MRIF mode, non atomic pending bit update */
> +
> +    /* MRIF pending bit address */
> +    addr = get_field(pte[0], RISCV_IOMMU_MSI_PTE_MRIF_ADDR) << 9;
> +    addr = addr | ((data & 0x7c0) >> 3);
> +
> +    trace_riscv_iommu_msi(s->parent_obj.id, PCI_BUS_NUM(ctx->devid),
> +                          PCI_SLOT(ctx->devid), PCI_FUNC(ctx->devid),
> +                          gpa, addr);
> +
> +    /* MRIF pending bit mask */
> +    data = 1ULL << (data & 0x03f);
> +    res = dma_memory_read(s->target_as, addr, &intn, sizeof(intn), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    intn = intn | data;
> +    res = dma_memory_write(s->target_as, addr, &intn, sizeof(intn), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
> +        goto err;
> +    }
> +
> +    /* Get MRIF enable bits */
> +    addr = addr + sizeof(intn);
> +    res = dma_memory_read(s->target_as, addr, &intn, sizeof(intn), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    if (!(intn & data)) {
> +        /* notification disabled, MRIF update completed. */
> +        return MEMTX_OK;
> +    }
> +
> +    /* Send notification message */
> +    addr = PPN_PHYS(get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NPPN));
> +    n190 = get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NID) |
> +          (get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NID_MSB) << 10);
> +
> +    res = dma_memory_write(s->target_as, addr, &n190, sizeof(n190), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
> +        goto err;
> +    }
> +
> +    return MEMTX_OK;
> +
> +err:
> +    riscv_iommu_report_fault(s, ctx, fault_type, cause,
> +                             !!ctx->pasid, 0, 0);
> +    return res;
> +}
> +
> +/*
> + * Check device context configuration as described by the
> + * riscv-iommu spec section "Device-context configuration
> + * checks".
> + */
> +static bool riscv_iommu_validate_device_ctx(RISCVIOMMUState *s,
> +                                            RISCVIOMMUContext *ctx)
> +{
> +    uint32_t msi_mode;
> +
> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_EN_PRI) &&
> +        ctx->tc & RISCV_IOMMU_DC_TC_PRPR) {
> +        return false;
> +    }
> +
> +    if (!(s->cap & RISCV_IOMMU_CAP_T2GPA) &&
> +        ctx->tc & RISCV_IOMMU_DC_TC_T2GPA) {
> +        return false;
> +    }
> +
> +    if (s->cap & RISCV_IOMMU_CAP_MSI_FLAT) {
> +        msi_mode = get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_MODE);
> +
> +        if (msi_mode != RISCV_IOMMU_DC_MSIPTP_MODE_OFF &&
> +            msi_mode != RISCV_IOMMU_DC_MSIPTP_MODE_FLAT) {
> +            return false;
> +        }
> +    }
> +
> +    /*
> +     * CAP_END is always zero (only one endianess). FCTL_BE is
> +     * always zero (little-endian accesses). Thus TC_SBE must
> +     * always be LE, i.e. zero.
> +     */
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_SBE) {
> +        return false;
> +    }
> +
> +    return true;
> +}
> +
> +/*
> + * RISC-V IOMMU Device Context Loopkup - Device Directory Tree Walk
> + *
> + * @s         : IOMMU Device State
> + * @ctx       : Device Translation Context with devid and pasid set.
> + * @return    : success or fault code.
> + */
> +static int riscv_iommu_ctx_fetch(RISCVIOMMUState *s, RISCVIOMMUContext *ctx)
> +{
> +    const uint64_t ddtp = s->ddtp;
> +    unsigned mode = get_field(ddtp, RISCV_IOMMU_DDTP_MODE);
> +    dma_addr_t addr = PPN_PHYS(get_field(ddtp, RISCV_IOMMU_DDTP_PPN));
> +    struct riscv_iommu_dc dc;
> +    /* Device Context format: 0: extended (64 bytes) | 1: base (32 bytes) */
> +    const int dc_fmt = !s->enable_msi;
> +    const size_t dc_len = sizeof(dc) >> dc_fmt;
> +    unsigned depth;
> +    uint64_t de;
> +
> +    switch (mode) {
> +    case RISCV_IOMMU_DDTP_MODE_OFF:
> +        return RISCV_IOMMU_FQ_CAUSE_DMA_DISABLED;
> +
> +    case RISCV_IOMMU_DDTP_MODE_BARE:
> +        /* mock up pass-through translation context */
> +        ctx->tc = RISCV_IOMMU_DC_TC_V;
> +        ctx->ta = 0;
> +        ctx->msiptp = 0;
> +        return 0;
> +
> +    case RISCV_IOMMU_DDTP_MODE_1LVL:
> +        depth = 0;
> +        break;
> +
> +    case RISCV_IOMMU_DDTP_MODE_2LVL:
> +        depth = 1;
> +        break;
> +
> +    case RISCV_IOMMU_DDTP_MODE_3LVL:
> +        depth = 2;
> +        break;
> +
> +    default:
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
> +    }
> +
> +    /*
> +     * Check supported device id width (in bits).
> +     * See IOMMU Specification, Chapter 6. Software guidelines.
> +     * - if extended device-context format is used:
> +     *   1LVL: 6, 2LVL: 15, 3LVL: 24
> +     * - if base device-context format is used:
> +     *   1LVL: 7, 2LVL: 16, 3LVL: 24
> +     */
> +    if (ctx->devid >= (1 << (depth * 9 + 6 + (dc_fmt && depth != 2)))) {
> +        return RISCV_IOMMU_FQ_CAUSE_TTYPE_BLOCKED;
> +    }
> +
> +    /* Device directory tree walk */
> +    for (; depth-- > 0; ) {
> +        /*
> +         * Select device id index bits based on device directory tree level
> +         * and device context format.
> +         * See IOMMU Specification, Chapter 2. Data Structures.
> +         * - if extended device-context format is used:
> +         *   device index: [23:15][14:6][5:0]
> +         * - if base device-context format is used:
> +         *   device index: [23:16][15:7][6:0]
> +         */
> +        const int split = depth * 9 + 6 + dc_fmt;
> +        addr |= ((ctx->devid >> split) << 3) & ~TARGET_PAGE_MASK;
> +        if (dma_memory_read(s->target_as, addr, &de, sizeof(de),
> +                            MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            return RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT;
> +        }
> +        le64_to_cpus(&de);
> +        if (!(de & RISCV_IOMMU_DDTE_VALID)) {
> +            /* invalid directory entry */
> +            return RISCV_IOMMU_FQ_CAUSE_DDT_INVALID;
> +        }
> +        if (de & ~(RISCV_IOMMU_DDTE_PPN | RISCV_IOMMU_DDTE_VALID)) {
> +            /* reserved bits set */
> +            return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
> +        }
> +        addr = PPN_PHYS(get_field(de, RISCV_IOMMU_DDTE_PPN));
> +    }
> +
> +    /* index into device context entry page */
> +    addr |= (ctx->devid * dc_len) & ~TARGET_PAGE_MASK;
> +
> +    memset(&dc, 0, sizeof(dc));
> +    if (dma_memory_read(s->target_as, addr, &dc, dc_len,
> +                        MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT;
> +    }
> +
> +    /* Set translation context. */
> +    ctx->tc = le64_to_cpu(dc.tc);
> +    ctx->ta = le64_to_cpu(dc.ta);
> +    ctx->msiptp = le64_to_cpu(dc.msiptp);
> +    ctx->msi_addr_mask = le64_to_cpu(dc.msi_addr_mask);
> +    ctx->msi_addr_pattern = le64_to_cpu(dc.msi_addr_pattern);
> +
> +    if (!riscv_iommu_validate_device_ctx(s, ctx)) {
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
> +    }
> +
> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_V)) {
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_INVALID;
> +    }
> +
> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_PDTV)) {
> +        if (ctx->pasid != RISCV_IOMMU_NOPASID) {
> +            /* PASID is disabled */
> +            return RISCV_IOMMU_FQ_CAUSE_TTYPE_BLOCKED;
> +        }
> +        return 0;
> +    }
> +
> +    /* FSC.TC.PDTV enabled */
> +    if (mode > RISCV_IOMMU_DC_FSC_PDTP_MODE_PD20) {
> +        /* Invalid PDTP.MODE */
> +        return RISCV_IOMMU_FQ_CAUSE_PDT_MISCONFIGURED;
> +    }
> +
> +    for (depth = mode - RISCV_IOMMU_DC_FSC_PDTP_MODE_PD8; depth-- > 0; ) {
> +        /*
> +         * Select process id index bits based on process directory tree
> +         * level. See IOMMU Specification, 2.2. Process-Directory-Table.
> +         */
> +        const int split = depth * 9 + 8;
> +        addr |= ((ctx->pasid >> split) << 3) & ~TARGET_PAGE_MASK;
> +        if (dma_memory_read(s->target_as, addr, &de, sizeof(de),
> +                            MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            return RISCV_IOMMU_FQ_CAUSE_PDT_LOAD_FAULT;
> +        }
> +        le64_to_cpus(&de);
> +        if (!(de & RISCV_IOMMU_PC_TA_V)) {
> +            return RISCV_IOMMU_FQ_CAUSE_PDT_INVALID;
> +        }
> +        addr = PPN_PHYS(get_field(de, RISCV_IOMMU_PC_FSC_PPN));
> +    }
> +
> +    /* Leaf entry in PDT */
> +    addr |= (ctx->pasid << 4) & ~TARGET_PAGE_MASK;
> +    if (dma_memory_read(s->target_as, addr, &dc.ta, sizeof(uint64_t) * 2,
> +                        MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +        return RISCV_IOMMU_FQ_CAUSE_PDT_LOAD_FAULT;
> +    }
> +
> +    /* Use FSC and TA from process directory entry. */
> +    ctx->ta = le64_to_cpu(dc.ta);
> +
> +    return 0;
> +}
> +
> +/* Translation Context cache support */
> +static gboolean __ctx_equal(gconstpointer v1, gconstpointer v2)
> +{
> +    RISCVIOMMUContext *c1 = (RISCVIOMMUContext *) v1;
> +    RISCVIOMMUContext *c2 = (RISCVIOMMUContext *) v2;
> +    return c1->devid == c2->devid && c1->pasid == c2->pasid;
> +}
> +
> +static guint __ctx_hash(gconstpointer v)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) v;
> +    /* Generate simple hash of (pasid, devid), assuming 24-bit wide devid */
> +    return (guint)(ctx->devid) + ((guint)(ctx->pasid) << 24);
> +}
> +
> +static void __ctx_inval_devid_pasid(gpointer key, gpointer value, gpointer data)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
> +    RISCVIOMMUContext *arg = (RISCVIOMMUContext *) data;
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V &&
> +        ctx->devid == arg->devid &&
> +        ctx->pasid == arg->pasid) {
> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
> +    }
> +}
> +
> +static void __ctx_inval_devid(gpointer key, gpointer value, gpointer data)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
> +    RISCVIOMMUContext *arg = (RISCVIOMMUContext *) data;
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V &&
> +        ctx->devid == arg->devid) {
> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
> +    }
> +}
> +
> +static void __ctx_inval_all(gpointer key, gpointer value, gpointer data)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V) {
> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
> +    }
> +}
> +
> +static void riscv_iommu_ctx_inval(RISCVIOMMUState *s, GHFunc func,
> +    uint32_t devid, uint32_t pasid)
> +{
> +    GHashTable *ctx_cache;
> +    RISCVIOMMUContext key = {
> +        .devid = devid,
> +        .pasid = pasid,
> +    };
> +    ctx_cache = g_hash_table_ref(s->ctx_cache);
> +    g_hash_table_foreach(ctx_cache, func, &key);
> +    g_hash_table_unref(ctx_cache);
> +}
> +
> +/* Find or allocate translation context for a given {device_id, process_id} */
> +static RISCVIOMMUContext *riscv_iommu_ctx(RISCVIOMMUState *s,
> +    unsigned devid, unsigned pasid, void **ref)
> +{
> +    GHashTable *ctx_cache;
> +    RISCVIOMMUContext *ctx;
> +    RISCVIOMMUContext key = {
> +        .devid = devid,
> +        .pasid = pasid,
> +    };
> +
> +    ctx_cache = g_hash_table_ref(s->ctx_cache);
> +    ctx = g_hash_table_lookup(ctx_cache, &key);
> +
> +    if (ctx && (ctx->tc & RISCV_IOMMU_DC_TC_V)) {
> +        *ref = ctx_cache;
> +        return ctx;
> +    }
> +
> +    if (g_hash_table_size(s->ctx_cache) >= LIMIT_CACHE_CTX) {
> +        ctx_cache = g_hash_table_new_full(__ctx_hash, __ctx_equal,
> +                                          g_free, NULL);
> +        g_hash_table_unref(qatomic_xchg(&s->ctx_cache, ctx_cache));
> +    }
> +
> +    ctx = g_new0(RISCVIOMMUContext, 1);
> +    ctx->devid = devid;
> +    ctx->pasid = pasid;
> +
> +    int fault = riscv_iommu_ctx_fetch(s, ctx);
> +    if (!fault) {
> +        g_hash_table_add(ctx_cache, ctx);
> +        *ref = ctx_cache;
> +        return ctx;
> +    }
> +
> +    g_hash_table_unref(ctx_cache);
> +    *ref = NULL;
> +
> +    riscv_iommu_report_fault(s, ctx, RISCV_IOMMU_FQ_TTYPE_UADDR_RD,
> +                             fault, !!pasid, 0, 0);
> +
> +    g_free(ctx);
> +    return NULL;
> +}
> +
> +static void riscv_iommu_ctx_put(RISCVIOMMUState *s, void *ref)
> +{
> +    if (ref) {
> +        g_hash_table_unref((GHashTable *)ref);
> +    }
> +}
> +
> +/* Find or allocate address space for a given device */
> +static AddressSpace *riscv_iommu_space(RISCVIOMMUState *s, uint32_t devid)
> +{
> +    RISCVIOMMUSpace *as;
> +
> +    /* FIXME: PCIe bus remapping for attached endpoints. */
> +    devid |= s->bus << 8;
> +
> +    qemu_mutex_lock(&s->core_lock);
> +    QLIST_FOREACH(as, &s->spaces, list) {
> +        if (as->devid == devid) {
> +            break;
> +        }
> +    }
> +    qemu_mutex_unlock(&s->core_lock);
> +
> +    if (as == NULL) {
> +        char name[64];
> +        as = g_new0(RISCVIOMMUSpace, 1);
> +
> +        as->iommu = s;
> +        as->devid = devid;
> +
> +        snprintf(name, sizeof(name), "riscv-iommu-%04x:%02x.%d-iova",
> +            PCI_BUS_NUM(as->devid), PCI_SLOT(as->devid), PCI_FUNC(as->devid));
> +
> +        /* IOVA address space, untranslated addresses */
> +        memory_region_init_iommu(&as->iova_mr, sizeof(as->iova_mr),
> +            TYPE_RISCV_IOMMU_MEMORY_REGION,
> +            OBJECT(as), "riscv_iommu", UINT64_MAX);
> +        address_space_init(&as->iova_as, MEMORY_REGION(&as->iova_mr), name);
> +
> +        qemu_mutex_lock(&s->core_lock);
> +        QLIST_INSERT_HEAD(&s->spaces, as, list);
> +        qemu_mutex_unlock(&s->core_lock);
> +
> +        trace_riscv_iommu_new(s->parent_obj.id, PCI_BUS_NUM(as->devid),
> +                PCI_SLOT(as->devid), PCI_FUNC(as->devid));
> +    }
> +    return &as->iova_as;
> +}
> +
> +static int riscv_iommu_translate(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
> +    IOMMUTLBEntry *iotlb)
> +{
> +    bool enable_pasid;
> +    bool enable_pri;
> +    int fault;
> +
> +    /*
> +     * TC[32] is reserved for custom extensions, used here to temporarily
> +     * enable automatic page-request generation for ATS queries.
> +     */
> +    enable_pri = (iotlb->perm == IOMMU_NONE) && (ctx->tc & BIT_ULL(32));
> +    enable_pasid = (ctx->tc & RISCV_IOMMU_DC_TC_PDTV);
> +
> +    /* Translate using device directory / page table information. */
> +    fault = riscv_iommu_spa_fetch(s, ctx, iotlb);
> +
> +    if (enable_pri && fault) {
> +        struct riscv_iommu_pq_record pr = {0};
> +        if (enable_pasid) {
> +            pr.hdr = set_field(RISCV_IOMMU_PREQ_HDR_PV,
> +                RISCV_IOMMU_PREQ_HDR_PID, ctx->pasid);
> +        }
> +        pr.hdr = set_field(pr.hdr, RISCV_IOMMU_PREQ_HDR_DID, ctx->devid);
> +        pr.payload = (iotlb->iova & TARGET_PAGE_MASK) |
> +                     RISCV_IOMMU_PREQ_PAYLOAD_M;
> +        riscv_iommu_pri(s, &pr);
> +        return fault;
> +    }
> +
> +    if (fault) {
> +        unsigned ttype;
> +
> +        if (iotlb->perm & IOMMU_RW) {
> +            ttype = RISCV_IOMMU_FQ_TTYPE_UADDR_WR;
> +        } else {
> +            ttype = RISCV_IOMMU_FQ_TTYPE_UADDR_RD;
> +        }
> +
> +        riscv_iommu_report_fault(s, ctx, ttype, fault, enable_pasid,
> +                                 iotlb->iova, iotlb->translated_addr);
> +        return fault;
> +    }
> +
> +    return 0;
> +}
> +
> +/* IOMMU Command Interface */
> +static MemTxResult riscv_iommu_iofence(RISCVIOMMUState *s, bool notify,
> +    uint64_t addr, uint32_t data)
> +{
> +    /*
> +     * ATS processing in this implementation of the IOMMU is synchronous,
> +     * no need to wait for completions here.
> +     */
> +    if (!notify) {
> +        return MEMTX_OK;
> +    }
> +
> +    return dma_memory_write(s->target_as, addr, &data, sizeof(data),
> +        MEMTXATTRS_UNSPECIFIED);
> +}
> +
> +static void riscv_iommu_process_ddtp(RISCVIOMMUState *s)
> +{
> +    uint64_t old_ddtp = s->ddtp;
> +    uint64_t new_ddtp = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_DDTP);
> +    unsigned new_mode = get_field(new_ddtp, RISCV_IOMMU_DDTP_MODE);
> +    unsigned old_mode = get_field(old_ddtp, RISCV_IOMMU_DDTP_MODE);
> +    bool ok = false;
> +
> +    /*
> +     * Check for allowed DDTP.MODE transitions:
> +     * {OFF, BARE}        -> {OFF, BARE, 1LVL, 2LVL, 3LVL}
> +     * {1LVL, 2LVL, 3LVL} -> {OFF, BARE}
> +     */
> +    if (new_mode == old_mode ||
> +        new_mode == RISCV_IOMMU_DDTP_MODE_OFF ||
> +        new_mode == RISCV_IOMMU_DDTP_MODE_BARE) {
> +        ok = true;
> +    } else if (new_mode == RISCV_IOMMU_DDTP_MODE_1LVL ||
> +               new_mode == RISCV_IOMMU_DDTP_MODE_2LVL ||
> +               new_mode == RISCV_IOMMU_DDTP_MODE_3LVL) {
> +        ok = old_mode == RISCV_IOMMU_DDTP_MODE_OFF ||
> +             old_mode == RISCV_IOMMU_DDTP_MODE_BARE;
> +    }
> +
> +    if (ok) {
> +        /* clear reserved and busy bits, report back sanitized version */
> +        new_ddtp = set_field(new_ddtp & RISCV_IOMMU_DDTP_PPN,
> +                             RISCV_IOMMU_DDTP_MODE, new_mode);
> +    } else {
> +        new_ddtp = old_ddtp;
> +    }
> +    s->ddtp = new_ddtp;
> +
> +    riscv_iommu_reg_set64(s, RISCV_IOMMU_REG_DDTP, new_ddtp);
> +}
> +
> +/* Command function and opcode field. */
> +#define RISCV_IOMMU_CMD(func, op) (((func) << 7) | (op))
> +
> +static void riscv_iommu_process_cq_tail(RISCVIOMMUState *s)
> +{
> +    struct riscv_iommu_command cmd;
> +    MemTxResult res;
> +    dma_addr_t addr;
> +    uint32_t tail, head, ctrl;
> +    uint64_t cmd_opcode;
> +    GHFunc func;
> +
> +    ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
> +    tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQT) & s->cq_mask;
> +    head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQH) & s->cq_mask;
> +
> +    /* Check for pending error or queue processing disabled */
> +    if (!(ctrl & RISCV_IOMMU_CQCSR_CQON) ||
> +        !!(ctrl & (RISCV_IOMMU_CQCSR_CMD_ILL | RISCV_IOMMU_CQCSR_CQMF))) {
> +        return;
> +    }
> +
> +    while (tail != head) {
> +        addr = s->cq_addr  + head * sizeof(cmd);
> +        res = dma_memory_read(s->target_as, addr, &cmd, sizeof(cmd),
> +                              MEMTXATTRS_UNSPECIFIED);
> +
> +        if (res != MEMTX_OK) {
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
> +                                  RISCV_IOMMU_CQCSR_CQMF, 0);
> +            goto fault;
> +        }
> +
> +        trace_riscv_iommu_cmd(s->parent_obj.id, cmd.dword0, cmd.dword1);
> +
> +        cmd_opcode = get_field(cmd.dword0,
> +                               RISCV_IOMMU_CMD_OPCODE | RISCV_IOMMU_CMD_FUNC);
> +
> +        switch (cmd_opcode) {
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOFENCE_FUNC_C,
> +                             RISCV_IOMMU_CMD_IOFENCE_OPCODE):
> +            res = riscv_iommu_iofence(s,
> +                cmd.dword0 & RISCV_IOMMU_CMD_IOFENCE_AV, cmd.dword1,
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IOFENCE_DATA));
> +
> +            if (res != MEMTX_OK) {
> +                riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
> +                                      RISCV_IOMMU_CQCSR_CQMF, 0);
> +                goto fault;
> +            }
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOTINVAL_FUNC_GVMA,
> +                             RISCV_IOMMU_CMD_IOTINVAL_OPCODE):
> +            if (cmd.dword0 & RISCV_IOMMU_CMD_IOTINVAL_PSCV) {
> +                /* illegal command arguments IOTINVAL.GVMA & PSCV == 1 */
> +                goto cmd_ill;
> +            }
> +            /* translation cache not implemented yet */
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOTINVAL_FUNC_VMA,
> +                             RISCV_IOMMU_CMD_IOTINVAL_OPCODE):
> +            /* translation cache not implemented yet */
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IODIR_FUNC_INVAL_DDT,
> +                             RISCV_IOMMU_CMD_IODIR_OPCODE):
> +            if (!(cmd.dword0 & RISCV_IOMMU_CMD_IODIR_DV)) {
> +                /* invalidate all device context cache mappings */
> +                func = __ctx_inval_all;
> +            } else {
> +                /* invalidate all device context matching DID */
> +                func = __ctx_inval_devid;
> +            }
> +            riscv_iommu_ctx_inval(s, func,
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_DID), 0);
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IODIR_FUNC_INVAL_PDT,
> +                             RISCV_IOMMU_CMD_IODIR_OPCODE):
> +            if (!(cmd.dword0 & RISCV_IOMMU_CMD_IODIR_DV)) {
> +                /* illegal command arguments IODIR_PDT & DV == 0 */
> +                goto cmd_ill;
> +            } else {
> +                func = __ctx_inval_devid_pasid;
> +            }
> +            riscv_iommu_ctx_inval(s, func,
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_DID),
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_PID));
> +            break;
> +
> +        default:
> +        cmd_ill:
> +            /* Invalid instruction, do not advance instruction index. */
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
> +                RISCV_IOMMU_CQCSR_CMD_ILL, 0);
> +            goto fault;
> +        }
> +
> +        /* Advance and update head pointer after command completes. */
> +        head = (head + 1) & s->cq_mask;
> +        riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_CQH, head);
> +    }
> +    return;
> +
> +fault:
> +    if (ctrl & RISCV_IOMMU_CQCSR_CIE) {
> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_CQ);
> +    }
> +}
> +
> +static void riscv_iommu_process_cq_control(RISCVIOMMUState *s)
> +{
> +    uint64_t base;
> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
> +    uint32_t ctrl_clr;
> +    bool enable = !!(ctrl_set & RISCV_IOMMU_CQCSR_CQEN);
> +    bool active = !!(ctrl_set & RISCV_IOMMU_CQCSR_CQON);
> +
> +    if (enable && !active) {
> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_CQB);
> +        s->cq_mask = (2ULL << get_field(base, RISCV_IOMMU_CQB_LOG2SZ)) - 1;
> +        s->cq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_CQB_PPN));
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQT], ~s->cq_mask);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_CQH], 0);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_CQT], 0);
> +        ctrl_set = RISCV_IOMMU_CQCSR_CQON;
> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY | RISCV_IOMMU_CQCSR_CQMF |
> +                   RISCV_IOMMU_CQCSR_CMD_ILL | RISCV_IOMMU_CQCSR_CMD_TO |
> +                   RISCV_IOMMU_CQCSR_FENCE_W_IP;
> +    } else if (!enable && active) {
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQT], ~0);
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY | RISCV_IOMMU_CQCSR_CQON;
> +    } else {
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR, ctrl_set, ctrl_clr);
> +}
> +
> +static void riscv_iommu_process_fq_control(RISCVIOMMUState *s)
> +{
> +    uint64_t base;
> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
> +    uint32_t ctrl_clr;
> +    bool enable = !!(ctrl_set & RISCV_IOMMU_FQCSR_FQEN);
> +    bool active = !!(ctrl_set & RISCV_IOMMU_FQCSR_FQON);
> +
> +    if (enable && !active) {
> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_FQB);
> +        s->fq_mask = (2ULL << get_field(base, RISCV_IOMMU_FQB_LOG2SZ)) - 1;
> +        s->fq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_FQB_PPN));
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQH], ~s->fq_mask);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_FQH], 0);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_FQT], 0);
> +        ctrl_set = RISCV_IOMMU_FQCSR_FQON;
> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY | RISCV_IOMMU_FQCSR_FQMF |
> +            RISCV_IOMMU_FQCSR_FQOF;
> +    } else if (!enable && active) {
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQH], ~0);
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY | RISCV_IOMMU_FQCSR_FQON;
> +    } else {
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR, ctrl_set, ctrl_clr);
> +}
> +
> +static void riscv_iommu_process_pq_control(RISCVIOMMUState *s)
> +{
> +    uint64_t base;
> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
> +    uint32_t ctrl_clr;
> +    bool enable = !!(ctrl_set & RISCV_IOMMU_PQCSR_PQEN);
> +    bool active = !!(ctrl_set & RISCV_IOMMU_PQCSR_PQON);
> +
> +    if (enable && !active) {
> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_PQB);
> +        s->pq_mask = (2ULL << get_field(base, RISCV_IOMMU_PQB_LOG2SZ)) - 1;
> +        s->pq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_PQB_PPN));
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQH], ~s->pq_mask);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_PQH], 0);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_PQT], 0);
> +        ctrl_set = RISCV_IOMMU_PQCSR_PQON;
> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY | RISCV_IOMMU_PQCSR_PQMF |
> +            RISCV_IOMMU_PQCSR_PQOF;
> +    } else if (!enable && active) {
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQH], ~0);
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY | RISCV_IOMMU_PQCSR_PQON;
> +    } else {
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR, ctrl_set, ctrl_clr);
> +}
> +
> +typedef void riscv_iommu_process_fn(RISCVIOMMUState *s);
> +
> +static void riscv_iommu_update_ipsr(RISCVIOMMUState *s, uint64_t data)
> +{
> +    uint32_t cqcsr, fqcsr, pqcsr;
> +    uint32_t ipsr_set = 0;
> +    uint32_t ipsr_clr = 0;
> +
> +    if (data & RISCV_IOMMU_IPSR_CIP) {
> +        cqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
> +
> +        if (cqcsr & RISCV_IOMMU_CQCSR_CIE &&
> +            (cqcsr & RISCV_IOMMU_CQCSR_FENCE_W_IP ||
> +             cqcsr & RISCV_IOMMU_CQCSR_CMD_ILL ||
> +             cqcsr & RISCV_IOMMU_CQCSR_CMD_TO ||
> +             cqcsr & RISCV_IOMMU_CQCSR_CQMF)) {
> +            ipsr_set |= RISCV_IOMMU_IPSR_CIP;
> +        } else {
> +            ipsr_clr |= RISCV_IOMMU_IPSR_CIP;
> +        }
> +    } else {
> +        ipsr_clr |= RISCV_IOMMU_IPSR_CIP;
> +    }
> +
> +    if (data & RISCV_IOMMU_IPSR_FIP) {
> +        fqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
> +
> +        if (fqcsr & RISCV_IOMMU_FQCSR_FIE &&
> +            (fqcsr & RISCV_IOMMU_FQCSR_FQOF ||
> +             fqcsr & RISCV_IOMMU_FQCSR_FQMF)) {
> +            ipsr_set |= RISCV_IOMMU_IPSR_FIP;
> +        } else {
> +            ipsr_clr |= RISCV_IOMMU_IPSR_FIP;
> +        }
> +    } else {
> +        ipsr_clr |= RISCV_IOMMU_IPSR_FIP;
> +    }
> +
> +    if (data & RISCV_IOMMU_IPSR_PIP) {
> +        pqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
> +
> +        if (pqcsr & RISCV_IOMMU_PQCSR_PIE &&
> +            (pqcsr & RISCV_IOMMU_PQCSR_PQOF ||
> +             pqcsr & RISCV_IOMMU_PQCSR_PQMF)) {
> +            ipsr_set |= RISCV_IOMMU_IPSR_PIP;
> +        } else {
> +            ipsr_clr |= RISCV_IOMMU_IPSR_PIP;
> +        }
> +    } else {
> +        ipsr_clr |= RISCV_IOMMU_IPSR_PIP;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_IPSR, ipsr_set, ipsr_clr);
> +}
> +
> +static MemTxResult riscv_iommu_mmio_write(void *opaque, hwaddr addr,
> +    uint64_t data, unsigned size, MemTxAttrs attrs)
> +{
> +    riscv_iommu_process_fn *process_fn = NULL;
> +    RISCVIOMMUState *s = opaque;
> +    uint32_t regb = addr & ~3;
> +    uint32_t busy = 0;
> +    uint64_t val = 0;
> +
> +    if ((addr & (size - 1)) != 0) {
> +        /* Unsupported MMIO alignment or access size */
> +        return MEMTX_ERROR;
> +    }
> +
> +    if (addr + size > RISCV_IOMMU_REG_MSI_CONFIG) {
> +        /* Unsupported MMIO access location. */
> +        return MEMTX_ACCESS_ERROR;
> +    }
> +
> +    /* Track actionable MMIO write. */
> +    switch (regb) {
> +    case RISCV_IOMMU_REG_DDTP:
> +    case RISCV_IOMMU_REG_DDTP + 4:
> +        process_fn = riscv_iommu_process_ddtp;
> +        regb = RISCV_IOMMU_REG_DDTP;
> +        busy = RISCV_IOMMU_DDTP_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_CQT:
> +        process_fn = riscv_iommu_process_cq_tail;
> +        break;
> +
> +    case RISCV_IOMMU_REG_CQCSR:
> +        process_fn = riscv_iommu_process_cq_control;
> +        busy = RISCV_IOMMU_CQCSR_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_FQCSR:
> +        process_fn = riscv_iommu_process_fq_control;
> +        busy = RISCV_IOMMU_FQCSR_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_PQCSR:
> +        process_fn = riscv_iommu_process_pq_control;
> +        busy = RISCV_IOMMU_PQCSR_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_IPSR:
> +        /*
> +         * IPSR has special procedures to update. Execute it
> +         * and exit.
> +         */
> +        if (size == 4) {
> +            uint32_t ro = ldl_le_p(&s->regs_ro[addr]);
> +            uint32_t wc = ldl_le_p(&s->regs_wc[addr]);
> +            uint32_t rw = ldl_le_p(&s->regs_rw[addr]);
> +            stl_le_p(&val, ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +        } else if (size == 8) {
> +            uint64_t ro = ldq_le_p(&s->regs_ro[addr]);
> +            uint64_t wc = ldq_le_p(&s->regs_wc[addr]);
> +            uint64_t rw = ldq_le_p(&s->regs_rw[addr]);
> +            stq_le_p(&val, ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +        }
> +
> +        riscv_iommu_update_ipsr(s, val);
> +
> +        return MEMTX_OK;
> +
> +    default:
> +        break;
> +    }
> +
> +    /*
> +     * Registers update might be not synchronized with core logic.
> +     * If system software updates register when relevant BUSY bit
> +     * is set IOMMU behavior of additional writes to the register
> +     * is UNSPECIFIED.
> +     */
> +    qemu_spin_lock(&s->regs_lock);
> +    if (size == 1) {
> +        uint8_t ro = s->regs_ro[addr];
> +        uint8_t wc = s->regs_wc[addr];
> +        uint8_t rw = s->regs_rw[addr];
> +        s->regs_rw[addr] = ((rw & ro) | (data & ~ro)) & ~(data & wc);
> +    } else if (size == 2) {
> +        uint16_t ro = lduw_le_p(&s->regs_ro[addr]);
> +        uint16_t wc = lduw_le_p(&s->regs_wc[addr]);
> +        uint16_t rw = lduw_le_p(&s->regs_rw[addr]);
> +        stw_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +    } else if (size == 4) {
> +        uint32_t ro = ldl_le_p(&s->regs_ro[addr]);
> +        uint32_t wc = ldl_le_p(&s->regs_wc[addr]);
> +        uint32_t rw = ldl_le_p(&s->regs_rw[addr]);
> +        stl_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +    } else if (size == 8) {
> +        uint64_t ro = ldq_le_p(&s->regs_ro[addr]);
> +        uint64_t wc = ldq_le_p(&s->regs_wc[addr]);
> +        uint64_t rw = ldq_le_p(&s->regs_rw[addr]);
> +        stq_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +    }
> +
> +    /* Busy flag update, MSB 4-byte register. */
> +    if (busy) {
> +        uint32_t rw = ldl_le_p(&s->regs_rw[regb]);
> +        stl_le_p(&s->regs_rw[regb], rw | busy);
> +    }
> +    qemu_spin_unlock(&s->regs_lock);
> +
> +    if (process_fn) {
> +        qemu_mutex_lock(&s->core_lock);
> +        process_fn(s);
> +        qemu_mutex_unlock(&s->core_lock);
> +    }
> +
> +    return MEMTX_OK;
> +}
> +
> +static MemTxResult riscv_iommu_mmio_read(void *opaque, hwaddr addr,
> +    uint64_t *data, unsigned size, MemTxAttrs attrs)
> +{
> +    RISCVIOMMUState *s = opaque;
> +    uint64_t val = -1;
> +    uint8_t *ptr;
> +
> +    if ((addr & (size - 1)) != 0) {
> +        /* Unsupported MMIO alignment. */
> +        return MEMTX_ERROR;
> +    }
> +
> +    if (addr + size > RISCV_IOMMU_REG_MSI_CONFIG) {
> +        return MEMTX_ACCESS_ERROR;
> +    }
> +
> +    ptr = &s->regs_rw[addr];
> +
> +    if (size == 1) {
> +        val = (uint64_t)*ptr;
> +    } else if (size == 2) {
> +        val = lduw_le_p(ptr);
> +    } else if (size == 4) {
> +        val = ldl_le_p(ptr);
> +    } else if (size == 8) {
> +        val = ldq_le_p(ptr);
> +    } else {
> +        return MEMTX_ERROR;
> +    }
> +
> +    *data = val;
> +
> +    return MEMTX_OK;
> +}
> +
> +static const MemoryRegionOps riscv_iommu_mmio_ops = {
> +    .read_with_attrs = riscv_iommu_mmio_read,
> +    .write_with_attrs = riscv_iommu_mmio_write,
> +    .endianness = DEVICE_NATIVE_ENDIAN,
> +    .impl = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +        .unaligned = false,
> +    },
> +    .valid = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +    }
> +};
> +
> +/*
> + * Translations matching MSI pattern check are redirected to "riscv-iommu-trap"
> + * memory region as untranslated address, for additional MSI/MRIF interception
> + * by IOMMU interrupt remapping implementation.
> + * Note: Device emulation code generating an MSI is expected to provide a valid
> + * memory transaction attributes with requested_id set.
> + */
> +static MemTxResult riscv_iommu_trap_write(void *opaque, hwaddr addr,
> +    uint64_t data, unsigned size, MemTxAttrs attrs)
> +{
> +    RISCVIOMMUState* s = (RISCVIOMMUState *)opaque;
> +    RISCVIOMMUContext *ctx;
> +    MemTxResult res;
> +    void *ref;
> +    uint32_t devid = attrs.requester_id;
> +
> +    if (attrs.unspecified) {
> +        return MEMTX_ACCESS_ERROR;
> +    }
> +
> +    /* FIXME: PCIe bus remapping for attached endpoints. */
> +    devid |= s->bus << 8;
> +
> +    ctx = riscv_iommu_ctx(s, devid, 0, &ref);
> +    if (ctx == NULL) {
> +        res = MEMTX_ACCESS_ERROR;
> +    } else {
> +        res = riscv_iommu_msi_write(s, ctx, addr, data, size, attrs);
> +    }
> +    riscv_iommu_ctx_put(s, ref);
> +    return res;
> +}
> +
> +static MemTxResult riscv_iommu_trap_read(void *opaque, hwaddr addr,
> +    uint64_t *data, unsigned size, MemTxAttrs attrs)
> +{
> +    return MEMTX_ACCESS_ERROR;
> +}
> +
> +static const MemoryRegionOps riscv_iommu_trap_ops = {
> +    .read_with_attrs = riscv_iommu_trap_read,
> +    .write_with_attrs = riscv_iommu_trap_write,
> +    .endianness = DEVICE_LITTLE_ENDIAN,
> +    .impl = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +        .unaligned = true,
> +    },
> +    .valid = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +    }
> +};
> +
> +static void riscv_iommu_realize(DeviceState *dev, Error **errp)
> +{
> +    RISCVIOMMUState *s = RISCV_IOMMU(dev);
> +
> +    s->cap = s->version & RISCV_IOMMU_CAP_VERSION;
> +    if (s->enable_msi) {
> +        s->cap |= RISCV_IOMMU_CAP_MSI_FLAT | RISCV_IOMMU_CAP_MSI_MRIF;
> +    }
> +    /* Report QEMU target physical address space limits */
> +    s->cap = set_field(s->cap, RISCV_IOMMU_CAP_PAS,
> +                       TARGET_PHYS_ADDR_SPACE_BITS);
> +
> +    /* TODO: method to report supported PASID bits */
> +    s->pasid_bits = 8; /* restricted to size of MemTxAttrs.pasid */
> +    s->cap |= RISCV_IOMMU_CAP_PD8;
> +
> +    /* Out-of-reset translation mode: OFF (DMA disabled) BARE (passthrough) */
> +    s->ddtp = set_field(0, RISCV_IOMMU_DDTP_MODE, s->enable_off ?
> +                        RISCV_IOMMU_DDTP_MODE_OFF : RISCV_IOMMU_DDTP_MODE_BARE);
> +
> +    /* register storage */
> +    s->regs_rw = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
> +    s->regs_ro = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
> +    s->regs_wc = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
> +
> +     /* Mark all registers read-only */
> +    memset(s->regs_ro, 0xff, RISCV_IOMMU_REG_SIZE);
> +
> +    /*
> +     * Register complete MMIO space, including MSI/PBA registers.
> +     * Note, PCIDevice implementation will add overlapping MR for MSI/PBA,
> +     * managed directly by the PCIDevice implementation.
> +     */
> +    memory_region_init_io(&s->regs_mr, OBJECT(dev), &riscv_iommu_mmio_ops, s,
> +        "riscv-iommu-regs", RISCV_IOMMU_REG_SIZE);
> +
> +    /* Set power-on register state */
> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_CAP], s->cap);
> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_FCTL], 0);
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_DDTP],
> +        ~(RISCV_IOMMU_DDTP_PPN | RISCV_IOMMU_DDTP_MODE));
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQB],
> +        ~(RISCV_IOMMU_CQB_LOG2SZ | RISCV_IOMMU_CQB_PPN));
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQB],
> +        ~(RISCV_IOMMU_FQB_LOG2SZ | RISCV_IOMMU_FQB_PPN));
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQB],
> +        ~(RISCV_IOMMU_PQB_LOG2SZ | RISCV_IOMMU_PQB_PPN));
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_CQCSR], RISCV_IOMMU_CQCSR_CQMF |
> +        RISCV_IOMMU_CQCSR_CMD_TO | RISCV_IOMMU_CQCSR_CMD_ILL);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQCSR], RISCV_IOMMU_CQCSR_CQON |
> +        RISCV_IOMMU_CQCSR_BUSY);
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_FQCSR], RISCV_IOMMU_FQCSR_FQMF |
> +        RISCV_IOMMU_FQCSR_FQOF);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQCSR], RISCV_IOMMU_FQCSR_FQON |
> +        RISCV_IOMMU_FQCSR_BUSY);
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_PQCSR], RISCV_IOMMU_PQCSR_PQMF |
> +        RISCV_IOMMU_PQCSR_PQOF);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQCSR], RISCV_IOMMU_PQCSR_PQON |
> +        RISCV_IOMMU_PQCSR_BUSY);
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_IPSR], ~0);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_IVEC], 0);
> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_DDTP], s->ddtp);
> +
> +    /* Memory region for downstream access, if specified. */
> +    if (s->target_mr) {
> +        s->target_as = g_new0(AddressSpace, 1);
> +        address_space_init(s->target_as, s->target_mr,
> +            "riscv-iommu-downstream");
> +    } else {
> +        /* Fallback to global system memory. */
> +        s->target_as = &address_space_memory;
> +    }
> +
> +    /* Memory region for untranslated MRIF/MSI writes */
> +    memory_region_init_io(&s->trap_mr, OBJECT(dev), &riscv_iommu_trap_ops, s,
> +            "riscv-iommu-trap", ~0ULL);
> +    address_space_init(&s->trap_as, &s->trap_mr, "riscv-iommu-trap-as");
> +
> +    /* Device translation context cache */
> +    s->ctx_cache = g_hash_table_new_full(__ctx_hash, __ctx_equal,
> +                                         g_free, NULL);
> +
> +    s->iommus.le_next = NULL;
> +    s->iommus.le_prev = NULL;
> +    QLIST_INIT(&s->spaces);
> +    qemu_mutex_init(&s->core_lock);
> +    qemu_spin_init(&s->regs_lock);
> +}
> +
> +static void riscv_iommu_unrealize(DeviceState *dev)
> +{
> +    RISCVIOMMUState *s = RISCV_IOMMU(dev);
> +
> +    qemu_mutex_destroy(&s->core_lock);
> +    g_hash_table_unref(s->ctx_cache);
> +}
> +
> +static Property riscv_iommu_properties[] = {
> +    DEFINE_PROP_UINT32("version", RISCVIOMMUState, version,
> +        RISCV_IOMMU_SPEC_DOT_VER),
> +    DEFINE_PROP_UINT32("bus", RISCVIOMMUState, bus, 0x0),
> +    DEFINE_PROP_BOOL("intremap", RISCVIOMMUState, enable_msi, TRUE),
> +    DEFINE_PROP_BOOL("off", RISCVIOMMUState, enable_off, TRUE),
> +    DEFINE_PROP_LINK("downstream-mr", RISCVIOMMUState, target_mr,
> +        TYPE_MEMORY_REGION, MemoryRegion *),
> +    DEFINE_PROP_END_OF_LIST(),
> +};
> +
> +static void riscv_iommu_class_init(ObjectClass *klass, void* data)
> +{
> +    DeviceClass *dc = DEVICE_CLASS(klass);
> +
> +    /* internal device for riscv-iommu-{pci/sys}, not user-creatable */
> +    dc->user_creatable = false;
> +    dc->realize = riscv_iommu_realize;
> +    dc->unrealize = riscv_iommu_unrealize;
> +    device_class_set_props(dc, riscv_iommu_properties);
> +}
> +
> +static const TypeInfo riscv_iommu_info = {
> +    .name = TYPE_RISCV_IOMMU,
> +    .parent = TYPE_DEVICE,
> +    .instance_size = sizeof(RISCVIOMMUState),
> +    .class_init = riscv_iommu_class_init,
> +};
> +
> +static const char *IOMMU_FLAG_STR[] = {
> +    "NA",
> +    "RO",
> +    "WR",
> +    "RW",
> +};
> +
> +/* RISC-V IOMMU Memory Region - Address Translation Space */
> +static IOMMUTLBEntry riscv_iommu_memory_region_translate(
> +    IOMMUMemoryRegion *iommu_mr, hwaddr addr,
> +    IOMMUAccessFlags flag, int iommu_idx)
> +{
> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
> +    RISCVIOMMUContext *ctx;
> +    void *ref;
> +    IOMMUTLBEntry iotlb = {
> +        .iova = addr,
> +        .target_as = as->iommu->target_as,
> +        .addr_mask = ~0ULL,
> +        .perm = flag,
> +    };
> +
> +    ctx = riscv_iommu_ctx(as->iommu, as->devid, iommu_idx, &ref);
> +    if (ctx == NULL) {
> +        /* Translation disabled or invalid. */
> +        iotlb.addr_mask = 0;
> +        iotlb.perm = IOMMU_NONE;
> +    } else if (riscv_iommu_translate(as->iommu, ctx, &iotlb)) {
> +        /* Translation disabled or fault reported. */
> +        iotlb.addr_mask = 0;
> +        iotlb.perm = IOMMU_NONE;
> +    }
> +
> +    /* Trace all dma translations with original access flags. */
> +    trace_riscv_iommu_dma(as->iommu->parent_obj.id, PCI_BUS_NUM(as->devid),
> +                          PCI_SLOT(as->devid), PCI_FUNC(as->devid), iommu_idx,
> +                          IOMMU_FLAG_STR[flag & IOMMU_RW], iotlb.iova,
> +                          iotlb.translated_addr);
> +
> +    riscv_iommu_ctx_put(as->iommu, ref);
> +
> +    return iotlb;
> +}
> +
> +static int riscv_iommu_memory_region_notify(
> +    IOMMUMemoryRegion *iommu_mr, IOMMUNotifierFlag old,
> +    IOMMUNotifierFlag new, Error **errp)
> +{
> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
> +
> +    if (old == IOMMU_NOTIFIER_NONE) {
> +        as->notifier = true;
> +        trace_riscv_iommu_notifier_add(iommu_mr->parent_obj.name);
> +    } else if (new == IOMMU_NOTIFIER_NONE) {
> +        as->notifier = false;
> +        trace_riscv_iommu_notifier_del(iommu_mr->parent_obj.name);
> +    }
> +
> +    return 0;
> +}
> +
> +static inline bool pci_is_iommu(PCIDevice *pdev)
> +{
> +    return pci_get_word(pdev->config + PCI_CLASS_DEVICE) == 0x0806;
> +}
> +
> +static AddressSpace *riscv_iommu_find_as(PCIBus *bus, void *opaque, int devfn)
> +{
> +    RISCVIOMMUState *s = (RISCVIOMMUState *) opaque;
> +    PCIDevice *pdev = pci_find_device(bus, pci_bus_num(bus), devfn);
> +    AddressSpace *as = NULL;
> +
> +    if (pdev && pci_is_iommu(pdev)) {
> +        return s->target_as;
> +    }
> +
> +    /* Find first registered IOMMU device */
> +    while (s->iommus.le_prev) {
> +        s = *(s->iommus.le_prev);
> +    }
> +
> +    /* Find first matching IOMMU */
> +    while (s != NULL && as == NULL) {
> +        as = riscv_iommu_space(s, PCI_BUILD_BDF(pci_bus_num(bus), devfn));
> +        s = s->iommus.le_next;
> +    }
> +
> +    return as ? as : &address_space_memory;
> +}
> +
> +static const PCIIOMMUOps riscv_iommu_ops = {
> +    .get_address_space = riscv_iommu_find_as,
> +};
> +
> +void riscv_iommu_pci_setup_iommu(RISCVIOMMUState *iommu, PCIBus *bus,
> +        Error **errp)
> +{
> +    if (bus->iommu_ops &&
> +        bus->iommu_ops->get_address_space == riscv_iommu_find_as) {
> +        /* Allow multiple IOMMUs on the same PCIe bus, link known devices */
> +        RISCVIOMMUState *last = (RISCVIOMMUState *)bus->iommu_opaque;
> +        QLIST_INSERT_AFTER(last, iommu, iommus);
> +    } else if (!bus->iommu_ops && !bus->iommu_opaque) {
> +        pci_setup_iommu(bus, &riscv_iommu_ops, iommu);
> +    } else {
> +        error_setg(errp, "can't register secondary IOMMU for PCI bus #%d",
> +            pci_bus_num(bus));
> +    }
> +}
> +
> +static int riscv_iommu_memory_region_index(IOMMUMemoryRegion *iommu_mr,
> +    MemTxAttrs attrs)
> +{
> +    return attrs.unspecified ? RISCV_IOMMU_NOPASID : (int)attrs.pasid;
> +}
> +
> +static int riscv_iommu_memory_region_index_len(IOMMUMemoryRegion *iommu_mr)
> +{
> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
> +    return 1 << as->iommu->pasid_bits;
> +}
> +
> +static void riscv_iommu_memory_region_init(ObjectClass *klass, void *data)
> +{
> +    IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_CLASS(klass);
> +
> +    imrc->translate = riscv_iommu_memory_region_translate;
> +    imrc->notify_flag_changed = riscv_iommu_memory_region_notify;
> +    imrc->attrs_to_index = riscv_iommu_memory_region_index;
> +    imrc->num_indexes = riscv_iommu_memory_region_index_len;
> +}
> +
> +static const TypeInfo riscv_iommu_memory_region_info = {
> +    .parent = TYPE_IOMMU_MEMORY_REGION,
> +    .name = TYPE_RISCV_IOMMU_MEMORY_REGION,
> +    .class_init = riscv_iommu_memory_region_init,
> +};
> +
> +static void riscv_iommu_register_mr_types(void)
> +{
> +    type_register_static(&riscv_iommu_memory_region_info);
> +    type_register_static(&riscv_iommu_info);
> +}
> +
> +type_init(riscv_iommu_register_mr_types);
> diff --git a/hw/riscv/riscv-iommu.h b/hw/riscv/riscv-iommu.h
> new file mode 100644
> index 0000000000..31d3907d33
> --- /dev/null
> +++ b/hw/riscv/riscv-iommu.h
> @@ -0,0 +1,141 @@
> +/*
> + * QEMU emulation of an RISC-V IOMMU
> + *
> + * Copyright (C) 2022-2023 Rivos Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#ifndef HW_RISCV_IOMMU_STATE_H
> +#define HW_RISCV_IOMMU_STATE_H
> +
> +#include "qemu/osdep.h"
> +#include "qom/object.h"
> +
> +#include "hw/riscv/iommu.h"
> +
> +struct RISCVIOMMUState {
> +    /*< private >*/
> +    DeviceState parent_obj;
> +
> +    /*< public >*/
> +    uint32_t version;     /* Reported interface version number */
> +    uint32_t pasid_bits;  /* process identifier width */
> +    uint32_t bus;         /* PCI bus mapping for non-root endpoints */
> +
> +    uint64_t cap;         /* IOMMU supported capabilities */
> +    uint64_t fctl;        /* IOMMU enabled features */
> +
> +    bool enable_off;      /* Enable out-of-reset OFF mode (DMA disabled) */
> +    bool enable_msi;      /* Enable MSI remapping */
> +
> +    /* IOMMU Internal State */
> +    uint64_t ddtp;        /* Validated Device Directory Tree Root Pointer */
> +
> +    dma_addr_t cq_addr;   /* Command queue base physical address */
> +    dma_addr_t fq_addr;   /* Fault/event queue base physical address */
> +    dma_addr_t pq_addr;   /* Page request queue base physical address */
> +
> +    uint32_t cq_mask;     /* Command queue index bit mask */
> +    uint32_t fq_mask;     /* Fault/event queue index bit mask */
> +    uint32_t pq_mask;     /* Page request queue index bit mask */
> +
> +    /* interrupt notifier */
> +    void (*notify)(RISCVIOMMUState *iommu, unsigned vector);
> +
> +    /* IOMMU State Machine */
> +    QemuThread core_proc; /* Background processing thread */
> +    QemuMutex core_lock;  /* Global IOMMU lock, used for cache/regs updates */
> +    QemuCond core_cond;   /* Background processing wake up signal */
> +    unsigned core_exec;   /* Processing thread execution actions */
> +
> +    /* IOMMU target address space */
> +    AddressSpace *target_as;
> +    MemoryRegion *target_mr;
> +
> +    /* MSI / MRIF access trap */
> +    AddressSpace trap_as;
> +    MemoryRegion trap_mr;
> +
> +    GHashTable *ctx_cache;          /* Device translation Context Cache */
> +
> +    /* MMIO Hardware Interface */
> +    MemoryRegion regs_mr;
> +    QemuSpin regs_lock;
> +    uint8_t *regs_rw;  /* register state (user write) */
> +    uint8_t *regs_wc;  /* write-1-to-clear mask */
> +    uint8_t *regs_ro;  /* read-only mask */
> +
> +    QLIST_ENTRY(RISCVIOMMUState) iommus;
> +    QLIST_HEAD(, RISCVIOMMUSpace) spaces;
> +};
> +
> +void riscv_iommu_pci_setup_iommu(RISCVIOMMUState *iommu, PCIBus *bus,
> +         Error **errp);
> +
> +/* private helpers */
> +
> +/* Register helper functions */
> +static inline uint32_t riscv_iommu_reg_mod32(RISCVIOMMUState *s,
> +    unsigned idx, uint32_t set, uint32_t clr)
> +{
> +    uint32_t val;
> +    qemu_spin_lock(&s->regs_lock);
> +    val = ldl_le_p(s->regs_rw + idx);
> +    stl_le_p(s->regs_rw + idx, (val & ~clr) | set);
> +    qemu_spin_unlock(&s->regs_lock);
> +    return val;
> +}
> +
> +static inline void riscv_iommu_reg_set32(RISCVIOMMUState *s,
> +    unsigned idx, uint32_t set)
> +{
> +    qemu_spin_lock(&s->regs_lock);
> +    stl_le_p(s->regs_rw + idx, set);
> +    qemu_spin_unlock(&s->regs_lock);
> +}
> +
> +static inline uint32_t riscv_iommu_reg_get32(RISCVIOMMUState *s,
> +    unsigned idx)
> +{
> +    return ldl_le_p(s->regs_rw + idx);
> +}
> +
> +static inline uint64_t riscv_iommu_reg_mod64(RISCVIOMMUState *s,
> +    unsigned idx, uint64_t set, uint64_t clr)
> +{
> +    uint64_t val;
> +    qemu_spin_lock(&s->regs_lock);
> +    val = ldq_le_p(s->regs_rw + idx);
> +    stq_le_p(s->regs_rw + idx, (val & ~clr) | set);
> +    qemu_spin_unlock(&s->regs_lock);
> +    return val;
> +}
> +
> +static inline void riscv_iommu_reg_set64(RISCVIOMMUState *s,
> +    unsigned idx, uint64_t set)
> +{
> +    qemu_spin_lock(&s->regs_lock);
> +    stq_le_p(s->regs_rw + idx, set);
> +    qemu_spin_unlock(&s->regs_lock);
> +}
> +
> +static inline uint64_t riscv_iommu_reg_get64(RISCVIOMMUState *s,
> +    unsigned idx)
> +{
> +    return ldq_le_p(s->regs_rw + idx);
> +}
> +
> +
> +
> +#endif
> diff --git a/hw/riscv/trace-events b/hw/riscv/trace-events
> new file mode 100644
> index 0000000000..42a97caffa
> --- /dev/null
> +++ b/hw/riscv/trace-events
> @@ -0,0 +1,11 @@
> +# See documentation at docs/devel/tracing.rst
> +
> +# riscv-iommu.c
> +riscv_iommu_new(const char *id, unsigned b, unsigned d, unsigned f) "%s: device attached %04x:%02x.%d"
> +riscv_iommu_flt(const char *id, unsigned b, unsigned d, unsigned f, uint64_t reason, uint64_t iova) "%s: fault %04x:%02x.%u reason: 0x%"PRIx64" iova: 0x%"PRIx64
> +riscv_iommu_pri(const char *id, unsigned b, unsigned d, unsigned f, uint64_t iova) "%s: page request %04x:%02x.%u iova: 0x%"PRIx64
> +riscv_iommu_dma(const char *id, unsigned b, unsigned d, unsigned f, unsigned pasid, const char *dir, uint64_t iova, uint64_t phys) "%s: translate %04x:%02x.%u #%u %s 0x%"PRIx64" -> 0x%"PRIx64
> +riscv_iommu_msi(const char *id, unsigned b, unsigned d, unsigned f, uint64_t iova, uint64_t phys) "%s: translate %04x:%02x.%u MSI 0x%"PRIx64" -> 0x%"PRIx64
> +riscv_iommu_cmd(const char *id, uint64_t l, uint64_t u) "%s: command 0x%"PRIx64" 0x%"PRIx64
> +riscv_iommu_notifier_add(const char *id) "%s: dev-iotlb notifier added"
> +riscv_iommu_notifier_del(const char *id) "%s: dev-iotlb notifier removed"
> diff --git a/hw/riscv/trace.h b/hw/riscv/trace.h
> new file mode 100644
> index 0000000000..8c0e3ca1f3
> --- /dev/null
> +++ b/hw/riscv/trace.h
> @@ -0,0 +1 @@
> +#include "trace/trace-hw_riscv.h"
> diff --git a/include/hw/riscv/iommu.h b/include/hw/riscv/iommu.h
> new file mode 100644
> index 0000000000..070ee69973
> --- /dev/null
> +++ b/include/hw/riscv/iommu.h
> @@ -0,0 +1,36 @@
> +/*
> + * QEMU emulation of an RISC-V IOMMU
> + *
> + * Copyright (C) 2022-2023 Rivos Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#ifndef HW_RISCV_IOMMU_H
> +#define HW_RISCV_IOMMU_H
> +
> +#include "qemu/osdep.h"
> +#include "qom/object.h"
> +
> +#define TYPE_RISCV_IOMMU "riscv-iommu"
> +OBJECT_DECLARE_SIMPLE_TYPE(RISCVIOMMUState, RISCV_IOMMU)
> +typedef struct RISCVIOMMUState RISCVIOMMUState;
> +
> +#define TYPE_RISCV_IOMMU_MEMORY_REGION "riscv-iommu-mr"
> +typedef struct RISCVIOMMUSpace RISCVIOMMUSpace;
> +
> +#define TYPE_RISCV_IOMMU_PCI "riscv-iommu-pci"
> +OBJECT_DECLARE_SIMPLE_TYPE(RISCVIOMMUStatePci, RISCV_IOMMU_PCI)
> +typedef struct RISCVIOMMUStatePci RISCVIOMMUStatePci;
> +
> +#endif
> diff --git a/meson.build b/meson.build
> index a9de71d450..8099d8271c 100644
> --- a/meson.build
> +++ b/meson.build
> @@ -3319,6 +3319,7 @@ if have_system
>       'hw/pci-host',
>       'hw/ppc',
>       'hw/rtc',
> +    'hw/riscv',
>       'hw/s390x',
>       'hw/scsi',
>       'hw/sd',

On 5/29/24 10:39 PM, Eric Cheng wrote:
> On 5/24/2024 1:39 AM, Daniel Henrique Barboza wrote:
>> From: Tomasz Jeznach <tjeznach@rivosinc.com>
>>
>> The RISC-V IOMMU specification is now ratified as-per the RISC-V
>> international process. The latest frozen specifcation can be found
>> at:
>>
>> https://github.com/riscv-non-isa/riscv-iommu/releases/download/v1.0/riscv-iommu.pdf
>>
>> Add the foundation of the device emulation for RISC-V IOMMU, which
>> includes an IOMMU that has no capabilities but MSI interrupt support and
>> fault queue interfaces. We'll add add more features incrementally in the
>                                ^^^  ^^^
> repeated 'add'

Fixed.

> 
>> next patches.
>>
>> Co-developed-by: Sebastien Boeuf <seb@rivosinc.com>
>> Signed-off-by: Sebastien Boeuf <seb@rivosinc.com>
>> Signed-off-by: Tomasz Jeznach <tjeznach@rivosinc.com>
>> Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
>> ---
>>   hw/riscv/Kconfig         |    4 +
>>   hw/riscv/meson.build     |    1 +
>>   hw/riscv/riscv-iommu.c   | 1602 ++++++++++++++++++++++++++++++++++++++
>>   hw/riscv/riscv-iommu.h   |  141 ++++
>>   hw/riscv/trace-events    |   11 +
>>   hw/riscv/trace.h         |    1 +
>>   include/hw/riscv/iommu.h |   36 +
>>   meson.build              |    1 +
>>   8 files changed, 1797 insertions(+)
>>   create mode 100644 hw/riscv/riscv-iommu.c
>>   create mode 100644 hw/riscv/riscv-iommu.h
>>   create mode 100644 hw/riscv/trace-events
>>   create mode 100644 hw/riscv/trace.h
>>   create mode 100644 include/hw/riscv/iommu.h
>>
>> diff --git a/hw/riscv/Kconfig b/hw/riscv/Kconfig
>> index a2030e3a6f..f69d6e3c8e 100644
>> --- a/hw/riscv/Kconfig
>> +++ b/hw/riscv/Kconfig
>> @@ -1,3 +1,6 @@
>> +config RISCV_IOMMU
>> +    bool
>> +
>>   config RISCV_NUMA
>>       bool
>> @@ -47,6 +50,7 @@ config RISCV_VIRT
>>       select SERIAL
>>       select RISCV_ACLINT
>>       select RISCV_APLIC
>> +    select RISCV_IOMMU
>>       select RISCV_IMSIC
>>       select SIFIVE_PLIC
>>       select SIFIVE_TEST
>> diff --git a/hw/riscv/meson.build b/hw/riscv/meson.build
>> index f872674093..cbc99c6e8e 100644
>> --- a/hw/riscv/meson.build
>> +++ b/hw/riscv/meson.build
>> @@ -10,5 +10,6 @@ riscv_ss.add(when: 'CONFIG_SIFIVE_U', if_true: files('sifive_u.c'))
>>   riscv_ss.add(when: 'CONFIG_SPIKE', if_true: files('spike.c'))
>>   riscv_ss.add(when: 'CONFIG_MICROCHIP_PFSOC', if_true: files('microchip_pfsoc.c'))
>>   riscv_ss.add(when: 'CONFIG_ACPI', if_true: files('virt-acpi-build.c'))
>> +riscv_ss.add(when: 'CONFIG_RISCV_IOMMU', if_true: files('riscv-iommu.c'))
>>   hw_arch += {'riscv': riscv_ss}
>> diff --git a/hw/riscv/riscv-iommu.c b/hw/riscv/riscv-iommu.c
>> new file mode 100644
>> index 0000000000..39b4ff1405
>> --- /dev/null
>> +++ b/hw/riscv/riscv-iommu.c
>> @@ -0,0 +1,1602 @@
>> +/*
>> + * QEMU emulation of an RISC-V IOMMU
>> + *
>> + * Copyright (C) 2021-2023, Rivos Inc.
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of the GNU General Public License as published by
>> + * the Free Software Foundation; either version 2 of the License.
>> + *
>> + * This program is distributed in the hope that it will be useful,
>> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
>> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>> + * GNU General Public License for more details.
>> + *
>> + * You should have received a copy of the GNU General Public License along
>> + * with this program; if not, see <http://www.gnu.org/licenses/>.
>> + */
>> +
>> +#include "qemu/osdep.h"
>> +#include "qom/object.h"
>> +#include "hw/pci/pci_bus.h"
>> +#include "hw/pci/pci_device.h"
>> +#include "hw/qdev-properties.h"
>> +#include "hw/riscv/riscv_hart.h"
>> +#include "migration/vmstate.h"
>> +#include "qapi/error.h"
>> +#include "qemu/timer.h"
>> +
>> +#include "cpu_bits.h"
>> +#include "riscv-iommu.h"
>> +#include "riscv-iommu-bits.h"
>> +#include "trace.h"
>> +
>> +#define LIMIT_CACHE_CTX               (1U << 7)
>> +#define LIMIT_CACHE_IOT               (1U << 20)
>> +
>> +/* Physical page number coversions */
>> +#define PPN_PHYS(ppn)                 ((ppn) << TARGET_PAGE_BITS)
>> +#define PPN_DOWN(phy)                 ((phy) >> TARGET_PAGE_BITS)
>> +
>> +typedef struct RISCVIOMMUContext RISCVIOMMUContext;
>> +typedef struct RISCVIOMMUEntry RISCVIOMMUEntry;
>> +
>> +/* Device assigned I/O address space */
>> +struct RISCVIOMMUSpace {
>> +    IOMMUMemoryRegion iova_mr;  /* IOVA memory region for attached device */
>> +    AddressSpace iova_as;       /* IOVA address space for attached device */
>> +    RISCVIOMMUState *iommu;     /* Managing IOMMU device state */
>> +    uint32_t devid;             /* Requester identifier, AKA device_id */
>> +    bool notifier;              /* IOMMU unmap notifier enabled */
>> +    QLIST_ENTRY(RISCVIOMMUSpace) list;
>> +};
>> +
>> +/* Device translation context state. */
>> +struct RISCVIOMMUContext {
>> +    uint64_t devid:24;          /* Requester Id, AKA device_id */
>> +    uint64_t pasid:20;          /* Process Address Space ID */
>> +    uint64_t __rfu:20;          /* reserved */
>> +    uint64_t tc;                /* Translation Control */
>> +    uint64_t ta;                /* Translation Attributes */
>> +    uint64_t msi_addr_mask;     /* MSI filtering - address mask */
>> +    uint64_t msi_addr_pattern;  /* MSI filtering - address pattern */
>> +    uint64_t msiptp;            /* MSI redirection page table pointer */
>> +};
> 
> Can we alias use (devid + pasid + __rfu) by union? so that can easily compare the key, esp. I assume functions like __ctx_equal() are on hot path.


I don't think so. The reason is that the device context can have different
process ids with the same devid. The spec mentions GPU as an example of a
device that could have multiple device contexts for each process using it.

In fact, we're checking for both attributes at the same time in functions
such as riscv_iommu_ctx_fetch().

> 
> And, pasid, is the term in PCI context. I suggest use more general name in spec: process_id.


The spec seems to use 'process_id' rather than PASID for this particular
use indeed. I'll rename it.



Thanks,

Daniel


> 
> e.g. below (just compiled, not tested)
> 
> diff --git a/hw/riscv/riscv-iommu.c b/hw/riscv/riscv-iommu.c
> index 1b34d226f9..74011c7f1f 100644
> --- a/hw/riscv/riscv-iommu.c
> +++ b/hw/riscv/riscv-iommu.c
> @@ -53,9 +53,12 @@ struct RISCVIOMMUSpace {
> 
>   /* Device translation context state. */
>   struct RISCVIOMMUContext {
> -    uint64_t devid:24;          /* Requester Id, AKA device_id */
> -    uint64_t pasid:20;          /* Process Address Space ID */
> -    uint64_t __rfu:20;          /* reserved */
> +    union {
> +        uint64_t devid:24,          /* Requester Id, AKA device_id */
> +                 pasid:20,          /* Process Address Space ID */
> +                 __rfu:20;          /* reserved */
> +        uint64_t key;
> +    };
>       uint64_t tc;                /* Translation Control */
>       uint64_t ta;                /* Translation Attributes */
>       uint64_t satp;              /* S-Stage address translation and protection */
> @@ -943,14 +946,14 @@ static gboolean __ctx_equal(gconstpointer v1, gconstpointer v2)
>   {
>       RISCVIOMMUContext *c1 = (RISCVIOMMUContext *) v1;
>       RISCVIOMMUContext *c2 = (RISCVIOMMUContext *) v2;
> -    return c1->devid == c2->devid && c1->pasid == c2->pasid;
> +    return c1->key == c2->key;
>   }
> 
>   static guint __ctx_hash(gconstpointer v)
>   {
>       RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) v;
>       /* Generate simple hash of (pasid, devid), assuming 24-bit wide devid */
> -    return (guint)(ctx->devid) + ((guint)(ctx->pasid) << 24);
> +    return (guint)ctx->key;
>   }
> 
>   static void __ctx_inval_devid_pasid(gpointer key, gpointer value, gpointer data)
> @@ -958,8 +961,7 @@ static void __ctx_inval_devid_pasid(gpointer key, gpointer value, gpointer data)
>       RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
>       RISCVIOMMUContext *arg = (RISCVIOMMUContext *) data;
>       if (ctx->tc & RISCV_IOMMU_DC_TC_V &&
> -        ctx->devid == arg->devid &&
> -        ctx->pasid == arg->pasid) {
> +        ctx->key == arg->key) {
>           ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
>       }
>   }
> @@ -989,6 +991,7 @@ static void riscv_iommu_ctx_inval(RISCVIOMMUState *s, GHFunc func,
>       RISCVIOMMUContext key = {
>           .devid = devid,
>           .pasid = pasid,
> +        .__rfu = 0,
>       };
>       ctx_cache = g_hash_table_ref(s->ctx_cache);
>       g_hash_table_foreach(ctx_cache, func, &key);
> @@ -1004,6 +1007,7 @@ static RISCVIOMMUContext *riscv_iommu_ctx(RISCVIOMMUState *s,
>       RISCVIOMMUContext key = {
>           .devid = devid,
>           .pasid = pasid,
> +        .__rfu = 0,
>       };
> 
>       ctx_cache = g_hash_table_ref(s->ctx_cache);
> 
> 
>> +
>> +/* IOMMU index for transactions without PASID specified. */
>> +#define RISCV_IOMMU_NOPASID 0
>> +
>> +static void riscv_iommu_notify(RISCVIOMMUState *s, int vec)
>> +{
>> +    const uint32_t fctl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FCTL);
>> +    uint32_t ipsr, ivec;
>> +
>> +    if (fctl & RISCV_IOMMU_FCTL_WSI || !s->notify) {
>> +        return;
>> +    }
>> +
>> +    ipsr = riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_IPSR, (1 << vec), 0);
>> +    ivec = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_IVEC);
>> +
>> +    if (!(ipsr & (1 << vec))) {
>> +        s->notify(s, (ivec >> (vec * 4)) & 0x0F);
>> +    }
>> +}
>> +
>> +static void riscv_iommu_fault(RISCVIOMMUState *s,
>> +                              struct riscv_iommu_fq_record *ev)
>> +{
>> +    uint32_t ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
>> +    uint32_t head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQH) & s->fq_mask;
>> +    uint32_t tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQT) & s->fq_mask;
>> +    uint32_t next = (tail + 1) & s->fq_mask;
>> +    uint32_t devid = get_field(ev->hdr, RISCV_IOMMU_FQ_HDR_DID);
>> +
>> +    trace_riscv_iommu_flt(s->parent_obj.id, PCI_BUS_NUM(devid), PCI_SLOT(devid),
>> +                          PCI_FUNC(devid), ev->hdr, ev->iotval);
>> +
>> +    if (!(ctrl & RISCV_IOMMU_FQCSR_FQON) ||
>> +        !!(ctrl & (RISCV_IOMMU_FQCSR_FQOF | RISCV_IOMMU_FQCSR_FQMF))) {
>> +        return;
>> +    }
>> +
>> +    if (head == next) {
>> +        riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR,
>> +                              RISCV_IOMMU_FQCSR_FQOF, 0);
>> +    } else {
>> +        dma_addr_t addr = s->fq_addr + tail * sizeof(*ev);
>> +        if (dma_memory_write(s->target_as, addr, ev, sizeof(*ev),
>> +                             MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
>> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR,
>> +                                  RISCV_IOMMU_FQCSR_FQMF, 0);
>> +        } else {
>> +            riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_FQT, next);
>> +        }
>> +    }
>> +
>> +    if (ctrl & RISCV_IOMMU_FQCSR_FIE) {
>> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_FQ);
>> +    }
>> +}
>> +
>> +static void riscv_iommu_pri(RISCVIOMMUState *s,
>> +    struct riscv_iommu_pq_record *pr)
>> +{
>> +    uint32_t ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
>> +    uint32_t head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQH) & s->pq_mask;
>> +    uint32_t tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQT) & s->pq_mask;
>> +    uint32_t next = (tail + 1) & s->pq_mask;
>> +    uint32_t devid = get_field(pr->hdr, RISCV_IOMMU_PREQ_HDR_DID);
>> +
>> +    trace_riscv_iommu_pri(s->parent_obj.id, PCI_BUS_NUM(devid), PCI_SLOT(devid),
>> +                          PCI_FUNC(devid), pr->payload);
>> +
>> +    if (!(ctrl & RISCV_IOMMU_PQCSR_PQON) ||
>> +        !!(ctrl & (RISCV_IOMMU_PQCSR_PQOF | RISCV_IOMMU_PQCSR_PQMF))) {
>> +        return;
>> +    }
>> +
>> +    if (head == next) {
>> +        riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR,
>> +                              RISCV_IOMMU_PQCSR_PQOF, 0);
>> +    } else {
>> +        dma_addr_t addr = s->pq_addr + tail * sizeof(*pr);
>> +        if (dma_memory_write(s->target_as, addr, pr, sizeof(*pr),
>> +                             MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
>> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR,
>> +                                  RISCV_IOMMU_PQCSR_PQMF, 0);
>> +        } else {
>> +            riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_PQT, next);
>> +        }
>> +    }
>> +
>> +    if (ctrl & RISCV_IOMMU_PQCSR_PIE) {
>> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_PQ);
>> +    }
>> +}
>> +
>> +/* Portable implementation of pext_u64, bit-mask extraction. */
>> +static uint64_t _pext_u64(uint64_t val, uint64_t ext)
>> +{
>> +    uint64_t ret = 0;
>> +    uint64_t rot = 1;
>> +
>> +    while (ext) {
>> +        if (ext & 1) {
>> +            if (val & 1) {
>> +                ret |= rot;
>> +            }
>> +            rot <<= 1;
>> +        }
>> +        val >>= 1;
>> +        ext >>= 1;
>> +    }
>> +
>> +    return ret;
>> +}
>> +
>> +/* Check if GPA matches MSI/MRIF pattern. */
>> +static bool riscv_iommu_msi_check(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
>> +    dma_addr_t gpa)
>> +{
>> +    if (get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_MODE) !=
>> +        RISCV_IOMMU_DC_MSIPTP_MODE_FLAT) {
>> +        return false; /* Invalid MSI/MRIF mode */
>> +    }
>> +
>> +    if ((PPN_DOWN(gpa) ^ ctx->msi_addr_pattern) & ~ctx->msi_addr_mask) {
>> +        return false; /* GPA not in MSI range defined by AIA IMSIC rules. */
>> +    }
>> +
>> +    return true;
>> +}
>> +
>> +/* RISCV IOMMU Address Translation Lookup - Page Table Walk */
>> +static int riscv_iommu_spa_fetch(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
>> +    IOMMUTLBEntry *iotlb)
>> +{
>> +    /* Early check for MSI address match when IOVA == GPA */
>> +    if (iotlb->perm & IOMMU_WO &&
>> +        riscv_iommu_msi_check(s, ctx, iotlb->iova)) {
>> +        iotlb->target_as = &s->trap_as;
>> +        iotlb->translated_addr = iotlb->iova;
>> +        iotlb->addr_mask = ~TARGET_PAGE_MASK;
>> +        return 0;
>> +    }
>> +
>> +    /* Exit early for pass-through mode. */
>> +    iotlb->translated_addr = iotlb->iova;
>> +    iotlb->addr_mask = ~TARGET_PAGE_MASK;
>> +    /* Allow R/W in pass-through mode */
>> +    iotlb->perm = IOMMU_RW;
>> +    return 0;
>> +}
>> +
>> +static void riscv_iommu_report_fault(RISCVIOMMUState *s,
>> +                                     RISCVIOMMUContext *ctx,
>> +                                     uint32_t fault_type, uint32_t cause,
>> +                                     bool pv,
>> +                                     uint64_t iotval, uint64_t iotval2)
>> +{
>> +    struct riscv_iommu_fq_record ev = { 0 };
>> +
>> +    if (ctx->tc & RISCV_IOMMU_DC_TC_DTF) {
>> +        switch (cause) {
>> +        case RISCV_IOMMU_FQ_CAUSE_DMA_DISABLED:
>> +        case RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT:
>> +        case RISCV_IOMMU_FQ_CAUSE_DDT_INVALID:
>> +        case RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED:
>> +        case RISCV_IOMMU_FQ_CAUSE_DDT_CORRUPTED:
>> +        case RISCV_IOMMU_FQ_CAUSE_INTERNAL_DP_ERROR:
>> +        case RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT:
>> +            break;
>> +        default:
>> +            /* DTF prevents reporting a fault for this given cause */
>> +            return;
>> +        }
>> +    }
>> +
>> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_CAUSE, cause);
>> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_TTYPE, fault_type);
>> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_DID, ctx->devid);
>> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_PV, true);
>> +
>> +    if (pv) {
>> +        ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_PID, ctx->pasid);
>> +    }
>> +
>> +    ev.iotval = iotval;
>> +    ev.iotval2 = iotval2;
>> +
>> +    riscv_iommu_fault(s, &ev);
>> +}
>> +
>> +/* Redirect MSI write for given GPA. */
>> +static MemTxResult riscv_iommu_msi_write(RISCVIOMMUState *s,
>> +    RISCVIOMMUContext *ctx, uint64_t gpa, uint64_t data,
>> +    unsigned size, MemTxAttrs attrs)
>> +{
>> +    MemTxResult res;
>> +    dma_addr_t addr;
>> +    uint64_t intn;
>> +    uint32_t n190;
>> +    uint64_t pte[2];
>> +    int fault_type = RISCV_IOMMU_FQ_TTYPE_UADDR_WR;
>> +    int cause;
>> +
>> +    if (!riscv_iommu_msi_check(s, ctx, gpa)) {
>> +        res = MEMTX_ACCESS_ERROR;
>> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
>> +        goto err;
>> +    }
>> +
>> +    /* Interrupt File Number */
>> +    intn = _pext_u64(PPN_DOWN(gpa), ctx->msi_addr_mask);
>> +    if (intn >= 256) {
>> +        /* Interrupt file number out of range */
>> +        res = MEMTX_ACCESS_ERROR;
>> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
>> +        goto err;
>> +    }
>> +
>> +    /* fetch MSI PTE */
>> +    addr = PPN_PHYS(get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_PPN));
>> +    addr = addr | (intn * sizeof(pte));
>> +    res = dma_memory_read(s->target_as, addr, &pte, sizeof(pte),
>> +            MEMTXATTRS_UNSPECIFIED);
>> +    if (res != MEMTX_OK) {
>> +        if (res == MEMTX_DECODE_ERROR) {
>> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_PT_CORRUPTED;
>> +        } else {
>> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
>> +        }
>> +        goto err;
>> +    }
>> +
>> +    le64_to_cpus(&pte[0]);
>> +    le64_to_cpus(&pte[1]);
>> +
>> +    if (!(pte[0] & RISCV_IOMMU_MSI_PTE_V) || (pte[0] & RISCV_IOMMU_MSI_PTE_C)) {
>> +        /*
>> +         * The spec mentions that: "If msipte.C == 1, then further
>> +         * processing to interpret the PTE is implementation
>> +         * defined.". We'll abort with cause = 262 for this
>> +         * case too.
>> +         */
>> +        res = MEMTX_ACCESS_ERROR;
>> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_INVALID;
>> +        goto err;
>> +    }
>> +
>> +    switch (get_field(pte[0], RISCV_IOMMU_MSI_PTE_M)) {
>> +    case RISCV_IOMMU_MSI_PTE_M_BASIC:
>> +        /* MSI Pass-through mode */
>> +        addr = PPN_PHYS(get_field(pte[0], RISCV_IOMMU_MSI_PTE_PPN));
>> +        addr = addr | (gpa & TARGET_PAGE_MASK);
>> +
>> +        trace_riscv_iommu_msi(s->parent_obj.id, PCI_BUS_NUM(ctx->devid),
>> +                              PCI_SLOT(ctx->devid), PCI_FUNC(ctx->devid),
>> +                              gpa, addr);
>> +
>> +        res = dma_memory_write(s->target_as, addr, &data, size, attrs);
>> +        if (res != MEMTX_OK) {
>> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
>> +            goto err;
>> +        }
>> +
>> +        return MEMTX_OK;
>> +    case RISCV_IOMMU_MSI_PTE_M_MRIF:
>> +        /* MRIF mode, continue. */
>> +        break;
>> +    default:
>> +        res = MEMTX_ACCESS_ERROR;
>> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_MISCONFIGURED;
>> +        goto err;
>> +    }
>> +
>> +    /*
>> +     * Report an error for interrupt identities exceeding the maximum allowed
>> +     * for an IMSIC interrupt file (2047) or destination address is not 32-bit
>> +     * aligned. See IOMMU Specification, Chapter 2.3. MSI page tables.
>> +     */
>> +    if ((data > 2047) || (gpa & 3)) {
>> +        res = MEMTX_ACCESS_ERROR;
>> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_MISCONFIGURED;
>> +        goto err;
>> +    }
>> +
>> +    /* MSI MRIF mode, non atomic pending bit update */
>> +
>> +    /* MRIF pending bit address */
>> +    addr = get_field(pte[0], RISCV_IOMMU_MSI_PTE_MRIF_ADDR) << 9;
>> +    addr = addr | ((data & 0x7c0) >> 3);
>> +
>> +    trace_riscv_iommu_msi(s->parent_obj.id, PCI_BUS_NUM(ctx->devid),
>> +                          PCI_SLOT(ctx->devid), PCI_FUNC(ctx->devid),
>> +                          gpa, addr);
>> +
>> +    /* MRIF pending bit mask */
>> +    data = 1ULL << (data & 0x03f);
>> +    res = dma_memory_read(s->target_as, addr, &intn, sizeof(intn), attrs);
>> +    if (res != MEMTX_OK) {
>> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
>> +        goto err;
>> +    }
>> +
>> +    intn = intn | data;
>> +    res = dma_memory_write(s->target_as, addr, &intn, sizeof(intn), attrs);
>> +    if (res != MEMTX_OK) {
>> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
>> +        goto err;
>> +    }
>> +
>> +    /* Get MRIF enable bits */
>> +    addr = addr + sizeof(intn);
>> +    res = dma_memory_read(s->target_as, addr, &intn, sizeof(intn), attrs);
>> +    if (res != MEMTX_OK) {
>> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
>> +        goto err;
>> +    }
>> +
>> +    if (!(intn & data)) {
>> +        /* notification disabled, MRIF update completed. */
>> +        return MEMTX_OK;
>> +    }
>> +
>> +    /* Send notification message */
>> +    addr = PPN_PHYS(get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NPPN));
>> +    n190 = get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NID) |
>> +          (get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NID_MSB) << 10);
>> +
>> +    res = dma_memory_write(s->target_as, addr, &n190, sizeof(n190), attrs);
>> +    if (res != MEMTX_OK) {
>> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
>> +        goto err;
>> +    }
>> +
>> +    return MEMTX_OK;
>> +
>> +err:
>> +    riscv_iommu_report_fault(s, ctx, fault_type, cause,
>> +                             !!ctx->pasid, 0, 0);
>> +    return res;
>> +}
>> +
>> +/*
>> + * Check device context configuration as described by the
>> + * riscv-iommu spec section "Device-context configuration
>> + * checks".
>> + */
>> +static bool riscv_iommu_validate_device_ctx(RISCVIOMMUState *s,
>> +                                            RISCVIOMMUContext *ctx)
>> +{
>> +    uint32_t msi_mode;
>> +
>> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_EN_PRI) &&
>> +        ctx->tc & RISCV_IOMMU_DC_TC_PRPR) {
>> +        return false;
>> +    }
>> +
>> +    if (!(s->cap & RISCV_IOMMU_CAP_T2GPA) &&
>> +        ctx->tc & RISCV_IOMMU_DC_TC_T2GPA) {
>> +        return false;
>> +    }
>> +
>> +    if (s->cap & RISCV_IOMMU_CAP_MSI_FLAT) {
>> +        msi_mode = get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_MODE);
>> +
>> +        if (msi_mode != RISCV_IOMMU_DC_MSIPTP_MODE_OFF &&
>> +            msi_mode != RISCV_IOMMU_DC_MSIPTP_MODE_FLAT) {
>> +            return false;
>> +        }
>> +    }
>> +
>> +    /*
>> +     * CAP_END is always zero (only one endianess). FCTL_BE is
>> +     * always zero (little-endian accesses). Thus TC_SBE must
>> +     * always be LE, i.e. zero.
>> +     */
>> +    if (ctx->tc & RISCV_IOMMU_DC_TC_SBE) {
>> +        return false;
>> +    }
>> +
>> +    return true;
>> +}
>> +
>> +/*
>> + * RISC-V IOMMU Device Context Loopkup - Device Directory Tree Walk
>> + *
>> + * @s         : IOMMU Device State
>> + * @ctx       : Device Translation Context with devid and pasid set.
>> + * @return    : success or fault code.
>> + */
>> +static int riscv_iommu_ctx_fetch(RISCVIOMMUState *s, RISCVIOMMUContext *ctx)
>> +{
>> +    const uint64_t ddtp = s->ddtp;
>> +    unsigned mode = get_field(ddtp, RISCV_IOMMU_DDTP_MODE);
>> +    dma_addr_t addr = PPN_PHYS(get_field(ddtp, RISCV_IOMMU_DDTP_PPN));
>> +    struct riscv_iommu_dc dc;
>> +    /* Device Context format: 0: extended (64 bytes) | 1: base (32 bytes) */
>> +    const int dc_fmt = !s->enable_msi;
>> +    const size_t dc_len = sizeof(dc) >> dc_fmt;
>> +    unsigned depth;
>> +    uint64_t de;
>> +
>> +    switch (mode) {
>> +    case RISCV_IOMMU_DDTP_MODE_OFF:
>> +        return RISCV_IOMMU_FQ_CAUSE_DMA_DISABLED;
>> +
>> +    case RISCV_IOMMU_DDTP_MODE_BARE:
>> +        /* mock up pass-through translation context */
>> +        ctx->tc = RISCV_IOMMU_DC_TC_V;
>> +        ctx->ta = 0;
>> +        ctx->msiptp = 0;
>> +        return 0;
>> +
>> +    case RISCV_IOMMU_DDTP_MODE_1LVL:
>> +        depth = 0;
>> +        break;
>> +
>> +    case RISCV_IOMMU_DDTP_MODE_2LVL:
>> +        depth = 1;
>> +        break;
>> +
>> +    case RISCV_IOMMU_DDTP_MODE_3LVL:
>> +        depth = 2;
>> +        break;
>> +
>> +    default:
>> +        return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
>> +    }
>> +
>> +    /*
>> +     * Check supported device id width (in bits).
>> +     * See IOMMU Specification, Chapter 6. Software guidelines.
>> +     * - if extended device-context format is used:
>> +     *   1LVL: 6, 2LVL: 15, 3LVL: 24
>> +     * - if base device-context format is used:
>> +     *   1LVL: 7, 2LVL: 16, 3LVL: 24
>> +     */
>> +    if (ctx->devid >= (1 << (depth * 9 + 6 + (dc_fmt && depth != 2)))) {
>> +        return RISCV_IOMMU_FQ_CAUSE_TTYPE_BLOCKED;
>> +    }
>> +
>> +    /* Device directory tree walk */
>> +    for (; depth-- > 0; ) {
>> +        /*
>> +         * Select device id index bits based on device directory tree level
>> +         * and device context format.
>> +         * See IOMMU Specification, Chapter 2. Data Structures.
>> +         * - if extended device-context format is used:
>> +         *   device index: [23:15][14:6][5:0]
>> +         * - if base device-context format is used:
>> +         *   device index: [23:16][15:7][6:0]
>> +         */
>> +        const int split = depth * 9 + 6 + dc_fmt;
>> +        addr |= ((ctx->devid >> split) << 3) & ~TARGET_PAGE_MASK;
>> +        if (dma_memory_read(s->target_as, addr, &de, sizeof(de),
>> +                            MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
>> +            return RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT;
>> +        }
>> +        le64_to_cpus(&de);
>> +        if (!(de & RISCV_IOMMU_DDTE_VALID)) {
>> +            /* invalid directory entry */
>> +            return RISCV_IOMMU_FQ_CAUSE_DDT_INVALID;
>> +        }
>> +        if (de & ~(RISCV_IOMMU_DDTE_PPN | RISCV_IOMMU_DDTE_VALID)) {
>> +            /* reserved bits set */
>> +            return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
>> +        }
>> +        addr = PPN_PHYS(get_field(de, RISCV_IOMMU_DDTE_PPN));
>> +    }
>> +
>> +    /* index into device context entry page */
>> +    addr |= (ctx->devid * dc_len) & ~TARGET_PAGE_MASK;
>> +
>> +    memset(&dc, 0, sizeof(dc));
>> +    if (dma_memory_read(s->target_as, addr, &dc, dc_len,
>> +                        MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
>> +        return RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT;
>> +    }
>> +
>> +    /* Set translation context. */
>> +    ctx->tc = le64_to_cpu(dc.tc);
>> +    ctx->ta = le64_to_cpu(dc.ta);
>> +    ctx->msiptp = le64_to_cpu(dc.msiptp);
>> +    ctx->msi_addr_mask = le64_to_cpu(dc.msi_addr_mask);
>> +    ctx->msi_addr_pattern = le64_to_cpu(dc.msi_addr_pattern);
>> +
>> +    if (!riscv_iommu_validate_device_ctx(s, ctx)) {
>> +        return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
>> +    }
>> +
>> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_V)) {
>> +        return RISCV_IOMMU_FQ_CAUSE_DDT_INVALID;
>> +    }
>> +
>> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_PDTV)) {
>> +        if (ctx->pasid != RISCV_IOMMU_NOPASID) {
>> +            /* PASID is disabled */
>> +            return RISCV_IOMMU_FQ_CAUSE_TTYPE_BLOCKED;
>> +        }
>> +        return 0;
>> +    }
>> +
>> +    /* FSC.TC.PDTV enabled */
>> +    if (mode > RISCV_IOMMU_DC_FSC_PDTP_MODE_PD20) {
>> +        /* Invalid PDTP.MODE */
>> +        return RISCV_IOMMU_FQ_CAUSE_PDT_MISCONFIGURED;
>> +    }
>> +
>> +    for (depth = mode - RISCV_IOMMU_DC_FSC_PDTP_MODE_PD8; depth-- > 0; ) {
>> +        /*
>> +         * Select process id index bits based on process directory tree
>> +         * level. See IOMMU Specification, 2.2. Process-Directory-Table.
>> +         */
>> +        const int split = depth * 9 + 8;
>> +        addr |= ((ctx->pasid >> split) << 3) & ~TARGET_PAGE_MASK;
>> +        if (dma_memory_read(s->target_as, addr, &de, sizeof(de),
>> +                            MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
>> +            return RISCV_IOMMU_FQ_CAUSE_PDT_LOAD_FAULT;
>> +        }
>> +        le64_to_cpus(&de);
>> +        if (!(de & RISCV_IOMMU_PC_TA_V)) {
>> +            return RISCV_IOMMU_FQ_CAUSE_PDT_INVALID;
>> +        }
>> +        addr = PPN_PHYS(get_field(de, RISCV_IOMMU_PC_FSC_PPN));
>> +    }
>> +
>> +    /* Leaf entry in PDT */
>> +    addr |= (ctx->pasid << 4) & ~TARGET_PAGE_MASK;
>> +    if (dma_memory_read(s->target_as, addr, &dc.ta, sizeof(uint64_t) * 2,
>> +                        MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
>> +        return RISCV_IOMMU_FQ_CAUSE_PDT_LOAD_FAULT;
>> +    }
>> +
>> +    /* Use FSC and TA from process directory entry. */
>> +    ctx->ta = le64_to_cpu(dc.ta);
>> +
>> +    return 0;
>> +}
>> +
>> +/* Translation Context cache support */
>> +static gboolean __ctx_equal(gconstpointer v1, gconstpointer v2)
>> +{
>> +    RISCVIOMMUContext *c1 = (RISCVIOMMUContext *) v1;
>> +    RISCVIOMMUContext *c2 = (RISCVIOMMUContext *) v2;
>> +    return c1->devid == c2->devid && c1->pasid == c2->pasid;
>> +}
>> +
>> +static guint __ctx_hash(gconstpointer v)
>> +{
>> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) v;
>> +    /* Generate simple hash of (pasid, devid), assuming 24-bit wide devid */
>> +    return (guint)(ctx->devid) + ((guint)(ctx->pasid) << 24);
>> +}
>> +
>> +static void __ctx_inval_devid_pasid(gpointer key, gpointer value, gpointer data)
>> +{
>> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
>> +    RISCVIOMMUContext *arg = (RISCVIOMMUContext *) data;
>> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V &&
>> +        ctx->devid == arg->devid &&
>> +        ctx->pasid == arg->pasid) {
>> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
>> +    }
>> +}
>> +
>> +static void __ctx_inval_devid(gpointer key, gpointer value, gpointer data)
>> +{
>> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
>> +    RISCVIOMMUContext *arg = (RISCVIOMMUContext *) data;
>> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V &&
>> +        ctx->devid == arg->devid) {
>> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
>> +    }
>> +}
>> +
>> +static void __ctx_inval_all(gpointer key, gpointer value, gpointer data)
>> +{
>> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
>> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V) {
>> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
>> +    }
>> +}
>> +
>> +static void riscv_iommu_ctx_inval(RISCVIOMMUState *s, GHFunc func,
>> +    uint32_t devid, uint32_t pasid)
>> +{
>> +    GHashTable *ctx_cache;
>> +    RISCVIOMMUContext key = {
>> +        .devid = devid,
>> +        .pasid = pasid,
>> +    };
>> +    ctx_cache = g_hash_table_ref(s->ctx_cache);
>> +    g_hash_table_foreach(ctx_cache, func, &key);
>> +    g_hash_table_unref(ctx_cache);
>> +}
>> +
>> +/* Find or allocate translation context for a given {device_id, process_id} */
>> +static RISCVIOMMUContext *riscv_iommu_ctx(RISCVIOMMUState *s,
>> +    unsigned devid, unsigned pasid, void **ref)
>> +{
>> +    GHashTable *ctx_cache;
>> +    RISCVIOMMUContext *ctx;
>> +    RISCVIOMMUContext key = {
>> +        .devid = devid,
>> +        .pasid = pasid,
>> +    };
>> +
>> +    ctx_cache = g_hash_table_ref(s->ctx_cache);
>> +    ctx = g_hash_table_lookup(ctx_cache, &key);
>> +
>> +    if (ctx && (ctx->tc & RISCV_IOMMU_DC_TC_V)) {
>> +        *ref = ctx_cache;
>> +        return ctx;
>> +    }
>> +
>> +    if (g_hash_table_size(s->ctx_cache) >= LIMIT_CACHE_CTX) {
>> +        ctx_cache = g_hash_table_new_full(__ctx_hash, __ctx_equal,
>> +                                          g_free, NULL);
>> +        g_hash_table_unref(qatomic_xchg(&s->ctx_cache, ctx_cache));
>> +    }
>> +
>> +    ctx = g_new0(RISCVIOMMUContext, 1);
>> +    ctx->devid = devid;
>> +    ctx->pasid = pasid;
>> +
>> +    int fault = riscv_iommu_ctx_fetch(s, ctx);
>> +    if (!fault) {
>> +        g_hash_table_add(ctx_cache, ctx);
>> +        *ref = ctx_cache;
>> +        return ctx;
>> +    }
>> +
>> +    g_hash_table_unref(ctx_cache);
>> +    *ref = NULL;
>> +
>> +    riscv_iommu_report_fault(s, ctx, RISCV_IOMMU_FQ_TTYPE_UADDR_RD,
>> +                             fault, !!pasid, 0, 0);
>> +
>> +    g_free(ctx);
>> +    return NULL;
>> +}
>> +
>> +static void riscv_iommu_ctx_put(RISCVIOMMUState *s, void *ref)
>> +{
>> +    if (ref) {
>> +        g_hash_table_unref((GHashTable *)ref);
>> +    }
>> +}
>> +
>> +/* Find or allocate address space for a given device */
>> +static AddressSpace *riscv_iommu_space(RISCVIOMMUState *s, uint32_t devid)
>> +{
>> +    RISCVIOMMUSpace *as;
>> +
>> +    /* FIXME: PCIe bus remapping for attached endpoints. */
>> +    devid |= s->bus << 8;
>> +
>> +    qemu_mutex_lock(&s->core_lock);
>> +    QLIST_FOREACH(as, &s->spaces, list) {
>> +        if (as->devid == devid) {
>> +            break;
>> +        }
>> +    }
>> +    qemu_mutex_unlock(&s->core_lock);
>> +
>> +    if (as == NULL) {
>> +        char name[64];
>> +        as = g_new0(RISCVIOMMUSpace, 1);
>> +
>> +        as->iommu = s;
>> +        as->devid = devid;
>> +
>> +        snprintf(name, sizeof(name), "riscv-iommu-%04x:%02x.%d-iova",
>> +            PCI_BUS_NUM(as->devid), PCI_SLOT(as->devid), PCI_FUNC(as->devid));
>> +
>> +        /* IOVA address space, untranslated addresses */
>> +        memory_region_init_iommu(&as->iova_mr, sizeof(as->iova_mr),
>> +            TYPE_RISCV_IOMMU_MEMORY_REGION,
>> +            OBJECT(as), "riscv_iommu", UINT64_MAX);
>> +        address_space_init(&as->iova_as, MEMORY_REGION(&as->iova_mr), name);
>> +
>> +        qemu_mutex_lock(&s->core_lock);
>> +        QLIST_INSERT_HEAD(&s->spaces, as, list);
>> +        qemu_mutex_unlock(&s->core_lock);
>> +
>> +        trace_riscv_iommu_new(s->parent_obj.id, PCI_BUS_NUM(as->devid),
>> +                PCI_SLOT(as->devid), PCI_FUNC(as->devid));
>> +    }
>> +    return &as->iova_as;
>> +}
>> +
>> +static int riscv_iommu_translate(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
>> +    IOMMUTLBEntry *iotlb)
>> +{
>> +    bool enable_pasid;
>> +    bool enable_pri;
>> +    int fault;
>> +
>> +    /*
>> +     * TC[32] is reserved for custom extensions, used here to temporarily
>> +     * enable automatic page-request generation for ATS queries.
>> +     */
>> +    enable_pri = (iotlb->perm == IOMMU_NONE) && (ctx->tc & BIT_ULL(32));
>> +    enable_pasid = (ctx->tc & RISCV_IOMMU_DC_TC_PDTV);
>> +
>> +    /* Translate using device directory / page table information. */
>> +    fault = riscv_iommu_spa_fetch(s, ctx, iotlb);
>> +
>> +    if (enable_pri && fault) {
>> +        struct riscv_iommu_pq_record pr = {0};
>> +        if (enable_pasid) {
>> +            pr.hdr = set_field(RISCV_IOMMU_PREQ_HDR_PV,
>> +                RISCV_IOMMU_PREQ_HDR_PID, ctx->pasid);
>> +        }
>> +        pr.hdr = set_field(pr.hdr, RISCV_IOMMU_PREQ_HDR_DID, ctx->devid);
>> +        pr.payload = (iotlb->iova & TARGET_PAGE_MASK) |
>> +                     RISCV_IOMMU_PREQ_PAYLOAD_M;
>> +        riscv_iommu_pri(s, &pr);
>> +        return fault;
>> +    }
>> +
>> +    if (fault) {
>> +        unsigned ttype;
>> +
>> +        if (iotlb->perm & IOMMU_RW) {
>> +            ttype = RISCV_IOMMU_FQ_TTYPE_UADDR_WR;
>> +        } else {
>> +            ttype = RISCV_IOMMU_FQ_TTYPE_UADDR_RD;
>> +        }
>> +
>> +        riscv_iommu_report_fault(s, ctx, ttype, fault, enable_pasid,
>> +                                 iotlb->iova, iotlb->translated_addr);
>> +        return fault;
>> +    }
>> +
>> +    return 0;
>> +}
>> +
>> +/* IOMMU Command Interface */
>> +static MemTxResult riscv_iommu_iofence(RISCVIOMMUState *s, bool notify,
>> +    uint64_t addr, uint32_t data)
>> +{
>> +    /*
>> +     * ATS processing in this implementation of the IOMMU is synchronous,
>> +     * no need to wait for completions here.
>> +     */
>> +    if (!notify) {
>> +        return MEMTX_OK;
>> +    }
>> +
>> +    return dma_memory_write(s->target_as, addr, &data, sizeof(data),
>> +        MEMTXATTRS_UNSPECIFIED);
>> +}
>> +
>> +static void riscv_iommu_process_ddtp(RISCVIOMMUState *s)
>> +{
>> +    uint64_t old_ddtp = s->ddtp;
>> +    uint64_t new_ddtp = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_DDTP);
>> +    unsigned new_mode = get_field(new_ddtp, RISCV_IOMMU_DDTP_MODE);
>> +    unsigned old_mode = get_field(old_ddtp, RISCV_IOMMU_DDTP_MODE);
>> +    bool ok = false;
>> +
>> +    /*
>> +     * Check for allowed DDTP.MODE transitions:
>> +     * {OFF, BARE}        -> {OFF, BARE, 1LVL, 2LVL, 3LVL}
>> +     * {1LVL, 2LVL, 3LVL} -> {OFF, BARE}
>> +     */
>> +    if (new_mode == old_mode ||
>> +        new_mode == RISCV_IOMMU_DDTP_MODE_OFF ||
>> +        new_mode == RISCV_IOMMU_DDTP_MODE_BARE) {
>> +        ok = true;
>> +    } else if (new_mode == RISCV_IOMMU_DDTP_MODE_1LVL ||
>> +               new_mode == RISCV_IOMMU_DDTP_MODE_2LVL ||
>> +               new_mode == RISCV_IOMMU_DDTP_MODE_3LVL) {
>> +        ok = old_mode == RISCV_IOMMU_DDTP_MODE_OFF ||
>> +             old_mode == RISCV_IOMMU_DDTP_MODE_BARE;
>> +    }
>> +
>> +    if (ok) {
>> +        /* clear reserved and busy bits, report back sanitized version */
>> +        new_ddtp = set_field(new_ddtp & RISCV_IOMMU_DDTP_PPN,
>> +                             RISCV_IOMMU_DDTP_MODE, new_mode);
>> +    } else {
>> +        new_ddtp = old_ddtp;
>> +    }
>> +    s->ddtp = new_ddtp;
>> +
>> +    riscv_iommu_reg_set64(s, RISCV_IOMMU_REG_DDTP, new_ddtp);
>> +}
>> +
>> +/* Command function and opcode field. */
>> +#define RISCV_IOMMU_CMD(func, op) (((func) << 7) | (op))
>> +
>> +static void riscv_iommu_process_cq_tail(RISCVIOMMUState *s)
>> +{
>> +    struct riscv_iommu_command cmd;
>> +    MemTxResult res;
>> +    dma_addr_t addr;
>> +    uint32_t tail, head, ctrl;
>> +    uint64_t cmd_opcode;
>> +    GHFunc func;
>> +
>> +    ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
>> +    tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQT) & s->cq_mask;
>> +    head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQH) & s->cq_mask;
>> +
>> +    /* Check for pending error or queue processing disabled */
>> +    if (!(ctrl & RISCV_IOMMU_CQCSR_CQON) ||
>> +        !!(ctrl & (RISCV_IOMMU_CQCSR_CMD_ILL | RISCV_IOMMU_CQCSR_CQMF))) {
>> +        return;
>> +    }
>> +
>> +    while (tail != head) {
>> +        addr = s->cq_addr  + head * sizeof(cmd);
>> +        res = dma_memory_read(s->target_as, addr, &cmd, sizeof(cmd),
>> +                              MEMTXATTRS_UNSPECIFIED);
>> +
>> +        if (res != MEMTX_OK) {
>> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
>> +                                  RISCV_IOMMU_CQCSR_CQMF, 0);
>> +            goto fault;
>> +        }
>> +
>> +        trace_riscv_iommu_cmd(s->parent_obj.id, cmd.dword0, cmd.dword1);
>> +
>> +        cmd_opcode = get_field(cmd.dword0,
>> +                               RISCV_IOMMU_CMD_OPCODE | RISCV_IOMMU_CMD_FUNC);
>> +
>> +        switch (cmd_opcode) {
>> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOFENCE_FUNC_C,
>> +                             RISCV_IOMMU_CMD_IOFENCE_OPCODE):
>> +            res = riscv_iommu_iofence(s,
>> +                cmd.dword0 & RISCV_IOMMU_CMD_IOFENCE_AV, cmd.dword1,
>> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IOFENCE_DATA));
>> +
>> +            if (res != MEMTX_OK) {
>> +                riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
>> +                                      RISCV_IOMMU_CQCSR_CQMF, 0);
>> +                goto fault;
>> +            }
>> +            break;
>> +
>> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOTINVAL_FUNC_GVMA,
>> +                             RISCV_IOMMU_CMD_IOTINVAL_OPCODE):
>> +            if (cmd.dword0 & RISCV_IOMMU_CMD_IOTINVAL_PSCV) {
>> +                /* illegal command arguments IOTINVAL.GVMA & PSCV == 1 */
>> +                goto cmd_ill;
>> +            }
>> +            /* translation cache not implemented yet */
>> +            break;
>> +
>> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOTINVAL_FUNC_VMA,
>> +                             RISCV_IOMMU_CMD_IOTINVAL_OPCODE):
>> +            /* translation cache not implemented yet */
>> +            break;
>> +
>> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IODIR_FUNC_INVAL_DDT,
>> +                             RISCV_IOMMU_CMD_IODIR_OPCODE):
>> +            if (!(cmd.dword0 & RISCV_IOMMU_CMD_IODIR_DV)) {
>> +                /* invalidate all device context cache mappings */
>> +                func = __ctx_inval_all;
>> +            } else {
>> +                /* invalidate all device context matching DID */
>> +                func = __ctx_inval_devid;
>> +            }
>> +            riscv_iommu_ctx_inval(s, func,
>> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_DID), 0);
>> +            break;
>> +
>> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IODIR_FUNC_INVAL_PDT,
>> +                             RISCV_IOMMU_CMD_IODIR_OPCODE):
>> +            if (!(cmd.dword0 & RISCV_IOMMU_CMD_IODIR_DV)) {
>> +                /* illegal command arguments IODIR_PDT & DV == 0 */
>> +                goto cmd_ill;
>> +            } else {
>> +                func = __ctx_inval_devid_pasid;
>> +            }
>> +            riscv_iommu_ctx_inval(s, func,
>> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_DID),
>> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_PID));
>> +            break;
>> +
>> +        default:
>> +        cmd_ill:
>> +            /* Invalid instruction, do not advance instruction index. */
>> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
>> +                RISCV_IOMMU_CQCSR_CMD_ILL, 0);
>> +            goto fault;
>> +        }
>> +
>> +        /* Advance and update head pointer after command completes. */
>> +        head = (head + 1) & s->cq_mask;
>> +        riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_CQH, head);
>> +    }
>> +    return;
>> +
>> +fault:
>> +    if (ctrl & RISCV_IOMMU_CQCSR_CIE) {
>> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_CQ);
>> +    }
>> +}
>> +
>> +static void riscv_iommu_process_cq_control(RISCVIOMMUState *s)
>> +{
>> +    uint64_t base;
>> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
>> +    uint32_t ctrl_clr;
>> +    bool enable = !!(ctrl_set & RISCV_IOMMU_CQCSR_CQEN);
>> +    bool active = !!(ctrl_set & RISCV_IOMMU_CQCSR_CQON);
>> +
>> +    if (enable && !active) {
>> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_CQB);
>> +        s->cq_mask = (2ULL << get_field(base, RISCV_IOMMU_CQB_LOG2SZ)) - 1;
>> +        s->cq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_CQB_PPN));
>> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQT], ~s->cq_mask);
>> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_CQH], 0);
>> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_CQT], 0);
>> +        ctrl_set = RISCV_IOMMU_CQCSR_CQON;
>> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY | RISCV_IOMMU_CQCSR_CQMF |
>> +                   RISCV_IOMMU_CQCSR_CMD_ILL | RISCV_IOMMU_CQCSR_CMD_TO |
>> +                   RISCV_IOMMU_CQCSR_FENCE_W_IP;
>> +    } else if (!enable && active) {
>> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQT], ~0);
>> +        ctrl_set = 0;
>> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY | RISCV_IOMMU_CQCSR_CQON;
>> +    } else {
>> +        ctrl_set = 0;
>> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY;
>> +    }
>> +
>> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR, ctrl_set, ctrl_clr);
>> +}
>> +
>> +static void riscv_iommu_process_fq_control(RISCVIOMMUState *s)
>> +{
>> +    uint64_t base;
>> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
>> +    uint32_t ctrl_clr;
>> +    bool enable = !!(ctrl_set & RISCV_IOMMU_FQCSR_FQEN);
>> +    bool active = !!(ctrl_set & RISCV_IOMMU_FQCSR_FQON);
>> +
>> +    if (enable && !active) {
>> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_FQB);
>> +        s->fq_mask = (2ULL << get_field(base, RISCV_IOMMU_FQB_LOG2SZ)) - 1;
>> +        s->fq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_FQB_PPN));
>> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQH], ~s->fq_mask);
>> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_FQH], 0);
>> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_FQT], 0);
>> +        ctrl_set = RISCV_IOMMU_FQCSR_FQON;
>> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY | RISCV_IOMMU_FQCSR_FQMF |
>> +            RISCV_IOMMU_FQCSR_FQOF;
>> +    } else if (!enable && active) {
>> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQH], ~0);
>> +        ctrl_set = 0;
>> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY | RISCV_IOMMU_FQCSR_FQON;
>> +    } else {
>> +        ctrl_set = 0;
>> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY;
>> +    }
>> +
>> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR, ctrl_set, ctrl_clr);
>> +}
>> +
>> +static void riscv_iommu_process_pq_control(RISCVIOMMUState *s)
>> +{
>> +    uint64_t base;
>> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
>> +    uint32_t ctrl_clr;
>> +    bool enable = !!(ctrl_set & RISCV_IOMMU_PQCSR_PQEN);
>> +    bool active = !!(ctrl_set & RISCV_IOMMU_PQCSR_PQON);
>> +
>> +    if (enable && !active) {
>> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_PQB);
>> +        s->pq_mask = (2ULL << get_field(base, RISCV_IOMMU_PQB_LOG2SZ)) - 1;
>> +        s->pq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_PQB_PPN));
>> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQH], ~s->pq_mask);
>> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_PQH], 0);
>> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_PQT], 0);
>> +        ctrl_set = RISCV_IOMMU_PQCSR_PQON;
>> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY | RISCV_IOMMU_PQCSR_PQMF |
>> +            RISCV_IOMMU_PQCSR_PQOF;
>> +    } else if (!enable && active) {
>> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQH], ~0);
>> +        ctrl_set = 0;
>> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY | RISCV_IOMMU_PQCSR_PQON;
>> +    } else {
>> +        ctrl_set = 0;
>> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY;
>> +    }
>> +
>> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR, ctrl_set, ctrl_clr);
>> +}
>> +
>> +typedef void riscv_iommu_process_fn(RISCVIOMMUState *s);
>> +
>> +static void riscv_iommu_update_ipsr(RISCVIOMMUState *s, uint64_t data)
>> +{
>> +    uint32_t cqcsr, fqcsr, pqcsr;
>> +    uint32_t ipsr_set = 0;
>> +    uint32_t ipsr_clr = 0;
>> +
>> +    if (data & RISCV_IOMMU_IPSR_CIP) {
>> +        cqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
>> +
>> +        if (cqcsr & RISCV_IOMMU_CQCSR_CIE &&
>> +            (cqcsr & RISCV_IOMMU_CQCSR_FENCE_W_IP ||
>> +             cqcsr & RISCV_IOMMU_CQCSR_CMD_ILL ||
>> +             cqcsr & RISCV_IOMMU_CQCSR_CMD_TO ||
>> +             cqcsr & RISCV_IOMMU_CQCSR_CQMF)) {
>> +            ipsr_set |= RISCV_IOMMU_IPSR_CIP;
>> +        } else {
>> +            ipsr_clr |= RISCV_IOMMU_IPSR_CIP;
>> +        }
>> +    } else {
>> +        ipsr_clr |= RISCV_IOMMU_IPSR_CIP;
>> +    }
>> +
>> +    if (data & RISCV_IOMMU_IPSR_FIP) {
>> +        fqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
>> +
>> +        if (fqcsr & RISCV_IOMMU_FQCSR_FIE &&
>> +            (fqcsr & RISCV_IOMMU_FQCSR_FQOF ||
>> +             fqcsr & RISCV_IOMMU_FQCSR_FQMF)) {
>> +            ipsr_set |= RISCV_IOMMU_IPSR_FIP;
>> +        } else {
>> +            ipsr_clr |= RISCV_IOMMU_IPSR_FIP;
>> +        }
>> +    } else {
>> +        ipsr_clr |= RISCV_IOMMU_IPSR_FIP;
>> +    }
>> +
>> +    if (data & RISCV_IOMMU_IPSR_PIP) {
>> +        pqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
>> +
>> +        if (pqcsr & RISCV_IOMMU_PQCSR_PIE &&
>> +            (pqcsr & RISCV_IOMMU_PQCSR_PQOF ||
>> +             pqcsr & RISCV_IOMMU_PQCSR_PQMF)) {
>> +            ipsr_set |= RISCV_IOMMU_IPSR_PIP;
>> +        } else {
>> +            ipsr_clr |= RISCV_IOMMU_IPSR_PIP;
>> +        }
>> +    } else {
>> +        ipsr_clr |= RISCV_IOMMU_IPSR_PIP;
>> +    }
>> +
>> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_IPSR, ipsr_set, ipsr_clr);
>> +}
>> +
>> +static MemTxResult riscv_iommu_mmio_write(void *opaque, hwaddr addr,
>> +    uint64_t data, unsigned size, MemTxAttrs attrs)
>> +{
>> +    riscv_iommu_process_fn *process_fn = NULL;
>> +    RISCVIOMMUState *s = opaque;
>> +    uint32_t regb = addr & ~3;
>> +    uint32_t busy = 0;
>> +    uint64_t val = 0;
>> +
>> +    if ((addr & (size - 1)) != 0) {
>> +        /* Unsupported MMIO alignment or access size */
>> +        return MEMTX_ERROR;
>> +    }
>> +
>> +    if (addr + size > RISCV_IOMMU_REG_MSI_CONFIG) {
>> +        /* Unsupported MMIO access location. */
>> +        return MEMTX_ACCESS_ERROR;
>> +    }
>> +
>> +    /* Track actionable MMIO write. */
>> +    switch (regb) {
>> +    case RISCV_IOMMU_REG_DDTP:
>> +    case RISCV_IOMMU_REG_DDTP + 4:
>> +        process_fn = riscv_iommu_process_ddtp;
>> +        regb = RISCV_IOMMU_REG_DDTP;
>> +        busy = RISCV_IOMMU_DDTP_BUSY;
>> +        break;
>> +
>> +    case RISCV_IOMMU_REG_CQT:
>> +        process_fn = riscv_iommu_process_cq_tail;
>> +        break;
>> +
>> +    case RISCV_IOMMU_REG_CQCSR:
>> +        process_fn = riscv_iommu_process_cq_control;
>> +        busy = RISCV_IOMMU_CQCSR_BUSY;
>> +        break;
>> +
>> +    case RISCV_IOMMU_REG_FQCSR:
>> +        process_fn = riscv_iommu_process_fq_control;
>> +        busy = RISCV_IOMMU_FQCSR_BUSY;
>> +        break;
>> +
>> +    case RISCV_IOMMU_REG_PQCSR:
>> +        process_fn = riscv_iommu_process_pq_control;
>> +        busy = RISCV_IOMMU_PQCSR_BUSY;
>> +        break;
>> +
>> +    case RISCV_IOMMU_REG_IPSR:
>> +        /*
>> +         * IPSR has special procedures to update. Execute it
>> +         * and exit.
>> +         */
>> +        if (size == 4) {
>> +            uint32_t ro = ldl_le_p(&s->regs_ro[addr]);
>> +            uint32_t wc = ldl_le_p(&s->regs_wc[addr]);
>> +            uint32_t rw = ldl_le_p(&s->regs_rw[addr]);
>> +            stl_le_p(&val, ((rw & ro) | (data & ~ro)) & ~(data & wc));
>> +        } else if (size == 8) {
>> +            uint64_t ro = ldq_le_p(&s->regs_ro[addr]);
>> +            uint64_t wc = ldq_le_p(&s->regs_wc[addr]);
>> +            uint64_t rw = ldq_le_p(&s->regs_rw[addr]);
>> +            stq_le_p(&val, ((rw & ro) | (data & ~ro)) & ~(data & wc));
>> +        }
>> +
>> +        riscv_iommu_update_ipsr(s, val);
>> +
>> +        return MEMTX_OK;
>> +
>> +    default:
>> +        break;
>> +    }
>> +
>> +    /*
>> +     * Registers update might be not synchronized with core logic.
>> +     * If system software updates register when relevant BUSY bit
>> +     * is set IOMMU behavior of additional writes to the register
>> +     * is UNSPECIFIED.
>> +     */
>> +    qemu_spin_lock(&s->regs_lock);
>> +    if (size == 1) {
>> +        uint8_t ro = s->regs_ro[addr];
>> +        uint8_t wc = s->regs_wc[addr];
>> +        uint8_t rw = s->regs_rw[addr];
>> +        s->regs_rw[addr] = ((rw & ro) | (data & ~ro)) & ~(data & wc);
>> +    } else if (size == 2) {
>> +        uint16_t ro = lduw_le_p(&s->regs_ro[addr]);
>> +        uint16_t wc = lduw_le_p(&s->regs_wc[addr]);
>> +        uint16_t rw = lduw_le_p(&s->regs_rw[addr]);
>> +        stw_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
>> +    } else if (size == 4) {
>> +        uint32_t ro = ldl_le_p(&s->regs_ro[addr]);
>> +        uint32_t wc = ldl_le_p(&s->regs_wc[addr]);
>> +        uint32_t rw = ldl_le_p(&s->regs_rw[addr]);
>> +        stl_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
>> +    } else if (size == 8) {
>> +        uint64_t ro = ldq_le_p(&s->regs_ro[addr]);
>> +        uint64_t wc = ldq_le_p(&s->regs_wc[addr]);
>> +        uint64_t rw = ldq_le_p(&s->regs_rw[addr]);
>> +        stq_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
>> +    }
>> +
>> +    /* Busy flag update, MSB 4-byte register. */
>> +    if (busy) {
>> +        uint32_t rw = ldl_le_p(&s->regs_rw[regb]);
>> +        stl_le_p(&s->regs_rw[regb], rw | busy);
>> +    }
>> +    qemu_spin_unlock(&s->regs_lock);
>> +
>> +    if (process_fn) {
>> +        qemu_mutex_lock(&s->core_lock);
>> +        process_fn(s);
>> +        qemu_mutex_unlock(&s->core_lock);
>> +    }
>> +
>> +    return MEMTX_OK;
>> +}
>> +
>> +static MemTxResult riscv_iommu_mmio_read(void *opaque, hwaddr addr,
>> +    uint64_t *data, unsigned size, MemTxAttrs attrs)
>> +{
>> +    RISCVIOMMUState *s = opaque;
>> +    uint64_t val = -1;
>> +    uint8_t *ptr;
>> +
>> +    if ((addr & (size - 1)) != 0) {
>> +        /* Unsupported MMIO alignment. */
>> +        return MEMTX_ERROR;
>> +    }
>> +
>> +    if (addr + size > RISCV_IOMMU_REG_MSI_CONFIG) {
>> +        return MEMTX_ACCESS_ERROR;
>> +    }
>> +
>> +    ptr = &s->regs_rw[addr];
>> +
>> +    if (size == 1) {
>> +        val = (uint64_t)*ptr;
>> +    } else if (size == 2) {
>> +        val = lduw_le_p(ptr);
>> +    } else if (size == 4) {
>> +        val = ldl_le_p(ptr);
>> +    } else if (size == 8) {
>> +        val = ldq_le_p(ptr);
>> +    } else {
>> +        return MEMTX_ERROR;
>> +    }
>> +
>> +    *data = val;
>> +
>> +    return MEMTX_OK;
>> +}
>> +
>> +static const MemoryRegionOps riscv_iommu_mmio_ops = {
>> +    .read_with_attrs = riscv_iommu_mmio_read,
>> +    .write_with_attrs = riscv_iommu_mmio_write,
>> +    .endianness = DEVICE_NATIVE_ENDIAN,
>> +    .impl = {
>> +        .min_access_size = 4,
>> +        .max_access_size = 8,
>> +        .unaligned = false,
>> +    },
>> +    .valid = {
>> +        .min_access_size = 4,
>> +        .max_access_size = 8,
>> +    }
>> +};
>> +
>> +/*
>> + * Translations matching MSI pattern check are redirected to "riscv-iommu-trap"
>> + * memory region as untranslated address, for additional MSI/MRIF interception
>> + * by IOMMU interrupt remapping implementation.
>> + * Note: Device emulation code generating an MSI is expected to provide a valid
>> + * memory transaction attributes with requested_id set.
>> + */
>> +static MemTxResult riscv_iommu_trap_write(void *opaque, hwaddr addr,
>> +    uint64_t data, unsigned size, MemTxAttrs attrs)
>> +{
>> +    RISCVIOMMUState* s = (RISCVIOMMUState *)opaque;
>> +    RISCVIOMMUContext *ctx;
>> +    MemTxResult res;
>> +    void *ref;
>> +    uint32_t devid = attrs.requester_id;
>> +
>> +    if (attrs.unspecified) {
>> +        return MEMTX_ACCESS_ERROR;
>> +    }
>> +
>> +    /* FIXME: PCIe bus remapping for attached endpoints. */
>> +    devid |= s->bus << 8;
>> +
>> +    ctx = riscv_iommu_ctx(s, devid, 0, &ref);
>> +    if (ctx == NULL) {
>> +        res = MEMTX_ACCESS_ERROR;
>> +    } else {
>> +        res = riscv_iommu_msi_write(s, ctx, addr, data, size, attrs);
>> +    }
>> +    riscv_iommu_ctx_put(s, ref);
>> +    return res;
>> +}
>> +
>> +static MemTxResult riscv_iommu_trap_read(void *opaque, hwaddr addr,
>> +    uint64_t *data, unsigned size, MemTxAttrs attrs)
>> +{
>> +    return MEMTX_ACCESS_ERROR;
>> +}
>> +
>> +static const MemoryRegionOps riscv_iommu_trap_ops = {
>> +    .read_with_attrs = riscv_iommu_trap_read,
>> +    .write_with_attrs = riscv_iommu_trap_write,
>> +    .endianness = DEVICE_LITTLE_ENDIAN,
>> +    .impl = {
>> +        .min_access_size = 4,
>> +        .max_access_size = 8,
>> +        .unaligned = true,
>> +    },
>> +    .valid = {
>> +        .min_access_size = 4,
>> +        .max_access_size = 8,
>> +    }
>> +};
>> +
>> +static void riscv_iommu_realize(DeviceState *dev, Error **errp)
>> +{
>> +    RISCVIOMMUState *s = RISCV_IOMMU(dev);
>> +
>> +    s->cap = s->version & RISCV_IOMMU_CAP_VERSION;
>> +    if (s->enable_msi) {
>> +        s->cap |= RISCV_IOMMU_CAP_MSI_FLAT | RISCV_IOMMU_CAP_MSI_MRIF;
>> +    }
>> +    /* Report QEMU target physical address space limits */
>> +    s->cap = set_field(s->cap, RISCV_IOMMU_CAP_PAS,
>> +                       TARGET_PHYS_ADDR_SPACE_BITS);
>> +
>> +    /* TODO: method to report supported PASID bits */
>> +    s->pasid_bits = 8; /* restricted to size of MemTxAttrs.pasid */
>> +    s->cap |= RISCV_IOMMU_CAP_PD8;
>> +
>> +    /* Out-of-reset translation mode: OFF (DMA disabled) BARE (passthrough) */
>> +    s->ddtp = set_field(0, RISCV_IOMMU_DDTP_MODE, s->enable_off ?
>> +                        RISCV_IOMMU_DDTP_MODE_OFF : RISCV_IOMMU_DDTP_MODE_BARE);
>> +
>> +    /* register storage */
>> +    s->regs_rw = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
>> +    s->regs_ro = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
>> +    s->regs_wc = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
>> +
>> +     /* Mark all registers read-only */
>> +    memset(s->regs_ro, 0xff, RISCV_IOMMU_REG_SIZE);
>> +
>> +    /*
>> +     * Register complete MMIO space, including MSI/PBA registers.
>> +     * Note, PCIDevice implementation will add overlapping MR for MSI/PBA,
>> +     * managed directly by the PCIDevice implementation.
>> +     */
>> +    memory_region_init_io(&s->regs_mr, OBJECT(dev), &riscv_iommu_mmio_ops, s,
>> +        "riscv-iommu-regs", RISCV_IOMMU_REG_SIZE);
>> +
>> +    /* Set power-on register state */
>> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_CAP], s->cap);
>> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_FCTL], 0);
>> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_DDTP],
>> +        ~(RISCV_IOMMU_DDTP_PPN | RISCV_IOMMU_DDTP_MODE));
>> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQB],
>> +        ~(RISCV_IOMMU_CQB_LOG2SZ | RISCV_IOMMU_CQB_PPN));
>> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQB],
>> +        ~(RISCV_IOMMU_FQB_LOG2SZ | RISCV_IOMMU_FQB_PPN));
>> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQB],
>> +        ~(RISCV_IOMMU_PQB_LOG2SZ | RISCV_IOMMU_PQB_PPN));
>> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_CQCSR], RISCV_IOMMU_CQCSR_CQMF |
>> +        RISCV_IOMMU_CQCSR_CMD_TO | RISCV_IOMMU_CQCSR_CMD_ILL);
>> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQCSR], RISCV_IOMMU_CQCSR_CQON |
>> +        RISCV_IOMMU_CQCSR_BUSY);
>> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_FQCSR], RISCV_IOMMU_FQCSR_FQMF |
>> +        RISCV_IOMMU_FQCSR_FQOF);
>> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQCSR], RISCV_IOMMU_FQCSR_FQON |
>> +        RISCV_IOMMU_FQCSR_BUSY);
>> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_PQCSR], RISCV_IOMMU_PQCSR_PQMF |
>> +        RISCV_IOMMU_PQCSR_PQOF);
>> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQCSR], RISCV_IOMMU_PQCSR_PQON |
>> +        RISCV_IOMMU_PQCSR_BUSY);
>> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_IPSR], ~0);
>> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_IVEC], 0);
>> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_DDTP], s->ddtp);
>> +
>> +    /* Memory region for downstream access, if specified. */
>> +    if (s->target_mr) {
>> +        s->target_as = g_new0(AddressSpace, 1);
>> +        address_space_init(s->target_as, s->target_mr,
>> +            "riscv-iommu-downstream");
>> +    } else {
>> +        /* Fallback to global system memory. */
>> +        s->target_as = &address_space_memory;
>> +    }
>> +
>> +    /* Memory region for untranslated MRIF/MSI writes */
>> +    memory_region_init_io(&s->trap_mr, OBJECT(dev), &riscv_iommu_trap_ops, s,
>> +            "riscv-iommu-trap", ~0ULL);
>> +    address_space_init(&s->trap_as, &s->trap_mr, "riscv-iommu-trap-as");
>> +
>> +    /* Device translation context cache */
>> +    s->ctx_cache = g_hash_table_new_full(__ctx_hash, __ctx_equal,
>> +                                         g_free, NULL);
>> +
>> +    s->iommus.le_next = NULL;
>> +    s->iommus.le_prev = NULL;
>> +    QLIST_INIT(&s->spaces);
>> +    qemu_mutex_init(&s->core_lock);
>> +    qemu_spin_init(&s->regs_lock);
>> +}
>> +
>> +static void riscv_iommu_unrealize(DeviceState *dev)
>> +{
>> +    RISCVIOMMUState *s = RISCV_IOMMU(dev);
>> +
>> +    qemu_mutex_destroy(&s->core_lock);
>> +    g_hash_table_unref(s->ctx_cache);
>> +}
>> +
>> +static Property riscv_iommu_properties[] = {
>> +    DEFINE_PROP_UINT32("version", RISCVIOMMUState, version,
>> +        RISCV_IOMMU_SPEC_DOT_VER),
>> +    DEFINE_PROP_UINT32("bus", RISCVIOMMUState, bus, 0x0),
>> +    DEFINE_PROP_BOOL("intremap", RISCVIOMMUState, enable_msi, TRUE),
>> +    DEFINE_PROP_BOOL("off", RISCVIOMMUState, enable_off, TRUE),
>> +    DEFINE_PROP_LINK("downstream-mr", RISCVIOMMUState, target_mr,
>> +        TYPE_MEMORY_REGION, MemoryRegion *),
>> +    DEFINE_PROP_END_OF_LIST(),
>> +};
>> +
>> +static void riscv_iommu_class_init(ObjectClass *klass, void* data)
>> +{
>> +    DeviceClass *dc = DEVICE_CLASS(klass);
>> +
>> +    /* internal device for riscv-iommu-{pci/sys}, not user-creatable */
>> +    dc->user_creatable = false;
>> +    dc->realize = riscv_iommu_realize;
>> +    dc->unrealize = riscv_iommu_unrealize;
>> +    device_class_set_props(dc, riscv_iommu_properties);
>> +}
>> +
>> +static const TypeInfo riscv_iommu_info = {
>> +    .name = TYPE_RISCV_IOMMU,
>> +    .parent = TYPE_DEVICE,
>> +    .instance_size = sizeof(RISCVIOMMUState),
>> +    .class_init = riscv_iommu_class_init,
>> +};
>> +
>> +static const char *IOMMU_FLAG_STR[] = {
>> +    "NA",
>> +    "RO",
>> +    "WR",
>> +    "RW",
>> +};
>> +
>> +/* RISC-V IOMMU Memory Region - Address Translation Space */
>> +static IOMMUTLBEntry riscv_iommu_memory_region_translate(
>> +    IOMMUMemoryRegion *iommu_mr, hwaddr addr,
>> +    IOMMUAccessFlags flag, int iommu_idx)
>> +{
>> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
>> +    RISCVIOMMUContext *ctx;
>> +    void *ref;
>> +    IOMMUTLBEntry iotlb = {
>> +        .iova = addr,
>> +        .target_as = as->iommu->target_as,
>> +        .addr_mask = ~0ULL,
>> +        .perm = flag,
>> +    };
>> +
>> +    ctx = riscv_iommu_ctx(as->iommu, as->devid, iommu_idx, &ref);
>> +    if (ctx == NULL) {
>> +        /* Translation disabled or invalid. */
>> +        iotlb.addr_mask = 0;
>> +        iotlb.perm = IOMMU_NONE;
>> +    } else if (riscv_iommu_translate(as->iommu, ctx, &iotlb)) {
>> +        /* Translation disabled or fault reported. */
>> +        iotlb.addr_mask = 0;
>> +        iotlb.perm = IOMMU_NONE;
>> +    }
>> +
>> +    /* Trace all dma translations with original access flags. */
>> +    trace_riscv_iommu_dma(as->iommu->parent_obj.id, PCI_BUS_NUM(as->devid),
>> +                          PCI_SLOT(as->devid), PCI_FUNC(as->devid), iommu_idx,
>> +                          IOMMU_FLAG_STR[flag & IOMMU_RW], iotlb.iova,
>> +                          iotlb.translated_addr);
>> +
>> +    riscv_iommu_ctx_put(as->iommu, ref);
>> +
>> +    return iotlb;
>> +}
>> +
>> +static int riscv_iommu_memory_region_notify(
>> +    IOMMUMemoryRegion *iommu_mr, IOMMUNotifierFlag old,
>> +    IOMMUNotifierFlag new, Error **errp)
>> +{
>> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
>> +
>> +    if (old == IOMMU_NOTIFIER_NONE) {
>> +        as->notifier = true;
>> +        trace_riscv_iommu_notifier_add(iommu_mr->parent_obj.name);
>> +    } else if (new == IOMMU_NOTIFIER_NONE) {
>> +        as->notifier = false;
>> +        trace_riscv_iommu_notifier_del(iommu_mr->parent_obj.name);
>> +    }
>> +
>> +    return 0;
>> +}
>> +
>> +static inline bool pci_is_iommu(PCIDevice *pdev)
>> +{
>> +    return pci_get_word(pdev->config + PCI_CLASS_DEVICE) == 0x0806;
>> +}
>> +
>> +static AddressSpace *riscv_iommu_find_as(PCIBus *bus, void *opaque, int devfn)
>> +{
>> +    RISCVIOMMUState *s = (RISCVIOMMUState *) opaque;
>> +    PCIDevice *pdev = pci_find_device(bus, pci_bus_num(bus), devfn);
>> +    AddressSpace *as = NULL;
>> +
>> +    if (pdev && pci_is_iommu(pdev)) {
>> +        return s->target_as;
>> +    }
>> +
>> +    /* Find first registered IOMMU device */
>> +    while (s->iommus.le_prev) {
>> +        s = *(s->iommus.le_prev);
>> +    }
>> +
>> +    /* Find first matching IOMMU */
>> +    while (s != NULL && as == NULL) {
>> +        as = riscv_iommu_space(s, PCI_BUILD_BDF(pci_bus_num(bus), devfn));
>> +        s = s->iommus.le_next;
>> +    }
>> +
>> +    return as ? as : &address_space_memory;
>> +}
>> +
>> +static const PCIIOMMUOps riscv_iommu_ops = {
>> +    .get_address_space = riscv_iommu_find_as,
>> +};
>> +
>> +void riscv_iommu_pci_setup_iommu(RISCVIOMMUState *iommu, PCIBus *bus,
>> +        Error **errp)
>> +{
>> +    if (bus->iommu_ops &&
>> +        bus->iommu_ops->get_address_space == riscv_iommu_find_as) {
>> +        /* Allow multiple IOMMUs on the same PCIe bus, link known devices */
>> +        RISCVIOMMUState *last = (RISCVIOMMUState *)bus->iommu_opaque;
>> +        QLIST_INSERT_AFTER(last, iommu, iommus);
>> +    } else if (!bus->iommu_ops && !bus->iommu_opaque) {
>> +        pci_setup_iommu(bus, &riscv_iommu_ops, iommu);
>> +    } else {
>> +        error_setg(errp, "can't register secondary IOMMU for PCI bus #%d",
>> +            pci_bus_num(bus));
>> +    }
>> +}
>> +
>> +static int riscv_iommu_memory_region_index(IOMMUMemoryRegion *iommu_mr,
>> +    MemTxAttrs attrs)
>> +{
>> +    return attrs.unspecified ? RISCV_IOMMU_NOPASID : (int)attrs.pasid;
>> +}
>> +
>> +static int riscv_iommu_memory_region_index_len(IOMMUMemoryRegion *iommu_mr)
>> +{
>> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
>> +    return 1 << as->iommu->pasid_bits;
>> +}
>> +
>> +static void riscv_iommu_memory_region_init(ObjectClass *klass, void *data)
>> +{
>> +    IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_CLASS(klass);
>> +
>> +    imrc->translate = riscv_iommu_memory_region_translate;
>> +    imrc->notify_flag_changed = riscv_iommu_memory_region_notify;
>> +    imrc->attrs_to_index = riscv_iommu_memory_region_index;
>> +    imrc->num_indexes = riscv_iommu_memory_region_index_len;
>> +}
>> +
>> +static const TypeInfo riscv_iommu_memory_region_info = {
>> +    .parent = TYPE_IOMMU_MEMORY_REGION,
>> +    .name = TYPE_RISCV_IOMMU_MEMORY_REGION,
>> +    .class_init = riscv_iommu_memory_region_init,
>> +};
>> +
>> +static void riscv_iommu_register_mr_types(void)
>> +{
>> +    type_register_static(&riscv_iommu_memory_region_info);
>> +    type_register_static(&riscv_iommu_info);
>> +}
>> +
>> +type_init(riscv_iommu_register_mr_types);
>> diff --git a/hw/riscv/riscv-iommu.h b/hw/riscv/riscv-iommu.h
>> new file mode 100644
>> index 0000000000..31d3907d33
>> --- /dev/null
>> +++ b/hw/riscv/riscv-iommu.h
>> @@ -0,0 +1,141 @@
>> +/*
>> + * QEMU emulation of an RISC-V IOMMU
>> + *
>> + * Copyright (C) 2022-2023 Rivos Inc.
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of the GNU General Public License as published by
>> + * the Free Software Foundation; either version 2 of the License.
>> + *
>> + * This program is distributed in the hope that it will be useful,
>> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
>> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>> + * GNU General Public License for more details.
>> + *
>> + * You should have received a copy of the GNU General Public License along
>> + * with this program; if not, see <http://www.gnu.org/licenses/>.
>> + */
>> +
>> +#ifndef HW_RISCV_IOMMU_STATE_H
>> +#define HW_RISCV_IOMMU_STATE_H
>> +
>> +#include "qemu/osdep.h"
>> +#include "qom/object.h"
>> +
>> +#include "hw/riscv/iommu.h"
>> +
>> +struct RISCVIOMMUState {
>> +    /*< private >*/
>> +    DeviceState parent_obj;
>> +
>> +    /*< public >*/
>> +    uint32_t version;     /* Reported interface version number */
>> +    uint32_t pasid_bits;  /* process identifier width */
>> +    uint32_t bus;         /* PCI bus mapping for non-root endpoints */
>> +
>> +    uint64_t cap;         /* IOMMU supported capabilities */
>> +    uint64_t fctl;        /* IOMMU enabled features */
>> +
>> +    bool enable_off;      /* Enable out-of-reset OFF mode (DMA disabled) */
>> +    bool enable_msi;      /* Enable MSI remapping */
>> +
>> +    /* IOMMU Internal State */
>> +    uint64_t ddtp;        /* Validated Device Directory Tree Root Pointer */
>> +
>> +    dma_addr_t cq_addr;   /* Command queue base physical address */
>> +    dma_addr_t fq_addr;   /* Fault/event queue base physical address */
>> +    dma_addr_t pq_addr;   /* Page request queue base physical address */
>> +
>> +    uint32_t cq_mask;     /* Command queue index bit mask */
>> +    uint32_t fq_mask;     /* Fault/event queue index bit mask */
>> +    uint32_t pq_mask;     /* Page request queue index bit mask */
>> +
>> +    /* interrupt notifier */
>> +    void (*notify)(RISCVIOMMUState *iommu, unsigned vector);
>> +
>> +    /* IOMMU State Machine */
>> +    QemuThread core_proc; /* Background processing thread */
>> +    QemuMutex core_lock;  /* Global IOMMU lock, used for cache/regs updates */
>> +    QemuCond core_cond;   /* Background processing wake up signal */
>> +    unsigned core_exec;   /* Processing thread execution actions */
>> +
>> +    /* IOMMU target address space */
>> +    AddressSpace *target_as;
>> +    MemoryRegion *target_mr;
>> +
>> +    /* MSI / MRIF access trap */
>> +    AddressSpace trap_as;
>> +    MemoryRegion trap_mr;
>> +
>> +    GHashTable *ctx_cache;          /* Device translation Context Cache */
>> +
>> +    /* MMIO Hardware Interface */
>> +    MemoryRegion regs_mr;
>> +    QemuSpin regs_lock;
>> +    uint8_t *regs_rw;  /* register state (user write) */
>> +    uint8_t *regs_wc;  /* write-1-to-clear mask */
>> +    uint8_t *regs_ro;  /* read-only mask */
>> +
>> +    QLIST_ENTRY(RISCVIOMMUState) iommus;
>> +    QLIST_HEAD(, RISCVIOMMUSpace) spaces;
>> +};
>> +
>> +void riscv_iommu_pci_setup_iommu(RISCVIOMMUState *iommu, PCIBus *bus,
>> +         Error **errp);
>> +
>> +/* private helpers */
>> +
>> +/* Register helper functions */
>> +static inline uint32_t riscv_iommu_reg_mod32(RISCVIOMMUState *s,
>> +    unsigned idx, uint32_t set, uint32_t clr)
>> +{
>> +    uint32_t val;
>> +    qemu_spin_lock(&s->regs_lock);
>> +    val = ldl_le_p(s->regs_rw + idx);
>> +    stl_le_p(s->regs_rw + idx, (val & ~clr) | set);
>> +    qemu_spin_unlock(&s->regs_lock);
>> +    return val;
>> +}
>> +
>> +static inline void riscv_iommu_reg_set32(RISCVIOMMUState *s,
>> +    unsigned idx, uint32_t set)
>> +{
>> +    qemu_spin_lock(&s->regs_lock);
>> +    stl_le_p(s->regs_rw + idx, set);
>> +    qemu_spin_unlock(&s->regs_lock);
>> +}
>> +
>> +static inline uint32_t riscv_iommu_reg_get32(RISCVIOMMUState *s,
>> +    unsigned idx)
>> +{
>> +    return ldl_le_p(s->regs_rw + idx);
>> +}
>> +
>> +static inline uint64_t riscv_iommu_reg_mod64(RISCVIOMMUState *s,
>> +    unsigned idx, uint64_t set, uint64_t clr)
>> +{
>> +    uint64_t val;
>> +    qemu_spin_lock(&s->regs_lock);
>> +    val = ldq_le_p(s->regs_rw + idx);
>> +    stq_le_p(s->regs_rw + idx, (val & ~clr) | set);
>> +    qemu_spin_unlock(&s->regs_lock);
>> +    return val;
>> +}
>> +
>> +static inline void riscv_iommu_reg_set64(RISCVIOMMUState *s,
>> +    unsigned idx, uint64_t set)
>> +{
>> +    qemu_spin_lock(&s->regs_lock);
>> +    stq_le_p(s->regs_rw + idx, set);
>> +    qemu_spin_unlock(&s->regs_lock);
>> +}
>> +
>> +static inline uint64_t riscv_iommu_reg_get64(RISCVIOMMUState *s,
>> +    unsigned idx)
>> +{
>> +    return ldq_le_p(s->regs_rw + idx);
>> +}
>> +
>> +
>> +
>> +#endif
>> diff --git a/hw/riscv/trace-events b/hw/riscv/trace-events
>> new file mode 100644
>> index 0000000000..42a97caffa
>> --- /dev/null
>> +++ b/hw/riscv/trace-events
>> @@ -0,0 +1,11 @@
>> +# See documentation at docs/devel/tracing.rst
>> +
>> +# riscv-iommu.c
>> +riscv_iommu_new(const char *id, unsigned b, unsigned d, unsigned f) "%s: device attached %04x:%02x.%d"
>> +riscv_iommu_flt(const char *id, unsigned b, unsigned d, unsigned f, uint64_t reason, uint64_t iova) "%s: fault %04x:%02x.%u reason: 0x%"PRIx64" iova: 0x%"PRIx64
>> +riscv_iommu_pri(const char *id, unsigned b, unsigned d, unsigned f, uint64_t iova) "%s: page request %04x:%02x.%u iova: 0x%"PRIx64
>> +riscv_iommu_dma(const char *id, unsigned b, unsigned d, unsigned f, unsigned pasid, const char *dir, uint64_t iova, uint64_t phys) "%s: translate %04x:%02x.%u #%u %s 0x%"PRIx64" -> 0x%"PRIx64
>> +riscv_iommu_msi(const char *id, unsigned b, unsigned d, unsigned f, uint64_t iova, uint64_t phys) "%s: translate %04x:%02x.%u MSI 0x%"PRIx64" -> 0x%"PRIx64
>> +riscv_iommu_cmd(const char *id, uint64_t l, uint64_t u) "%s: command 0x%"PRIx64" 0x%"PRIx64
>> +riscv_iommu_notifier_add(const char *id) "%s: dev-iotlb notifier added"
>> +riscv_iommu_notifier_del(const char *id) "%s: dev-iotlb notifier removed"
>> diff --git a/hw/riscv/trace.h b/hw/riscv/trace.h
>> new file mode 100644
>> index 0000000000..8c0e3ca1f3
>> --- /dev/null
>> +++ b/hw/riscv/trace.h
>> @@ -0,0 +1 @@
>> +#include "trace/trace-hw_riscv.h"
>> diff --git a/include/hw/riscv/iommu.h b/include/hw/riscv/iommu.h
>> new file mode 100644
>> index 0000000000..070ee69973
>> --- /dev/null
>> +++ b/include/hw/riscv/iommu.h
>> @@ -0,0 +1,36 @@
>> +/*
>> + * QEMU emulation of an RISC-V IOMMU
>> + *
>> + * Copyright (C) 2022-2023 Rivos Inc.
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of the GNU General Public License as published by
>> + * the Free Software Foundation; either version 2 of the License.
>> + *
>> + * This program is distributed in the hope that it will be useful,
>> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
>> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>> + * GNU General Public License for more details.
>> + *
>> + * You should have received a copy of the GNU General Public License along
>> + * with this program; if not, see <http://www.gnu.org/licenses/>.
>> + */
>> +
>> +#ifndef HW_RISCV_IOMMU_H
>> +#define HW_RISCV_IOMMU_H
>> +
>> +#include "qemu/osdep.h"
>> +#include "qom/object.h"
>> +
>> +#define TYPE_RISCV_IOMMU "riscv-iommu"
>> +OBJECT_DECLARE_SIMPLE_TYPE(RISCVIOMMUState, RISCV_IOMMU)
>> +typedef struct RISCVIOMMUState RISCVIOMMUState;
>> +
>> +#define TYPE_RISCV_IOMMU_MEMORY_REGION "riscv-iommu-mr"
>> +typedef struct RISCVIOMMUSpace RISCVIOMMUSpace;
>> +
>> +#define TYPE_RISCV_IOMMU_PCI "riscv-iommu-pci"
>> +OBJECT_DECLARE_SIMPLE_TYPE(RISCVIOMMUStatePci, RISCV_IOMMU_PCI)
>> +typedef struct RISCVIOMMUStatePci RISCVIOMMUStatePci;
>> +
>> +#endif
>> diff --git a/meson.build b/meson.build
>> index a9de71d450..8099d8271c 100644
>> --- a/meson.build
>> +++ b/meson.build
>> @@ -3319,6 +3319,7 @@ if have_system
>>       'hw/pci-host',
>>       'hw/ppc',
>>       'hw/rtc',
>> +    'hw/riscv',
>>       'hw/s390x',
>>       'hw/scsi',
>>       'hw/sd',
>

Hi Daniel,

On 2024/5/24 上午 01:39, Daniel Henrique Barboza wrote:
> From: Tomasz Jeznach <tjeznach@rivosinc.com>
>
> The RISC-V IOMMU specification is now ratified as-per the RISC-V
> international process. The latest frozen specifcation can be found
> at:
>
> https://github.com/riscv-non-isa/riscv-iommu/releases/download/v1.0/riscv-iommu.pdf
>
> Add the foundation of the device emulation for RISC-V IOMMU, which
> includes an IOMMU that has no capabilities but MSI interrupt support and
> fault queue interfaces. We'll add add more features incrementally in the
> next patches.
>
> Co-developed-by: Sebastien Boeuf <seb@rivosinc.com>
> Signed-off-by: Sebastien Boeuf <seb@rivosinc.com>
> Signed-off-by: Tomasz Jeznach <tjeznach@rivosinc.com>
> Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
> ---
>   hw/riscv/Kconfig         |    4 +
>   hw/riscv/meson.build     |    1 +
>   hw/riscv/riscv-iommu.c   | 1602 ++++++++++++++++++++++++++++++++++++++
>   hw/riscv/riscv-iommu.h   |  141 ++++
>   hw/riscv/trace-events    |   11 +
>   hw/riscv/trace.h         |    1 +
>   include/hw/riscv/iommu.h |   36 +
>   meson.build              |    1 +
>   8 files changed, 1797 insertions(+)
>   create mode 100644 hw/riscv/riscv-iommu.c
>   create mode 100644 hw/riscv/riscv-iommu.h
>   create mode 100644 hw/riscv/trace-events
>   create mode 100644 hw/riscv/trace.h
>   create mode 100644 include/hw/riscv/iommu.h
>
> diff --git a/hw/riscv/Kconfig b/hw/riscv/Kconfig
> index a2030e3a6f..f69d6e3c8e 100644
> --- a/hw/riscv/Kconfig
> +++ b/hw/riscv/Kconfig
> @@ -1,3 +1,6 @@
> +config RISCV_IOMMU
> +    bool
> +
>   config RISCV_NUMA
>       bool
>   
> @@ -47,6 +50,7 @@ config RISCV_VIRT
>       select SERIAL
>       select RISCV_ACLINT
>       select RISCV_APLIC
> +    select RISCV_IOMMU
>       select RISCV_IMSIC
>       select SIFIVE_PLIC
>       select SIFIVE_TEST
> diff --git a/hw/riscv/meson.build b/hw/riscv/meson.build
> index f872674093..cbc99c6e8e 100644
> --- a/hw/riscv/meson.build
> +++ b/hw/riscv/meson.build
> @@ -10,5 +10,6 @@ riscv_ss.add(when: 'CONFIG_SIFIVE_U', if_true: files('sifive_u.c'))
>   riscv_ss.add(when: 'CONFIG_SPIKE', if_true: files('spike.c'))
>   riscv_ss.add(when: 'CONFIG_MICROCHIP_PFSOC', if_true: files('microchip_pfsoc.c'))
>   riscv_ss.add(when: 'CONFIG_ACPI', if_true: files('virt-acpi-build.c'))
> +riscv_ss.add(when: 'CONFIG_RISCV_IOMMU', if_true: files('riscv-iommu.c'))
>   
>   hw_arch += {'riscv': riscv_ss}
> diff --git a/hw/riscv/riscv-iommu.c b/hw/riscv/riscv-iommu.c
> new file mode 100644
> index 0000000000..39b4ff1405
> --- /dev/null
> +++ b/hw/riscv/riscv-iommu.c
> @@ -0,0 +1,1602 @@
> +/*
> + * QEMU emulation of an RISC-V IOMMU
> + *
> + * Copyright (C) 2021-2023, Rivos Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#include "qemu/osdep.h"
> +#include "qom/object.h"
> +#include "hw/pci/pci_bus.h"
> +#include "hw/pci/pci_device.h"
> +#include "hw/qdev-properties.h"
> +#include "hw/riscv/riscv_hart.h"
> +#include "migration/vmstate.h"
> +#include "qapi/error.h"
> +#include "qemu/timer.h"
> +
> +#include "cpu_bits.h"
> +#include "riscv-iommu.h"
> +#include "riscv-iommu-bits.h"
> +#include "trace.h"
> +
> +#define LIMIT_CACHE_CTX               (1U << 7)
> +#define LIMIT_CACHE_IOT               (1U << 20)
> +
> +/* Physical page number coversions */
> +#define PPN_PHYS(ppn)                 ((ppn) << TARGET_PAGE_BITS)
> +#define PPN_DOWN(phy)                 ((phy) >> TARGET_PAGE_BITS)
> +
> +typedef struct RISCVIOMMUContext RISCVIOMMUContext;
> +typedef struct RISCVIOMMUEntry RISCVIOMMUEntry;
> +
> +/* Device assigned I/O address space */
> +struct RISCVIOMMUSpace {
> +    IOMMUMemoryRegion iova_mr;  /* IOVA memory region for attached device */
> +    AddressSpace iova_as;       /* IOVA address space for attached device */
> +    RISCVIOMMUState *iommu;     /* Managing IOMMU device state */
> +    uint32_t devid;             /* Requester identifier, AKA device_id */
> +    bool notifier;              /* IOMMU unmap notifier enabled */
> +    QLIST_ENTRY(RISCVIOMMUSpace) list;
> +};
> +
> +/* Device translation context state. */
> +struct RISCVIOMMUContext {
> +    uint64_t devid:24;          /* Requester Id, AKA device_id */
> +    uint64_t pasid:20;          /* Process Address Space ID */
> +    uint64_t __rfu:20;          /* reserved */
> +    uint64_t tc;                /* Translation Control */
> +    uint64_t ta;                /* Translation Attributes */
> +    uint64_t msi_addr_mask;     /* MSI filtering - address mask */
> +    uint64_t msi_addr_pattern;  /* MSI filtering - address pattern */
> +    uint64_t msiptp;            /* MSI redirection page table pointer */
> +};
> +
> +/* IOMMU index for transactions without PASID specified. */
> +#define RISCV_IOMMU_NOPASID 0
> +
> +static void riscv_iommu_notify(RISCVIOMMUState *s, int vec)
> +{
> +    const uint32_t fctl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FCTL);
> +    uint32_t ipsr, ivec;
> +
> +    if (fctl & RISCV_IOMMU_FCTL_WSI || !s->notify) {
> +        return;
> +    }
> +
> +    ipsr = riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_IPSR, (1 << vec), 0);
> +    ivec = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_IVEC);
> +
> +    if (!(ipsr & (1 << vec))) {
> +        s->notify(s, (ivec >> (vec * 4)) & 0x0F);
> +    }
> +}
> +
> +static void riscv_iommu_fault(RISCVIOMMUState *s,
> +                              struct riscv_iommu_fq_record *ev)
> +{
> +    uint32_t ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
> +    uint32_t head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQH) & s->fq_mask;
> +    uint32_t tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQT) & s->fq_mask;
> +    uint32_t next = (tail + 1) & s->fq_mask;
> +    uint32_t devid = get_field(ev->hdr, RISCV_IOMMU_FQ_HDR_DID);
> +
> +    trace_riscv_iommu_flt(s->parent_obj.id, PCI_BUS_NUM(devid), PCI_SLOT(devid),
> +                          PCI_FUNC(devid), ev->hdr, ev->iotval);
> +
> +    if (!(ctrl & RISCV_IOMMU_FQCSR_FQON) ||
> +        !!(ctrl & (RISCV_IOMMU_FQCSR_FQOF | RISCV_IOMMU_FQCSR_FQMF))) {
> +        return;
> +    }
> +
> +    if (head == next) {
> +        riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR,
> +                              RISCV_IOMMU_FQCSR_FQOF, 0);
> +    } else {
> +        dma_addr_t addr = s->fq_addr + tail * sizeof(*ev);
> +        if (dma_memory_write(s->target_as, addr, ev, sizeof(*ev),
> +                             MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR,
> +                                  RISCV_IOMMU_FQCSR_FQMF, 0);
> +        } else {
> +            riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_FQT, next);
> +        }
> +    }
> +
> +    if (ctrl & RISCV_IOMMU_FQCSR_FIE) {
> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_FQ);
> +    }
> +}
> +
> +static void riscv_iommu_pri(RISCVIOMMUState *s,
> +    struct riscv_iommu_pq_record *pr)
> +{
> +    uint32_t ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
> +    uint32_t head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQH) & s->pq_mask;
> +    uint32_t tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQT) & s->pq_mask;
> +    uint32_t next = (tail + 1) & s->pq_mask;
> +    uint32_t devid = get_field(pr->hdr, RISCV_IOMMU_PREQ_HDR_DID);
> +
> +    trace_riscv_iommu_pri(s->parent_obj.id, PCI_BUS_NUM(devid), PCI_SLOT(devid),
> +                          PCI_FUNC(devid), pr->payload);
> +
> +    if (!(ctrl & RISCV_IOMMU_PQCSR_PQON) ||
> +        !!(ctrl & (RISCV_IOMMU_PQCSR_PQOF | RISCV_IOMMU_PQCSR_PQMF))) {
> +        return;
> +    }
> +
> +    if (head == next) {
> +        riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR,
> +                              RISCV_IOMMU_PQCSR_PQOF, 0);
> +    } else {
> +        dma_addr_t addr = s->pq_addr + tail * sizeof(*pr);
> +        if (dma_memory_write(s->target_as, addr, pr, sizeof(*pr),
> +                             MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR,
> +                                  RISCV_IOMMU_PQCSR_PQMF, 0);
> +        } else {
> +            riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_PQT, next);
> +        }
> +    }
> +
> +    if (ctrl & RISCV_IOMMU_PQCSR_PIE) {
> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_PQ);
> +    }
> +}
> +
> +/* Portable implementation of pext_u64, bit-mask extraction. */
> +static uint64_t _pext_u64(uint64_t val, uint64_t ext)
> +{
> +    uint64_t ret = 0;
> +    uint64_t rot = 1;
> +
> +    while (ext) {
> +        if (ext & 1) {
> +            if (val & 1) {
> +                ret |= rot;
> +            }
> +            rot <<= 1;
> +        }
> +        val >>= 1;
> +        ext >>= 1;
> +    }
> +
> +    return ret;
> +}
> +
> +/* Check if GPA matches MSI/MRIF pattern. */
> +static bool riscv_iommu_msi_check(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
> +    dma_addr_t gpa)
> +{
> +    if (get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_MODE) !=
> +        RISCV_IOMMU_DC_MSIPTP_MODE_FLAT) {
> +        return false; /* Invalid MSI/MRIF mode */
> +    }
> +
> +    if ((PPN_DOWN(gpa) ^ ctx->msi_addr_pattern) & ~ctx->msi_addr_mask) {
> +        return false; /* GPA not in MSI range defined by AIA IMSIC rules. */
> +    }
> +
> +    return true;
> +}
> +
> +/* RISCV IOMMU Address Translation Lookup - Page Table Walk */
> +static int riscv_iommu_spa_fetch(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
> +    IOMMUTLBEntry *iotlb)
> +{
> +    /* Early check for MSI address match when IOVA == GPA */
> +    if (iotlb->perm & IOMMU_WO &&
> +        riscv_iommu_msi_check(s, ctx, iotlb->iova)) {
> +        iotlb->target_as = &s->trap_as;
> +        iotlb->translated_addr = iotlb->iova;
> +        iotlb->addr_mask = ~TARGET_PAGE_MASK;
> +        return 0;
> +    }
> +
> +    /* Exit early for pass-through mode. */
> +    iotlb->translated_addr = iotlb->iova;
> +    iotlb->addr_mask = ~TARGET_PAGE_MASK;
> +    /* Allow R/W in pass-through mode */
> +    iotlb->perm = IOMMU_RW;
> +    return 0;
> +}
> +
> +static void riscv_iommu_report_fault(RISCVIOMMUState *s,
> +                                     RISCVIOMMUContext *ctx,
> +                                     uint32_t fault_type, uint32_t cause,
> +                                     bool pv,
> +                                     uint64_t iotval, uint64_t iotval2)
> +{
> +    struct riscv_iommu_fq_record ev = { 0 };
> +
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_DTF) {
> +        switch (cause) {
> +        case RISCV_IOMMU_FQ_CAUSE_DMA_DISABLED:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_INVALID:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_CORRUPTED:
> +        case RISCV_IOMMU_FQ_CAUSE_INTERNAL_DP_ERROR:
> +        case RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT:
> +            break;
> +        default:
> +            /* DTF prevents reporting a fault for this given cause */
> +            return;
> +        }
> +    }
> +
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_CAUSE, cause);
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_TTYPE, fault_type);
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_DID, ctx->devid);
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_PV, true);
> +
> +    if (pv) {
> +        ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_PID, ctx->pasid);
> +    }
> +
> +    ev.iotval = iotval;
> +    ev.iotval2 = iotval2;
> +
> +    riscv_iommu_fault(s, &ev);
> +}
> +
> +/* Redirect MSI write for given GPA. */
> +static MemTxResult riscv_iommu_msi_write(RISCVIOMMUState *s,
> +    RISCVIOMMUContext *ctx, uint64_t gpa, uint64_t data,
> +    unsigned size, MemTxAttrs attrs)
> +{
> +    MemTxResult res;
> +    dma_addr_t addr;
> +    uint64_t intn;
> +    uint32_t n190;
> +    uint64_t pte[2];
> +    int fault_type = RISCV_IOMMU_FQ_TTYPE_UADDR_WR;
> +    int cause;
> +
> +    if (!riscv_iommu_msi_check(s, ctx, gpa)) {
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    /* Interrupt File Number */
> +    intn = _pext_u64(PPN_DOWN(gpa), ctx->msi_addr_mask);
> +    if (intn >= 256) {
> +        /* Interrupt file number out of range */
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    /* fetch MSI PTE */
> +    addr = PPN_PHYS(get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_PPN));
> +    addr = addr | (intn * sizeof(pte));
> +    res = dma_memory_read(s->target_as, addr, &pte, sizeof(pte),
> +            MEMTXATTRS_UNSPECIFIED);
> +    if (res != MEMTX_OK) {
> +        if (res == MEMTX_DECODE_ERROR) {
> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_PT_CORRUPTED;
> +        } else {
> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        }
> +        goto err;
> +    }
> +
> +    le64_to_cpus(&pte[0]);
> +    le64_to_cpus(&pte[1]);
> +
> +    if (!(pte[0] & RISCV_IOMMU_MSI_PTE_V) || (pte[0] & RISCV_IOMMU_MSI_PTE_C)) {
> +        /*
> +         * The spec mentions that: "If msipte.C == 1, then further
> +         * processing to interpret the PTE is implementation
> +         * defined.". We'll abort with cause = 262 for this
> +         * case too.
> +         */
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_INVALID;
> +        goto err;
> +    }
> +
> +    switch (get_field(pte[0], RISCV_IOMMU_MSI_PTE_M)) {
> +    case RISCV_IOMMU_MSI_PTE_M_BASIC:
> +        /* MSI Pass-through mode */
> +        addr = PPN_PHYS(get_field(pte[0], RISCV_IOMMU_MSI_PTE_PPN));
> +        addr = addr | (gpa & TARGET_PAGE_MASK);
> +
> +        trace_riscv_iommu_msi(s->parent_obj.id, PCI_BUS_NUM(ctx->devid),
> +                              PCI_SLOT(ctx->devid), PCI_FUNC(ctx->devid),
> +                              gpa, addr);
> +
> +        res = dma_memory_write(s->target_as, addr, &data, size, attrs);
> +        if (res != MEMTX_OK) {
> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
> +            goto err;
> +        }
> +
> +        return MEMTX_OK;
> +    case RISCV_IOMMU_MSI_PTE_M_MRIF:
> +        /* MRIF mode, continue. */
> +        break;
> +    default:
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_MISCONFIGURED;
> +        goto err;
> +    }
> +
> +    /*
> +     * Report an error for interrupt identities exceeding the maximum allowed
> +     * for an IMSIC interrupt file (2047) or destination address is not 32-bit
> +     * aligned. See IOMMU Specification, Chapter 2.3. MSI page tables.
> +     */
> +    if ((data > 2047) || (gpa & 3)) {
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_MISCONFIGURED;
> +        goto err;
> +    }
> +
> +    /* MSI MRIF mode, non atomic pending bit update */
> +
> +    /* MRIF pending bit address */
> +    addr = get_field(pte[0], RISCV_IOMMU_MSI_PTE_MRIF_ADDR) << 9;
> +    addr = addr | ((data & 0x7c0) >> 3);
> +
> +    trace_riscv_iommu_msi(s->parent_obj.id, PCI_BUS_NUM(ctx->devid),
> +                          PCI_SLOT(ctx->devid), PCI_FUNC(ctx->devid),
> +                          gpa, addr);
> +
> +    /* MRIF pending bit mask */
> +    data = 1ULL << (data & 0x03f);
> +    res = dma_memory_read(s->target_as, addr, &intn, sizeof(intn), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    intn = intn | data;
> +    res = dma_memory_write(s->target_as, addr, &intn, sizeof(intn), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
> +        goto err;
> +    }
> +
> +    /* Get MRIF enable bits */
> +    addr = addr + sizeof(intn);
> +    res = dma_memory_read(s->target_as, addr, &intn, sizeof(intn), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    if (!(intn & data)) {
> +        /* notification disabled, MRIF update completed. */
> +        return MEMTX_OK;
> +    }
> +
> +    /* Send notification message */
> +    addr = PPN_PHYS(get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NPPN));
> +    n190 = get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NID) |
> +          (get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NID_MSB) << 10);
> +
> +    res = dma_memory_write(s->target_as, addr, &n190, sizeof(n190), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
> +        goto err;
> +    }
> +
> +    return MEMTX_OK;
> +
> +err:
> +    riscv_iommu_report_fault(s, ctx, fault_type, cause,
> +                             !!ctx->pasid, 0, 0);
> +    return res;
> +}
> +
> +/*
> + * Check device context configuration as described by the
> + * riscv-iommu spec section "Device-context configuration
> + * checks".
> + */
> +static bool riscv_iommu_validate_device_ctx(RISCVIOMMUState *s,
> +                                            RISCVIOMMUContext *ctx)
> +{
> +    uint32_t msi_mode;
> +
> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_EN_PRI) &&
> +        ctx->tc & RISCV_IOMMU_DC_TC_PRPR) {
> +        return false;
> +    }
> +
> +    if (!(s->cap & RISCV_IOMMU_CAP_T2GPA) &&
> +        ctx->tc & RISCV_IOMMU_DC_TC_T2GPA) {
> +        return false;
> +    }
> +
> +    if (s->cap & RISCV_IOMMU_CAP_MSI_FLAT) {
> +        msi_mode = get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_MODE);
> +
> +        if (msi_mode != RISCV_IOMMU_DC_MSIPTP_MODE_OFF &&
> +            msi_mode != RISCV_IOMMU_DC_MSIPTP_MODE_FLAT) {
> +            return false;
> +        }
> +    }
> +
> +    /*
> +     * CAP_END is always zero (only one endianess). FCTL_BE is
> +     * always zero (little-endian accesses). Thus TC_SBE must
> +     * always be LE, i.e. zero.
> +     */
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_SBE) {
> +        return false;
> +    }
> +
> +    return true;
> +}
> +
> +/*
> + * RISC-V IOMMU Device Context Loopkup - Device Directory Tree Walk
> + *
> + * @s         : IOMMU Device State
> + * @ctx       : Device Translation Context with devid and pasid set.
> + * @return    : success or fault code.
> + */
> +static int riscv_iommu_ctx_fetch(RISCVIOMMUState *s, RISCVIOMMUContext *ctx)
> +{
> +    const uint64_t ddtp = s->ddtp;
> +    unsigned mode = get_field(ddtp, RISCV_IOMMU_DDTP_MODE);
> +    dma_addr_t addr = PPN_PHYS(get_field(ddtp, RISCV_IOMMU_DDTP_PPN));
> +    struct riscv_iommu_dc dc;
> +    /* Device Context format: 0: extended (64 bytes) | 1: base (32 bytes) */
> +    const int dc_fmt = !s->enable_msi;
> +    const size_t dc_len = sizeof(dc) >> dc_fmt;
> +    unsigned depth;
> +    uint64_t de;
> +
> +    switch (mode) {
> +    case RISCV_IOMMU_DDTP_MODE_OFF:
> +        return RISCV_IOMMU_FQ_CAUSE_DMA_DISABLED;
> +
> +    case RISCV_IOMMU_DDTP_MODE_BARE:
> +        /* mock up pass-through translation context */
> +        ctx->tc = RISCV_IOMMU_DC_TC_V;
> +        ctx->ta = 0;
> +        ctx->msiptp = 0;
> +        return 0;
> +
> +    case RISCV_IOMMU_DDTP_MODE_1LVL:
> +        depth = 0;
> +        break;
> +
> +    case RISCV_IOMMU_DDTP_MODE_2LVL:
> +        depth = 1;
> +        break;
> +
> +    case RISCV_IOMMU_DDTP_MODE_3LVL:
> +        depth = 2;
> +        break;
> +
> +    default:
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
> +    }
> +
> +    /*
> +     * Check supported device id width (in bits).
> +     * See IOMMU Specification, Chapter 6. Software guidelines.
> +     * - if extended device-context format is used:
> +     *   1LVL: 6, 2LVL: 15, 3LVL: 24
> +     * - if base device-context format is used:
> +     *   1LVL: 7, 2LVL: 16, 3LVL: 24
> +     */
> +    if (ctx->devid >= (1 << (depth * 9 + 6 + (dc_fmt && depth != 2)))) {
> +        return RISCV_IOMMU_FQ_CAUSE_TTYPE_BLOCKED;
> +    }
> +
> +    /* Device directory tree walk */
> +    for (; depth-- > 0; ) {
> +        /*
> +         * Select device id index bits based on device directory tree level
> +         * and device context format.
> +         * See IOMMU Specification, Chapter 2. Data Structures.
> +         * - if extended device-context format is used:
> +         *   device index: [23:15][14:6][5:0]
> +         * - if base device-context format is used:
> +         *   device index: [23:16][15:7][6:0]
> +         */
> +        const int split = depth * 9 + 6 + dc_fmt;
> +        addr |= ((ctx->devid >> split) << 3) & ~TARGET_PAGE_MASK;
> +        if (dma_memory_read(s->target_as, addr, &de, sizeof(de),
> +                            MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            return RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT;
> +        }
> +        le64_to_cpus(&de);
> +        if (!(de & RISCV_IOMMU_DDTE_VALID)) {
> +            /* invalid directory entry */
> +            return RISCV_IOMMU_FQ_CAUSE_DDT_INVALID;
> +        }
> +        if (de & ~(RISCV_IOMMU_DDTE_PPN | RISCV_IOMMU_DDTE_VALID)) {
> +            /* reserved bits set */
> +            return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
> +        }
> +        addr = PPN_PHYS(get_field(de, RISCV_IOMMU_DDTE_PPN));
> +    }
> +
> +    /* index into device context entry page */
> +    addr |= (ctx->devid * dc_len) & ~TARGET_PAGE_MASK;
> +
> +    memset(&dc, 0, sizeof(dc));
> +    if (dma_memory_read(s->target_as, addr, &dc, dc_len,
> +                        MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT;
> +    }
> +
> +    /* Set translation context. */
> +    ctx->tc = le64_to_cpu(dc.tc);
> +    ctx->ta = le64_to_cpu(dc.ta);
> +    ctx->msiptp = le64_to_cpu(dc.msiptp);
> +    ctx->msi_addr_mask = le64_to_cpu(dc.msi_addr_mask);
> +    ctx->msi_addr_pattern = le64_to_cpu(dc.msi_addr_pattern);
> +
> +    if (!riscv_iommu_validate_device_ctx(s, ctx)) {
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
> +    }
> +
> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_V)) {
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_INVALID;
> +    }
> +
> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_PDTV)) {
> +        if (ctx->pasid != RISCV_IOMMU_NOPASID) {
> +            /* PASID is disabled */
> +            return RISCV_IOMMU_FQ_CAUSE_TTYPE_BLOCKED;
> +        }
> +        return 0;
> +    }
> +
> +    /* FSC.TC.PDTV enabled */
> +    if (mode > RISCV_IOMMU_DC_FSC_PDTP_MODE_PD20) {
> +        /* Invalid PDTP.MODE */
> +        return RISCV_IOMMU_FQ_CAUSE_PDT_MISCONFIGURED;
> +    }
> +
> +    for (depth = mode - RISCV_IOMMU_DC_FSC_PDTP_MODE_PD8; depth-- > 0; ) {
> +        /*
> +         * Select process id index bits based on process directory tree
> +         * level. See IOMMU Specification, 2.2. Process-Directory-Table.
> +         */
> +        const int split = depth * 9 + 8;
> +        addr |= ((ctx->pasid >> split) << 3) & ~TARGET_PAGE_MASK;
> +        if (dma_memory_read(s->target_as, addr, &de, sizeof(de),
> +                            MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            return RISCV_IOMMU_FQ_CAUSE_PDT_LOAD_FAULT;
> +        }
> +        le64_to_cpus(&de);
> +        if (!(de & RISCV_IOMMU_PC_TA_V)) {
> +            return RISCV_IOMMU_FQ_CAUSE_PDT_INVALID;
> +        }
> +        addr = PPN_PHYS(get_field(de, RISCV_IOMMU_PC_FSC_PPN));
> +    }
> +
> +    /* Leaf entry in PDT */
> +    addr |= (ctx->pasid << 4) & ~TARGET_PAGE_MASK;
> +    if (dma_memory_read(s->target_as, addr, &dc.ta, sizeof(uint64_t) * 2,
> +                        MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +        return RISCV_IOMMU_FQ_CAUSE_PDT_LOAD_FAULT;
> +    }
> +
> +    /* Use FSC and TA from process directory entry. */
> +    ctx->ta = le64_to_cpu(dc.ta);
> +
> +    return 0;
> +}
> +
> +/* Translation Context cache support */
> +static gboolean __ctx_equal(gconstpointer v1, gconstpointer v2)
> +{
> +    RISCVIOMMUContext *c1 = (RISCVIOMMUContext *) v1;
> +    RISCVIOMMUContext *c2 = (RISCVIOMMUContext *) v2;
> +    return c1->devid == c2->devid && c1->pasid == c2->pasid;
> +}
> +
> +static guint __ctx_hash(gconstpointer v)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) v;
> +    /* Generate simple hash of (pasid, devid), assuming 24-bit wide devid */
> +    return (guint)(ctx->devid) + ((guint)(ctx->pasid) << 24);
> +}
> +
> +static void __ctx_inval_devid_pasid(gpointer key, gpointer value, gpointer data)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
> +    RISCVIOMMUContext *arg = (RISCVIOMMUContext *) data;
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V &&
> +        ctx->devid == arg->devid &&
> +        ctx->pasid == arg->pasid) {
> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
> +    }
> +}
> +
> +static void __ctx_inval_devid(gpointer key, gpointer value, gpointer data)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
> +    RISCVIOMMUContext *arg = (RISCVIOMMUContext *) data;
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V &&
> +        ctx->devid == arg->devid) {
> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
> +    }
> +}
> +
> +static void __ctx_inval_all(gpointer key, gpointer value, gpointer data)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V) {
> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
> +    }
> +}
> +
> +static void riscv_iommu_ctx_inval(RISCVIOMMUState *s, GHFunc func,
> +    uint32_t devid, uint32_t pasid)
> +{
> +    GHashTable *ctx_cache;
> +    RISCVIOMMUContext key = {
> +        .devid = devid,
> +        .pasid = pasid,
> +    };
> +    ctx_cache = g_hash_table_ref(s->ctx_cache);
> +    g_hash_table_foreach(ctx_cache, func, &key);
> +    g_hash_table_unref(ctx_cache);
> +}
> +
> +/* Find or allocate translation context for a given {device_id, process_id} */
> +static RISCVIOMMUContext *riscv_iommu_ctx(RISCVIOMMUState *s,
> +    unsigned devid, unsigned pasid, void **ref)
> +{
> +    GHashTable *ctx_cache;
> +    RISCVIOMMUContext *ctx;
> +    RISCVIOMMUContext key = {
> +        .devid = devid,
> +        .pasid = pasid,
> +    };
> +
> +    ctx_cache = g_hash_table_ref(s->ctx_cache);
> +    ctx = g_hash_table_lookup(ctx_cache, &key);
> +
> +    if (ctx && (ctx->tc & RISCV_IOMMU_DC_TC_V)) {
> +        *ref = ctx_cache;
> +        return ctx;
> +    }
> +
> +    if (g_hash_table_size(s->ctx_cache) >= LIMIT_CACHE_CTX) {
> +        ctx_cache = g_hash_table_new_full(__ctx_hash, __ctx_equal,
> +                                          g_free, NULL);
> +        g_hash_table_unref(qatomic_xchg(&s->ctx_cache, ctx_cache));
> +    }
> +
> +    ctx = g_new0(RISCVIOMMUContext, 1);
> +    ctx->devid = devid;
> +    ctx->pasid = pasid;
> +
> +    int fault = riscv_iommu_ctx_fetch(s, ctx);
> +    if (!fault) {
> +        g_hash_table_add(ctx_cache, ctx);
> +        *ref = ctx_cache;
> +        return ctx;
> +    }
> +
> +    g_hash_table_unref(ctx_cache);
> +    *ref = NULL;
> +
> +    riscv_iommu_report_fault(s, ctx, RISCV_IOMMU_FQ_TTYPE_UADDR_RD,
> +                             fault, !!pasid, 0, 0);
> +
> +    g_free(ctx);
> +    return NULL;
> +}
> +
> +static void riscv_iommu_ctx_put(RISCVIOMMUState *s, void *ref)
> +{
> +    if (ref) {
> +        g_hash_table_unref((GHashTable *)ref);
> +    }
> +}
> +
> +/* Find or allocate address space for a given device */
> +static AddressSpace *riscv_iommu_space(RISCVIOMMUState *s, uint32_t devid)
> +{
> +    RISCVIOMMUSpace *as;
> +
> +    /* FIXME: PCIe bus remapping for attached endpoints. */
> +    devid |= s->bus << 8;
> +
> +    qemu_mutex_lock(&s->core_lock);
> +    QLIST_FOREACH(as, &s->spaces, list) {
> +        if (as->devid == devid) {
> +            break;
> +        }
> +    }
> +    qemu_mutex_unlock(&s->core_lock);
> +
> +    if (as == NULL) {
> +        char name[64];
> +        as = g_new0(RISCVIOMMUSpace, 1);
> +
> +        as->iommu = s;
> +        as->devid = devid;
> +
> +        snprintf(name, sizeof(name), "riscv-iommu-%04x:%02x.%d-iova",
> +            PCI_BUS_NUM(as->devid), PCI_SLOT(as->devid), PCI_FUNC(as->devid));
> +
> +        /* IOVA address space, untranslated addresses */
> +        memory_region_init_iommu(&as->iova_mr, sizeof(as->iova_mr),
> +            TYPE_RISCV_IOMMU_MEMORY_REGION,
> +            OBJECT(as), "riscv_iommu", UINT64_MAX);
> +        address_space_init(&as->iova_as, MEMORY_REGION(&as->iova_mr), name);
> +
> +        qemu_mutex_lock(&s->core_lock);
> +        QLIST_INSERT_HEAD(&s->spaces, as, list);
> +        qemu_mutex_unlock(&s->core_lock);
> +
> +        trace_riscv_iommu_new(s->parent_obj.id, PCI_BUS_NUM(as->devid),
> +                PCI_SLOT(as->devid), PCI_FUNC(as->devid));
> +    }
> +    return &as->iova_as;
> +}
> +
> +static int riscv_iommu_translate(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
> +    IOMMUTLBEntry *iotlb)
> +{
> +    bool enable_pasid;
> +    bool enable_pri;
> +    int fault;
> +
> +    /*
> +     * TC[32] is reserved for custom extensions, used here to temporarily
> +     * enable automatic page-request generation for ATS queries.
> +     */
> +    enable_pri = (iotlb->perm == IOMMU_NONE) && (ctx->tc & BIT_ULL(32));
> +    enable_pasid = (ctx->tc & RISCV_IOMMU_DC_TC_PDTV);
> +
> +    /* Translate using device directory / page table information. */
> +    fault = riscv_iommu_spa_fetch(s, ctx, iotlb);
> +
> +    if (enable_pri && fault) {
> +        struct riscv_iommu_pq_record pr = {0};
> +        if (enable_pasid) {
> +            pr.hdr = set_field(RISCV_IOMMU_PREQ_HDR_PV,
> +                RISCV_IOMMU_PREQ_HDR_PID, ctx->pasid);
> +        }
> +        pr.hdr = set_field(pr.hdr, RISCV_IOMMU_PREQ_HDR_DID, ctx->devid);
> +        pr.payload = (iotlb->iova & TARGET_PAGE_MASK) |
> +                     RISCV_IOMMU_PREQ_PAYLOAD_M;
> +        riscv_iommu_pri(s, &pr);
> +        return fault;
> +    }
> +
> +    if (fault) {
> +        unsigned ttype;
> +
> +        if (iotlb->perm & IOMMU_RW) {
> +            ttype = RISCV_IOMMU_FQ_TTYPE_UADDR_WR;
> +        } else {
> +            ttype = RISCV_IOMMU_FQ_TTYPE_UADDR_RD;
> +        }
> +
> +        riscv_iommu_report_fault(s, ctx, ttype, fault, enable_pasid,
> +                                 iotlb->iova, iotlb->translated_addr);
> +        return fault;
> +    }
> +
> +    return 0;
> +}
> +
> +/* IOMMU Command Interface */
> +static MemTxResult riscv_iommu_iofence(RISCVIOMMUState *s, bool notify,
> +    uint64_t addr, uint32_t data)
> +{
> +    /*
> +     * ATS processing in this implementation of the IOMMU is synchronous,
> +     * no need to wait for completions here.
> +     */
> +    if (!notify) {
> +        return MEMTX_OK;
> +    }
> +
> +    return dma_memory_write(s->target_as, addr, &data, sizeof(data),
> +        MEMTXATTRS_UNSPECIFIED);
> +}
> +
> +static void riscv_iommu_process_ddtp(RISCVIOMMUState *s)
> +{
> +    uint64_t old_ddtp = s->ddtp;
> +    uint64_t new_ddtp = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_DDTP);
> +    unsigned new_mode = get_field(new_ddtp, RISCV_IOMMU_DDTP_MODE);
> +    unsigned old_mode = get_field(old_ddtp, RISCV_IOMMU_DDTP_MODE);
> +    bool ok = false;
> +
> +    /*
> +     * Check for allowed DDTP.MODE transitions:
> +     * {OFF, BARE}        -> {OFF, BARE, 1LVL, 2LVL, 3LVL}
> +     * {1LVL, 2LVL, 3LVL} -> {OFF, BARE}
> +     */
> +    if (new_mode == old_mode ||
> +        new_mode == RISCV_IOMMU_DDTP_MODE_OFF ||
> +        new_mode == RISCV_IOMMU_DDTP_MODE_BARE) {
> +        ok = true;
> +    } else if (new_mode == RISCV_IOMMU_DDTP_MODE_1LVL ||
> +               new_mode == RISCV_IOMMU_DDTP_MODE_2LVL ||
> +               new_mode == RISCV_IOMMU_DDTP_MODE_3LVL) {
> +        ok = old_mode == RISCV_IOMMU_DDTP_MODE_OFF ||
> +             old_mode == RISCV_IOMMU_DDTP_MODE_BARE;
> +    }
> +
> +    if (ok) {
> +        /* clear reserved and busy bits, report back sanitized version */
> +        new_ddtp = set_field(new_ddtp & RISCV_IOMMU_DDTP_PPN,
> +                             RISCV_IOMMU_DDTP_MODE, new_mode);
> +    } else {
> +        new_ddtp = old_ddtp;
> +    }
> +    s->ddtp = new_ddtp;
> +
> +    riscv_iommu_reg_set64(s, RISCV_IOMMU_REG_DDTP, new_ddtp);
> +}
> +
> +/* Command function and opcode field. */
> +#define RISCV_IOMMU_CMD(func, op) (((func) << 7) | (op))
> +
> +static void riscv_iommu_process_cq_tail(RISCVIOMMUState *s)
> +{
> +    struct riscv_iommu_command cmd;
> +    MemTxResult res;
> +    dma_addr_t addr;
> +    uint32_t tail, head, ctrl;
> +    uint64_t cmd_opcode;
> +    GHFunc func;
> +
> +    ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
> +    tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQT) & s->cq_mask;
> +    head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQH) & s->cq_mask;
> +
> +    /* Check for pending error or queue processing disabled */
> +    if (!(ctrl & RISCV_IOMMU_CQCSR_CQON) ||
> +        !!(ctrl & (RISCV_IOMMU_CQCSR_CMD_ILL | RISCV_IOMMU_CQCSR_CQMF))) {
> +        return;
> +    }
> +
> +    while (tail != head) {
> +        addr = s->cq_addr  + head * sizeof(cmd);
> +        res = dma_memory_read(s->target_as, addr, &cmd, sizeof(cmd),
> +                              MEMTXATTRS_UNSPECIFIED);
> +
> +        if (res != MEMTX_OK) {
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
> +                                  RISCV_IOMMU_CQCSR_CQMF, 0);
> +            goto fault;
> +        }
> +
> +        trace_riscv_iommu_cmd(s->parent_obj.id, cmd.dword0, cmd.dword1);
> +
> +        cmd_opcode = get_field(cmd.dword0,
> +                               RISCV_IOMMU_CMD_OPCODE | RISCV_IOMMU_CMD_FUNC);
> +
> +        switch (cmd_opcode) {
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOFENCE_FUNC_C,
> +                             RISCV_IOMMU_CMD_IOFENCE_OPCODE):
> +            res = riscv_iommu_iofence(s,
> +                cmd.dword0 & RISCV_IOMMU_CMD_IOFENCE_AV, cmd.dword1,
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IOFENCE_DATA));
> +
> +            if (res != MEMTX_OK) {
> +                riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
> +                                      RISCV_IOMMU_CQCSR_CQMF, 0);
> +                goto fault;
> +            }
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOTINVAL_FUNC_GVMA,
> +                             RISCV_IOMMU_CMD_IOTINVAL_OPCODE):
> +            if (cmd.dword0 & RISCV_IOMMU_CMD_IOTINVAL_PSCV) {
> +                /* illegal command arguments IOTINVAL.GVMA & PSCV == 1 */
> +                goto cmd_ill;
> +            }
> +            /* translation cache not implemented yet */
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOTINVAL_FUNC_VMA,
> +                             RISCV_IOMMU_CMD_IOTINVAL_OPCODE):
> +            /* translation cache not implemented yet */
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IODIR_FUNC_INVAL_DDT,
> +                             RISCV_IOMMU_CMD_IODIR_OPCODE):
> +            if (!(cmd.dword0 & RISCV_IOMMU_CMD_IODIR_DV)) {
> +                /* invalidate all device context cache mappings */
> +                func = __ctx_inval_all;
> +            } else {
> +                /* invalidate all device context matching DID */
> +                func = __ctx_inval_devid;
> +            }
> +            riscv_iommu_ctx_inval(s, func,
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_DID), 0);
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IODIR_FUNC_INVAL_PDT,
> +                             RISCV_IOMMU_CMD_IODIR_OPCODE):
> +            if (!(cmd.dword0 & RISCV_IOMMU_CMD_IODIR_DV)) {
> +                /* illegal command arguments IODIR_PDT & DV == 0 */
> +                goto cmd_ill;
> +            } else {
> +                func = __ctx_inval_devid_pasid;
> +            }
> +            riscv_iommu_ctx_inval(s, func,
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_DID),
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_PID));
> +            break;
> +
> +        default:
> +        cmd_ill:
> +            /* Invalid instruction, do not advance instruction index. */
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
> +                RISCV_IOMMU_CQCSR_CMD_ILL, 0);
> +            goto fault;
> +        }
> +
> +        /* Advance and update head pointer after command completes. */
> +        head = (head + 1) & s->cq_mask;
> +        riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_CQH, head);
> +    }
> +    return;
> +
> +fault:
> +    if (ctrl & RISCV_IOMMU_CQCSR_CIE) {
> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_CQ);
> +    }
> +}
> +
> +static void riscv_iommu_process_cq_control(RISCVIOMMUState *s)
> +{
> +    uint64_t base;
> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
> +    uint32_t ctrl_clr;
> +    bool enable = !!(ctrl_set & RISCV_IOMMU_CQCSR_CQEN);
> +    bool active = !!(ctrl_set & RISCV_IOMMU_CQCSR_CQON);
> +
> +    if (enable && !active) {
> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_CQB);
> +        s->cq_mask = (2ULL << get_field(base, RISCV_IOMMU_CQB_LOG2SZ)) - 1;
> +        s->cq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_CQB_PPN));
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQT], ~s->cq_mask);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_CQH], 0);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_CQT], 0);
> +        ctrl_set = RISCV_IOMMU_CQCSR_CQON;
> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY | RISCV_IOMMU_CQCSR_CQMF |
> +                   RISCV_IOMMU_CQCSR_CMD_ILL | RISCV_IOMMU_CQCSR_CMD_TO |
> +                   RISCV_IOMMU_CQCSR_FENCE_W_IP;
> +    } else if (!enable && active) {
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQT], ~0);
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY | RISCV_IOMMU_CQCSR_CQON;
> +    } else {
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR, ctrl_set, ctrl_clr);
> +}
> +
> +static void riscv_iommu_process_fq_control(RISCVIOMMUState *s)
> +{
> +    uint64_t base;
> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
> +    uint32_t ctrl_clr;
> +    bool enable = !!(ctrl_set & RISCV_IOMMU_FQCSR_FQEN);
> +    bool active = !!(ctrl_set & RISCV_IOMMU_FQCSR_FQON);
> +
> +    if (enable && !active) {
> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_FQB);
> +        s->fq_mask = (2ULL << get_field(base, RISCV_IOMMU_FQB_LOG2SZ)) - 1;
> +        s->fq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_FQB_PPN));
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQH], ~s->fq_mask);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_FQH], 0);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_FQT], 0);
> +        ctrl_set = RISCV_IOMMU_FQCSR_FQON;
> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY | RISCV_IOMMU_FQCSR_FQMF |
> +            RISCV_IOMMU_FQCSR_FQOF;
> +    } else if (!enable && active) {
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQH], ~0);
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY | RISCV_IOMMU_FQCSR_FQON;
> +    } else {
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR, ctrl_set, ctrl_clr);
> +}
> +
> +static void riscv_iommu_process_pq_control(RISCVIOMMUState *s)
> +{
> +    uint64_t base;
> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
> +    uint32_t ctrl_clr;
> +    bool enable = !!(ctrl_set & RISCV_IOMMU_PQCSR_PQEN);
> +    bool active = !!(ctrl_set & RISCV_IOMMU_PQCSR_PQON);
> +
> +    if (enable && !active) {
> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_PQB);
> +        s->pq_mask = (2ULL << get_field(base, RISCV_IOMMU_PQB_LOG2SZ)) - 1;
> +        s->pq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_PQB_PPN));
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQH], ~s->pq_mask);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_PQH], 0);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_PQT], 0);
> +        ctrl_set = RISCV_IOMMU_PQCSR_PQON;
> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY | RISCV_IOMMU_PQCSR_PQMF |
> +            RISCV_IOMMU_PQCSR_PQOF;
> +    } else if (!enable && active) {
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQH], ~0);
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY | RISCV_IOMMU_PQCSR_PQON;
> +    } else {
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR, ctrl_set, ctrl_clr);
> +}
> +
> +typedef void riscv_iommu_process_fn(RISCVIOMMUState *s);
> +
> +static void riscv_iommu_update_ipsr(RISCVIOMMUState *s, uint64_t data)
> +{
> +    uint32_t cqcsr, fqcsr, pqcsr;
> +    uint32_t ipsr_set = 0;
> +    uint32_t ipsr_clr = 0;
> +
> +    if (data & RISCV_IOMMU_IPSR_CIP) {
> +        cqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
> +
> +        if (cqcsr & RISCV_IOMMU_CQCSR_CIE &&
> +            (cqcsr & RISCV_IOMMU_CQCSR_FENCE_W_IP ||
> +             cqcsr & RISCV_IOMMU_CQCSR_CMD_ILL ||
> +             cqcsr & RISCV_IOMMU_CQCSR_CMD_TO ||
> +             cqcsr & RISCV_IOMMU_CQCSR_CQMF)) {
> +            ipsr_set |= RISCV_IOMMU_IPSR_CIP;
> +        } else {
> +            ipsr_clr |= RISCV_IOMMU_IPSR_CIP;
> +        }
> +    } else {
> +        ipsr_clr |= RISCV_IOMMU_IPSR_CIP;
> +    }
> +
> +    if (data & RISCV_IOMMU_IPSR_FIP) {
> +        fqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
> +
> +        if (fqcsr & RISCV_IOMMU_FQCSR_FIE &&
> +            (fqcsr & RISCV_IOMMU_FQCSR_FQOF ||
> +             fqcsr & RISCV_IOMMU_FQCSR_FQMF)) {
> +            ipsr_set |= RISCV_IOMMU_IPSR_FIP;
> +        } else {
> +            ipsr_clr |= RISCV_IOMMU_IPSR_FIP;
> +        }
> +    } else {
> +        ipsr_clr |= RISCV_IOMMU_IPSR_FIP;
> +    }
> +
> +    if (data & RISCV_IOMMU_IPSR_PIP) {
> +        pqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
> +
> +        if (pqcsr & RISCV_IOMMU_PQCSR_PIE &&
> +            (pqcsr & RISCV_IOMMU_PQCSR_PQOF ||
> +             pqcsr & RISCV_IOMMU_PQCSR_PQMF)) {
> +            ipsr_set |= RISCV_IOMMU_IPSR_PIP;
> +        } else {
> +            ipsr_clr |= RISCV_IOMMU_IPSR_PIP;
> +        }
> +    } else {
> +        ipsr_clr |= RISCV_IOMMU_IPSR_PIP;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_IPSR, ipsr_set, ipsr_clr);
> +}
> +
> +static MemTxResult riscv_iommu_mmio_write(void *opaque, hwaddr addr,
> +    uint64_t data, unsigned size, MemTxAttrs attrs)
> +{
> +    riscv_iommu_process_fn *process_fn = NULL;
> +    RISCVIOMMUState *s = opaque;
> +    uint32_t regb = addr & ~3;
> +    uint32_t busy = 0;
> +    uint64_t val = 0;
> +
> +    if ((addr & (size - 1)) != 0) {
> +        /* Unsupported MMIO alignment or access size */
> +        return MEMTX_ERROR;
> +    }
> +
> +    if (addr + size > RISCV_IOMMU_REG_MSI_CONFIG) {
> +        /* Unsupported MMIO access location. */
> +        return MEMTX_ACCESS_ERROR;
> +    }
> +
> +    /* Track actionable MMIO write. */
> +    switch (regb) {
> +    case RISCV_IOMMU_REG_DDTP:
> +    case RISCV_IOMMU_REG_DDTP + 4:
> +        process_fn = riscv_iommu_process_ddtp;
> +        regb = RISCV_IOMMU_REG_DDTP;
> +        busy = RISCV_IOMMU_DDTP_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_CQT:
> +        process_fn = riscv_iommu_process_cq_tail;
> +        break;
> +
> +    case RISCV_IOMMU_REG_CQCSR:
> +        process_fn = riscv_iommu_process_cq_control;
> +        busy = RISCV_IOMMU_CQCSR_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_FQCSR:
> +        process_fn = riscv_iommu_process_fq_control;
> +        busy = RISCV_IOMMU_FQCSR_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_PQCSR:
> +        process_fn = riscv_iommu_process_pq_control;
> +        busy = RISCV_IOMMU_PQCSR_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_IPSR:
> +        /*
> +         * IPSR has special procedures to update. Execute it
> +         * and exit.
> +         */
> +        if (size == 4) {
> +            uint32_t ro = ldl_le_p(&s->regs_ro[addr]);
> +            uint32_t wc = ldl_le_p(&s->regs_wc[addr]);
> +            uint32_t rw = ldl_le_p(&s->regs_rw[addr]);
> +            stl_le_p(&val, ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +        } else if (size == 8) {
> +            uint64_t ro = ldq_le_p(&s->regs_ro[addr]);
> +            uint64_t wc = ldq_le_p(&s->regs_wc[addr]);
> +            uint64_t rw = ldq_le_p(&s->regs_rw[addr]);
> +            stq_le_p(&val, ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +        }
> +
> +        riscv_iommu_update_ipsr(s, val);
> +
> +        return MEMTX_OK;
> +
> +    default:
> +        break;
> +    }
> +
> +    /*
> +     * Registers update might be not synchronized with core logic.
> +     * If system software updates register when relevant BUSY bit
> +     * is set IOMMU behavior of additional writes to the register
> +     * is UNSPECIFIED.
> +     */
> +    qemu_spin_lock(&s->regs_lock);
> +    if (size == 1) {
> +        uint8_t ro = s->regs_ro[addr];
> +        uint8_t wc = s->regs_wc[addr];
> +        uint8_t rw = s->regs_rw[addr];
> +        s->regs_rw[addr] = ((rw & ro) | (data & ~ro)) & ~(data & wc);
> +    } else if (size == 2) {
> +        uint16_t ro = lduw_le_p(&s->regs_ro[addr]);
> +        uint16_t wc = lduw_le_p(&s->regs_wc[addr]);
> +        uint16_t rw = lduw_le_p(&s->regs_rw[addr]);
> +        stw_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +    } else if (size == 4) {
> +        uint32_t ro = ldl_le_p(&s->regs_ro[addr]);
> +        uint32_t wc = ldl_le_p(&s->regs_wc[addr]);
> +        uint32_t rw = ldl_le_p(&s->regs_rw[addr]);
> +        stl_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +    } else if (size == 8) {
> +        uint64_t ro = ldq_le_p(&s->regs_ro[addr]);
> +        uint64_t wc = ldq_le_p(&s->regs_wc[addr]);
> +        uint64_t rw = ldq_le_p(&s->regs_rw[addr]);
> +        stq_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +    }
> +
> +    /* Busy flag update, MSB 4-byte register. */
> +    if (busy) {
> +        uint32_t rw = ldl_le_p(&s->regs_rw[regb]);
> +        stl_le_p(&s->regs_rw[regb], rw | busy);
> +    }
> +    qemu_spin_unlock(&s->regs_lock);
> +
> +    if (process_fn) {
> +        qemu_mutex_lock(&s->core_lock);
> +        process_fn(s);
> +        qemu_mutex_unlock(&s->core_lock);
> +    }
> +
> +    return MEMTX_OK;
> +}
> +
> +static MemTxResult riscv_iommu_mmio_read(void *opaque, hwaddr addr,
> +    uint64_t *data, unsigned size, MemTxAttrs attrs)
> +{
> +    RISCVIOMMUState *s = opaque;
> +    uint64_t val = -1;
> +    uint8_t *ptr;
> +
> +    if ((addr & (size - 1)) != 0) {
> +        /* Unsupported MMIO alignment. */
> +        return MEMTX_ERROR;
> +    }
> +
> +    if (addr + size > RISCV_IOMMU_REG_MSI_CONFIG) {
> +        return MEMTX_ACCESS_ERROR;
> +    }
> +
> +    ptr = &s->regs_rw[addr];
> +
> +    if (size == 1) {
> +        val = (uint64_t)*ptr;
> +    } else if (size == 2) {
> +        val = lduw_le_p(ptr);
> +    } else if (size == 4) {
> +        val = ldl_le_p(ptr);
> +    } else if (size == 8) {
> +        val = ldq_le_p(ptr);
> +    } else {
> +        return MEMTX_ERROR;
> +    }
> +
> +    *data = val;
> +
> +    return MEMTX_OK;
> +}
> +
> +static const MemoryRegionOps riscv_iommu_mmio_ops = {
> +    .read_with_attrs = riscv_iommu_mmio_read,
> +    .write_with_attrs = riscv_iommu_mmio_write,
> +    .endianness = DEVICE_NATIVE_ENDIAN,
> +    .impl = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +        .unaligned = false,
> +    },
> +    .valid = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +    }
> +};
> +
> +/*
> + * Translations matching MSI pattern check are redirected to "riscv-iommu-trap"
> + * memory region as untranslated address, for additional MSI/MRIF interception
> + * by IOMMU interrupt remapping implementation.
> + * Note: Device emulation code generating an MSI is expected to provide a valid
> + * memory transaction attributes with requested_id set.
> + */
> +static MemTxResult riscv_iommu_trap_write(void *opaque, hwaddr addr,
> +    uint64_t data, unsigned size, MemTxAttrs attrs)
> +{
> +    RISCVIOMMUState* s = (RISCVIOMMUState *)opaque;
> +    RISCVIOMMUContext *ctx;
> +    MemTxResult res;
> +    void *ref;
> +    uint32_t devid = attrs.requester_id;
> +
> +    if (attrs.unspecified) {
> +        return MEMTX_ACCESS_ERROR;
> +    }
> +
> +    /* FIXME: PCIe bus remapping for attached endpoints. */
> +    devid |= s->bus << 8;
> +
> +    ctx = riscv_iommu_ctx(s, devid, 0, &ref);
> +    if (ctx == NULL) {
> +        res = MEMTX_ACCESS_ERROR;
> +    } else {
> +        res = riscv_iommu_msi_write(s, ctx, addr, data, size, attrs);
> +    }
> +    riscv_iommu_ctx_put(s, ref);
> +    return res;
> +}
> +
> +static MemTxResult riscv_iommu_trap_read(void *opaque, hwaddr addr,
> +    uint64_t *data, unsigned size, MemTxAttrs attrs)
> +{
> +    return MEMTX_ACCESS_ERROR;
> +}
> +
> +static const MemoryRegionOps riscv_iommu_trap_ops = {
> +    .read_with_attrs = riscv_iommu_trap_read,
> +    .write_with_attrs = riscv_iommu_trap_write,
> +    .endianness = DEVICE_LITTLE_ENDIAN,
> +    .impl = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +        .unaligned = true,
> +    },
> +    .valid = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +    }
> +};
> +
> +static void riscv_iommu_realize(DeviceState *dev, Error **errp)
> +{
> +    RISCVIOMMUState *s = RISCV_IOMMU(dev);
> +
> +    s->cap = s->version & RISCV_IOMMU_CAP_VERSION;
> +    if (s->enable_msi) {
> +        s->cap |= RISCV_IOMMU_CAP_MSI_FLAT | RISCV_IOMMU_CAP_MSI_MRIF;
> +    }
> +    /* Report QEMU target physical address space limits */
> +    s->cap = set_field(s->cap, RISCV_IOMMU_CAP_PAS,
> +                       TARGET_PHYS_ADDR_SPACE_BITS);
> +
> +    /* TODO: method to report supported PASID bits */
> +    s->pasid_bits = 8; /* restricted to size of MemTxAttrs.pasid */
> +    s->cap |= RISCV_IOMMU_CAP_PD8;
> +
> +    /* Out-of-reset translation mode: OFF (DMA disabled) BARE (passthrough) */
> +    s->ddtp = set_field(0, RISCV_IOMMU_DDTP_MODE, s->enable_off ?
> +                        RISCV_IOMMU_DDTP_MODE_OFF : RISCV_IOMMU_DDTP_MODE_BARE);
> +
> +    /* register storage */
> +    s->regs_rw = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
> +    s->regs_ro = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
> +    s->regs_wc = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
> +
> +     /* Mark all registers read-only */
> +    memset(s->regs_ro, 0xff, RISCV_IOMMU_REG_SIZE);
> +
> +    /*
> +     * Register complete MMIO space, including MSI/PBA registers.
> +     * Note, PCIDevice implementation will add overlapping MR for MSI/PBA,
> +     * managed directly by the PCIDevice implementation.
> +     */
> +    memory_region_init_io(&s->regs_mr, OBJECT(dev), &riscv_iommu_mmio_ops, s,
> +        "riscv-iommu-regs", RISCV_IOMMU_REG_SIZE);
> +
> +    /* Set power-on register state */
> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_CAP], s->cap);
> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_FCTL], 0);
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_DDTP],
> +        ~(RISCV_IOMMU_DDTP_PPN | RISCV_IOMMU_DDTP_MODE));
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQB],
> +        ~(RISCV_IOMMU_CQB_LOG2SZ | RISCV_IOMMU_CQB_PPN));
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQB],
> +        ~(RISCV_IOMMU_FQB_LOG2SZ | RISCV_IOMMU_FQB_PPN));
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQB],
> +        ~(RISCV_IOMMU_PQB_LOG2SZ | RISCV_IOMMU_PQB_PPN));
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_CQCSR], RISCV_IOMMU_CQCSR_CQMF |
> +        RISCV_IOMMU_CQCSR_CMD_TO | RISCV_IOMMU_CQCSR_CMD_ILL);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQCSR], RISCV_IOMMU_CQCSR_CQON |
> +        RISCV_IOMMU_CQCSR_BUSY);
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_FQCSR], RISCV_IOMMU_FQCSR_FQMF |
> +        RISCV_IOMMU_FQCSR_FQOF);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQCSR], RISCV_IOMMU_FQCSR_FQON |
> +        RISCV_IOMMU_FQCSR_BUSY);
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_PQCSR], RISCV_IOMMU_PQCSR_PQMF |
> +        RISCV_IOMMU_PQCSR_PQOF);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQCSR], RISCV_IOMMU_PQCSR_PQON |
> +        RISCV_IOMMU_PQCSR_BUSY);
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_IPSR], ~0);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_IVEC], 0);
> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_DDTP], s->ddtp);
> +
> +    /* Memory region for downstream access, if specified. */
> +    if (s->target_mr) {
> +        s->target_as = g_new0(AddressSpace, 1);
> +        address_space_init(s->target_as, s->target_mr,
> +            "riscv-iommu-downstream");
> +    } else {
> +        /* Fallback to global system memory. */
> +        s->target_as = &address_space_memory;
> +    }
> +
> +    /* Memory region for untranslated MRIF/MSI writes */
> +    memory_region_init_io(&s->trap_mr, OBJECT(dev), &riscv_iommu_trap_ops, s,
> +            "riscv-iommu-trap", ~0ULL);
> +    address_space_init(&s->trap_as, &s->trap_mr, "riscv-iommu-trap-as");
> +
> +    /* Device translation context cache */
> +    s->ctx_cache = g_hash_table_new_full(__ctx_hash, __ctx_equal,
> +                                         g_free, NULL);
> +
> +    s->iommus.le_next = NULL;
> +    s->iommus.le_prev = NULL;
> +    QLIST_INIT(&s->spaces);
> +    qemu_mutex_init(&s->core_lock);
> +    qemu_spin_init(&s->regs_lock);
> +}
> +
> +static void riscv_iommu_unrealize(DeviceState *dev)
> +{
> +    RISCVIOMMUState *s = RISCV_IOMMU(dev);
> +
> +    qemu_mutex_destroy(&s->core_lock);
> +    g_hash_table_unref(s->ctx_cache);
> +}
> +
> +static Property riscv_iommu_properties[] = {
> +    DEFINE_PROP_UINT32("version", RISCVIOMMUState, version,
> +        RISCV_IOMMU_SPEC_DOT_VER),
> +    DEFINE_PROP_UINT32("bus", RISCVIOMMUState, bus, 0x0),
> +    DEFINE_PROP_BOOL("intremap", RISCVIOMMUState, enable_msi, TRUE),
> +    DEFINE_PROP_BOOL("off", RISCVIOMMUState, enable_off, TRUE),
> +    DEFINE_PROP_LINK("downstream-mr", RISCVIOMMUState, target_mr,
> +        TYPE_MEMORY_REGION, MemoryRegion *),
> +    DEFINE_PROP_END_OF_LIST(),
> +};
> +
> +static void riscv_iommu_class_init(ObjectClass *klass, void* data)
> +{
> +    DeviceClass *dc = DEVICE_CLASS(klass);
> +
> +    /* internal device for riscv-iommu-{pci/sys}, not user-creatable */
> +    dc->user_creatable = false;
> +    dc->realize = riscv_iommu_realize;
> +    dc->unrealize = riscv_iommu_unrealize;
> +    device_class_set_props(dc, riscv_iommu_properties);
> +}
> +
> +static const TypeInfo riscv_iommu_info = {
> +    .name = TYPE_RISCV_IOMMU,
> +    .parent = TYPE_DEVICE,
> +    .instance_size = sizeof(RISCVIOMMUState),
> +    .class_init = riscv_iommu_class_init,
> +};
> +
> +static const char *IOMMU_FLAG_STR[] = {
> +    "NA",
> +    "RO",
> +    "WR",
> +    "RW",
> +};
> +
> +/* RISC-V IOMMU Memory Region - Address Translation Space */
> +static IOMMUTLBEntry riscv_iommu_memory_region_translate(
> +    IOMMUMemoryRegion *iommu_mr, hwaddr addr,
> +    IOMMUAccessFlags flag, int iommu_idx)
> +{
> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
> +    RISCVIOMMUContext *ctx;
> +    void *ref;
> +    IOMMUTLBEntry iotlb = {
> +        .iova = addr,
> +        .target_as = as->iommu->target_as,
> +        .addr_mask = ~0ULL,
> +        .perm = flag,
> +    };
> +
> +    ctx = riscv_iommu_ctx(as->iommu, as->devid, iommu_idx, &ref);
> +    if (ctx == NULL) {
> +        /* Translation disabled or invalid. */
> +        iotlb.addr_mask = 0;
> +        iotlb.perm = IOMMU_NONE;
> +    } else if (riscv_iommu_translate(as->iommu, ctx, &iotlb)) {
> +        /* Translation disabled or fault reported. */
> +        iotlb.addr_mask = 0;
> +        iotlb.perm = IOMMU_NONE;
> +    }
> +
> +    /* Trace all dma translations with original access flags. */
> +    trace_riscv_iommu_dma(as->iommu->parent_obj.id, PCI_BUS_NUM(as->devid),
> +                          PCI_SLOT(as->devid), PCI_FUNC(as->devid), iommu_idx,
> +                          IOMMU_FLAG_STR[flag & IOMMU_RW], iotlb.iova,
> +                          iotlb.translated_addr);
> +
> +    riscv_iommu_ctx_put(as->iommu, ref);
> +
> +    return iotlb;
> +}
> +
> +static int riscv_iommu_memory_region_notify(
> +    IOMMUMemoryRegion *iommu_mr, IOMMUNotifierFlag old,
> +    IOMMUNotifierFlag new, Error **errp)
> +{
> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
> +
> +    if (old == IOMMU_NOTIFIER_NONE) {
> +        as->notifier = true;
> +        trace_riscv_iommu_notifier_add(iommu_mr->parent_obj.name);
> +    } else if (new == IOMMU_NOTIFIER_NONE) {
> +        as->notifier = false;
> +        trace_riscv_iommu_notifier_del(iommu_mr->parent_obj.name);
> +    }
> +
> +    return 0;
> +}
> +
> +static inline bool pci_is_iommu(PCIDevice *pdev)
> +{
> +    return pci_get_word(pdev->config + PCI_CLASS_DEVICE) == 0x0806;
> +}
> +
> +static AddressSpace *riscv_iommu_find_as(PCIBus *bus, void *opaque, int devfn)
> +{
> +    RISCVIOMMUState *s = (RISCVIOMMUState *) opaque;
> +    PCIDevice *pdev = pci_find_device(bus, pci_bus_num(bus), devfn);
> +    AddressSpace *as = NULL;
> +
> +    if (pdev && pci_is_iommu(pdev)) {
> +        return s->target_as;
> +    }
> +
> +    /* Find first registered IOMMU device */
> +    while (s->iommus.le_prev) {
> +        s = *(s->iommus.le_prev);
> +    }
> +
> +    /* Find first matching IOMMU */
> +    while (s != NULL && as == NULL) {
> +        as = riscv_iommu_space(s, PCI_BUILD_BDF(pci_bus_num(bus), devfn));
> +        s = s->iommus.le_next;
> +    }
> +
> +    return as ? as : &address_space_memory;
> +}
> +
> +static const PCIIOMMUOps riscv_iommu_ops = {
> +    .get_address_space = riscv_iommu_find_as,
> +};
> +
> +void riscv_iommu_pci_setup_iommu(RISCVIOMMUState *iommu, PCIBus *bus,
> +        Error **errp)
> +{
> +    if (bus->iommu_ops &&
> +        bus->iommu_ops->get_address_space == riscv_iommu_find_as) {
> +        /* Allow multiple IOMMUs on the same PCIe bus, link known devices */
> +        RISCVIOMMUState *last = (RISCVIOMMUState *)bus->iommu_opaque;
> +        QLIST_INSERT_AFTER(last, iommu, iommus);
> +    } else if (!bus->iommu_ops && !bus->iommu_opaque) {
> +        pci_setup_iommu(bus, &riscv_iommu_ops, iommu);

We use Designware PCIe with RISCV IOMMU internally and there is a 
problem that we would like to point out.

Both hw/riscv/riscv-iommu.c and hw/pci-host/designware.c utilize 
pci_setup_iommu(). When pci_setup_iommu() is invoked in 
hw/riscv/riscv-iommu.c, the iommu_ops set by Designware PCIe host is 
lost, which results in incorrect translation as the PCIe translation 
logic is overwritten and lost.

I think it may be a better choice to expose a memory region property in 
each PCIe host for the purpose of specifying the target memory region 
that the PCIe host should send requests to. By doing this, the 
Designware PCIe host can finish its translation and directs the request 
to the IOMMU memory region of the RISCV IOMMU.

The below code, based on riscv_iommu_v3, exposes a target memory region 
property in Designware PCIe and directs inbound requests to the target 
memory which can be specified to be the IOMMU memory region of RISCV IOMMU.

diff --git a/hw/pci-host/designware.c b/hw/pci-host/designware.c
index c25d50f1c6..6b6d4ac1aa 100644
--- a/hw/pci-host/designware.c
+++ b/hw/pci-host/designware.c
@@ -435,7 +435,7 @@ static void designware_pcie_root_realize(PCIDevice 
*dev, Error **errp)
          viewport->cr[0]   = DESIGNWARE_PCIE_ATU_TYPE_MEM;

          source      = &host->pci.address_space_root;
-        destination = get_system_memory();
+        destination = host->target_mr;
          direction   = "Inbound";

          /*
@@ -713,6 +713,10 @@ static void 
designware_pcie_host_realize(DeviceState *dev, Error **errp)
                         "pcie-bus-address-space");
      pci_setup_iommu(pci->bus, &designware_iommu_ops, s);

+    if (!s->target_mr) {
+        s->target_mr = get_system_memory();
+    }
+
      qdev_realize(DEVICE(&s->root), BUS(pci->bus), &error_fatal);
  }

@@ -730,6 +734,12 @@ static const VMStateDescription 
vmstate_designware_pcie_host = {
      }
  };

+static Property designware_pcie_host_properties[] = {
+    DEFINE_PROP_LINK("target-mr", DesignwarePCIEHost, target_mr,
+                     TYPE_MEMORY_REGION, MemoryRegion *),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
  static void designware_pcie_host_class_init(ObjectClass *klass, void 
*data)
  {
      DeviceClass *dc = DEVICE_CLASS(klass);
@@ -740,6 +750,7 @@ static void 
designware_pcie_host_class_init(ObjectClass *klass, void *data)
      set_bit(DEVICE_CATEGORY_BRIDGE, dc->categories);
      dc->fw_name = "pci";
      dc->vmsd = &vmstate_designware_pcie_host;
+    device_class_set_props(dc, designware_pcie_host_properties);
  }

  static void designware_pcie_host_init(Object *obj)
diff --git a/include/hw/pci-host/designware.h 
b/include/hw/pci-host/designware.h
index 908f3d946b..2530eacbb0 100644
--- a/include/hw/pci-host/designware.h
+++ b/include/hw/pci-host/designware.h
@@ -91,6 +91,7 @@ struct DesignwarePCIEHost {
      } pci;

      MemoryRegion mmio;
+    MemoryRegion *target_mr;
  };

  #endif /* DESIGNWARE_H */

We also need to specify the requester id (BDF) in the memory attribute 
when sending requests to the IOMMU memory region in order to distinguish 
the source endpoint, since all endpoints under Designware PCIe host 
write to the same IOMMU memory region.

diff --git a/include/hw/pci/pci_device.h b/include/hw/pci/pci_device.h
index d3dd0f64b2..1fc64a2d1f 100644
--- a/include/hw/pci/pci_device.h
+++ b/include/hw/pci/pci_device.h
@@ -249,8 +249,11 @@ static inline MemTxResult pci_dma_rw(PCIDevice 
*dev, dma_addr_t addr,
  static inline MemTxResult pci_dma_read(PCIDevice *dev, dma_addr_t addr,
                                         void *buf, dma_addr_t len)
  {
+    MemTxAttrs attrs = {};
+    attrs.requester_id = pci_requester_id(dev);
+
      return pci_dma_rw(dev, addr, buf, len,
-                      DMA_DIRECTION_TO_DEVICE, MEMTXATTRS_UNSPECIFIED);
+                      DMA_DIRECTION_TO_DEVICE, attrs);
  }

  /**
@@ -268,8 +271,11 @@ static inline MemTxResult pci_dma_read(PCIDevice 
*dev, dma_addr_t addr,
  static inline MemTxResult pci_dma_write(PCIDevice *dev, dma_addr_t addr,
                                          const void *buf, dma_addr_t len)
  {
+    MemTxAttrs attrs = {};
+    attrs.requester_id = pci_requester_id(dev);
+
      return pci_dma_rw(dev, addr, (void *) buf, len,
-                      DMA_DIRECTION_FROM_DEVICE, MEMTXATTRS_UNSPECIFIED);
+                      DMA_DIRECTION_FROM_DEVICE, attrs);
  }

  #define PCI_DMA_DEFINE_LDST(_l, _s, _bits) \
@@ -313,8 +319,11 @@ PCI_DMA_DEFINE_LDST(q_be, q_be, 64);
  static inline void *pci_dma_map(PCIDevice *dev, dma_addr_t addr,
                                  dma_addr_t *plen, DMADirection dir)
  {
+    MemTxAttrs attrs = {};
+    attrs.requester_id = pci_requester_id(dev);
+
      return dma_memory_map(pci_get_address_space(dev), addr, plen, dir,
-                          MEMTXATTRS_UNSPECIFIED);
+                          attrs);
  }

We hope not to call pci_setup_iommu() in hw/riscv/riscv-iommu.c to avoid 
conflicts of iommu_ops. Do you have any suggestion on this issue?

Thanks.

> +    } else {
> +        error_setg(errp, "can't register secondary IOMMU for PCI bus #%d",
> +            pci_bus_num(bus));
> +    }
> +}
> +
> +static int riscv_iommu_memory_region_index(IOMMUMemoryRegion *iommu_mr,
> +    MemTxAttrs attrs)
> +{
> +    return attrs.unspecified ? RISCV_IOMMU_NOPASID : (int)attrs.pasid;
> +}
> +
> +static int riscv_iommu_memory_region_index_len(IOMMUMemoryRegion *iommu_mr)
> +{
> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
> +    return 1 << as->iommu->pasid_bits;
> +}
> +
> +static void riscv_iommu_memory_region_init(ObjectClass *klass, void *data)
> +{
> +    IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_CLASS(klass);
> +
> +    imrc->translate = riscv_iommu_memory_region_translate;
> +    imrc->notify_flag_changed = riscv_iommu_memory_region_notify;
> +    imrc->attrs_to_index = riscv_iommu_memory_region_index;
> +    imrc->num_indexes = riscv_iommu_memory_region_index_len;
> +}
> +
> +static const TypeInfo riscv_iommu_memory_region_info = {
> +    .parent = TYPE_IOMMU_MEMORY_REGION,
> +    .name = TYPE_RISCV_IOMMU_MEMORY_REGION,
> +    .class_init = riscv_iommu_memory_region_init,
> +};
> +
> +static void riscv_iommu_register_mr_types(void)
> +{
> +    type_register_static(&riscv_iommu_memory_region_info);
> +    type_register_static(&riscv_iommu_info);
> +}
> +
> +type_init(riscv_iommu_register_mr_types);
> diff --git a/hw/riscv/riscv-iommu.h b/hw/riscv/riscv-iommu.h
> new file mode 100644
> index 0000000000..31d3907d33
> --- /dev/null
> +++ b/hw/riscv/riscv-iommu.h
> @@ -0,0 +1,141 @@
> +/*
> + * QEMU emulation of an RISC-V IOMMU
> + *
> + * Copyright (C) 2022-2023 Rivos Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#ifndef HW_RISCV_IOMMU_STATE_H
> +#define HW_RISCV_IOMMU_STATE_H
> +
> +#include "qemu/osdep.h"
> +#include "qom/object.h"
> +
> +#include "hw/riscv/iommu.h"
> +
> +struct RISCVIOMMUState {
> +    /*< private >*/
> +    DeviceState parent_obj;
> +
> +    /*< public >*/
> +    uint32_t version;     /* Reported interface version number */
> +    uint32_t pasid_bits;  /* process identifier width */
> +    uint32_t bus;         /* PCI bus mapping for non-root endpoints */
> +
> +    uint64_t cap;         /* IOMMU supported capabilities */
> +    uint64_t fctl;        /* IOMMU enabled features */
> +
> +    bool enable_off;      /* Enable out-of-reset OFF mode (DMA disabled) */
> +    bool enable_msi;      /* Enable MSI remapping */
> +
> +    /* IOMMU Internal State */
> +    uint64_t ddtp;        /* Validated Device Directory Tree Root Pointer */
> +
> +    dma_addr_t cq_addr;   /* Command queue base physical address */
> +    dma_addr_t fq_addr;   /* Fault/event queue base physical address */
> +    dma_addr_t pq_addr;   /* Page request queue base physical address */
> +
> +    uint32_t cq_mask;     /* Command queue index bit mask */
> +    uint32_t fq_mask;     /* Fault/event queue index bit mask */
> +    uint32_t pq_mask;     /* Page request queue index bit mask */
> +
> +    /* interrupt notifier */
> +    void (*notify)(RISCVIOMMUState *iommu, unsigned vector);
> +
> +    /* IOMMU State Machine */
> +    QemuThread core_proc; /* Background processing thread */
> +    QemuMutex core_lock;  /* Global IOMMU lock, used for cache/regs updates */
> +    QemuCond core_cond;   /* Background processing wake up signal */
> +    unsigned core_exec;   /* Processing thread execution actions */
> +
> +    /* IOMMU target address space */
> +    AddressSpace *target_as;
> +    MemoryRegion *target_mr;
> +
> +    /* MSI / MRIF access trap */
> +    AddressSpace trap_as;
> +    MemoryRegion trap_mr;
> +
> +    GHashTable *ctx_cache;          /* Device translation Context Cache */
> +
> +    /* MMIO Hardware Interface */
> +    MemoryRegion regs_mr;
> +    QemuSpin regs_lock;
> +    uint8_t *regs_rw;  /* register state (user write) */
> +    uint8_t *regs_wc;  /* write-1-to-clear mask */
> +    uint8_t *regs_ro;  /* read-only mask */
> +
> +    QLIST_ENTRY(RISCVIOMMUState) iommus;
> +    QLIST_HEAD(, RISCVIOMMUSpace) spaces;
> +};
> +
> +void riscv_iommu_pci_setup_iommu(RISCVIOMMUState *iommu, PCIBus *bus,
> +         Error **errp);
> +
> +/* private helpers */
> +
> +/* Register helper functions */
> +static inline uint32_t riscv_iommu_reg_mod32(RISCVIOMMUState *s,
> +    unsigned idx, uint32_t set, uint32_t clr)
> +{
> +    uint32_t val;
> +    qemu_spin_lock(&s->regs_lock);
> +    val = ldl_le_p(s->regs_rw + idx);
> +    stl_le_p(s->regs_rw + idx, (val & ~clr) | set);
> +    qemu_spin_unlock(&s->regs_lock);
> +    return val;
> +}
> +
> +static inline void riscv_iommu_reg_set32(RISCVIOMMUState *s,
> +    unsigned idx, uint32_t set)
> +{
> +    qemu_spin_lock(&s->regs_lock);
> +    stl_le_p(s->regs_rw + idx, set);
> +    qemu_spin_unlock(&s->regs_lock);
> +}
> +
> +static inline uint32_t riscv_iommu_reg_get32(RISCVIOMMUState *s,
> +    unsigned idx)
> +{
> +    return ldl_le_p(s->regs_rw + idx);
> +}
> +
> +static inline uint64_t riscv_iommu_reg_mod64(RISCVIOMMUState *s,
> +    unsigned idx, uint64_t set, uint64_t clr)
> +{
> +    uint64_t val;
> +    qemu_spin_lock(&s->regs_lock);
> +    val = ldq_le_p(s->regs_rw + idx);
> +    stq_le_p(s->regs_rw + idx, (val & ~clr) | set);
> +    qemu_spin_unlock(&s->regs_lock);
> +    return val;
> +}
> +
> +static inline void riscv_iommu_reg_set64(RISCVIOMMUState *s,
> +    unsigned idx, uint64_t set)
> +{
> +    qemu_spin_lock(&s->regs_lock);
> +    stq_le_p(s->regs_rw + idx, set);
> +    qemu_spin_unlock(&s->regs_lock);
> +}
> +
> +static inline uint64_t riscv_iommu_reg_get64(RISCVIOMMUState *s,
> +    unsigned idx)
> +{
> +    return ldq_le_p(s->regs_rw + idx);
> +}
> +
> +
> +
> +#endif
> diff --git a/hw/riscv/trace-events b/hw/riscv/trace-events
> new file mode 100644
> index 0000000000..42a97caffa
> --- /dev/null
> +++ b/hw/riscv/trace-events
> @@ -0,0 +1,11 @@
> +# See documentation at docs/devel/tracing.rst
> +
> +# riscv-iommu.c
> +riscv_iommu_new(const char *id, unsigned b, unsigned d, unsigned f) "%s: device attached %04x:%02x.%d"
> +riscv_iommu_flt(const char *id, unsigned b, unsigned d, unsigned f, uint64_t reason, uint64_t iova) "%s: fault %04x:%02x.%u reason: 0x%"PRIx64" iova: 0x%"PRIx64
> +riscv_iommu_pri(const char *id, unsigned b, unsigned d, unsigned f, uint64_t iova) "%s: page request %04x:%02x.%u iova: 0x%"PRIx64
> +riscv_iommu_dma(const char *id, unsigned b, unsigned d, unsigned f, unsigned pasid, const char *dir, uint64_t iova, uint64_t phys) "%s: translate %04x:%02x.%u #%u %s 0x%"PRIx64" -> 0x%"PRIx64
> +riscv_iommu_msi(const char *id, unsigned b, unsigned d, unsigned f, uint64_t iova, uint64_t phys) "%s: translate %04x:%02x.%u MSI 0x%"PRIx64" -> 0x%"PRIx64
> +riscv_iommu_cmd(const char *id, uint64_t l, uint64_t u) "%s: command 0x%"PRIx64" 0x%"PRIx64
> +riscv_iommu_notifier_add(const char *id) "%s: dev-iotlb notifier added"
> +riscv_iommu_notifier_del(const char *id) "%s: dev-iotlb notifier removed"
> diff --git a/hw/riscv/trace.h b/hw/riscv/trace.h
> new file mode 100644
> index 0000000000..8c0e3ca1f3
> --- /dev/null
> +++ b/hw/riscv/trace.h
> @@ -0,0 +1 @@
> +#include "trace/trace-hw_riscv.h"
> diff --git a/include/hw/riscv/iommu.h b/include/hw/riscv/iommu.h
> new file mode 100644
> index 0000000000..070ee69973
> --- /dev/null
> +++ b/include/hw/riscv/iommu.h
> @@ -0,0 +1,36 @@
> +/*
> + * QEMU emulation of an RISC-V IOMMU
> + *
> + * Copyright (C) 2022-2023 Rivos Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#ifndef HW_RISCV_IOMMU_H
> +#define HW_RISCV_IOMMU_H
> +
> +#include "qemu/osdep.h"
> +#include "qom/object.h"
> +
> +#define TYPE_RISCV_IOMMU "riscv-iommu"
> +OBJECT_DECLARE_SIMPLE_TYPE(RISCVIOMMUState, RISCV_IOMMU)
> +typedef struct RISCVIOMMUState RISCVIOMMUState;
> +
> +#define TYPE_RISCV_IOMMU_MEMORY_REGION "riscv-iommu-mr"
> +typedef struct RISCVIOMMUSpace RISCVIOMMUSpace;
> +
> +#define TYPE_RISCV_IOMMU_PCI "riscv-iommu-pci"
> +OBJECT_DECLARE_SIMPLE_TYPE(RISCVIOMMUStatePci, RISCV_IOMMU_PCI)
> +typedef struct RISCVIOMMUStatePci RISCVIOMMUStatePci;
> +
> +#endif
> diff --git a/meson.build b/meson.build
> index a9de71d450..8099d8271c 100644
> --- a/meson.build
> +++ b/meson.build
> @@ -3319,6 +3319,7 @@ if have_system
>       'hw/pci-host',
>       'hw/ppc',
>       'hw/rtc',
> +    'hw/riscv',
>       'hw/s390x',
>       'hw/scsi',
>       'hw/sd',

Hi Jason,

(CCing designware folks and Cedric)

On 6/11/24 1:15 PM, Jason Chien wrote:
> Hi Daniel,
> 
> On 2024/5/24 上午 01:39, Daniel Henrique Barboza wrote:
>> From: Tomasz Jeznach <tjeznach@rivosinc.com>
>>
>> The RISC-V IOMMU specification is now ratified as-per the RISC-V
>> international process. The latest frozen specifcation can be found
>> at:
>>
>> https://github.com/riscv-non-isa/riscv-iommu/releases/download/v1.0/riscv-iommu.pdf
>>
>> Add the foundation of the device emulation for RISC-V IOMMU, which
>> includes an IOMMU that has no capabilities but MSI interrupt support and
>> fault queue interfaces. We'll add add more features incrementally in the
>> next patches.
>>
>> Co-developed-by: Sebastien Boeuf <seb@rivosinc.com>
>> Signed-off-by: Sebastien Boeuf <seb@rivosinc.com>
>> Signed-off-by: Tomasz Jeznach <tjeznach@rivosinc.com>
>> Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
>> ---

[...]

>> +static const PCIIOMMUOps riscv_iommu_ops = {
>> +    .get_address_space = riscv_iommu_find_as,
>> +};
>> +
>> +void riscv_iommu_pci_setup_iommu(RISCVIOMMUState *iommu, PCIBus *bus,
>> +        Error **errp)
>> +{
>> +    if (bus->iommu_ops &&
>> +        bus->iommu_ops->get_address_space == riscv_iommu_find_as) {
>> +        /* Allow multiple IOMMUs on the same PCIe bus, link known devices */
>> +        RISCVIOMMUState *last = (RISCVIOMMUState *)bus->iommu_opaque;
>> +        QLIST_INSERT_AFTER(last, iommu, iommus);
>> +    } else if (!bus->iommu_ops && !bus->iommu_opaque) {
>> +        pci_setup_iommu(bus, &riscv_iommu_ops, iommu);
> 
> We use Designware PCIe with RISCV IOMMU internally and there is a problem that we would like to point out.
> 
> Both hw/riscv/riscv-iommu.c and hw/pci-host/designware.c utilize pci_setup_iommu(). When pci_setup_iommu() is invoked in hw/riscv/riscv-iommu.c, the iommu_ops set by Designware PCIe host is lost, which results in incorrect translation as the PCIe translation logic is overwritten and lost.
> 
> I think it may be a better choice to expose a memory region property in each PCIe host for the purpose of specifying the target memory region that the PCIe host should send requests to. By doing this, the Designware PCIe host can finish its translation and directs the request to the IOMMU memory region of the RISCV IOMMU.
> 
> The below code, based on riscv_iommu_v3, exposes a target memory region property in Designware PCIe and directs inbound requests to the target memory which can be specified to be the IOMMU memory region of RISCV IOMMU.
> 
> diff --git a/hw/pci-host/designware.c b/hw/pci-host/designware.c
> index c25d50f1c6..6b6d4ac1aa 100644
> --- a/hw/pci-host/designware.c
> +++ b/hw/pci-host/designware.c
> @@ -435,7 +435,7 @@ static void designware_pcie_root_realize(PCIDevice *dev, Error **errp)
>           viewport->cr[0]   = DESIGNWARE_PCIE_ATU_TYPE_MEM;
> 
>           source      = &host->pci.address_space_root;
> -        destination = get_system_memory();
> +        destination = host->target_mr;
>           direction   = "Inbound";
> 
>           /*
> @@ -713,6 +713,10 @@ static void designware_pcie_host_realize(DeviceState *dev, Error **errp)
>                          "pcie-bus-address-space");
>       pci_setup_iommu(pci->bus, &designware_iommu_ops, s);
> 
> +    if (!s->target_mr) {
> +        s->target_mr = get_system_memory();
> +    }
> +
>       qdev_realize(DEVICE(&s->root), BUS(pci->bus), &error_fatal);
>   }
> 
> @@ -730,6 +734,12 @@ static const VMStateDescription vmstate_designware_pcie_host = {
>       }
>   };
> 
> +static Property designware_pcie_host_properties[] = {
> +    DEFINE_PROP_LINK("target-mr", DesignwarePCIEHost, target_mr,
> +                     TYPE_MEMORY_REGION, MemoryRegion *),
> +    DEFINE_PROP_END_OF_LIST(),
> +};
> +
>   static void designware_pcie_host_class_init(ObjectClass *klass, void *data)
>   {
>       DeviceClass *dc = DEVICE_CLASS(klass);
> @@ -740,6 +750,7 @@ static void designware_pcie_host_class_init(ObjectClass *klass, void *data)
>       set_bit(DEVICE_CATEGORY_BRIDGE, dc->categories);
>       dc->fw_name = "pci";
>       dc->vmsd = &vmstate_designware_pcie_host;
> +    device_class_set_props(dc, designware_pcie_host_properties);
>   }
> 
>   static void designware_pcie_host_init(Object *obj)
> diff --git a/include/hw/pci-host/designware.h b/include/hw/pci-host/designware.h
> index 908f3d946b..2530eacbb0 100644
> --- a/include/hw/pci-host/designware.h
> +++ b/include/hw/pci-host/designware.h
> @@ -91,6 +91,7 @@ struct DesignwarePCIEHost {
>       } pci;
> 
>       MemoryRegion mmio;
> +    MemoryRegion *target_mr;
>   };
> 
>   #endif /* DESIGNWARE_H */
> 
> We also need to specify the requester id (BDF) in the memory attribute when sending requests to the IOMMU memory region in order to distinguish the source endpoint, since all endpoints under Designware PCIe host write to the same IOMMU memory region.
> 
> diff --git a/include/hw/pci/pci_device.h b/include/hw/pci/pci_device.h
> index d3dd0f64b2..1fc64a2d1f 100644
> --- a/include/hw/pci/pci_device.h
> +++ b/include/hw/pci/pci_device.h
> @@ -249,8 +249,11 @@ static inline MemTxResult pci_dma_rw(PCIDevice *dev, dma_addr_t addr,
>   static inline MemTxResult pci_dma_read(PCIDevice *dev, dma_addr_t addr,
>                                          void *buf, dma_addr_t len)
>   {
> +    MemTxAttrs attrs = {};
> +    attrs.requester_id = pci_requester_id(dev);
> +
>       return pci_dma_rw(dev, addr, buf, len,
> -                      DMA_DIRECTION_TO_DEVICE, MEMTXATTRS_UNSPECIFIED);
> +                      DMA_DIRECTION_TO_DEVICE, attrs);
>   }
> 
>   /**
> @@ -268,8 +271,11 @@ static inline MemTxResult pci_dma_read(PCIDevice *dev, dma_addr_t addr,
>   static inline MemTxResult pci_dma_write(PCIDevice *dev, dma_addr_t addr,
>                                           const void *buf, dma_addr_t len)
>   {
> +    MemTxAttrs attrs = {};
> +    attrs.requester_id = pci_requester_id(dev);
> +
>       return pci_dma_rw(dev, addr, (void *) buf, len,
> -                      DMA_DIRECTION_FROM_DEVICE, MEMTXATTRS_UNSPECIFIED);
> +                      DMA_DIRECTION_FROM_DEVICE, attrs);
>   }
> 
>   #define PCI_DMA_DEFINE_LDST(_l, _s, _bits) \
> @@ -313,8 +319,11 @@ PCI_DMA_DEFINE_LDST(q_be, q_be, 64);
>   static inline void *pci_dma_map(PCIDevice *dev, dma_addr_t addr,
>                                   dma_addr_t *plen, DMADirection dir)
>   {
> +    MemTxAttrs attrs = {};
> +    attrs.requester_id = pci_requester_id(dev);
> +
>       return dma_memory_map(pci_get_address_space(dev), addr, plen, dir,
> -                          MEMTXATTRS_UNSPECIFIED);
> +                          attrs);
>   }
> 
> We hope not to call pci_setup_iommu() in hw/riscv/riscv-iommu.c to avoid conflicts of iommu_ops. Do you have any suggestion on this issue?

As far as the riscv-iommu changes might go, I'm ok with adding more properties
in the device to customize whether it creates its own iommu_ops (like it's
done today) or whether we should skip it since another entity might provide
it.

I can't comment much on the Designware changes. It seems sensible to me but
I'm not acquainted with how the Designware pci-host works. I couldn't find
the docs for it either in a (lazy) search I just did.

I appreciate if someone more knowledgeable with the Designware device can
comment.


Thanks,

Daniel


> 
> Thanks.
> 
>> +    } else {
>> +        error_setg(errp, "can't register secondary IOMMU for PCI bus #%d",
>> +            pci_bus_num(bus));
>> +    }
>> +}
>> +
>> +static int riscv_iommu_memory_region_index(IOMMUMemoryRegion *iommu_mr,
>> +    MemTxAttrs attrs)
>> +{
>> +    return attrs.unspecified ? RISCV_IOMMU_NOPASID : (int)attrs.pasid;
>> +}
>> +
>> +static int riscv_iommu_memory_region_index_len(IOMMUMemoryRegion *iommu_mr)
>> +{
>> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
>> +    return 1 << as->iommu->pasid_bits;
>> +}
>> +
>> +static void riscv_iommu_memory_region_init(ObjectClass *klass, void *data)
>> +{
>> +    IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_CLASS(klass);
>> +
>> +    imrc->translate = riscv_iommu_memory_region_translate;
>> +    imrc->notify_flag_changed = riscv_iommu_memory_region_notify;
>> +    imrc->attrs_to_index = riscv_iommu_memory_region_index;
>> +    imrc->num_indexes = riscv_iommu_memory_region_index_len;
>> +}
>> +
>> +static const TypeInfo riscv_iommu_memory_region_info = {
>> +    .parent = TYPE_IOMMU_MEMORY_REGION,
>> +    .name = TYPE_RISCV_IOMMU_MEMORY_REGION,
>> +    .class_init = riscv_iommu_memory_region_init,
>> +};
>> +
>> +static void riscv_iommu_register_mr_types(void)
>> +{
>> +    type_register_static(&riscv_iommu_memory_region_info);
>> +    type_register_static(&riscv_iommu_info);
>> +}
>> +
>> +type_init(riscv_iommu_register_mr_types);
>> diff --git a/hw/riscv/riscv-iommu.h b/hw/riscv/riscv-iommu.h
>> new file mode 100644
>> index 0000000000..31d3907d33
>> --- /dev/null
>> +++ b/hw/riscv/riscv-iommu.h
>> @@ -0,0 +1,141 @@
>> +/*
>> + * QEMU emulation of an RISC-V IOMMU
>> + *
>> + * Copyright (C) 2022-2023 Rivos Inc.
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of the GNU General Public License as published by
>> + * the Free Software Foundation; either version 2 of the License.
>> + *
>> + * This program is distributed in the hope that it will be useful,
>> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
>> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>> + * GNU General Public License for more details.
>> + *
>> + * You should have received a copy of the GNU General Public License along
>> + * with this program; if not, see <http://www.gnu.org/licenses/>.
>> + */
>> +
>> +#ifndef HW_RISCV_IOMMU_STATE_H
>> +#define HW_RISCV_IOMMU_STATE_H
>> +
>> +#include "qemu/osdep.h"
>> +#include "qom/object.h"
>> +
>> +#include "hw/riscv/iommu.h"
>> +
>> +struct RISCVIOMMUState {
>> +    /*< private >*/
>> +    DeviceState parent_obj;
>> +
>> +    /*< public >*/
>> +    uint32_t version;     /* Reported interface version number */
>> +    uint32_t pasid_bits;  /* process identifier width */
>> +    uint32_t bus;         /* PCI bus mapping for non-root endpoints */
>> +
>> +    uint64_t cap;         /* IOMMU supported capabilities */
>> +    uint64_t fctl;        /* IOMMU enabled features */
>> +
>> +    bool enable_off;      /* Enable out-of-reset OFF mode (DMA disabled) */
>> +    bool enable_msi;      /* Enable MSI remapping */
>> +
>> +    /* IOMMU Internal State */
>> +    uint64_t ddtp;        /* Validated Device Directory Tree Root Pointer */
>> +
>> +    dma_addr_t cq_addr;   /* Command queue base physical address */
>> +    dma_addr_t fq_addr;   /* Fault/event queue base physical address */
>> +    dma_addr_t pq_addr;   /* Page request queue base physical address */
>> +
>> +    uint32_t cq_mask;     /* Command queue index bit mask */
>> +    uint32_t fq_mask;     /* Fault/event queue index bit mask */
>> +    uint32_t pq_mask;     /* Page request queue index bit mask */
>> +
>> +    /* interrupt notifier */
>> +    void (*notify)(RISCVIOMMUState *iommu, unsigned vector);
>> +
>> +    /* IOMMU State Machine */
>> +    QemuThread core_proc; /* Background processing thread */
>> +    QemuMutex core_lock;  /* Global IOMMU lock, used for cache/regs updates */
>> +    QemuCond core_cond;   /* Background processing wake up signal */
>> +    unsigned core_exec;   /* Processing thread execution actions */
>> +
>> +    /* IOMMU target address space */
>> +    AddressSpace *target_as;
>> +    MemoryRegion *target_mr;
>> +
>> +    /* MSI / MRIF access trap */
>> +    AddressSpace trap_as;
>> +    MemoryRegion trap_mr;
>> +
>> +    GHashTable *ctx_cache;          /* Device translation Context Cache */
>> +
>> +    /* MMIO Hardware Interface */
>> +    MemoryRegion regs_mr;
>> +    QemuSpin regs_lock;
>> +    uint8_t *regs_rw;  /* register state (user write) */
>> +    uint8_t *regs_wc;  /* write-1-to-clear mask */
>> +    uint8_t *regs_ro;  /* read-only mask */
>> +
>> +    QLIST_ENTRY(RISCVIOMMUState) iommus;
>> +    QLIST_HEAD(, RISCVIOMMUSpace) spaces;
>> +};
>> +
>> +void riscv_iommu_pci_setup_iommu(RISCVIOMMUState *iommu, PCIBus *bus,
>> +         Error **errp);
>> +
>> +/* private helpers */
>> +
>> +/* Register helper functions */
>> +static inline uint32_t riscv_iommu_reg_mod32(RISCVIOMMUState *s,
>> +    unsigned idx, uint32_t set, uint32_t clr)
>> +{
>> +    uint32_t val;
>> +    qemu_spin_lock(&s->regs_lock);
>> +    val = ldl_le_p(s->regs_rw + idx);
>> +    stl_le_p(s->regs_rw + idx, (val & ~clr) | set);
>> +    qemu_spin_unlock(&s->regs_lock);
>> +    return val;
>> +}
>> +
>> +static inline void riscv_iommu_reg_set32(RISCVIOMMUState *s,
>> +    unsigned idx, uint32_t set)
>> +{
>> +    qemu_spin_lock(&s->regs_lock);
>> +    stl_le_p(s->regs_rw + idx, set);
>> +    qemu_spin_unlock(&s->regs_lock);
>> +}
>> +
>> +static inline uint32_t riscv_iommu_reg_get32(RISCVIOMMUState *s,
>> +    unsigned idx)
>> +{
>> +    return ldl_le_p(s->regs_rw + idx);
>> +}
>> +
>> +static inline uint64_t riscv_iommu_reg_mod64(RISCVIOMMUState *s,
>> +    unsigned idx, uint64_t set, uint64_t clr)
>> +{
>> +    uint64_t val;
>> +    qemu_spin_lock(&s->regs_lock);
>> +    val = ldq_le_p(s->regs_rw + idx);
>> +    stq_le_p(s->regs_rw + idx, (val & ~clr) | set);
>> +    qemu_spin_unlock(&s->regs_lock);
>> +    return val;
>> +}
>> +
>> +static inline void riscv_iommu_reg_set64(RISCVIOMMUState *s,
>> +    unsigned idx, uint64_t set)
>> +{
>> +    qemu_spin_lock(&s->regs_lock);
>> +    stq_le_p(s->regs_rw + idx, set);
>> +    qemu_spin_unlock(&s->regs_lock);
>> +}
>> +
>> +static inline uint64_t riscv_iommu_reg_get64(RISCVIOMMUState *s,
>> +    unsigned idx)
>> +{
>> +    return ldq_le_p(s->regs_rw + idx);
>> +}
>> +
>> +
>> +
>> +#endif
>> diff --git a/hw/riscv/trace-events b/hw/riscv/trace-events
>> new file mode 100644
>> index 0000000000..42a97caffa
>> --- /dev/null
>> +++ b/hw/riscv/trace-events
>> @@ -0,0 +1,11 @@
>> +# See documentation at docs/devel/tracing.rst
>> +
>> +# riscv-iommu.c
>> +riscv_iommu_new(const char *id, unsigned b, unsigned d, unsigned f) "%s: device attached %04x:%02x.%d"
>> +riscv_iommu_flt(const char *id, unsigned b, unsigned d, unsigned f, uint64_t reason, uint64_t iova) "%s: fault %04x:%02x.%u reason: 0x%"PRIx64" iova: 0x%"PRIx64
>> +riscv_iommu_pri(const char *id, unsigned b, unsigned d, unsigned f, uint64_t iova) "%s: page request %04x:%02x.%u iova: 0x%"PRIx64
>> +riscv_iommu_dma(const char *id, unsigned b, unsigned d, unsigned f, unsigned pasid, const char *dir, uint64_t iova, uint64_t phys) "%s: translate %04x:%02x.%u #%u %s 0x%"PRIx64" -> 0x%"PRIx64
>> +riscv_iommu_msi(const char *id, unsigned b, unsigned d, unsigned f, uint64_t iova, uint64_t phys) "%s: translate %04x:%02x.%u MSI 0x%"PRIx64" -> 0x%"PRIx64
>> +riscv_iommu_cmd(const char *id, uint64_t l, uint64_t u) "%s: command 0x%"PRIx64" 0x%"PRIx64
>> +riscv_iommu_notifier_add(const char *id) "%s: dev-iotlb notifier added"
>> +riscv_iommu_notifier_del(const char *id) "%s: dev-iotlb notifier removed"
>> diff --git a/hw/riscv/trace.h b/hw/riscv/trace.h
>> new file mode 100644
>> index 0000000000..8c0e3ca1f3
>> --- /dev/null
>> +++ b/hw/riscv/trace.h
>> @@ -0,0 +1 @@
>> +#include "trace/trace-hw_riscv.h"
>> diff --git a/include/hw/riscv/iommu.h b/include/hw/riscv/iommu.h
>> new file mode 100644
>> index 0000000000..070ee69973
>> --- /dev/null
>> +++ b/include/hw/riscv/iommu.h
>> @@ -0,0 +1,36 @@
>> +/*
>> + * QEMU emulation of an RISC-V IOMMU
>> + *
>> + * Copyright (C) 2022-2023 Rivos Inc.
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of the GNU General Public License as published by
>> + * the Free Software Foundation; either version 2 of the License.
>> + *
>> + * This program is distributed in the hope that it will be useful,
>> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
>> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>> + * GNU General Public License for more details.
>> + *
>> + * You should have received a copy of the GNU General Public License along
>> + * with this program; if not, see <http://www.gnu.org/licenses/>.
>> + */
>> +
>> +#ifndef HW_RISCV_IOMMU_H
>> +#define HW_RISCV_IOMMU_H
>> +
>> +#include "qemu/osdep.h"
>> +#include "qom/object.h"
>> +
>> +#define TYPE_RISCV_IOMMU "riscv-iommu"
>> +OBJECT_DECLARE_SIMPLE_TYPE(RISCVIOMMUState, RISCV_IOMMU)
>> +typedef struct RISCVIOMMUState RISCVIOMMUState;
>> +
>> +#define TYPE_RISCV_IOMMU_MEMORY_REGION "riscv-iommu-mr"
>> +typedef struct RISCVIOMMUSpace RISCVIOMMUSpace;
>> +
>> +#define TYPE_RISCV_IOMMU_PCI "riscv-iommu-pci"
>> +OBJECT_DECLARE_SIMPLE_TYPE(RISCVIOMMUStatePci, RISCV_IOMMU_PCI)
>> +typedef struct RISCVIOMMUStatePci RISCVIOMMUStatePci;
>> +
>> +#endif
>> diff --git a/meson.build b/meson.build
>> index a9de71d450..8099d8271c 100644
>> --- a/meson.build
>> +++ b/meson.build
>> @@ -3319,6 +3319,7 @@ if have_system
>>       'hw/pci-host',
>>       'hw/ppc',
>>       'hw/rtc',
>> +    'hw/riscv',
>>       'hw/s390x',
>>       'hw/scsi',
>>       'hw/sd',

Hi Daniel,

On 2024/5/24 上午 01:39, Daniel Henrique Barboza wrote:
> From: Tomasz Jeznach <tjeznach@rivosinc.com>
>
> The RISC-V IOMMU specification is now ratified as-per the RISC-V
> international process. The latest frozen specifcation can be found
> at:
>
> https://github.com/riscv-non-isa/riscv-iommu/releases/download/v1.0/riscv-iommu.pdf
>
> Add the foundation of the device emulation for RISC-V IOMMU, which
> includes an IOMMU that has no capabilities but MSI interrupt support and
> fault queue interfaces. We'll add add more features incrementally in the
> next patches.
>
> Co-developed-by: Sebastien Boeuf <seb@rivosinc.com>
> Signed-off-by: Sebastien Boeuf <seb@rivosinc.com>
> Signed-off-by: Tomasz Jeznach <tjeznach@rivosinc.com>
> Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
> ---
>   hw/riscv/Kconfig         |    4 +
>   hw/riscv/meson.build     |    1 +
>   hw/riscv/riscv-iommu.c   | 1602 ++++++++++++++++++++++++++++++++++++++
>   hw/riscv/riscv-iommu.h   |  141 ++++
>   hw/riscv/trace-events    |   11 +
>   hw/riscv/trace.h         |    1 +
>   include/hw/riscv/iommu.h |   36 +
>   meson.build              |    1 +
>   8 files changed, 1797 insertions(+)
>   create mode 100644 hw/riscv/riscv-iommu.c
>   create mode 100644 hw/riscv/riscv-iommu.h
>   create mode 100644 hw/riscv/trace-events
>   create mode 100644 hw/riscv/trace.h
>   create mode 100644 include/hw/riscv/iommu.h
>
> diff --git a/hw/riscv/Kconfig b/hw/riscv/Kconfig
> index a2030e3a6f..f69d6e3c8e 100644
> --- a/hw/riscv/Kconfig
> +++ b/hw/riscv/Kconfig
> @@ -1,3 +1,6 @@
> +config RISCV_IOMMU
> +    bool
> +
>   config RISCV_NUMA
>       bool
>   
> @@ -47,6 +50,7 @@ config RISCV_VIRT
>       select SERIAL
>       select RISCV_ACLINT
>       select RISCV_APLIC
> +    select RISCV_IOMMU
>       select RISCV_IMSIC
>       select SIFIVE_PLIC
>       select SIFIVE_TEST
> diff --git a/hw/riscv/meson.build b/hw/riscv/meson.build
> index f872674093..cbc99c6e8e 100644
> --- a/hw/riscv/meson.build
> +++ b/hw/riscv/meson.build
> @@ -10,5 +10,6 @@ riscv_ss.add(when: 'CONFIG_SIFIVE_U', if_true: files('sifive_u.c'))
>   riscv_ss.add(when: 'CONFIG_SPIKE', if_true: files('spike.c'))
>   riscv_ss.add(when: 'CONFIG_MICROCHIP_PFSOC', if_true: files('microchip_pfsoc.c'))
>   riscv_ss.add(when: 'CONFIG_ACPI', if_true: files('virt-acpi-build.c'))
> +riscv_ss.add(when: 'CONFIG_RISCV_IOMMU', if_true: files('riscv-iommu.c'))
>   
>   hw_arch += {'riscv': riscv_ss}
> diff --git a/hw/riscv/riscv-iommu.c b/hw/riscv/riscv-iommu.c
> new file mode 100644
> index 0000000000..39b4ff1405
> --- /dev/null
> +++ b/hw/riscv/riscv-iommu.c
> @@ -0,0 +1,1602 @@
> +/*
> + * QEMU emulation of an RISC-V IOMMU
> + *
> + * Copyright (C) 2021-2023, Rivos Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#include "qemu/osdep.h"
> +#include "qom/object.h"
> +#include "hw/pci/pci_bus.h"
> +#include "hw/pci/pci_device.h"
> +#include "hw/qdev-properties.h"
> +#include "hw/riscv/riscv_hart.h"
> +#include "migration/vmstate.h"
> +#include "qapi/error.h"
> +#include "qemu/timer.h"
> +
> +#include "cpu_bits.h"
> +#include "riscv-iommu.h"
> +#include "riscv-iommu-bits.h"
> +#include "trace.h"
> +
> +#define LIMIT_CACHE_CTX               (1U << 7)
> +#define LIMIT_CACHE_IOT               (1U << 20)
> +
> +/* Physical page number coversions */
> +#define PPN_PHYS(ppn)                 ((ppn) << TARGET_PAGE_BITS)
> +#define PPN_DOWN(phy)                 ((phy) >> TARGET_PAGE_BITS)
> +
> +typedef struct RISCVIOMMUContext RISCVIOMMUContext;
> +typedef struct RISCVIOMMUEntry RISCVIOMMUEntry;
> +
> +/* Device assigned I/O address space */
> +struct RISCVIOMMUSpace {
> +    IOMMUMemoryRegion iova_mr;  /* IOVA memory region for attached device */
> +    AddressSpace iova_as;       /* IOVA address space for attached device */
> +    RISCVIOMMUState *iommu;     /* Managing IOMMU device state */
> +    uint32_t devid;             /* Requester identifier, AKA device_id */
> +    bool notifier;              /* IOMMU unmap notifier enabled */
> +    QLIST_ENTRY(RISCVIOMMUSpace) list;
> +};
> +
> +/* Device translation context state. */
> +struct RISCVIOMMUContext {
> +    uint64_t devid:24;          /* Requester Id, AKA device_id */
> +    uint64_t pasid:20;          /* Process Address Space ID */
> +    uint64_t __rfu:20;          /* reserved */
> +    uint64_t tc;                /* Translation Control */
> +    uint64_t ta;                /* Translation Attributes */
> +    uint64_t msi_addr_mask;     /* MSI filtering - address mask */
> +    uint64_t msi_addr_pattern;  /* MSI filtering - address pattern */
> +    uint64_t msiptp;            /* MSI redirection page table pointer */
> +};
> +
> +/* IOMMU index for transactions without PASID specified. */
> +#define RISCV_IOMMU_NOPASID 0
> +
> +static void riscv_iommu_notify(RISCVIOMMUState *s, int vec)
> +{
> +    const uint32_t fctl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FCTL);
> +    uint32_t ipsr, ivec;
> +
> +    if (fctl & RISCV_IOMMU_FCTL_WSI || !s->notify) {
> +        return;
> +    }
> +
> +    ipsr = riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_IPSR, (1 << vec), 0);
> +    ivec = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_IVEC);
> +
> +    if (!(ipsr & (1 << vec))) {
> +        s->notify(s, (ivec >> (vec * 4)) & 0x0F);
> +    }
> +}
> +
> +static void riscv_iommu_fault(RISCVIOMMUState *s,
> +                              struct riscv_iommu_fq_record *ev)
> +{
> +    uint32_t ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
> +    uint32_t head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQH) & s->fq_mask;
> +    uint32_t tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQT) & s->fq_mask;
> +    uint32_t next = (tail + 1) & s->fq_mask;
> +    uint32_t devid = get_field(ev->hdr, RISCV_IOMMU_FQ_HDR_DID);
> +
> +    trace_riscv_iommu_flt(s->parent_obj.id, PCI_BUS_NUM(devid), PCI_SLOT(devid),
> +                          PCI_FUNC(devid), ev->hdr, ev->iotval);
> +
> +    if (!(ctrl & RISCV_IOMMU_FQCSR_FQON) ||
> +        !!(ctrl & (RISCV_IOMMU_FQCSR_FQOF | RISCV_IOMMU_FQCSR_FQMF))) {
> +        return;
> +    }
> +
> +    if (head == next) {
> +        riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR,
> +                              RISCV_IOMMU_FQCSR_FQOF, 0);
> +    } else {
> +        dma_addr_t addr = s->fq_addr + tail * sizeof(*ev);
> +        if (dma_memory_write(s->target_as, addr, ev, sizeof(*ev),
> +                             MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR,
> +                                  RISCV_IOMMU_FQCSR_FQMF, 0);
> +        } else {
> +            riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_FQT, next);
> +        }
> +    }
> +
> +    if (ctrl & RISCV_IOMMU_FQCSR_FIE) {
> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_FQ);
> +    }
> +}
> +
> +static void riscv_iommu_pri(RISCVIOMMUState *s,
> +    struct riscv_iommu_pq_record *pr)
> +{
> +    uint32_t ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
> +    uint32_t head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQH) & s->pq_mask;
> +    uint32_t tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQT) & s->pq_mask;
> +    uint32_t next = (tail + 1) & s->pq_mask;
> +    uint32_t devid = get_field(pr->hdr, RISCV_IOMMU_PREQ_HDR_DID);
> +
> +    trace_riscv_iommu_pri(s->parent_obj.id, PCI_BUS_NUM(devid), PCI_SLOT(devid),
> +                          PCI_FUNC(devid), pr->payload);
> +
> +    if (!(ctrl & RISCV_IOMMU_PQCSR_PQON) ||
> +        !!(ctrl & (RISCV_IOMMU_PQCSR_PQOF | RISCV_IOMMU_PQCSR_PQMF))) {
> +        return;
> +    }
> +
> +    if (head == next) {
> +        riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR,
> +                              RISCV_IOMMU_PQCSR_PQOF, 0);
> +    } else {
> +        dma_addr_t addr = s->pq_addr + tail * sizeof(*pr);
> +        if (dma_memory_write(s->target_as, addr, pr, sizeof(*pr),
> +                             MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR,
> +                                  RISCV_IOMMU_PQCSR_PQMF, 0);
> +        } else {
> +            riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_PQT, next);
> +        }
> +    }
> +
> +    if (ctrl & RISCV_IOMMU_PQCSR_PIE) {
> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_PQ);
> +    }
> +}
> +
> +/* Portable implementation of pext_u64, bit-mask extraction. */
> +static uint64_t _pext_u64(uint64_t val, uint64_t ext)
> +{
> +    uint64_t ret = 0;
> +    uint64_t rot = 1;
> +
> +    while (ext) {
> +        if (ext & 1) {
> +            if (val & 1) {
> +                ret |= rot;
> +            }
> +            rot <<= 1;
> +        }
> +        val >>= 1;
> +        ext >>= 1;
> +    }
> +
> +    return ret;
> +}
> +
> +/* Check if GPA matches MSI/MRIF pattern. */
> +static bool riscv_iommu_msi_check(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
> +    dma_addr_t gpa)
> +{
> +    if (get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_MODE) !=
> +        RISCV_IOMMU_DC_MSIPTP_MODE_FLAT) {
> +        return false; /* Invalid MSI/MRIF mode */
> +    }
> +
> +    if ((PPN_DOWN(gpa) ^ ctx->msi_addr_pattern) & ~ctx->msi_addr_mask) {
> +        return false; /* GPA not in MSI range defined by AIA IMSIC rules. */
> +    }
> +
> +    return true;
> +}
> +
> +/* RISCV IOMMU Address Translation Lookup - Page Table Walk */
> +static int riscv_iommu_spa_fetch(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
> +    IOMMUTLBEntry *iotlb)
> +{
> +    /* Early check for MSI address match when IOVA == GPA */
> +    if (iotlb->perm & IOMMU_WO &&
> +        riscv_iommu_msi_check(s, ctx, iotlb->iova)) {
> +        iotlb->target_as = &s->trap_as;
> +        iotlb->translated_addr = iotlb->iova;
> +        iotlb->addr_mask = ~TARGET_PAGE_MASK;
> +        return 0;
> +    }
> +
> +    /* Exit early for pass-through mode. */
> +    iotlb->translated_addr = iotlb->iova;
> +    iotlb->addr_mask = ~TARGET_PAGE_MASK;
> +    /* Allow R/W in pass-through mode */
> +    iotlb->perm = IOMMU_RW;
> +    return 0;
> +}
> +
> +static void riscv_iommu_report_fault(RISCVIOMMUState *s,
> +                                     RISCVIOMMUContext *ctx,
> +                                     uint32_t fault_type, uint32_t cause,
> +                                     bool pv,
> +                                     uint64_t iotval, uint64_t iotval2)
> +{
> +    struct riscv_iommu_fq_record ev = { 0 };
> +
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_DTF) {
> +        switch (cause) {
> +        case RISCV_IOMMU_FQ_CAUSE_DMA_DISABLED:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_INVALID:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED:
> +        case RISCV_IOMMU_FQ_CAUSE_DDT_CORRUPTED:
> +        case RISCV_IOMMU_FQ_CAUSE_INTERNAL_DP_ERROR:
> +        case RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT:
> +            break;
> +        default:
> +            /* DTF prevents reporting a fault for this given cause */
> +            return;
> +        }
> +    }
> +
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_CAUSE, cause);
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_TTYPE, fault_type);
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_DID, ctx->devid);
> +    ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_PV, true);
> +
> +    if (pv) {
> +        ev.hdr = set_field(ev.hdr, RISCV_IOMMU_FQ_HDR_PID, ctx->pasid);
> +    }
> +
> +    ev.iotval = iotval;
> +    ev.iotval2 = iotval2;
> +
> +    riscv_iommu_fault(s, &ev);
> +}
> +
> +/* Redirect MSI write for given GPA. */
> +static MemTxResult riscv_iommu_msi_write(RISCVIOMMUState *s,
> +    RISCVIOMMUContext *ctx, uint64_t gpa, uint64_t data,
> +    unsigned size, MemTxAttrs attrs)
> +{
> +    MemTxResult res;
> +    dma_addr_t addr;
> +    uint64_t intn;
> +    uint32_t n190;
> +    uint64_t pte[2];
> +    int fault_type = RISCV_IOMMU_FQ_TTYPE_UADDR_WR;
> +    int cause;
> +
> +    if (!riscv_iommu_msi_check(s, ctx, gpa)) {
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    /* Interrupt File Number */
> +    intn = _pext_u64(PPN_DOWN(gpa), ctx->msi_addr_mask);
> +    if (intn >= 256) {
> +        /* Interrupt file number out of range */
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    /* fetch MSI PTE */
> +    addr = PPN_PHYS(get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_PPN));
> +    addr = addr | (intn * sizeof(pte));
> +    res = dma_memory_read(s->target_as, addr, &pte, sizeof(pte),
> +            MEMTXATTRS_UNSPECIFIED);
> +    if (res != MEMTX_OK) {
> +        if (res == MEMTX_DECODE_ERROR) {
> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_PT_CORRUPTED;
> +        } else {
> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        }
> +        goto err;
> +    }
> +
> +    le64_to_cpus(&pte[0]);
> +    le64_to_cpus(&pte[1]);
> +
> +    if (!(pte[0] & RISCV_IOMMU_MSI_PTE_V) || (pte[0] & RISCV_IOMMU_MSI_PTE_C)) {
> +        /*
> +         * The spec mentions that: "If msipte.C == 1, then further
> +         * processing to interpret the PTE is implementation
> +         * defined.". We'll abort with cause = 262 for this
> +         * case too.
> +         */
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_INVALID;
> +        goto err;
> +    }
> +
> +    switch (get_field(pte[0], RISCV_IOMMU_MSI_PTE_M)) {
> +    case RISCV_IOMMU_MSI_PTE_M_BASIC:
> +        /* MSI Pass-through mode */
> +        addr = PPN_PHYS(get_field(pte[0], RISCV_IOMMU_MSI_PTE_PPN));
> +        addr = addr | (gpa & TARGET_PAGE_MASK);
> +
> +        trace_riscv_iommu_msi(s->parent_obj.id, PCI_BUS_NUM(ctx->devid),
> +                              PCI_SLOT(ctx->devid), PCI_FUNC(ctx->devid),
> +                              gpa, addr);
> +
> +        res = dma_memory_write(s->target_as, addr, &data, size, attrs);
> +        if (res != MEMTX_OK) {
> +            cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
> +            goto err;
> +        }
> +
> +        return MEMTX_OK;
> +    case RISCV_IOMMU_MSI_PTE_M_MRIF:
> +        /* MRIF mode, continue. */
> +        break;
> +    default:
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_MISCONFIGURED;
> +        goto err;
> +    }
> +
> +    /*
> +     * Report an error for interrupt identities exceeding the maximum allowed
> +     * for an IMSIC interrupt file (2047) or destination address is not 32-bit
> +     * aligned. See IOMMU Specification, Chapter 2.3. MSI page tables.
> +     */
> +    if ((data > 2047) || (gpa & 3)) {
> +        res = MEMTX_ACCESS_ERROR;
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_MISCONFIGURED;
> +        goto err;
> +    }
> +
> +    /* MSI MRIF mode, non atomic pending bit update */
> +
> +    /* MRIF pending bit address */
> +    addr = get_field(pte[0], RISCV_IOMMU_MSI_PTE_MRIF_ADDR) << 9;
> +    addr = addr | ((data & 0x7c0) >> 3);
> +
> +    trace_riscv_iommu_msi(s->parent_obj.id, PCI_BUS_NUM(ctx->devid),
> +                          PCI_SLOT(ctx->devid), PCI_FUNC(ctx->devid),
> +                          gpa, addr);
> +
> +    /* MRIF pending bit mask */
> +    data = 1ULL << (data & 0x03f);
> +    res = dma_memory_read(s->target_as, addr, &intn, sizeof(intn), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    intn = intn | data;
> +    res = dma_memory_write(s->target_as, addr, &intn, sizeof(intn), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
> +        goto err;
> +    }
> +
> +    /* Get MRIF enable bits */
> +    addr = addr + sizeof(intn);
> +    res = dma_memory_read(s->target_as, addr, &intn, sizeof(intn), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_LOAD_FAULT;
> +        goto err;
> +    }
> +
> +    if (!(intn & data)) {
> +        /* notification disabled, MRIF update completed. */
> +        return MEMTX_OK;
> +    }
> +
> +    /* Send notification message */
> +    addr = PPN_PHYS(get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NPPN));
> +    n190 = get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NID) |
> +          (get_field(pte[1], RISCV_IOMMU_MSI_MRIF_NID_MSB) << 10);
> +
> +    res = dma_memory_write(s->target_as, addr, &n190, sizeof(n190), attrs);
> +    if (res != MEMTX_OK) {
> +        cause = RISCV_IOMMU_FQ_CAUSE_MSI_WR_FAULT;
> +        goto err;
> +    }
> +
> +    return MEMTX_OK;
> +
> +err:
> +    riscv_iommu_report_fault(s, ctx, fault_type, cause,
> +                             !!ctx->pasid, 0, 0);
> +    return res;
> +}
> +
> +/*
> + * Check device context configuration as described by the
> + * riscv-iommu spec section "Device-context configuration
> + * checks".
> + */
> +static bool riscv_iommu_validate_device_ctx(RISCVIOMMUState *s,
> +                                            RISCVIOMMUContext *ctx)
> +{
> +    uint32_t msi_mode;
> +
> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_EN_PRI) &&
> +        ctx->tc & RISCV_IOMMU_DC_TC_PRPR) {
> +        return false;
> +    }
> +
> +    if (!(s->cap & RISCV_IOMMU_CAP_T2GPA) &&
> +        ctx->tc & RISCV_IOMMU_DC_TC_T2GPA) {
> +        return false;
> +    }
> +
> +    if (s->cap & RISCV_IOMMU_CAP_MSI_FLAT) {
> +        msi_mode = get_field(ctx->msiptp, RISCV_IOMMU_DC_MSIPTP_MODE);
> +
> +        if (msi_mode != RISCV_IOMMU_DC_MSIPTP_MODE_OFF &&
> +            msi_mode != RISCV_IOMMU_DC_MSIPTP_MODE_FLAT) {
> +            return false;
> +        }
> +    }
> +
> +    /*
> +     * CAP_END is always zero (only one endianess). FCTL_BE is
> +     * always zero (little-endian accesses). Thus TC_SBE must
> +     * always be LE, i.e. zero.
> +     */
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_SBE) {
> +        return false;
> +    }
> +
> +    return true;
> +}
> +
> +/*
> + * RISC-V IOMMU Device Context Loopkup - Device Directory Tree Walk
> + *
> + * @s         : IOMMU Device State
> + * @ctx       : Device Translation Context with devid and pasid set.
> + * @return    : success or fault code.
> + */
> +static int riscv_iommu_ctx_fetch(RISCVIOMMUState *s, RISCVIOMMUContext *ctx)
> +{
> +    const uint64_t ddtp = s->ddtp;
> +    unsigned mode = get_field(ddtp, RISCV_IOMMU_DDTP_MODE);
> +    dma_addr_t addr = PPN_PHYS(get_field(ddtp, RISCV_IOMMU_DDTP_PPN));
> +    struct riscv_iommu_dc dc;
> +    /* Device Context format: 0: extended (64 bytes) | 1: base (32 bytes) */
> +    const int dc_fmt = !s->enable_msi;
> +    const size_t dc_len = sizeof(dc) >> dc_fmt;
> +    unsigned depth;
> +    uint64_t de;
> +
> +    switch (mode) {
> +    case RISCV_IOMMU_DDTP_MODE_OFF:
> +        return RISCV_IOMMU_FQ_CAUSE_DMA_DISABLED;
> +
> +    case RISCV_IOMMU_DDTP_MODE_BARE:
> +        /* mock up pass-through translation context */
> +        ctx->tc = RISCV_IOMMU_DC_TC_V;
> +        ctx->ta = 0;
> +        ctx->msiptp = 0;
> +        return 0;
> +
> +    case RISCV_IOMMU_DDTP_MODE_1LVL:
> +        depth = 0;
> +        break;
> +
> +    case RISCV_IOMMU_DDTP_MODE_2LVL:
> +        depth = 1;
> +        break;
> +
> +    case RISCV_IOMMU_DDTP_MODE_3LVL:
> +        depth = 2;
> +        break;
> +
> +    default:
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
> +    }
> +
> +    /*
> +     * Check supported device id width (in bits).
> +     * See IOMMU Specification, Chapter 6. Software guidelines.
> +     * - if extended device-context format is used:
> +     *   1LVL: 6, 2LVL: 15, 3LVL: 24
> +     * - if base device-context format is used:
> +     *   1LVL: 7, 2LVL: 16, 3LVL: 24
> +     */
> +    if (ctx->devid >= (1 << (depth * 9 + 6 + (dc_fmt && depth != 2)))) {
> +        return RISCV_IOMMU_FQ_CAUSE_TTYPE_BLOCKED;
> +    }
> +
> +    /* Device directory tree walk */
> +    for (; depth-- > 0; ) {
> +        /*
> +         * Select device id index bits based on device directory tree level
> +         * and device context format.
> +         * See IOMMU Specification, Chapter 2. Data Structures.
> +         * - if extended device-context format is used:
> +         *   device index: [23:15][14:6][5:0]
> +         * - if base device-context format is used:
> +         *   device index: [23:16][15:7][6:0]
> +         */
> +        const int split = depth * 9 + 6 + dc_fmt;
> +        addr |= ((ctx->devid >> split) << 3) & ~TARGET_PAGE_MASK;
> +        if (dma_memory_read(s->target_as, addr, &de, sizeof(de),
> +                            MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            return RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT;
> +        }
> +        le64_to_cpus(&de);
> +        if (!(de & RISCV_IOMMU_DDTE_VALID)) {
> +            /* invalid directory entry */
> +            return RISCV_IOMMU_FQ_CAUSE_DDT_INVALID;
> +        }
> +        if (de & ~(RISCV_IOMMU_DDTE_PPN | RISCV_IOMMU_DDTE_VALID)) {
> +            /* reserved bits set */
> +            return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
> +        }
> +        addr = PPN_PHYS(get_field(de, RISCV_IOMMU_DDTE_PPN));
> +    }
> +
> +    /* index into device context entry page */
> +    addr |= (ctx->devid * dc_len) & ~TARGET_PAGE_MASK;
> +
> +    memset(&dc, 0, sizeof(dc));
> +    if (dma_memory_read(s->target_as, addr, &dc, dc_len,
> +                        MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_LOAD_FAULT;
> +    }
> +
> +    /* Set translation context. */
> +    ctx->tc = le64_to_cpu(dc.tc);
> +    ctx->ta = le64_to_cpu(dc.ta);
> +    ctx->msiptp = le64_to_cpu(dc.msiptp);
> +    ctx->msi_addr_mask = le64_to_cpu(dc.msi_addr_mask);
> +    ctx->msi_addr_pattern = le64_to_cpu(dc.msi_addr_pattern);
> +
> +    if (!riscv_iommu_validate_device_ctx(s, ctx)) {
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_MISCONFIGURED;
> +    }
> +
> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_V)) {
> +        return RISCV_IOMMU_FQ_CAUSE_DDT_INVALID;
> +    }
According to section 2.3.1(item 9 and item 10), DC.tc.V == 0 should be 
checked before riscv_iommu_validate_device_ctx() is checked.
> +
> +    if (!(ctx->tc & RISCV_IOMMU_DC_TC_PDTV)) {
> +        if (ctx->pasid != RISCV_IOMMU_NOPASID) {
> +            /* PASID is disabled */
> +            return RISCV_IOMMU_FQ_CAUSE_TTYPE_BLOCKED;
> +        }
> +        return 0;
> +    }
> +
> +    /* FSC.TC.PDTV enabled */
> +    if (mode > RISCV_IOMMU_DC_FSC_PDTP_MODE_PD20) {
> +        /* Invalid PDTP.MODE */
> +        return RISCV_IOMMU_FQ_CAUSE_PDT_MISCONFIGURED;
> +    }
> +
> +    for (depth = mode - RISCV_IOMMU_DC_FSC_PDTP_MODE_PD8; depth-- > 0; ) {
> +        /*
> +         * Select process id index bits based on process directory tree
> +         * level. See IOMMU Specification, 2.2. Process-Directory-Table.
> +         */
> +        const int split = depth * 9 + 8;
> +        addr |= ((ctx->pasid >> split) << 3) & ~TARGET_PAGE_MASK;
> +        if (dma_memory_read(s->target_as, addr, &de, sizeof(de),
> +                            MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +            return RISCV_IOMMU_FQ_CAUSE_PDT_LOAD_FAULT;
> +        }
> +        le64_to_cpus(&de);
> +        if (!(de & RISCV_IOMMU_PC_TA_V)) {
> +            return RISCV_IOMMU_FQ_CAUSE_PDT_INVALID;
> +        }
> +        addr = PPN_PHYS(get_field(de, RISCV_IOMMU_PC_FSC_PPN));
> +    }
> +
> +    /* Leaf entry in PDT */
> +    addr |= (ctx->pasid << 4) & ~TARGET_PAGE_MASK;
> +    if (dma_memory_read(s->target_as, addr, &dc.ta, sizeof(uint64_t) * 2,
> +                        MEMTXATTRS_UNSPECIFIED) != MEMTX_OK) {
> +        return RISCV_IOMMU_FQ_CAUSE_PDT_LOAD_FAULT;
> +    }
> +
> +    /* Use FSC and TA from process directory entry. */
> +    ctx->ta = le64_to_cpu(dc.ta);
> +

According to section 2.3.2:

10. If PC.ta.V == 0, stop and report "PDT entry not valid" (cause = 266).
11. If the PC is misconfigured as determined by rules outlined in 
Section 2.2.4 then stop and report "PDT entry misconfigured" (cause = 267).

> +    return 0;
> +}
> +
> +/* Translation Context cache support */
> +static gboolean __ctx_equal(gconstpointer v1, gconstpointer v2)
> +{
> +    RISCVIOMMUContext *c1 = (RISCVIOMMUContext *) v1;
> +    RISCVIOMMUContext *c2 = (RISCVIOMMUContext *) v2;
> +    return c1->devid == c2->devid && c1->pasid == c2->pasid;
> +}
> +
> +static guint __ctx_hash(gconstpointer v)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) v;
> +    /* Generate simple hash of (pasid, devid), assuming 24-bit wide devid */
> +    return (guint)(ctx->devid) + ((guint)(ctx->pasid) << 24);
> +}
> +
> +static void __ctx_inval_devid_pasid(gpointer key, gpointer value, gpointer data)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
> +    RISCVIOMMUContext *arg = (RISCVIOMMUContext *) data;
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V &&
> +        ctx->devid == arg->devid &&
> +        ctx->pasid == arg->pasid) {
> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
> +    }
> +}
> +
> +static void __ctx_inval_devid(gpointer key, gpointer value, gpointer data)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
> +    RISCVIOMMUContext *arg = (RISCVIOMMUContext *) data;
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V &&
> +        ctx->devid == arg->devid) {
> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
> +    }
> +}
> +
> +static void __ctx_inval_all(gpointer key, gpointer value, gpointer data)
> +{
> +    RISCVIOMMUContext *ctx = (RISCVIOMMUContext *) value;
> +    if (ctx->tc & RISCV_IOMMU_DC_TC_V) {
> +        ctx->tc &= ~RISCV_IOMMU_DC_TC_V;
> +    }
> +}
> +
> +static void riscv_iommu_ctx_inval(RISCVIOMMUState *s, GHFunc func,
> +    uint32_t devid, uint32_t pasid)
> +{
> +    GHashTable *ctx_cache;
> +    RISCVIOMMUContext key = {
> +        .devid = devid,
> +        .pasid = pasid,
> +    };
> +    ctx_cache = g_hash_table_ref(s->ctx_cache);
> +    g_hash_table_foreach(ctx_cache, func, &key);
> +    g_hash_table_unref(ctx_cache);
> +}
> +
> +/* Find or allocate translation context for a given {device_id, process_id} */
> +static RISCVIOMMUContext *riscv_iommu_ctx(RISCVIOMMUState *s,
> +    unsigned devid, unsigned pasid, void **ref)
> +{
> +    GHashTable *ctx_cache;
> +    RISCVIOMMUContext *ctx;
> +    RISCVIOMMUContext key = {
> +        .devid = devid,
> +        .pasid = pasid,
> +    };
> +
> +    ctx_cache = g_hash_table_ref(s->ctx_cache);
> +    ctx = g_hash_table_lookup(ctx_cache, &key);
> +
> +    if (ctx && (ctx->tc & RISCV_IOMMU_DC_TC_V)) {
> +        *ref = ctx_cache;
> +        return ctx;
Is it possible that ddtp.iommu_mode is set to off or bare which requires 
no translation? It looks like that the returned ctx will be used to 
perform translation in riscv_iommu_translate().
> +    }
> +
> +    if (g_hash_table_size(s->ctx_cache) >= LIMIT_CACHE_CTX) {
> +        ctx_cache = g_hash_table_new_full(__ctx_hash, __ctx_equal,
> +                                          g_free, NULL);
> +        g_hash_table_unref(qatomic_xchg(&s->ctx_cache, ctx_cache));
> +    }
> +
> +    ctx = g_new0(RISCVIOMMUContext, 1);
> +    ctx->devid = devid;
> +    ctx->pasid = pasid;
> +
> +    int fault = riscv_iommu_ctx_fetch(s, ctx);
> +    if (!fault) {
> +        g_hash_table_add(ctx_cache, ctx);
> +        *ref = ctx_cache;
> +        return ctx;
> +    }
> +
> +    g_hash_table_unref(ctx_cache);
> +    *ref = NULL;
> +
> +    riscv_iommu_report_fault(s, ctx, RISCV_IOMMU_FQ_TTYPE_UADDR_RD,
> +                             fault, !!pasid, 0, 0);
> +
> +    g_free(ctx);
> +    return NULL;
> +}
> +
> +static void riscv_iommu_ctx_put(RISCVIOMMUState *s, void *ref)
> +{
> +    if (ref) {
> +        g_hash_table_unref((GHashTable *)ref);
> +    }
> +}
> +
> +/* Find or allocate address space for a given device */
> +static AddressSpace *riscv_iommu_space(RISCVIOMMUState *s, uint32_t devid)
> +{
> +    RISCVIOMMUSpace *as;
> +
> +    /* FIXME: PCIe bus remapping for attached endpoints. */
> +    devid |= s->bus << 8;
> +
> +    qemu_mutex_lock(&s->core_lock);
> +    QLIST_FOREACH(as, &s->spaces, list) {
> +        if (as->devid == devid) {
> +            break;
> +        }
> +    }
> +    qemu_mutex_unlock(&s->core_lock);
> +
> +    if (as == NULL) {
> +        char name[64];
> +        as = g_new0(RISCVIOMMUSpace, 1);
> +
> +        as->iommu = s;
> +        as->devid = devid;
> +
> +        snprintf(name, sizeof(name), "riscv-iommu-%04x:%02x.%d-iova",
> +            PCI_BUS_NUM(as->devid), PCI_SLOT(as->devid), PCI_FUNC(as->devid));
> +
> +        /* IOVA address space, untranslated addresses */
> +        memory_region_init_iommu(&as->iova_mr, sizeof(as->iova_mr),
> +            TYPE_RISCV_IOMMU_MEMORY_REGION,
> +            OBJECT(as), "riscv_iommu", UINT64_MAX);
> +        address_space_init(&as->iova_as, MEMORY_REGION(&as->iova_mr), name);
> +
> +        qemu_mutex_lock(&s->core_lock);
> +        QLIST_INSERT_HEAD(&s->spaces, as, list);
> +        qemu_mutex_unlock(&s->core_lock);
> +
> +        trace_riscv_iommu_new(s->parent_obj.id, PCI_BUS_NUM(as->devid),
> +                PCI_SLOT(as->devid), PCI_FUNC(as->devid));
> +    }
> +    return &as->iova_as;
> +}
> +
> +static int riscv_iommu_translate(RISCVIOMMUState *s, RISCVIOMMUContext *ctx,
> +    IOMMUTLBEntry *iotlb)
> +{
> +    bool enable_pasid;
> +    bool enable_pri;
> +    int fault;
> +
> +    /*
> +     * TC[32] is reserved for custom extensions, used here to temporarily
> +     * enable automatic page-request generation for ATS queries.
> +     */
> +    enable_pri = (iotlb->perm == IOMMU_NONE) && (ctx->tc & BIT_ULL(32));
> +    enable_pasid = (ctx->tc & RISCV_IOMMU_DC_TC_PDTV);
> +
> +    /* Translate using device directory / page table information. */
> +    fault = riscv_iommu_spa_fetch(s, ctx, iotlb);
> +
> +    if (enable_pri && fault) {
> +        struct riscv_iommu_pq_record pr = {0};
> +        if (enable_pasid) {
> +            pr.hdr = set_field(RISCV_IOMMU_PREQ_HDR_PV,
> +                RISCV_IOMMU_PREQ_HDR_PID, ctx->pasid);
> +        }
> +        pr.hdr = set_field(pr.hdr, RISCV_IOMMU_PREQ_HDR_DID, ctx->devid);
> +        pr.payload = (iotlb->iova & TARGET_PAGE_MASK) |
> +                     RISCV_IOMMU_PREQ_PAYLOAD_M;
> +        riscv_iommu_pri(s, &pr);
> +        return fault;
> +    }
> +
> +    if (fault) {
> +        unsigned ttype;
> +
> +        if (iotlb->perm & IOMMU_RW) {
> +            ttype = RISCV_IOMMU_FQ_TTYPE_UADDR_WR;
> +        } else {
> +            ttype = RISCV_IOMMU_FQ_TTYPE_UADDR_RD;
> +        }
> +
> +        riscv_iommu_report_fault(s, ctx, ttype, fault, enable_pasid,
> +                                 iotlb->iova, iotlb->translated_addr);
> +        return fault;
> +    }
> +
> +    return 0;
> +}
> +
> +/* IOMMU Command Interface */
> +static MemTxResult riscv_iommu_iofence(RISCVIOMMUState *s, bool notify,
> +    uint64_t addr, uint32_t data)
> +{
> +    /*
> +     * ATS processing in this implementation of the IOMMU is synchronous,
> +     * no need to wait for completions here.
> +     */
> +    if (!notify) {
> +        return MEMTX_OK;
> +    }
> +
> +    return dma_memory_write(s->target_as, addr, &data, sizeof(data),
> +        MEMTXATTRS_UNSPECIFIED);
> +}
> +
> +static void riscv_iommu_process_ddtp(RISCVIOMMUState *s)
> +{
> +    uint64_t old_ddtp = s->ddtp;
> +    uint64_t new_ddtp = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_DDTP);
> +    unsigned new_mode = get_field(new_ddtp, RISCV_IOMMU_DDTP_MODE);
> +    unsigned old_mode = get_field(old_ddtp, RISCV_IOMMU_DDTP_MODE);
> +    bool ok = false;
> +
> +    /*
> +     * Check for allowed DDTP.MODE transitions:
> +     * {OFF, BARE}        -> {OFF, BARE, 1LVL, 2LVL, 3LVL}
> +     * {1LVL, 2LVL, 3LVL} -> {OFF, BARE}
> +     */
> +    if (new_mode == old_mode ||
> +        new_mode == RISCV_IOMMU_DDTP_MODE_OFF ||
> +        new_mode == RISCV_IOMMU_DDTP_MODE_BARE) {
> +        ok = true;
> +    } else if (new_mode == RISCV_IOMMU_DDTP_MODE_1LVL ||
> +               new_mode == RISCV_IOMMU_DDTP_MODE_2LVL ||
> +               new_mode == RISCV_IOMMU_DDTP_MODE_3LVL) {
> +        ok = old_mode == RISCV_IOMMU_DDTP_MODE_OFF ||
> +             old_mode == RISCV_IOMMU_DDTP_MODE_BARE;
> +    }
> +
> +    if (ok) {
> +        /* clear reserved and busy bits, report back sanitized version */
> +        new_ddtp = set_field(new_ddtp & RISCV_IOMMU_DDTP_PPN,
> +                             RISCV_IOMMU_DDTP_MODE, new_mode);
> +    } else {
> +        new_ddtp = old_ddtp;
> +    }
> +    s->ddtp = new_ddtp;
> +
> +    riscv_iommu_reg_set64(s, RISCV_IOMMU_REG_DDTP, new_ddtp);
> +}
> +
> +/* Command function and opcode field. */
> +#define RISCV_IOMMU_CMD(func, op) (((func) << 7) | (op))
> +
> +static void riscv_iommu_process_cq_tail(RISCVIOMMUState *s)
> +{
> +    struct riscv_iommu_command cmd;
> +    MemTxResult res;
> +    dma_addr_t addr;
> +    uint32_t tail, head, ctrl;
> +    uint64_t cmd_opcode;
> +    GHFunc func;
> +
> +    ctrl = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
> +    tail = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQT) & s->cq_mask;
> +    head = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQH) & s->cq_mask;
> +
> +    /* Check for pending error or queue processing disabled */
> +    if (!(ctrl & RISCV_IOMMU_CQCSR_CQON) ||
> +        !!(ctrl & (RISCV_IOMMU_CQCSR_CMD_ILL | RISCV_IOMMU_CQCSR_CQMF))) {
> +        return;
> +    }
> +
> +    while (tail != head) {
> +        addr = s->cq_addr  + head * sizeof(cmd);
> +        res = dma_memory_read(s->target_as, addr, &cmd, sizeof(cmd),
> +                              MEMTXATTRS_UNSPECIFIED);
> +
> +        if (res != MEMTX_OK) {
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
> +                                  RISCV_IOMMU_CQCSR_CQMF, 0);
> +            goto fault;
> +        }
> +
> +        trace_riscv_iommu_cmd(s->parent_obj.id, cmd.dword0, cmd.dword1);
> +
> +        cmd_opcode = get_field(cmd.dword0,
> +                               RISCV_IOMMU_CMD_OPCODE | RISCV_IOMMU_CMD_FUNC);
> +
> +        switch (cmd_opcode) {
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOFENCE_FUNC_C,
> +                             RISCV_IOMMU_CMD_IOFENCE_OPCODE):
> +            res = riscv_iommu_iofence(s,
> +                cmd.dword0 & RISCV_IOMMU_CMD_IOFENCE_AV, cmd.dword1,
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IOFENCE_DATA));
> +
> +            if (res != MEMTX_OK) {
> +                riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
> +                                      RISCV_IOMMU_CQCSR_CQMF, 0);
> +                goto fault;
> +            }
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOTINVAL_FUNC_GVMA,
> +                             RISCV_IOMMU_CMD_IOTINVAL_OPCODE):
> +            if (cmd.dword0 & RISCV_IOMMU_CMD_IOTINVAL_PSCV) {
> +                /* illegal command arguments IOTINVAL.GVMA & PSCV == 1 */
> +                goto cmd_ill;
> +            }
> +            /* translation cache not implemented yet */
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IOTINVAL_FUNC_VMA,
> +                             RISCV_IOMMU_CMD_IOTINVAL_OPCODE):
> +            /* translation cache not implemented yet */
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IODIR_FUNC_INVAL_DDT,
> +                             RISCV_IOMMU_CMD_IODIR_OPCODE):
> +            if (!(cmd.dword0 & RISCV_IOMMU_CMD_IODIR_DV)) {
> +                /* invalidate all device context cache mappings */
> +                func = __ctx_inval_all;
> +            } else {
> +                /* invalidate all device context matching DID */
> +                func = __ctx_inval_devid;
> +            }
> +            riscv_iommu_ctx_inval(s, func,
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_DID), 0);
> +            break;
> +
> +        case RISCV_IOMMU_CMD(RISCV_IOMMU_CMD_IODIR_FUNC_INVAL_PDT,
> +                             RISCV_IOMMU_CMD_IODIR_OPCODE):
> +            if (!(cmd.dword0 & RISCV_IOMMU_CMD_IODIR_DV)) {
> +                /* illegal command arguments IODIR_PDT & DV == 0 */
> +                goto cmd_ill;
> +            } else {
> +                func = __ctx_inval_devid_pasid;
> +            }
> +            riscv_iommu_ctx_inval(s, func,
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_DID),
> +                get_field(cmd.dword0, RISCV_IOMMU_CMD_IODIR_PID));
> +            break;
> +
> +        default:
> +        cmd_ill:
> +            /* Invalid instruction, do not advance instruction index. */
> +            riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR,
> +                RISCV_IOMMU_CQCSR_CMD_ILL, 0);
> +            goto fault;
> +        }
> +
> +        /* Advance and update head pointer after command completes. */
> +        head = (head + 1) & s->cq_mask;
> +        riscv_iommu_reg_set32(s, RISCV_IOMMU_REG_CQH, head);
> +    }
> +    return;
> +
> +fault:
> +    if (ctrl & RISCV_IOMMU_CQCSR_CIE) {
> +        riscv_iommu_notify(s, RISCV_IOMMU_INTR_CQ);
> +    }
> +}
> +
> +static void riscv_iommu_process_cq_control(RISCVIOMMUState *s)
> +{
> +    uint64_t base;
> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
> +    uint32_t ctrl_clr;
> +    bool enable = !!(ctrl_set & RISCV_IOMMU_CQCSR_CQEN);
> +    bool active = !!(ctrl_set & RISCV_IOMMU_CQCSR_CQON);
> +
> +    if (enable && !active) {
> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_CQB);
> +        s->cq_mask = (2ULL << get_field(base, RISCV_IOMMU_CQB_LOG2SZ)) - 1;
> +        s->cq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_CQB_PPN));
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQT], ~s->cq_mask);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_CQH], 0);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_CQT], 0);
> +        ctrl_set = RISCV_IOMMU_CQCSR_CQON;
> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY | RISCV_IOMMU_CQCSR_CQMF |
> +                   RISCV_IOMMU_CQCSR_CMD_ILL | RISCV_IOMMU_CQCSR_CMD_TO |
> +                   RISCV_IOMMU_CQCSR_FENCE_W_IP;
> +    } else if (!enable && active) {
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQT], ~0);
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY | RISCV_IOMMU_CQCSR_CQON;
> +    } else {
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_CQCSR_BUSY;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_CQCSR, ctrl_set, ctrl_clr);
> +}
> +
> +static void riscv_iommu_process_fq_control(RISCVIOMMUState *s)
> +{
> +    uint64_t base;
> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
> +    uint32_t ctrl_clr;
> +    bool enable = !!(ctrl_set & RISCV_IOMMU_FQCSR_FQEN);
> +    bool active = !!(ctrl_set & RISCV_IOMMU_FQCSR_FQON);
> +
> +    if (enable && !active) {
> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_FQB);
> +        s->fq_mask = (2ULL << get_field(base, RISCV_IOMMU_FQB_LOG2SZ)) - 1;
> +        s->fq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_FQB_PPN));
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQH], ~s->fq_mask);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_FQH], 0);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_FQT], 0);
> +        ctrl_set = RISCV_IOMMU_FQCSR_FQON;
> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY | RISCV_IOMMU_FQCSR_FQMF |
> +            RISCV_IOMMU_FQCSR_FQOF;
> +    } else if (!enable && active) {
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQH], ~0);
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY | RISCV_IOMMU_FQCSR_FQON;
> +    } else {
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_FQCSR_BUSY;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_FQCSR, ctrl_set, ctrl_clr);
> +}
> +
> +static void riscv_iommu_process_pq_control(RISCVIOMMUState *s)
> +{
> +    uint64_t base;
> +    uint32_t ctrl_set = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
> +    uint32_t ctrl_clr;
> +    bool enable = !!(ctrl_set & RISCV_IOMMU_PQCSR_PQEN);
> +    bool active = !!(ctrl_set & RISCV_IOMMU_PQCSR_PQON);
> +
> +    if (enable && !active) {
> +        base = riscv_iommu_reg_get64(s, RISCV_IOMMU_REG_PQB);
> +        s->pq_mask = (2ULL << get_field(base, RISCV_IOMMU_PQB_LOG2SZ)) - 1;
> +        s->pq_addr = PPN_PHYS(get_field(base, RISCV_IOMMU_PQB_PPN));
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQH], ~s->pq_mask);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_PQH], 0);
> +        stl_le_p(&s->regs_rw[RISCV_IOMMU_REG_PQT], 0);
> +        ctrl_set = RISCV_IOMMU_PQCSR_PQON;
> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY | RISCV_IOMMU_PQCSR_PQMF |
> +            RISCV_IOMMU_PQCSR_PQOF;
> +    } else if (!enable && active) {
> +        stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQH], ~0);
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY | RISCV_IOMMU_PQCSR_PQON;
> +    } else {
> +        ctrl_set = 0;
> +        ctrl_clr = RISCV_IOMMU_PQCSR_BUSY;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_PQCSR, ctrl_set, ctrl_clr);
> +}
> +
> +typedef void riscv_iommu_process_fn(RISCVIOMMUState *s);
> +
> +static void riscv_iommu_update_ipsr(RISCVIOMMUState *s, uint64_t data)
> +{
> +    uint32_t cqcsr, fqcsr, pqcsr;
> +    uint32_t ipsr_set = 0;
> +    uint32_t ipsr_clr = 0;
> +
> +    if (data & RISCV_IOMMU_IPSR_CIP) {
> +        cqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_CQCSR);
> +
> +        if (cqcsr & RISCV_IOMMU_CQCSR_CIE &&
> +            (cqcsr & RISCV_IOMMU_CQCSR_FENCE_W_IP ||
> +             cqcsr & RISCV_IOMMU_CQCSR_CMD_ILL ||
> +             cqcsr & RISCV_IOMMU_CQCSR_CMD_TO ||
> +             cqcsr & RISCV_IOMMU_CQCSR_CQMF)) {
> +            ipsr_set |= RISCV_IOMMU_IPSR_CIP;
> +        } else {
> +            ipsr_clr |= RISCV_IOMMU_IPSR_CIP;
> +        }
> +    } else {
> +        ipsr_clr |= RISCV_IOMMU_IPSR_CIP;
> +    }
> +
> +    if (data & RISCV_IOMMU_IPSR_FIP) {
> +        fqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_FQCSR);
> +
> +        if (fqcsr & RISCV_IOMMU_FQCSR_FIE &&
> +            (fqcsr & RISCV_IOMMU_FQCSR_FQOF ||
> +             fqcsr & RISCV_IOMMU_FQCSR_FQMF)) {
> +            ipsr_set |= RISCV_IOMMU_IPSR_FIP;
> +        } else {
> +            ipsr_clr |= RISCV_IOMMU_IPSR_FIP;
> +        }
> +    } else {
> +        ipsr_clr |= RISCV_IOMMU_IPSR_FIP;
> +    }
> +
> +    if (data & RISCV_IOMMU_IPSR_PIP) {
> +        pqcsr = riscv_iommu_reg_get32(s, RISCV_IOMMU_REG_PQCSR);
> +
> +        if (pqcsr & RISCV_IOMMU_PQCSR_PIE &&
> +            (pqcsr & RISCV_IOMMU_PQCSR_PQOF ||
> +             pqcsr & RISCV_IOMMU_PQCSR_PQMF)) {
> +            ipsr_set |= RISCV_IOMMU_IPSR_PIP;
> +        } else {
> +            ipsr_clr |= RISCV_IOMMU_IPSR_PIP;
> +        }
> +    } else {
> +        ipsr_clr |= RISCV_IOMMU_IPSR_PIP;
> +    }
> +
> +    riscv_iommu_reg_mod32(s, RISCV_IOMMU_REG_IPSR, ipsr_set, ipsr_clr);
> +}
> +
> +static MemTxResult riscv_iommu_mmio_write(void *opaque, hwaddr addr,
> +    uint64_t data, unsigned size, MemTxAttrs attrs)
> +{
> +    riscv_iommu_process_fn *process_fn = NULL;
> +    RISCVIOMMUState *s = opaque;
> +    uint32_t regb = addr & ~3;
> +    uint32_t busy = 0;
> +    uint64_t val = 0;
> +
> +    if ((addr & (size - 1)) != 0) {
> +        /* Unsupported MMIO alignment or access size */
> +        return MEMTX_ERROR;
> +    }
> +
> +    if (addr + size > RISCV_IOMMU_REG_MSI_CONFIG) {
> +        /* Unsupported MMIO access location. */
> +        return MEMTX_ACCESS_ERROR;
> +    }
> +
> +    /* Track actionable MMIO write. */
> +    switch (regb) {
> +    case RISCV_IOMMU_REG_DDTP:
> +    case RISCV_IOMMU_REG_DDTP + 4:
> +        process_fn = riscv_iommu_process_ddtp;
> +        regb = RISCV_IOMMU_REG_DDTP;
> +        busy = RISCV_IOMMU_DDTP_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_CQT:
> +        process_fn = riscv_iommu_process_cq_tail;
> +        break;
> +
> +    case RISCV_IOMMU_REG_CQCSR:
> +        process_fn = riscv_iommu_process_cq_control;
> +        busy = RISCV_IOMMU_CQCSR_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_FQCSR:
> +        process_fn = riscv_iommu_process_fq_control;
> +        busy = RISCV_IOMMU_FQCSR_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_PQCSR:
> +        process_fn = riscv_iommu_process_pq_control;
> +        busy = RISCV_IOMMU_PQCSR_BUSY;
> +        break;
> +
> +    case RISCV_IOMMU_REG_IPSR:
> +        /*
> +         * IPSR has special procedures to update. Execute it
> +         * and exit.
> +         */
> +        if (size == 4) {
> +            uint32_t ro = ldl_le_p(&s->regs_ro[addr]);
> +            uint32_t wc = ldl_le_p(&s->regs_wc[addr]);
> +            uint32_t rw = ldl_le_p(&s->regs_rw[addr]);
> +            stl_le_p(&val, ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +        } else if (size == 8) {
> +            uint64_t ro = ldq_le_p(&s->regs_ro[addr]);
> +            uint64_t wc = ldq_le_p(&s->regs_wc[addr]);
> +            uint64_t rw = ldq_le_p(&s->regs_rw[addr]);
> +            stq_le_p(&val, ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +        }
> +
> +        riscv_iommu_update_ipsr(s, val);
> +
> +        return MEMTX_OK;
> +
> +    default:
> +        break;
> +    }
> +
> +    /*
> +     * Registers update might be not synchronized with core logic.
> +     * If system software updates register when relevant BUSY bit
> +     * is set IOMMU behavior of additional writes to the register
> +     * is UNSPECIFIED.
> +     */
> +    qemu_spin_lock(&s->regs_lock);
> +    if (size == 1) {
> +        uint8_t ro = s->regs_ro[addr];
> +        uint8_t wc = s->regs_wc[addr];
> +        uint8_t rw = s->regs_rw[addr];
> +        s->regs_rw[addr] = ((rw & ro) | (data & ~ro)) & ~(data & wc);
> +    } else if (size == 2) {
> +        uint16_t ro = lduw_le_p(&s->regs_ro[addr]);
> +        uint16_t wc = lduw_le_p(&s->regs_wc[addr]);
> +        uint16_t rw = lduw_le_p(&s->regs_rw[addr]);
> +        stw_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +    } else if (size == 4) {
> +        uint32_t ro = ldl_le_p(&s->regs_ro[addr]);
> +        uint32_t wc = ldl_le_p(&s->regs_wc[addr]);
> +        uint32_t rw = ldl_le_p(&s->regs_rw[addr]);
> +        stl_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +    } else if (size == 8) {
> +        uint64_t ro = ldq_le_p(&s->regs_ro[addr]);
> +        uint64_t wc = ldq_le_p(&s->regs_wc[addr]);
> +        uint64_t rw = ldq_le_p(&s->regs_rw[addr]);
> +        stq_le_p(&s->regs_rw[addr], ((rw & ro) | (data & ~ro)) & ~(data & wc));
> +    }
> +
> +    /* Busy flag update, MSB 4-byte register. */
> +    if (busy) {
> +        uint32_t rw = ldl_le_p(&s->regs_rw[regb]);
> +        stl_le_p(&s->regs_rw[regb], rw | busy);
> +    }
> +    qemu_spin_unlock(&s->regs_lock);
> +
> +    if (process_fn) {
> +        qemu_mutex_lock(&s->core_lock);
> +        process_fn(s);
> +        qemu_mutex_unlock(&s->core_lock);
> +    }
> +
> +    return MEMTX_OK;
> +}
> +
> +static MemTxResult riscv_iommu_mmio_read(void *opaque, hwaddr addr,
> +    uint64_t *data, unsigned size, MemTxAttrs attrs)
> +{
> +    RISCVIOMMUState *s = opaque;
> +    uint64_t val = -1;
> +    uint8_t *ptr;
> +
> +    if ((addr & (size - 1)) != 0) {
> +        /* Unsupported MMIO alignment. */
> +        return MEMTX_ERROR;
> +    }
> +
> +    if (addr + size > RISCV_IOMMU_REG_MSI_CONFIG) {
> +        return MEMTX_ACCESS_ERROR;
> +    }
> +
> +    ptr = &s->regs_rw[addr];
> +
> +    if (size == 1) {
> +        val = (uint64_t)*ptr;
> +    } else if (size == 2) {
> +        val = lduw_le_p(ptr);
> +    } else if (size == 4) {
> +        val = ldl_le_p(ptr);
> +    } else if (size == 8) {
> +        val = ldq_le_p(ptr);
> +    } else {
> +        return MEMTX_ERROR;
> +    }
> +
> +    *data = val;
> +
> +    return MEMTX_OK;
> +}
> +
> +static const MemoryRegionOps riscv_iommu_mmio_ops = {
> +    .read_with_attrs = riscv_iommu_mmio_read,
> +    .write_with_attrs = riscv_iommu_mmio_write,
> +    .endianness = DEVICE_NATIVE_ENDIAN,
> +    .impl = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +        .unaligned = false,
> +    },
> +    .valid = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +    }
> +};
> +
> +/*
> + * Translations matching MSI pattern check are redirected to "riscv-iommu-trap"
> + * memory region as untranslated address, for additional MSI/MRIF interception
> + * by IOMMU interrupt remapping implementation.
> + * Note: Device emulation code generating an MSI is expected to provide a valid
> + * memory transaction attributes with requested_id set.
> + */
> +static MemTxResult riscv_iommu_trap_write(void *opaque, hwaddr addr,
> +    uint64_t data, unsigned size, MemTxAttrs attrs)
> +{
> +    RISCVIOMMUState* s = (RISCVIOMMUState *)opaque;
> +    RISCVIOMMUContext *ctx;
> +    MemTxResult res;
> +    void *ref;
> +    uint32_t devid = attrs.requester_id;
> +
> +    if (attrs.unspecified) {
> +        return MEMTX_ACCESS_ERROR;
> +    }
> +
> +    /* FIXME: PCIe bus remapping for attached endpoints. */
> +    devid |= s->bus << 8;
> +
> +    ctx = riscv_iommu_ctx(s, devid, 0, &ref);
> +    if (ctx == NULL) {
> +        res = MEMTX_ACCESS_ERROR;
> +    } else {
> +        res = riscv_iommu_msi_write(s, ctx, addr, data, size, attrs);
> +    }
> +    riscv_iommu_ctx_put(s, ref);
> +    return res;
> +}
> +
> +static MemTxResult riscv_iommu_trap_read(void *opaque, hwaddr addr,
> +    uint64_t *data, unsigned size, MemTxAttrs attrs)
> +{
> +    return MEMTX_ACCESS_ERROR;
> +}
> +
> +static const MemoryRegionOps riscv_iommu_trap_ops = {
> +    .read_with_attrs = riscv_iommu_trap_read,
> +    .write_with_attrs = riscv_iommu_trap_write,
> +    .endianness = DEVICE_LITTLE_ENDIAN,
> +    .impl = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +        .unaligned = true,
> +    },
> +    .valid = {
> +        .min_access_size = 4,
> +        .max_access_size = 8,
> +    }
> +};
> +
> +static void riscv_iommu_realize(DeviceState *dev, Error **errp)
> +{
> +    RISCVIOMMUState *s = RISCV_IOMMU(dev);
> +
> +    s->cap = s->version & RISCV_IOMMU_CAP_VERSION;
> +    if (s->enable_msi) {
> +        s->cap |= RISCV_IOMMU_CAP_MSI_FLAT | RISCV_IOMMU_CAP_MSI_MRIF;
> +    }
> +    /* Report QEMU target physical address space limits */
> +    s->cap = set_field(s->cap, RISCV_IOMMU_CAP_PAS,
> +                       TARGET_PHYS_ADDR_SPACE_BITS);
> +
> +    /* TODO: method to report supported PASID bits */
> +    s->pasid_bits = 8; /* restricted to size of MemTxAttrs.pasid */
> +    s->cap |= RISCV_IOMMU_CAP_PD8;
> +
> +    /* Out-of-reset translation mode: OFF (DMA disabled) BARE (passthrough) */
> +    s->ddtp = set_field(0, RISCV_IOMMU_DDTP_MODE, s->enable_off ?
> +                        RISCV_IOMMU_DDTP_MODE_OFF : RISCV_IOMMU_DDTP_MODE_BARE);
> +
> +    /* register storage */
> +    s->regs_rw = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
> +    s->regs_ro = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
> +    s->regs_wc = g_new0(uint8_t, RISCV_IOMMU_REG_SIZE);
> +
> +     /* Mark all registers read-only */
> +    memset(s->regs_ro, 0xff, RISCV_IOMMU_REG_SIZE);
> +
> +    /*
> +     * Register complete MMIO space, including MSI/PBA registers.
> +     * Note, PCIDevice implementation will add overlapping MR for MSI/PBA,
> +     * managed directly by the PCIDevice implementation.
> +     */
> +    memory_region_init_io(&s->regs_mr, OBJECT(dev), &riscv_iommu_mmio_ops, s,
> +        "riscv-iommu-regs", RISCV_IOMMU_REG_SIZE);
> +
> +    /* Set power-on register state */
> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_CAP], s->cap);
> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_FCTL], 0);
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_DDTP],
> +        ~(RISCV_IOMMU_DDTP_PPN | RISCV_IOMMU_DDTP_MODE));
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQB],
> +        ~(RISCV_IOMMU_CQB_LOG2SZ | RISCV_IOMMU_CQB_PPN));
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQB],
> +        ~(RISCV_IOMMU_FQB_LOG2SZ | RISCV_IOMMU_FQB_PPN));
> +    stq_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQB],
> +        ~(RISCV_IOMMU_PQB_LOG2SZ | RISCV_IOMMU_PQB_PPN));
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_CQCSR], RISCV_IOMMU_CQCSR_CQMF |
> +        RISCV_IOMMU_CQCSR_CMD_TO | RISCV_IOMMU_CQCSR_CMD_ILL);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_CQCSR], RISCV_IOMMU_CQCSR_CQON |
> +        RISCV_IOMMU_CQCSR_BUSY);
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_FQCSR], RISCV_IOMMU_FQCSR_FQMF |
> +        RISCV_IOMMU_FQCSR_FQOF);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_FQCSR], RISCV_IOMMU_FQCSR_FQON |
> +        RISCV_IOMMU_FQCSR_BUSY);
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_PQCSR], RISCV_IOMMU_PQCSR_PQMF |
> +        RISCV_IOMMU_PQCSR_PQOF);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_PQCSR], RISCV_IOMMU_PQCSR_PQON |
> +        RISCV_IOMMU_PQCSR_BUSY);
> +    stl_le_p(&s->regs_wc[RISCV_IOMMU_REG_IPSR], ~0);
> +    stl_le_p(&s->regs_ro[RISCV_IOMMU_REG_IVEC], 0);
> +    stq_le_p(&s->regs_rw[RISCV_IOMMU_REG_DDTP], s->ddtp);
> +
> +    /* Memory region for downstream access, if specified. */
> +    if (s->target_mr) {
> +        s->target_as = g_new0(AddressSpace, 1);
> +        address_space_init(s->target_as, s->target_mr,
> +            "riscv-iommu-downstream");
> +    } else {
> +        /* Fallback to global system memory. */
> +        s->target_as = &address_space_memory;
> +    }
> +
> +    /* Memory region for untranslated MRIF/MSI writes */
> +    memory_region_init_io(&s->trap_mr, OBJECT(dev), &riscv_iommu_trap_ops, s,
> +            "riscv-iommu-trap", ~0ULL);
> +    address_space_init(&s->trap_as, &s->trap_mr, "riscv-iommu-trap-as");
> +
> +    /* Device translation context cache */
> +    s->ctx_cache = g_hash_table_new_full(__ctx_hash, __ctx_equal,
> +                                         g_free, NULL);
> +
> +    s->iommus.le_next = NULL;
> +    s->iommus.le_prev = NULL;
> +    QLIST_INIT(&s->spaces);
> +    qemu_mutex_init(&s->core_lock);
> +    qemu_spin_init(&s->regs_lock);
> +}
> +
> +static void riscv_iommu_unrealize(DeviceState *dev)
> +{
> +    RISCVIOMMUState *s = RISCV_IOMMU(dev);
> +
> +    qemu_mutex_destroy(&s->core_lock);
> +    g_hash_table_unref(s->ctx_cache);
> +}
> +
> +static Property riscv_iommu_properties[] = {
> +    DEFINE_PROP_UINT32("version", RISCVIOMMUState, version,
> +        RISCV_IOMMU_SPEC_DOT_VER),
> +    DEFINE_PROP_UINT32("bus", RISCVIOMMUState, bus, 0x0),
> +    DEFINE_PROP_BOOL("intremap", RISCVIOMMUState, enable_msi, TRUE),
> +    DEFINE_PROP_BOOL("off", RISCVIOMMUState, enable_off, TRUE),
> +    DEFINE_PROP_LINK("downstream-mr", RISCVIOMMUState, target_mr,
> +        TYPE_MEMORY_REGION, MemoryRegion *),
> +    DEFINE_PROP_END_OF_LIST(),
> +};
> +
> +static void riscv_iommu_class_init(ObjectClass *klass, void* data)
> +{
> +    DeviceClass *dc = DEVICE_CLASS(klass);
> +
> +    /* internal device for riscv-iommu-{pci/sys}, not user-creatable */
> +    dc->user_creatable = false;
> +    dc->realize = riscv_iommu_realize;
> +    dc->unrealize = riscv_iommu_unrealize;
> +    device_class_set_props(dc, riscv_iommu_properties);
> +}
> +
> +static const TypeInfo riscv_iommu_info = {
> +    .name = TYPE_RISCV_IOMMU,
> +    .parent = TYPE_DEVICE,
> +    .instance_size = sizeof(RISCVIOMMUState),
> +    .class_init = riscv_iommu_class_init,
> +};
> +
> +static const char *IOMMU_FLAG_STR[] = {
> +    "NA",
> +    "RO",
> +    "WR",
> +    "RW",
> +};
> +
> +/* RISC-V IOMMU Memory Region - Address Translation Space */
> +static IOMMUTLBEntry riscv_iommu_memory_region_translate(
> +    IOMMUMemoryRegion *iommu_mr, hwaddr addr,
> +    IOMMUAccessFlags flag, int iommu_idx)
> +{
> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
> +    RISCVIOMMUContext *ctx;
> +    void *ref;
> +    IOMMUTLBEntry iotlb = {
> +        .iova = addr,
> +        .target_as = as->iommu->target_as,
> +        .addr_mask = ~0ULL,
> +        .perm = flag,
> +    };
> +
> +    ctx = riscv_iommu_ctx(as->iommu, as->devid, iommu_idx, &ref);
> +    if (ctx == NULL) {
> +        /* Translation disabled or invalid. */
> +        iotlb.addr_mask = 0;
> +        iotlb.perm = IOMMU_NONE;
> +    } else if (riscv_iommu_translate(as->iommu, ctx, &iotlb)) {
> +        /* Translation disabled or fault reported. */
> +        iotlb.addr_mask = 0;
> +        iotlb.perm = IOMMU_NONE;
> +    }
> +
> +    /* Trace all dma translations with original access flags. */
> +    trace_riscv_iommu_dma(as->iommu->parent_obj.id, PCI_BUS_NUM(as->devid),
> +                          PCI_SLOT(as->devid), PCI_FUNC(as->devid), iommu_idx,
> +                          IOMMU_FLAG_STR[flag & IOMMU_RW], iotlb.iova,
> +                          iotlb.translated_addr);
> +
> +    riscv_iommu_ctx_put(as->iommu, ref);
> +
> +    return iotlb;
> +}
> +
> +static int riscv_iommu_memory_region_notify(
> +    IOMMUMemoryRegion *iommu_mr, IOMMUNotifierFlag old,
> +    IOMMUNotifierFlag new, Error **errp)
> +{
> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
> +
> +    if (old == IOMMU_NOTIFIER_NONE) {
> +        as->notifier = true;
> +        trace_riscv_iommu_notifier_add(iommu_mr->parent_obj.name);
> +    } else if (new == IOMMU_NOTIFIER_NONE) {
> +        as->notifier = false;
> +        trace_riscv_iommu_notifier_del(iommu_mr->parent_obj.name);
> +    }
> +
> +    return 0;
> +}
> +
> +static inline bool pci_is_iommu(PCIDevice *pdev)
> +{
> +    return pci_get_word(pdev->config + PCI_CLASS_DEVICE) == 0x0806;
> +}
> +
> +static AddressSpace *riscv_iommu_find_as(PCIBus *bus, void *opaque, int devfn)
> +{
> +    RISCVIOMMUState *s = (RISCVIOMMUState *) opaque;
> +    PCIDevice *pdev = pci_find_device(bus, pci_bus_num(bus), devfn);
> +    AddressSpace *as = NULL;
> +
> +    if (pdev && pci_is_iommu(pdev)) {
> +        return s->target_as;
> +    }
> +
> +    /* Find first registered IOMMU device */
> +    while (s->iommus.le_prev) {
> +        s = *(s->iommus.le_prev);
> +    }
> +
> +    /* Find first matching IOMMU */
> +    while (s != NULL && as == NULL) {
> +        as = riscv_iommu_space(s, PCI_BUILD_BDF(pci_bus_num(bus), devfn));
> +        s = s->iommus.le_next;
> +    }
> +
> +    return as ? as : &address_space_memory;
> +}
> +
> +static const PCIIOMMUOps riscv_iommu_ops = {
> +    .get_address_space = riscv_iommu_find_as,
> +};
> +
> +void riscv_iommu_pci_setup_iommu(RISCVIOMMUState *iommu, PCIBus *bus,
> +        Error **errp)
> +{
> +    if (bus->iommu_ops &&
> +        bus->iommu_ops->get_address_space == riscv_iommu_find_as) {
> +        /* Allow multiple IOMMUs on the same PCIe bus, link known devices */
> +        RISCVIOMMUState *last = (RISCVIOMMUState *)bus->iommu_opaque;
> +        QLIST_INSERT_AFTER(last, iommu, iommus);
> +    } else if (!bus->iommu_ops && !bus->iommu_opaque) {
> +        pci_setup_iommu(bus, &riscv_iommu_ops, iommu);
> +    } else {
> +        error_setg(errp, "can't register secondary IOMMU for PCI bus #%d",
> +            pci_bus_num(bus));
> +    }
> +}
> +
> +static int riscv_iommu_memory_region_index(IOMMUMemoryRegion *iommu_mr,
> +    MemTxAttrs attrs)
> +{
> +    return attrs.unspecified ? RISCV_IOMMU_NOPASID : (int)attrs.pasid;
> +}
> +
> +static int riscv_iommu_memory_region_index_len(IOMMUMemoryRegion *iommu_mr)
> +{
> +    RISCVIOMMUSpace *as = container_of(iommu_mr, RISCVIOMMUSpace, iova_mr);
> +    return 1 << as->iommu->pasid_bits;
> +}
> +
> +static void riscv_iommu_memory_region_init(ObjectClass *klass, void *data)
> +{
> +    IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_CLASS(klass);
> +
> +    imrc->translate = riscv_iommu_memory_region_translate;
> +    imrc->notify_flag_changed = riscv_iommu_memory_region_notify;
> +    imrc->attrs_to_index = riscv_iommu_memory_region_index;
> +    imrc->num_indexes = riscv_iommu_memory_region_index_len;
> +}
> +
> +static const TypeInfo riscv_iommu_memory_region_info = {
> +    .parent = TYPE_IOMMU_MEMORY_REGION,
> +    .name = TYPE_RISCV_IOMMU_MEMORY_REGION,
> +    .class_init = riscv_iommu_memory_region_init,
> +};
> +
> +static void riscv_iommu_register_mr_types(void)
> +{
> +    type_register_static(&riscv_iommu_memory_region_info);
> +    type_register_static(&riscv_iommu_info);
> +}
> +
> +type_init(riscv_iommu_register_mr_types);
> diff --git a/hw/riscv/riscv-iommu.h b/hw/riscv/riscv-iommu.h
> new file mode 100644
> index 0000000000..31d3907d33
> --- /dev/null
> +++ b/hw/riscv/riscv-iommu.h
> @@ -0,0 +1,141 @@
> +/*
> + * QEMU emulation of an RISC-V IOMMU
> + *
> + * Copyright (C) 2022-2023 Rivos Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#ifndef HW_RISCV_IOMMU_STATE_H
> +#define HW_RISCV_IOMMU_STATE_H
> +
> +#include "qemu/osdep.h"
> +#include "qom/object.h"
> +
> +#include "hw/riscv/iommu.h"
> +
> +struct RISCVIOMMUState {
> +    /*< private >*/
> +    DeviceState parent_obj;
> +
> +    /*< public >*/
> +    uint32_t version;     /* Reported interface version number */
> +    uint32_t pasid_bits;  /* process identifier width */
> +    uint32_t bus;         /* PCI bus mapping for non-root endpoints */
> +
> +    uint64_t cap;         /* IOMMU supported capabilities */
> +    uint64_t fctl;        /* IOMMU enabled features */
> +
> +    bool enable_off;      /* Enable out-of-reset OFF mode (DMA disabled) */
> +    bool enable_msi;      /* Enable MSI remapping */
> +
> +    /* IOMMU Internal State */
> +    uint64_t ddtp;        /* Validated Device Directory Tree Root Pointer */
> +
> +    dma_addr_t cq_addr;   /* Command queue base physical address */
> +    dma_addr_t fq_addr;   /* Fault/event queue base physical address */
> +    dma_addr_t pq_addr;   /* Page request queue base physical address */
> +
> +    uint32_t cq_mask;     /* Command queue index bit mask */
> +    uint32_t fq_mask;     /* Fault/event queue index bit mask */
> +    uint32_t pq_mask;     /* Page request queue index bit mask */
> +
> +    /* interrupt notifier */
> +    void (*notify)(RISCVIOMMUState *iommu, unsigned vector);
> +
> +    /* IOMMU State Machine */
> +    QemuThread core_proc; /* Background processing thread */
> +    QemuMutex core_lock;  /* Global IOMMU lock, used for cache/regs updates */
> +    QemuCond core_cond;   /* Background processing wake up signal */
> +    unsigned core_exec;   /* Processing thread execution actions */
> +
> +    /* IOMMU target address space */
> +    AddressSpace *target_as;
> +    MemoryRegion *target_mr;
> +
> +    /* MSI / MRIF access trap */
> +    AddressSpace trap_as;
> +    MemoryRegion trap_mr;
> +
> +    GHashTable *ctx_cache;          /* Device translation Context Cache */
> +
> +    /* MMIO Hardware Interface */
> +    MemoryRegion regs_mr;
> +    QemuSpin regs_lock;
> +    uint8_t *regs_rw;  /* register state (user write) */
> +    uint8_t *regs_wc;  /* write-1-to-clear mask */
> +    uint8_t *regs_ro;  /* read-only mask */
> +
> +    QLIST_ENTRY(RISCVIOMMUState) iommus;
> +    QLIST_HEAD(, RISCVIOMMUSpace) spaces;
> +};
> +
> +void riscv_iommu_pci_setup_iommu(RISCVIOMMUState *iommu, PCIBus *bus,
> +         Error **errp);
> +
> +/* private helpers */
> +
> +/* Register helper functions */
> +static inline uint32_t riscv_iommu_reg_mod32(RISCVIOMMUState *s,
> +    unsigned idx, uint32_t set, uint32_t clr)
> +{
> +    uint32_t val;
> +    qemu_spin_lock(&s->regs_lock);
> +    val = ldl_le_p(s->regs_rw + idx);
> +    stl_le_p(s->regs_rw + idx, (val & ~clr) | set);
> +    qemu_spin_unlock(&s->regs_lock);
> +    return val;
> +}
> +
> +static inline void riscv_iommu_reg_set32(RISCVIOMMUState *s,
> +    unsigned idx, uint32_t set)
> +{
> +    qemu_spin_lock(&s->regs_lock);
> +    stl_le_p(s->regs_rw + idx, set);
> +    qemu_spin_unlock(&s->regs_lock);
> +}
> +
> +static inline uint32_t riscv_iommu_reg_get32(RISCVIOMMUState *s,
> +    unsigned idx)
> +{
> +    return ldl_le_p(s->regs_rw + idx);
> +}
> +
> +static inline uint64_t riscv_iommu_reg_mod64(RISCVIOMMUState *s,
> +    unsigned idx, uint64_t set, uint64_t clr)
> +{
> +    uint64_t val;
> +    qemu_spin_lock(&s->regs_lock);
> +    val = ldq_le_p(s->regs_rw + idx);
> +    stq_le_p(s->regs_rw + idx, (val & ~clr) | set);
> +    qemu_spin_unlock(&s->regs_lock);
> +    return val;
> +}
> +
> +static inline void riscv_iommu_reg_set64(RISCVIOMMUState *s,
> +    unsigned idx, uint64_t set)
> +{
> +    qemu_spin_lock(&s->regs_lock);
> +    stq_le_p(s->regs_rw + idx, set);
> +    qemu_spin_unlock(&s->regs_lock);
> +}
> +
> +static inline uint64_t riscv_iommu_reg_get64(RISCVIOMMUState *s,
> +    unsigned idx)
> +{
> +    return ldq_le_p(s->regs_rw + idx);
> +}
> +
> +
> +
> +#endif
> diff --git a/hw/riscv/trace-events b/hw/riscv/trace-events
> new file mode 100644
> index 0000000000..42a97caffa
> --- /dev/null
> +++ b/hw/riscv/trace-events
> @@ -0,0 +1,11 @@
> +# See documentation at docs/devel/tracing.rst
> +
> +# riscv-iommu.c
> +riscv_iommu_new(const char *id, unsigned b, unsigned d, unsigned f) "%s: device attached %04x:%02x.%d"
> +riscv_iommu_flt(const char *id, unsigned b, unsigned d, unsigned f, uint64_t reason, uint64_t iova) "%s: fault %04x:%02x.%u reason: 0x%"PRIx64" iova: 0x%"PRIx64
> +riscv_iommu_pri(const char *id, unsigned b, unsigned d, unsigned f, uint64_t iova) "%s: page request %04x:%02x.%u iova: 0x%"PRIx64
> +riscv_iommu_dma(const char *id, unsigned b, unsigned d, unsigned f, unsigned pasid, const char *dir, uint64_t iova, uint64_t phys) "%s: translate %04x:%02x.%u #%u %s 0x%"PRIx64" -> 0x%"PRIx64
> +riscv_iommu_msi(const char *id, unsigned b, unsigned d, unsigned f, uint64_t iova, uint64_t phys) "%s: translate %04x:%02x.%u MSI 0x%"PRIx64" -> 0x%"PRIx64
> +riscv_iommu_cmd(const char *id, uint64_t l, uint64_t u) "%s: command 0x%"PRIx64" 0x%"PRIx64
> +riscv_iommu_notifier_add(const char *id) "%s: dev-iotlb notifier added"
> +riscv_iommu_notifier_del(const char *id) "%s: dev-iotlb notifier removed"
> diff --git a/hw/riscv/trace.h b/hw/riscv/trace.h
> new file mode 100644
> index 0000000000..8c0e3ca1f3
> --- /dev/null
> +++ b/hw/riscv/trace.h
> @@ -0,0 +1 @@
> +#include "trace/trace-hw_riscv.h"
> diff --git a/include/hw/riscv/iommu.h b/include/hw/riscv/iommu.h
> new file mode 100644
> index 0000000000..070ee69973
> --- /dev/null
> +++ b/include/hw/riscv/iommu.h
> @@ -0,0 +1,36 @@
> +/*
> + * QEMU emulation of an RISC-V IOMMU
> + *
> + * Copyright (C) 2022-2023 Rivos Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#ifndef HW_RISCV_IOMMU_H
> +#define HW_RISCV_IOMMU_H
> +
> +#include "qemu/osdep.h"
> +#include "qom/object.h"
> +
> +#define TYPE_RISCV_IOMMU "riscv-iommu"
> +OBJECT_DECLARE_SIMPLE_TYPE(RISCVIOMMUState, RISCV_IOMMU)
> +typedef struct RISCVIOMMUState RISCVIOMMUState;
> +
> +#define TYPE_RISCV_IOMMU_MEMORY_REGION "riscv-iommu-mr"
> +typedef struct RISCVIOMMUSpace RISCVIOMMUSpace;
> +
> +#define TYPE_RISCV_IOMMU_PCI "riscv-iommu-pci"
> +OBJECT_DECLARE_SIMPLE_TYPE(RISCVIOMMUStatePci, RISCV_IOMMU_PCI)
> +typedef struct RISCVIOMMUStatePci RISCVIOMMUStatePci;
> +
> +#endif
> diff --git a/meson.build b/meson.build
> index a9de71d450..8099d8271c 100644
> --- a/meson.build
> +++ b/meson.build
> @@ -3319,6 +3319,7 @@ if have_system
>       'hw/pci-host',
>       'hw/ppc',
>       'hw/rtc',
> +    'hw/riscv',
>       'hw/s390x',
>       'hw/scsi',
>       'hw/sd',