[RFC PATCH 0/2] Add region id support for PCI

[RFC PATCH 0/2] Add region id support for PCI

[Mykyta Poturai]

On Renesas R-Car platforms, the Region ID (RGID) feature encodes an
identifier into the upper bits of the physical address for every CPU memory
access. This address manipulations can mostly be done via system and domain
congirutaion, but for PCI bar it is not possible as some BARs are 32bit only.

This series introduces a platform hook that lets the PCI host
bridge driver rewrite BAR addresses at vPCI init time, and provides the
R-Car Gen4 implementation that adds the configured RGID into each
address.

This is sent as RFC to get early feedback on the approach: 
- Is a pci_ops callback the right place for this kind of address fixup?
- Is platform Kconfig a good place for RGID related options or is it better to
  put them in some other place.

Mykyta Poturai (2):
  pci: Allow platforms to modify BAR adresses
  plat/rcar: Add region id support for PCI

 xen/arch/arm/include/asm/pci.h    |  3 +++
 xen/arch/arm/include/asm/vpci.h   |  9 +++++++++
 xen/arch/arm/pci/pci-host-rcar4.c | 15 +++++++++++++++
 xen/arch/arm/platforms/Kconfig    | 21 +++++++++++++++++++++
 xen/arch/arm/vpci.c               | 12 ++++++++++++
 xen/arch/x86/include/asm/vpci.h   |  6 ++++++
 xen/drivers/vpci/header.c         |  2 ++
 7 files changed, 68 insertions(+)

-- 
2.51.2

[RFC PATCH 2/2] plat/rcar: Add region id support for PCI

[Mykyta Poturai]

With Region ID enabled, all CPU memory accesses need to have rgid bits
set in the physical address. This creates a problem for PCI BAR
accesses, as it would require all BARs to be 64bit. Implement fixup_bar
callback to add rgid bits to the address before mapping it to the
guests.

Add Kconfig options to enable region id support and set the rgid value
and physical address space size.

Signed-off-by: Mykyta Poturai <mykyta_poturai@epam.com>
---
 xen/arch/arm/pci/pci-host-rcar4.c | 15 +++++++++++++++
 xen/arch/arm/platforms/Kconfig    | 21 +++++++++++++++++++++
 2 files changed, 36 insertions(+)

diff --git a/xen/arch/arm/pci/pci-host-rcar4.c b/xen/arch/arm/pci/pci-host-rcar4.c
index e1e8eb0ee1..50400c04c4 100644
--- a/xen/arch/arm/pci/pci-host-rcar4.c
+++ b/xen/arch/arm/pci/pci-host-rcar4.c
@@ -43,6 +43,12 @@ struct rcar4_pcie_priv {
     DECLARE_BITMAP(osid_regs, NUM_OSID_REGS);
 };
 
+#define ULL(X) _AC(X, ULL)
+#define MADDR_RGID(a)    (ULL(a) << CONFIG_RCAR_PA_BITS)
+#define MADDR_PA_MASK    ((1ULL << CONFIG_RCAR_PA_BITS) - 1)
+
+#define MADDR_ENCODE_RGID(a)    (MADDR_RGID(CONFIG_RCAR_RGID) | (a))
+
 /*
  * PCI host bridges often have different ways to access the root and child
  * bus config spaces:
@@ -61,6 +67,14 @@ static int __init rcar4_child_cfg_reg_index(struct dt_device_node *np)
     return dt_property_match_string(np, "reg-names", "config");
 }
 
+static void rcar4_pcie_fixup_bar(struct pci_host_bridge *bridge,
+                                 unsigned int bar_num,
+                                 paddr_t *addr)
+{
+    if ( IS_ENABLED(CONFIG_RCAR_REGION_ID_SUPPORT) )
+        *addr = MADDR_ENCODE_RGID(*addr);
+}
+
 /* ECAM ops */
 static const struct pci_ecam_ops rcar4_pcie_ops = {
     .bus_shift  = 20,
@@ -71,6 +85,7 @@ static const struct pci_ecam_ops rcar4_pcie_ops = {
         .write                  = pci_generic_config_write,
         .need_p2m_hwdom_mapping = pci_ecam_need_p2m_hwdom_mapping,
         .init_bus_range         = pci_generic_init_bus_range,
+        .fixup_bar              = rcar4_pcie_fixup_bar,
     }
 };
 
diff --git a/xen/arch/arm/platforms/Kconfig b/xen/arch/arm/platforms/Kconfig
index 888d0b85d5..db096952c8 100644
--- a/xen/arch/arm/platforms/Kconfig
+++ b/xen/arch/arm/platforms/Kconfig
@@ -64,6 +64,27 @@ config NO_PLAT
 
 endchoice
 
+menu "RCar Region ID Support"
+    visible if RCAR4
+
+config RCAR_REGION_ID_SUPPORT
+	bool "Renesas Region ID support for R-Car Gen4 platforms" if EXPERT
+	depends on RCAR4
+	help
+	  Enable experimental Region ID support for R-Car Gen4 platforms
+
+config RCAR_RGID
+	int "Region ID encoded in physical address"
+	depends on RCAR_REGION_ID_SUPPORT
+	default 0
+
+config RCAR_PA_BITS
+	int "Physical address space size"
+	depends on RCAR_REGION_ID_SUPPORT
+	default 36
+
+endmenu
+
 config ALL64_PLAT
 	bool
 	default (ALL_PLAT && ARM_64)
-- 
2.51.2

Re: [RFC PATCH 2/2] plat/rcar: Add region id support for PCI

[Jan Beulich]

On 21.04.2026 09:57, Mykyta Poturai wrote:
> With Region ID enabled, all CPU memory accesses need to have rgid bits
> set in the physical address. This creates a problem for PCI BAR
> accesses, as it would require all BARs to be 64bit.

I don't understand this. CPU accesses of addresses read / derived from
BARs can be massaged in any way the OS likes. That doesn't require the
BARs to be 64-bit. Are you trying to arrange for RGID to be transparent
to guests?

Jan

Re: [RFC PATCH 2/2] plat/rcar: Add region id support for PCI

[Mykyta Poturai]

On 4/21/26 11:28, Jan Beulich wrote:
> On 21.04.2026 09:57, Mykyta Poturai wrote:
>> With Region ID enabled, all CPU memory accesses need to have rgid bits
>> set in the physical address. This creates a problem for PCI BAR
>> accesses, as it would require all BARs to be 64bit.
> 
> I don't understand this. CPU accesses of addresses read / derived from
> BARs can be massaged in any way the OS likes. That doesn't require the
> BARs to be 64-bit. Are you trying to arrange for RGID to be transparent
> to guests?
> 
> Jan

Yes, making it transparent to guests is the goal.

-- 
Mykyta

Re: [RFC PATCH 2/2] plat/rcar: Add region id support for PCI

[Julien Grall]

Hi Mykyta,

On 21/04/2026 08:57, Mykyta Poturai wrote:
> diff --git a/xen/arch/arm/platforms/Kconfig b/xen/arch/arm/platforms/Kconfig
> index 888d0b85d5..db096952c8 100644
> --- a/xen/arch/arm/platforms/Kconfig
> +++ b/xen/arch/arm/platforms/Kconfig
> @@ -64,6 +64,27 @@ config NO_PLAT
>   
>   endchoice
>   
> +menu "RCar Region ID Support"
> +    visible if RCAR4
> +
> +config RCAR_REGION_ID_SUPPORT
> +	bool "Renesas Region ID support for R-Car Gen4 platforms" if EXPERT
> +	depends on RCAR4
> +	help
> +	  Enable experimental Region ID support for R-Car Gen4 platforms
> +
> +config RCAR_RGID
> +	int "Region ID encoded in physical address"
> +	depends on RCAR_REGION_ID_SUPPORT
> +	default 0
> +
> +config RCAR_PA_BITS
> +	int "Physical address space size"
> +	depends on RCAR_REGION_ID_SUPPORT
> +	default 36

The 3 configs above implies that the value will change per-board. Is 
that correct? If so, this should be described in the firmware table so a 
single Xen binary can boot on all the platforms.

Cheers,

-- 
Julien Grall

Re: [RFC PATCH 2/2] plat/rcar: Add region id support for PCI

[Oleksandr Tyshchenko]

On 4/21/26 10:57, Mykyta Poturai wrote:

Hello Mykyta

> With Region ID enabled, all CPU memory accesses need to have rgid bits
> set in the physical address. This creates a problem for PCI BAR
> accesses, as it would require all BARs to be 64bit. Implement fixup_bar
> callback to add rgid bits to the address before mapping it to the
> guests.

For context. When Region ID support is enabled, the RGID value is 
encoded in the upper bits of physical addresses for both RAM and device 
MMIO in host device tree. From Xen's perspective, these are just regular 
physical addresses, and ideally Xen should not need to be RGID-aware at 
all. But I understand what this series is trying to solve.

It is worth mentioning that when Region ID support is enabled, there 
will be no 32-bit addresses in the host device tree anymore, since 
encoding the RGID effectively moves every address beyond the 4GB 
boundary (unless the RGID is 0). This might break any code in Xen that 
relies on 32-bit address space for some reason, e.g. 
HOST_ITS_WORKAROUND_32BIT_ADDR which requires 32-bit addresses for 
in-memory data structures (refer gicv3_its_enable_quirk_gen4()).


> 
> Add Kconfig options to enable region id support and set the rgid value
> and physical address space size.
> 
> Signed-off-by: Mykyta Poturai <mykyta_poturai@epam.com>
> ---
>   xen/arch/arm/pci/pci-host-rcar4.c | 15 +++++++++++++++
>   xen/arch/arm/platforms/Kconfig    | 21 +++++++++++++++++++++
>   2 files changed, 36 insertions(+)
> 
> diff --git a/xen/arch/arm/pci/pci-host-rcar4.c b/xen/arch/arm/pci/pci-host-rcar4.c
> index e1e8eb0ee1..50400c04c4 100644
> --- a/xen/arch/arm/pci/pci-host-rcar4.c
> +++ b/xen/arch/arm/pci/pci-host-rcar4.c
> @@ -43,6 +43,12 @@ struct rcar4_pcie_priv {
>       DECLARE_BITMAP(osid_regs, NUM_OSID_REGS);
>   };
>   
> +#define ULL(X) _AC(X, ULL)
> +#define MADDR_RGID(a)    (ULL(a) << CONFIG_RCAR_PA_BITS)
> +#define MADDR_PA_MASK    ((1ULL << CONFIG_RCAR_PA_BITS) - 1)

NIT: MADDR_PA_MASK is defined but never used ...

> +
> +#define MADDR_ENCODE_RGID(a)    (MADDR_RGID(CONFIG_RCAR_RGID) | (a))
> +
>   /*
>    * PCI host bridges often have different ways to access the root and child
>    * bus config spaces:
> @@ -61,6 +67,14 @@ static int __init rcar4_child_cfg_reg_index(struct dt_device_node *np)
>       return dt_property_match_string(np, "reg-names", "config");
>   }
>   
> +static void rcar4_pcie_fixup_bar(struct pci_host_bridge *bridge,
> +                                 unsigned int bar_num,
> +                                 paddr_t *addr)
> +{
> +    if ( IS_ENABLED(CONFIG_RCAR_REGION_ID_SUPPORT) )
> +        *addr = MADDR_ENCODE_RGID(*addr);
> +}
> +
>   /* ECAM ops */
>   static const struct pci_ecam_ops rcar4_pcie_ops = {
>       .bus_shift  = 20,
> @@ -71,6 +85,7 @@ static const struct pci_ecam_ops rcar4_pcie_ops = {
>           .write                  = pci_generic_config_write,
>           .need_p2m_hwdom_mapping = pci_ecam_need_p2m_hwdom_mapping,
>           .init_bus_range         = pci_generic_init_bus_range,
> +        .fixup_bar              = rcar4_pcie_fixup_bar,
>       }
>   };
>   
> diff --git a/xen/arch/arm/platforms/Kconfig b/xen/arch/arm/platforms/Kconfig
> index 888d0b85d5..db096952c8 100644
> --- a/xen/arch/arm/platforms/Kconfig
> +++ b/xen/arch/arm/platforms/Kconfig
> @@ -64,6 +64,27 @@ config NO_PLAT
>   
>   endchoice
>   
> +menu "RCar Region ID Support"
> +    visible if RCAR4
> +
> +config RCAR_REGION_ID_SUPPORT
> +	bool "Renesas Region ID support for R-Car Gen4 platforms" if EXPERT
> +	depends on RCAR4
> +	help
> +	  Enable experimental Region ID support for R-Car Gen4 platforms
> +
> +config RCAR_RGID
> +	int "Region ID encoded in physical address"
> +	depends on RCAR_REGION_ID_SUPPORT
> +	default 0
> +
> +config RCAR_PA_BITS
> +	int "Physical address space size"
> +	depends on RCAR_REGION_ID_SUPPORT
> +	default 36

Options RCAR_RGID and RCAR_PA_BITS lack range constraints. A 
misconfigured options might cause undefined shift behavior at compile 
time. However, I see a valid request in a separate email to consider 
using a firmware table to pass these values, so just ignore this comment.

*****

I was wondering whether the RGID could be extracted at runtime from an
address that already carries it. Do addresses in the host device tree 
(e.g., device MMIO ranges, host bridge windows, etc) already contain 
RGID bits? If so, then in rcar4_pcie_fixup_bar() you could obtain the 
RGID from the bridge instance that is already passed as a parameter.

Something like below:

#define MADDR_GET_RGID(addr)  ((addr) >> CONFIG_RCAR_PA_BITS)

paddr_t rgid = 
MADDR_GET_RGID(bridge->some_hw_addr_as_it_is_specified_in_dt);
*addr = (rgid << CONFIG_RCAR_PA_BITS) | (*addr & MADDR_PA_MASK);

This would eliminate CONFIG_RCAR_RGID entirely by deducing it from
existing platform data rather than requiring it to be passed explicitly
via the device tree.

As for CONFIG_RCAR_PA_BITS, I think it could remain as-is since it is
constant across an SoC generation, unless I am mistaken. As far as I am 
aware, on Gen4 the RGID occupies 4 bits (0–15) starting at bit 36, and 
on Gen5 it occupies 5 bits (0–31) starting at bit 43.
Since Xen can identify the hardware it is running on (thanks to the 
compatible string), the appropriate #define-s could be applied.

Would the above approach work, or is it better to pass all these values 
explicitly?

Please note, I am not 100% sure whether this is the right and reliable 
approach, but it should be considered.



> +
> +endmenu
> +
>   config ALL64_PLAT
>   	bool
>   	default (ALL_PLAT && ARM_64)

[RFC PATCH 1/2] pci: Allow platforms to modify BAR adresses

[Mykyta Poturai]

This patch is a preparatory work for adding Region ID support on Renesas
R-Car series boards. Add new host bridge op "fixup_bar" that allows
platforms to modify BAR addresses before they are mapped.

Because x86 don't have support for PCI Host Bridge drivers, add another
level of indirection in form of platform_pci_fixup_bar() function, that
will call host bridge op on ARM and do nothing on x86.

Signed-off-by: Mykyta Poturai <mykyta_poturai@epam.com>
---
 xen/arch/arm/include/asm/pci.h  |  3 +++
 xen/arch/arm/include/asm/vpci.h |  9 +++++++++
 xen/arch/arm/vpci.c             | 12 ++++++++++++
 xen/arch/x86/include/asm/vpci.h |  6 ++++++
 xen/drivers/vpci/header.c       |  2 ++
 5 files changed, 32 insertions(+)

diff --git a/xen/arch/arm/include/asm/pci.h b/xen/arch/arm/include/asm/pci.h
index 7c3211823f..398a4eb746 100644
--- a/xen/arch/arm/include/asm/pci.h
+++ b/xen/arch/arm/include/asm/pci.h
@@ -80,6 +80,9 @@ struct pci_ops {
     void (*init_bus_range)(struct dt_device_node *dev,
                            struct pci_host_bridge *bridge,
                            struct pci_config_window *cfg);
+    void (*fixup_bar)(struct pci_host_bridge *bridge,
+                      unsigned int bar_num,
+                      paddr_t *addr);
 };
 
 /*
diff --git a/xen/arch/arm/include/asm/vpci.h b/xen/arch/arm/include/asm/vpci.h
index 0cc6f5a105..f5c817a51c 100644
--- a/xen/arch/arm/include/asm/vpci.h
+++ b/xen/arch/arm/include/asm/vpci.h
@@ -16,6 +16,10 @@ struct vpci_arch_msix_entry {
 
 int domain_vpci_init(struct domain *d);
 unsigned int domain_vpci_get_num_mmio_handlers(struct domain *d);
+
+void platform_pci_fixup_bar(const struct pci_dev *pdev, unsigned int bar_num,
+                            paddr_t *addr);
+
 #else
 static inline int domain_vpci_init(struct domain *d)
 {
@@ -26,6 +30,11 @@ static inline unsigned int domain_vpci_get_num_mmio_handlers(struct domain *d)
 {
     return 0;
 }
+
+static inline void platform_pci_fixup_bar(const struct pci_dev *pdev,
+                                          unsigned int bar_num,
+                                          paddr_t *addr)
+{}
 #endif /* CONFIG_HAS_VPCI */
 
 #endif /* ARM_VPCI_H */
diff --git a/xen/arch/arm/vpci.c b/xen/arch/arm/vpci.c
index d41aa383a8..ec6efec22e 100644
--- a/xen/arch/arm/vpci.c
+++ b/xen/arch/arm/vpci.c
@@ -189,6 +189,18 @@ unsigned int domain_vpci_get_num_mmio_handlers(struct domain *d)
     return 1;
 }
 
+void platform_pci_fixup_bar(const struct pci_dev *pdev,
+                                          unsigned int bar_num,
+                                          paddr_t *addr)
+{
+    struct pci_host_bridge *bridge = pci_find_host_bridge(pdev->sbdf.seg, pdev->sbdf.bus);
+
+    if ( bridge->ops->fixup_bar )
+    {
+        bridge->ops->fixup_bar(bridge, bar_num, addr);
+    }
+}
+
 /*
  * Local variables:
  * mode: C
diff --git a/xen/arch/x86/include/asm/vpci.h b/xen/arch/x86/include/asm/vpci.h
index c501ff1709..a05b70abbf 100644
--- a/xen/arch/x86/include/asm/vpci.h
+++ b/xen/arch/x86/include/asm/vpci.h
@@ -16,6 +16,12 @@ struct vpci_arch_msix_entry {
     int pirq;
 };
 
+/* X86 does not require PCI BAR modifications */
+static inline void platform_pci_fixup_bar(const struct pci_dev *pdev,
+                                          unsigned int bar_num,
+                                          paddr_t *addr)
+{}
+
 #endif /* X86_VPCI_H */
 
 /*
diff --git a/xen/drivers/vpci/header.c b/xen/drivers/vpci/header.c
index a760d8c32f..d89e43354c 100644
--- a/xen/drivers/vpci/header.c
+++ b/xen/drivers/vpci/header.c
@@ -882,6 +882,8 @@ int vpci_init_header(struct pci_dev *pdev)
         bars[i].size = size;
         bars[i].prefetchable = val & PCI_BASE_ADDRESS_MEM_PREFETCH;
 
+        platform_pci_fixup_bar(pdev, i, &bars[i].addr);
+
         rc = vpci_add_register(pdev->vpci,
                                is_hwdom ? vpci_hw_read32 : guest_mem_bar_read,
                                is_hwdom ? bar_write : guest_mem_bar_write,
-- 
2.51.2

Re: [RFC PATCH 1/2] pci: Allow platforms to modify BAR adresses

[Jan Beulich]

On 21.04.2026 09:57, Mykyta Poturai wrote:
> This patch is a preparatory work for adding Region ID support on Renesas
> R-Car series boards. Add new host bridge op "fixup_bar" that allows
> platforms to modify BAR addresses before they are mapped.
> 
> Because x86 don't have support for PCI Host Bridge drivers, add another
> level of indirection in form of platform_pci_fixup_bar() function, that
> will call host bridge op on ARM and do nothing on x86.
> 
> Signed-off-by: Mykyta Poturai <mykyta_poturai@epam.com>
> ---
>  xen/arch/arm/include/asm/pci.h  |  3 +++
>  xen/arch/arm/include/asm/vpci.h |  9 +++++++++
>  xen/arch/arm/vpci.c             | 12 ++++++++++++
>  xen/arch/x86/include/asm/vpci.h |  6 ++++++
>  xen/drivers/vpci/header.c       |  2 ++
>  5 files changed, 32 insertions(+)
> 
> diff --git a/xen/arch/arm/include/asm/pci.h b/xen/arch/arm/include/asm/pci.h
> index 7c3211823f..398a4eb746 100644
> --- a/xen/arch/arm/include/asm/pci.h
> +++ b/xen/arch/arm/include/asm/pci.h
> @@ -80,6 +80,9 @@ struct pci_ops {
>      void (*init_bus_range)(struct dt_device_node *dev,
>                             struct pci_host_bridge *bridge,
>                             struct pci_config_window *cfg);
> +    void (*fixup_bar)(struct pci_host_bridge *bridge,
> +                      unsigned int bar_num,
> +                      paddr_t *addr);

I'm not an Arm maintainer, but if such a hook - used only when a certain
CONFIG_* is active, as per patch 2 - was introduced in common or x86 code,
I'd ask for the hook to also be conditional. Doing so avoids accidental use
without ...

> --- a/xen/arch/arm/vpci.c
> +++ b/xen/arch/arm/vpci.c
> @@ -189,6 +189,18 @@ unsigned int domain_vpci_get_num_mmio_handlers(struct domain *d)
>      return 1;
>  }
>  
> +void platform_pci_fixup_bar(const struct pci_dev *pdev,
> +                                          unsigned int bar_num,
> +                                          paddr_t *addr)
> +{
> +    struct pci_host_bridge *bridge = pci_find_host_bridge(pdev->sbdf.seg, pdev->sbdf.bus);
> +
> +    if ( bridge->ops->fixup_bar )
> +    {
> +        bridge->ops->fixup_bar(bridge, bar_num, addr);
> +    }

... such a conditional around it.

Jan

Re: [RFC PATCH 1/2] pci: Allow platforms to modify BAR adresses

[Roger Pau Monné]

On Tue, Apr 21, 2026 at 07:57:14AM +0000, Mykyta Poturai wrote:
> This patch is a preparatory work for adding Region ID support on Renesas
> R-Car series boards. Add new host bridge op "fixup_bar" that allows
> platforms to modify BAR addresses before they are mapped.
> 
> Because x86 don't have support for PCI Host Bridge drivers, add another
> level of indirection in form of platform_pci_fixup_bar() function, that
> will call host bridge op on ARM and do nothing on x86.
> 
> Signed-off-by: Mykyta Poturai <mykyta_poturai@epam.com>
> ---
>  xen/arch/arm/include/asm/pci.h  |  3 +++
>  xen/arch/arm/include/asm/vpci.h |  9 +++++++++
>  xen/arch/arm/vpci.c             | 12 ++++++++++++
>  xen/arch/x86/include/asm/vpci.h |  6 ++++++
>  xen/drivers/vpci/header.c       |  2 ++
>  5 files changed, 32 insertions(+)
> 
> diff --git a/xen/arch/arm/include/asm/pci.h b/xen/arch/arm/include/asm/pci.h
> index 7c3211823f..398a4eb746 100644
> --- a/xen/arch/arm/include/asm/pci.h
> +++ b/xen/arch/arm/include/asm/pci.h
> @@ -80,6 +80,9 @@ struct pci_ops {
>      void (*init_bus_range)(struct dt_device_node *dev,
>                             struct pci_host_bridge *bridge,
>                             struct pci_config_window *cfg);
> +    void (*fixup_bar)(struct pci_host_bridge *bridge,
> +                      unsigned int bar_num,
> +                      paddr_t *addr);
>  };
>  
>  /*
> diff --git a/xen/arch/arm/include/asm/vpci.h b/xen/arch/arm/include/asm/vpci.h
> index 0cc6f5a105..f5c817a51c 100644
> --- a/xen/arch/arm/include/asm/vpci.h
> +++ b/xen/arch/arm/include/asm/vpci.h
> @@ -16,6 +16,10 @@ struct vpci_arch_msix_entry {
>  
>  int domain_vpci_init(struct domain *d);
>  unsigned int domain_vpci_get_num_mmio_handlers(struct domain *d);
> +
> +void platform_pci_fixup_bar(const struct pci_dev *pdev, unsigned int bar_num,
> +                            paddr_t *addr);
> +
>  #else
>  static inline int domain_vpci_init(struct domain *d)
>  {
> @@ -26,6 +30,11 @@ static inline unsigned int domain_vpci_get_num_mmio_handlers(struct domain *d)
>  {
>      return 0;
>  }
> +
> +static inline void platform_pci_fixup_bar(const struct pci_dev *pdev,
> +                                          unsigned int bar_num,
> +                                          paddr_t *addr)
> +{}
>  #endif /* CONFIG_HAS_VPCI */
>  
>  #endif /* ARM_VPCI_H */
> diff --git a/xen/arch/arm/vpci.c b/xen/arch/arm/vpci.c
> index d41aa383a8..ec6efec22e 100644
> --- a/xen/arch/arm/vpci.c
> +++ b/xen/arch/arm/vpci.c
> @@ -189,6 +189,18 @@ unsigned int domain_vpci_get_num_mmio_handlers(struct domain *d)
>      return 1;
>  }
>  
> +void platform_pci_fixup_bar(const struct pci_dev *pdev,
> +                                          unsigned int bar_num,
> +                                          paddr_t *addr)
> +{
> +    struct pci_host_bridge *bridge = pci_find_host_bridge(pdev->sbdf.seg, pdev->sbdf.bus);
> +
> +    if ( bridge->ops->fixup_bar )
> +    {
> +        bridge->ops->fixup_bar(bridge, bar_num, addr);
> +    }
> +}
> +
>  /*
>   * Local variables:
>   * mode: C
> diff --git a/xen/arch/x86/include/asm/vpci.h b/xen/arch/x86/include/asm/vpci.h
> index c501ff1709..a05b70abbf 100644
> --- a/xen/arch/x86/include/asm/vpci.h
> +++ b/xen/arch/x86/include/asm/vpci.h
> @@ -16,6 +16,12 @@ struct vpci_arch_msix_entry {
>      int pirq;
>  };
>  
> +/* X86 does not require PCI BAR modifications */
> +static inline void platform_pci_fixup_bar(const struct pci_dev *pdev,
> +                                          unsigned int bar_num,
> +                                          paddr_t *addr)
> +{}
> +
>  #endif /* X86_VPCI_H */
>  
>  /*
> diff --git a/xen/drivers/vpci/header.c b/xen/drivers/vpci/header.c
> index a760d8c32f..d89e43354c 100644
> --- a/xen/drivers/vpci/header.c
> +++ b/xen/drivers/vpci/header.c
> @@ -882,6 +882,8 @@ int vpci_init_header(struct pci_dev *pdev)
>          bars[i].size = size;
>          bars[i].prefetchable = val & PCI_BASE_ADDRESS_MEM_PREFETCH;
>  
> +        platform_pci_fixup_bar(pdev, i, &bars[i].addr);

You do this for vpci_init_header(), but don't you also need to do it
for bar_write(), in case dom0 decides to relocate the BAR?

Also - I would assume that both the firmware and dom0 are aware of
this limitation, and hence wonder why you need this fixup in Xen?

Thanks, Roger.