diff mbox series

[v4,07/11] pc: Allow instantiating a virtio-iommu device

Message ID 20211001173358.863017-8-jean-philippe@linaro.org
State New
Headers show
Series virtio-iommu: Add ACPI support | expand

Commit Message

Jean-Philippe Brucker Oct. 1, 2021, 5:33 p.m. UTC 
Allow instantiating a virtio-iommu device by adding an ACPI Virtual I/O
Translation table (VIOT), which describes the relation between the
virtio-iommu and the endpoints it manages.

Add a hotplug handler for virtio-iommu on x86 and set the necessary
reserved region property. On x86, the [0xfee00000, 0xfeefffff] DMA
region is reserved for MSIs. DMA transactions to this range either
trigger IRQ remapping in the IOMMU or bypasses IOMMU translation.

Although virtio-iommu does not support IRQ remapping it must be informed
of the reserved region so that it can forward DMA transactions targeting
this region.

Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>

---
 include/hw/i386/pc.h |  2 ++
 hw/i386/acpi-build.c |  5 +++++
 hw/i386/pc.c         | 24 ++++++++++++++++++++++--
 hw/i386/Kconfig      |  1 +
 4 files changed, 30 insertions(+), 2 deletions(-)

-- 
2.33.0

Comments

Eric Auger Oct. 5, 2021, 7:18 p.m. UTC | #1 
Hi Jean,
On 10/1/21 7:33 PM, Jean-Philippe Brucker wrote:
> Allow instantiating a virtio-iommu device by adding an ACPI Virtual I/O

> Translation table (VIOT), which describes the relation between the

> virtio-iommu and the endpoints it manages.

>

> Add a hotplug handler for virtio-iommu on x86 and set the necessary

> reserved region property. On x86, the [0xfee00000, 0xfeefffff] DMA

> region is reserved for MSIs. DMA transactions to this range either

> trigger IRQ remapping in the IOMMU or bypasses IOMMU translation.

>

> Although virtio-iommu does not support IRQ remapping it must be informed

> of the reserved region so that it can forward DMA transactions targeting

> this region.

>

> Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>

Reviewed-by: Eric Auger <eric.auger@redhat.com>


tested by a protecting a virtio-net-pci device plugged onto a pxb-pcie
and setting
default-bus-bypass-iommu=true on pcie.0.

As described in the cover letter,
without [PATCH 0/3] virtio-iommu: Support VIRTIO_IOMMU_F_BYPASS_CONFIG
the ahci emits some failure if it is protected by the virtio-iommu:

qemu-system-x86_64: virtio_iommu_translate sid=250 is not known!!
qemu-system-x86_64: no buffer available in event queue to report event
qemu-system-x86_64: AHCI: Failed to start FIS receive engine: bad FIS
receive buffer address
../..
Invalid access at addr 0x7FFA6900, size 4, region '(null)', reason: rejected

But this is expected.

So feel free to add
Tested-by: Eric Auger <eric.auger@redhat.com>


Thanks

Eric

> ---

>  include/hw/i386/pc.h |  2 ++

>  hw/i386/acpi-build.c |  5 +++++

>  hw/i386/pc.c         | 24 ++++++++++++++++++++++--

>  hw/i386/Kconfig      |  1 +

>  4 files changed, 30 insertions(+), 2 deletions(-)

>

> diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h

> index 82cf7b7e30..f3ba1ee4c0 100644

> --- a/include/hw/i386/pc.h

> +++ b/include/hw/i386/pc.h

> @@ -45,6 +45,8 @@ typedef struct PCMachineState {

>      bool pit_enabled;

>      bool hpet_enabled;

>      bool default_bus_bypass_iommu;

> +    bool virtio_iommu;

> +    uint16_t virtio_iommu_bdf;

>      uint64_t max_fw_size;

>  

>      /* ACPI Memory hotplug IO base address */

> diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c

> index d1c28440f4..4e46585709 100644

> --- a/hw/i386/acpi-build.c

> +++ b/hw/i386/acpi-build.c

> @@ -71,6 +71,7 @@

>  

>  #include "hw/acpi/ipmi.h"

>  #include "hw/acpi/hmat.h"

> +#include "hw/acpi/viot.h"

>  

>  /* These are used to size the ACPI tables for -M pc-i440fx-1.7 and

>   * -M pc-i440fx-2.0.  Even if the actual amount of AML generated grows

> @@ -2593,6 +2594,10 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine)

>              build_dmar_q35(tables_blob, tables->linker, x86ms->oem_id,

>                             x86ms->oem_table_id);

>          }

> +    } else if (pcms->virtio_iommu) {

> +        acpi_add_table(table_offsets, tables_blob);

> +        build_viot(machine, tables_blob, tables->linker, pcms->virtio_iommu_bdf,

> +                   x86ms->oem_id, x86ms->oem_table_id);

>      }

>      if (machine->nvdimms_state->is_enabled) {

>          nvdimm_build_acpi(table_offsets, tables_blob, tables->linker,

> diff --git a/hw/i386/pc.c b/hw/i386/pc.c

> index 789ccb6ef4..31710bc4fb 100644

> --- a/hw/i386/pc.c

> +++ b/hw/i386/pc.c

> @@ -83,6 +83,7 @@

>  #include "hw/i386/intel_iommu.h"

>  #include "hw/net/ne2000-isa.h"

>  #include "standard-headers/asm-x86/bootparam.h"

> +#include "hw/virtio/virtio-iommu.h"

>  #include "hw/virtio/virtio-pmem-pci.h"

>  #include "hw/virtio/virtio-mem-pci.h"

>  #include "hw/mem/memory-device.h"

> @@ -1367,8 +1368,11 @@ static void pc_virtio_md_pci_unplug(HotplugHandler *hotplug_dev,

>  static void pc_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,

>                                            DeviceState *dev, Error **errp)

>  {

> -    if (object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) &&

> -        x86_iommu_get_default()) {

> +    PCMachineState *pcms = PC_MACHINE(hotplug_dev);

> +

> +    if ((object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) ||

> +         object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) &&

> +        (x86_iommu_get_default() || pcms->virtio_iommu)) {

>          error_setg(errp, "QEMU does not support multiple vIOMMUs "

>                     "for x86 yet.");

>          return;

> @@ -1381,6 +1385,15 @@ static void pc_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,

>      } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||

>                 object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI)) {

>          pc_virtio_md_pci_pre_plug(hotplug_dev, dev, errp);

> +    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) {

> +        /* Declare the APIC range as the reserved MSI region */

> +        char *resv_prop_str = g_strdup_printf("0xfee00000:0xfeefffff:%d",

> +                                              VIRTIO_IOMMU_RESV_MEM_T_MSI);

> +

> +        object_property_set_uint(OBJECT(dev), "len-reserved-regions", 1, errp);

> +        object_property_set_str(OBJECT(dev), "reserved-regions[0]",

> +                                resv_prop_str, errp);

> +        g_free(resv_prop_str);

>      }

>  }

>  

> @@ -1394,6 +1407,12 @@ static void pc_machine_device_plug_cb(HotplugHandler *hotplug_dev,

>      } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||

>                 object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI)) {

>          pc_virtio_md_pci_plug(hotplug_dev, dev, errp);

> +    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) {

> +        PCMachineState *pcms = PC_MACHINE(hotplug_dev);

> +        PCIDevice *pdev = PCI_DEVICE(dev);

> +

> +        pcms->virtio_iommu = true;

> +        pcms->virtio_iommu_bdf = pci_get_bdf(pdev);

>      }

>  }

>  

> @@ -1436,6 +1455,7 @@ static HotplugHandler *pc_get_hotplug_handler(MachineState *machine,

>          object_dynamic_cast(OBJECT(dev), TYPE_CPU) ||

>          object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||

>          object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI) ||

> +        object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI) ||

>          object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE)) {

>          return HOTPLUG_HANDLER(machine);

>      }

> diff --git a/hw/i386/Kconfig b/hw/i386/Kconfig

> index ddedcef0b2..13db05d557 100644

> --- a/hw/i386/Kconfig

> +++ b/hw/i386/Kconfig

> @@ -54,6 +54,7 @@ config PC_ACPI

>      select ACPI_X86

>      select ACPI_CPU_HOTPLUG

>      select ACPI_MEMORY_HOTPLUG

> +    select ACPI_VIOT

>      select SMBUS_EEPROM

>      select PFLASH_CFI01

>      depends on ACPI_SMBUS
Igor Mammedov Oct. 6, 2021, 7:19 a.m. UTC | #2 
On Fri,  1 Oct 2021 18:33:55 +0100
Jean-Philippe Brucker <jean-philippe@linaro.org> wrote:

> Allow instantiating a virtio-iommu device by adding an ACPI Virtual I/O

> Translation table (VIOT), which describes the relation between the

> virtio-iommu and the endpoints it manages.

> 

> Add a hotplug handler for virtio-iommu on x86 and set the necessary

> reserved region property. On x86, the [0xfee00000, 0xfeefffff] DMA

> region is reserved for MSIs. DMA transactions to this range either

> trigger IRQ remapping in the IOMMU or bypasses IOMMU translation.

> 

> Although virtio-iommu does not support IRQ remapping it must be informed

> of the reserved region so that it can forward DMA transactions targeting

> this region.

> 

> Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>

> ---

>  include/hw/i386/pc.h |  2 ++

>  hw/i386/acpi-build.c |  5 +++++

>  hw/i386/pc.c         | 24 ++++++++++++++++++++++--

>  hw/i386/Kconfig      |  1 +

>  4 files changed, 30 insertions(+), 2 deletions(-)

> 

> diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h

> index 82cf7b7e30..f3ba1ee4c0 100644

> --- a/include/hw/i386/pc.h

> +++ b/include/hw/i386/pc.h

> @@ -45,6 +45,8 @@ typedef struct PCMachineState {

>      bool pit_enabled;

>      bool hpet_enabled;

>      bool default_bus_bypass_iommu;

> +    bool virtio_iommu;

> +    uint16_t virtio_iommu_bdf;

>      uint64_t max_fw_size;

>  

>      /* ACPI Memory hotplug IO base address */

> diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c

> index d1c28440f4..4e46585709 100644

> --- a/hw/i386/acpi-build.c

> +++ b/hw/i386/acpi-build.c

> @@ -71,6 +71,7 @@

>  

>  #include "hw/acpi/ipmi.h"

>  #include "hw/acpi/hmat.h"

> +#include "hw/acpi/viot.h"

>  

>  /* These are used to size the ACPI tables for -M pc-i440fx-1.7 and

>   * -M pc-i440fx-2.0.  Even if the actual amount of AML generated grows

> @@ -2593,6 +2594,10 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine)

>              build_dmar_q35(tables_blob, tables->linker, x86ms->oem_id,

>                             x86ms->oem_table_id);

>          }

> +    } else if (pcms->virtio_iommu) {

> +        acpi_add_table(table_offsets, tables_blob);

> +        build_viot(machine, tables_blob, tables->linker, pcms->virtio_iommu_bdf,

I'd drop PCMachineState::virtio_iommu_bdf and s/pcms->virtio_iommu_bdf/pci_get_bdf(iommu)/

> +                   x86ms->oem_id, x86ms->oem_table_id);

>      }

>      if (machine->nvdimms_state->is_enabled) {

>          nvdimm_build_acpi(table_offsets, tables_blob, tables->linker,

> diff --git a/hw/i386/pc.c b/hw/i386/pc.c

> index 789ccb6ef4..31710bc4fb 100644

> --- a/hw/i386/pc.c

> +++ b/hw/i386/pc.c

> @@ -83,6 +83,7 @@

>  #include "hw/i386/intel_iommu.h"

>  #include "hw/net/ne2000-isa.h"

>  #include "standard-headers/asm-x86/bootparam.h"

> +#include "hw/virtio/virtio-iommu.h"

>  #include "hw/virtio/virtio-pmem-pci.h"

>  #include "hw/virtio/virtio-mem-pci.h"

>  #include "hw/mem/memory-device.h"

> @@ -1367,8 +1368,11 @@ static void pc_virtio_md_pci_unplug(HotplugHandler *hotplug_dev,

>  static void pc_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,

>                                            DeviceState *dev, Error **errp)

>  {

> -    if (object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) &&

> -        x86_iommu_get_default()) {

> +    PCMachineState *pcms = PC_MACHINE(hotplug_dev);

> +

> +    if ((object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) ||

> +         object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) &&

> +        (x86_iommu_get_default() || pcms->virtio_iommu)) {


this check is getting uglier,
may be instead of introducing pcms->virtio_iommu boolean, better approach
would be adding 'Device* PCMachineState::iommu' and setting it to IOMMU
so the check would reduce to:
      if ((object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) ||
           object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)))
      {
          if (pcms->iommu)
            err
          else set pcms->iommu in plug handler or here
      }
      
that also will let to cleanup/get rid of x86_iommu_[s|g]et_default()
and x86_iommu_default 'global'.
Maybe replace previous patch with one that would remove
x86_iommu_[s|g]et_default().

>          error_setg(errp, "QEMU does not support multiple vIOMMUs "

>                     "for x86 yet.");

>          return;

> @@ -1381,6 +1385,15 @@ static void pc_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,

>      } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||

>                 object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI)) {

>          pc_virtio_md_pci_pre_plug(hotplug_dev, dev, errp);

> +    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) {

> +        /* Declare the APIC range as the reserved MSI region */

> +        char *resv_prop_str = g_strdup_printf("0xfee00000:0xfeefffff:%d",

> +                                              VIRTIO_IOMMU_RESV_MEM_T_MSI);

> +

> +        object_property_set_uint(OBJECT(dev), "len-reserved-regions", 1, errp);

> +        object_property_set_str(OBJECT(dev), "reserved-regions[0]",

> +                                resv_prop_str, errp);

> +        g_free(resv_prop_str);

>      }

>  }

>  

> @@ -1394,6 +1407,12 @@ static void pc_machine_device_plug_cb(HotplugHandler *hotplug_dev,

>      } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||

>                 object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI)) {

>          pc_virtio_md_pci_plug(hotplug_dev, dev, errp);

> +    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) {

> +        PCMachineState *pcms = PC_MACHINE(hotplug_dev);

> +        PCIDevice *pdev = PCI_DEVICE(dev);

> +

> +        pcms->virtio_iommu = true;

> +        pcms->virtio_iommu_bdf = pci_get_bdf(pdev);

>      }

>  }

>  

> @@ -1436,6 +1455,7 @@ static HotplugHandler *pc_get_hotplug_handler(MachineState *machine,

>          object_dynamic_cast(OBJECT(dev), TYPE_CPU) ||

>          object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||

>          object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI) ||

> +        object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI) ||

>          object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE)) {

>          return HOTPLUG_HANDLER(machine);

>      }

> diff --git a/hw/i386/Kconfig b/hw/i386/Kconfig

> index ddedcef0b2..13db05d557 100644

> --- a/hw/i386/Kconfig

> +++ b/hw/i386/Kconfig

> @@ -54,6 +54,7 @@ config PC_ACPI

>      select ACPI_X86

>      select ACPI_CPU_HOTPLUG

>      select ACPI_MEMORY_HOTPLUG

> +    select ACPI_VIOT

>      select SMBUS_EEPROM

>      select PFLASH_CFI01

>      depends on ACPI_SMBUS
Michael S. Tsirkin Oct. 8, 2021, 10:46 a.m. UTC | #3 
On Fri, Oct 01, 2021 at 06:33:55PM +0100, Jean-Philippe Brucker wrote:
> Allow instantiating a virtio-iommu device by adding an ACPI Virtual I/O

> Translation table (VIOT), which describes the relation between the

> virtio-iommu and the endpoints it manages.

> 

> Add a hotplug handler for virtio-iommu on x86 and set the necessary

> reserved region property. On x86, the [0xfee00000, 0xfeefffff] DMA

> region is reserved for MSIs. DMA transactions to this range either

> trigger IRQ remapping in the IOMMU or bypasses IOMMU translation.

> 

> Although virtio-iommu does not support IRQ remapping it must be informed

> of the reserved region so that it can forward DMA transactions targeting

> this region.

> 

> Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>


If you like, we can start by merging the pc bits when they are ready.
These are not widely used so have less of a chance to break someone's
setup.

> ---

>  include/hw/i386/pc.h |  2 ++

>  hw/i386/acpi-build.c |  5 +++++

>  hw/i386/pc.c         | 24 ++++++++++++++++++++++--

>  hw/i386/Kconfig      |  1 +

>  4 files changed, 30 insertions(+), 2 deletions(-)

> 

> diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h

> index 82cf7b7e30..f3ba1ee4c0 100644

> --- a/include/hw/i386/pc.h

> +++ b/include/hw/i386/pc.h

> @@ -45,6 +45,8 @@ typedef struct PCMachineState {

>      bool pit_enabled;

>      bool hpet_enabled;

>      bool default_bus_bypass_iommu;

> +    bool virtio_iommu;

> +    uint16_t virtio_iommu_bdf;

>      uint64_t max_fw_size;

>  

>      /* ACPI Memory hotplug IO base address */

> diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c

> index d1c28440f4..4e46585709 100644

> --- a/hw/i386/acpi-build.c

> +++ b/hw/i386/acpi-build.c

> @@ -71,6 +71,7 @@

>  

>  #include "hw/acpi/ipmi.h"

>  #include "hw/acpi/hmat.h"

> +#include "hw/acpi/viot.h"

>  

>  /* These are used to size the ACPI tables for -M pc-i440fx-1.7 and

>   * -M pc-i440fx-2.0.  Even if the actual amount of AML generated grows

> @@ -2593,6 +2594,10 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine)

>              build_dmar_q35(tables_blob, tables->linker, x86ms->oem_id,

>                             x86ms->oem_table_id);

>          }

> +    } else if (pcms->virtio_iommu) {

> +        acpi_add_table(table_offsets, tables_blob);

> +        build_viot(machine, tables_blob, tables->linker, pcms->virtio_iommu_bdf,

> +                   x86ms->oem_id, x86ms->oem_table_id);

>      }

>      if (machine->nvdimms_state->is_enabled) {

>          nvdimm_build_acpi(table_offsets, tables_blob, tables->linker,

> diff --git a/hw/i386/pc.c b/hw/i386/pc.c

> index 789ccb6ef4..31710bc4fb 100644

> --- a/hw/i386/pc.c

> +++ b/hw/i386/pc.c

> @@ -83,6 +83,7 @@

>  #include "hw/i386/intel_iommu.h"

>  #include "hw/net/ne2000-isa.h"

>  #include "standard-headers/asm-x86/bootparam.h"

> +#include "hw/virtio/virtio-iommu.h"

>  #include "hw/virtio/virtio-pmem-pci.h"

>  #include "hw/virtio/virtio-mem-pci.h"

>  #include "hw/mem/memory-device.h"

> @@ -1367,8 +1368,11 @@ static void pc_virtio_md_pci_unplug(HotplugHandler *hotplug_dev,

>  static void pc_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,

>                                            DeviceState *dev, Error **errp)

>  {

> -    if (object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) &&

> -        x86_iommu_get_default()) {

> +    PCMachineState *pcms = PC_MACHINE(hotplug_dev);

> +

> +    if ((object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) ||

> +         object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) &&

> +        (x86_iommu_get_default() || pcms->virtio_iommu)) {

>          error_setg(errp, "QEMU does not support multiple vIOMMUs "

>                     "for x86 yet.");

>          return;

> @@ -1381,6 +1385,15 @@ static void pc_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,

>      } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||

>                 object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI)) {

>          pc_virtio_md_pci_pre_plug(hotplug_dev, dev, errp);

> +    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) {

> +        /* Declare the APIC range as the reserved MSI region */

> +        char *resv_prop_str = g_strdup_printf("0xfee00000:0xfeefffff:%d",

> +                                              VIRTIO_IOMMU_RESV_MEM_T_MSI);

> +

> +        object_property_set_uint(OBJECT(dev), "len-reserved-regions", 1, errp);

> +        object_property_set_str(OBJECT(dev), "reserved-regions[0]",

> +                                resv_prop_str, errp);

> +        g_free(resv_prop_str);

>      }

>  }

>  

> @@ -1394,6 +1407,12 @@ static void pc_machine_device_plug_cb(HotplugHandler *hotplug_dev,

>      } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||

>                 object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI)) {

>          pc_virtio_md_pci_plug(hotplug_dev, dev, errp);

> +    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) {

> +        PCMachineState *pcms = PC_MACHINE(hotplug_dev);

> +        PCIDevice *pdev = PCI_DEVICE(dev);

> +

> +        pcms->virtio_iommu = true;

> +        pcms->virtio_iommu_bdf = pci_get_bdf(pdev);

>      }

>  }

>  

> @@ -1436,6 +1455,7 @@ static HotplugHandler *pc_get_hotplug_handler(MachineState *machine,

>          object_dynamic_cast(OBJECT(dev), TYPE_CPU) ||

>          object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||

>          object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI) ||

> +        object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI) ||

>          object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE)) {

>          return HOTPLUG_HANDLER(machine);

>      }

> diff --git a/hw/i386/Kconfig b/hw/i386/Kconfig

> index ddedcef0b2..13db05d557 100644

> --- a/hw/i386/Kconfig

> +++ b/hw/i386/Kconfig

> @@ -54,6 +54,7 @@ config PC_ACPI

>      select ACPI_X86

>      select ACPI_CPU_HOTPLUG

>      select ACPI_MEMORY_HOTPLUG

> +    select ACPI_VIOT

>      select SMBUS_EEPROM

>      select PFLASH_CFI01

>      depends on ACPI_SMBUS

> -- 

> 2.33.0
Jean-Philippe Brucker Oct. 8, 2021, 3:24 p.m. UTC | #4 
On Wed, Oct 06, 2021 at 09:19:54AM +0200, Igor Mammedov wrote:
> > @@ -1367,8 +1368,11 @@ static void pc_virtio_md_pci_unplug(HotplugHandler *hotplug_dev,

> >  static void pc_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,

> >                                            DeviceState *dev, Error **errp)

> >  {

> > -    if (object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) &&

> > -        x86_iommu_get_default()) {

> > +    PCMachineState *pcms = PC_MACHINE(hotplug_dev);

> > +

> > +    if ((object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) ||

> > +         object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) &&

> > +        (x86_iommu_get_default() || pcms->virtio_iommu)) {

> 

> this check is getting uglier,

> may be instead of introducing pcms->virtio_iommu boolean, better approach

> would be adding 'Device* PCMachineState::iommu' and setting it to IOMMU

> so the check would reduce to:

>       if ((object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) ||

>            object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)))

>       {

>           if (pcms->iommu)

>             err

>           else set pcms->iommu in plug handler or here

>       }

>       

> that also will let to cleanup/get rid of x86_iommu_[s|g]et_default()

> and x86_iommu_default 'global'.

> Maybe replace previous patch with one that would remove

> x86_iommu_[s|g]et_default().


Ok, I can't figure out a nice way to do this at the moment, will think
more about it. Callers of x86_iommu_get_default() still need a helper, and
most but not all assume that the returned object is for IRQ remapping. At
least adding Device *iommu to pcms should be nicer, but not sure about
removing the x86_iommu_get/set_default helpers.

Thanks,
Jean
diff mbox series

Patch

diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index 82cf7b7e30..f3ba1ee4c0 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -45,6 +45,8 @@  typedef struct PCMachineState {
     bool pit_enabled;
     bool hpet_enabled;
     bool default_bus_bypass_iommu;
+    bool virtio_iommu;
+    uint16_t virtio_iommu_bdf;
     uint64_t max_fw_size;
 
     /* ACPI Memory hotplug IO base address */
diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c
index d1c28440f4..4e46585709 100644
--- a/hw/i386/acpi-build.c
+++ b/hw/i386/acpi-build.c
@@ -71,6 +71,7 @@ 
 
 #include "hw/acpi/ipmi.h"
 #include "hw/acpi/hmat.h"
+#include "hw/acpi/viot.h"
 
 /* These are used to size the ACPI tables for -M pc-i440fx-1.7 and
  * -M pc-i440fx-2.0.  Even if the actual amount of AML generated grows
@@ -2593,6 +2594,10 @@  void acpi_build(AcpiBuildTables *tables, MachineState *machine)
             build_dmar_q35(tables_blob, tables->linker, x86ms->oem_id,
                            x86ms->oem_table_id);
         }
+    } else if (pcms->virtio_iommu) {
+        acpi_add_table(table_offsets, tables_blob);
+        build_viot(machine, tables_blob, tables->linker, pcms->virtio_iommu_bdf,
+                   x86ms->oem_id, x86ms->oem_table_id);
     }
     if (machine->nvdimms_state->is_enabled) {
         nvdimm_build_acpi(table_offsets, tables_blob, tables->linker,
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index 789ccb6ef4..31710bc4fb 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -83,6 +83,7 @@ 
 #include "hw/i386/intel_iommu.h"
 #include "hw/net/ne2000-isa.h"
 #include "standard-headers/asm-x86/bootparam.h"
+#include "hw/virtio/virtio-iommu.h"
 #include "hw/virtio/virtio-pmem-pci.h"
 #include "hw/virtio/virtio-mem-pci.h"
 #include "hw/mem/memory-device.h"
@@ -1367,8 +1368,11 @@  static void pc_virtio_md_pci_unplug(HotplugHandler *hotplug_dev,
 static void pc_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,
                                           DeviceState *dev, Error **errp)
 {
-    if (object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) &&
-        x86_iommu_get_default()) {
+    PCMachineState *pcms = PC_MACHINE(hotplug_dev);
+
+    if ((object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE) ||
+         object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) &&
+        (x86_iommu_get_default() || pcms->virtio_iommu)) {
         error_setg(errp, "QEMU does not support multiple vIOMMUs "
                    "for x86 yet.");
         return;
@@ -1381,6 +1385,15 @@  static void pc_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,
     } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||
                object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI)) {
         pc_virtio_md_pci_pre_plug(hotplug_dev, dev, errp);
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) {
+        /* Declare the APIC range as the reserved MSI region */
+        char *resv_prop_str = g_strdup_printf("0xfee00000:0xfeefffff:%d",
+                                              VIRTIO_IOMMU_RESV_MEM_T_MSI);
+
+        object_property_set_uint(OBJECT(dev), "len-reserved-regions", 1, errp);
+        object_property_set_str(OBJECT(dev), "reserved-regions[0]",
+                                resv_prop_str, errp);
+        g_free(resv_prop_str);
     }
 }
 
@@ -1394,6 +1407,12 @@  static void pc_machine_device_plug_cb(HotplugHandler *hotplug_dev,
     } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||
                object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI)) {
         pc_virtio_md_pci_plug(hotplug_dev, dev, errp);
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) {
+        PCMachineState *pcms = PC_MACHINE(hotplug_dev);
+        PCIDevice *pdev = PCI_DEVICE(dev);
+
+        pcms->virtio_iommu = true;
+        pcms->virtio_iommu_bdf = pci_get_bdf(pdev);
     }
 }
 
@@ -1436,6 +1455,7 @@  static HotplugHandler *pc_get_hotplug_handler(MachineState *machine,
         object_dynamic_cast(OBJECT(dev), TYPE_CPU) ||
         object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_PMEM_PCI) ||
         object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MEM_PCI) ||
+        object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI) ||
         object_dynamic_cast(OBJECT(dev), TYPE_X86_IOMMU_DEVICE)) {
         return HOTPLUG_HANDLER(machine);
     }
diff --git a/hw/i386/Kconfig b/hw/i386/Kconfig
index ddedcef0b2..13db05d557 100644
--- a/hw/i386/Kconfig
+++ b/hw/i386/Kconfig
@@ -54,6 +54,7 @@  config PC_ACPI
     select ACPI_X86
     select ACPI_CPU_HOTPLUG
     select ACPI_MEMORY_HOTPLUG
+    select ACPI_VIOT
     select SMBUS_EEPROM
     select PFLASH_CFI01
     depends on ACPI_SMBUS