[PATCH v2 0/2] PCI: Unify domain emulation

[PATCH v2 0/2] PCI: Unify domain emulation

[Dan Williams]

Changes since v1 [1]:
- Rebase on v6.18-rc2
- Support callers supplying both a hint and a range for the Hyper-V hint
  + fallback case (Michael)
- Add comment explaining domain number 0 vs Gen1 VMs, and domain values
  greater than U16_MAX concerns (Michael)
- Leave the VMD status quo comment about requesting domain numbers >
  U16_MAX

[1]: http://lore.kernel.org/20250716160835.680486-1-dan.j.williams@intel.com

The PCI/TSM effort created a sample driver to test ABI flows
(samples/devsec/ [2]). Suzuki observed that it only worked on x86 due to
its dependency on CONFIG_PCI_DOMAINS_GENERIC=n. I.e. an unfortunate
restriction for what should be an architecture agnostic test framework.

Introduce a new pci_bus_find_emul_domain_nr() helper that all "soft"
host-bridge drivers can share and hide the CONFIG_PCI_DOMAINS_GENERIC
details behind that helper.

[2]: https://git.kernel.org/pub/scm/linux/kernel/git/devsec/tsm.git/commit/?id=0e16ce0b9c64

Dan Williams (2):
  PCI: Enable host bridge emulation for PCI_DOMAINS_GENERIC platforms
  PCI: vmd: Switch to pci_bus_find_emul_domain_nr()

 include/linux/pci.h                 |  7 ++++
 drivers/pci/controller/pci-hyperv.c | 62 +++++------------------------
 drivers/pci/controller/vmd.c        | 40 ++++++++-----------
 drivers/pci/pci.c                   | 24 ++++++++++-
 drivers/pci/probe.c                 |  8 +++-
 5 files changed, 63 insertions(+), 78 deletions(-)


base-commit: 211ddde0823f1442e4ad052a2f30f050145ccada
-- 
2.51.0

[PATCH v2 1/2] PCI: Enable host bridge emulation for PCI_DOMAINS_GENERIC platforms

[Dan Williams]

The ability to emulate a host bridge is useful not only for hardware PCI
controllers like CONFIG_VMD, or virtual PCI controllers like
CONFIG_PCI_HYPERV, but also for test and development scenarios like
CONFIG_SAMPLES_DEVSEC [1].

One stumbling block for defining CONFIG_SAMPLES_DEVSEC, a sample
implementation of a platform TSM for PCI Device Security, is the need to
accommodate PCI_DOMAINS_GENERIC architectures alongside x86 [2].

In support of supplementing the existing CONFIG_PCI_BRIDGE_EMUL
infrastructure for host bridges:

* Introduce pci_bus_find_emul_domain_nr() as a common way to find a free
  PCI domain number whether that is to reuse the existing dynamic
  allocation code in the !ACPI case, or to assign an unused domain above
  the last ACPI segment.

* Convert pci-hyperv to the new allocator so that the PCI core can
  unconditionally assume that bridge->domain_nr != PCI_DOMAIN_NR_NOT_SET
  is the dynamically allocated case.

A follow on patch can also convert vmd to the new scheme. Currently vmd
is limited to CONFIG_PCI_DOMAINS_GENERIC=n (x86) so, unlike pci-hyperv,
it does not immediately conflict with this new
pci_bus_find_emul_domain_nr() mechanism.

Link: http://lore.kernel.org/174107249038.1288555.12362100502109498455.stgit@dwillia2-xfh.jf.intel.com [1]
Reported-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Closes: http://lore.kernel.org/20250311144601.145736-3-suzuki.poulose@arm.com
Cc: Lorenzo Pieralisi <lpieralisi@kernel.org>
Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Cc: Rob Herring <robh@kernel.org>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: "K. Y. Srinivasan" <kys@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Cc: Wei Liu <wei.liu@kernel.org>
Cc: Dexuan Cui <decui@microsoft.com>
Tested-by: Suzuki K Poulose <suzuki.poulose@arm.com>
[michael: maintain compatibility with userspace that expects 16-bit ids]
Cc: Michael Kelley <mhklinux@outlook.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 include/linux/pci.h                 |  7 ++++
 drivers/pci/controller/pci-hyperv.c | 62 +++++------------------------
 drivers/pci/pci.c                   | 24 ++++++++++-
 drivers/pci/probe.c                 |  8 +++-
 4 files changed, 46 insertions(+), 55 deletions(-)

diff --git a/include/linux/pci.h b/include/linux/pci.h
index d1fdf81fbe1e..1ef1535802b0 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -1956,10 +1956,17 @@ DEFINE_GUARD(pci_dev, struct pci_dev *, pci_dev_lock(_T), pci_dev_unlock(_T))
  */
 #ifdef CONFIG_PCI_DOMAINS
 extern int pci_domains_supported;
+int pci_bus_find_emul_domain_nr(u32 hint, u32 min, u32 max);
+void pci_bus_release_emul_domain_nr(int domain_nr);
 #else
 enum { pci_domains_supported = 0 };
 static inline int pci_domain_nr(struct pci_bus *bus) { return 0; }
 static inline int pci_proc_domain(struct pci_bus *bus) { return 0; }
+static inline int pci_bus_find_emul_domain_nr(u32 hint, u32 min, u32 max)
+{
+	return 0;
+}
+static inline void pci_bus_release_emul_domain_nr(int domain_nr) { }
 #endif /* CONFIG_PCI_DOMAINS */
 
 /*
diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c
index 146b43981b27..64f68efaf547 100644
--- a/drivers/pci/controller/pci-hyperv.c
+++ b/drivers/pci/controller/pci-hyperv.c
@@ -3696,48 +3696,6 @@ static int hv_send_resources_released(struct hv_device *hdev)
 	return 0;
 }
 
-#define HVPCI_DOM_MAP_SIZE (64 * 1024)
-static DECLARE_BITMAP(hvpci_dom_map, HVPCI_DOM_MAP_SIZE);
-
-/*
- * PCI domain number 0 is used by emulated devices on Gen1 VMs, so define 0
- * as invalid for passthrough PCI devices of this driver.
- */
-#define HVPCI_DOM_INVALID 0
-
-/**
- * hv_get_dom_num() - Get a valid PCI domain number
- * Check if the PCI domain number is in use, and return another number if
- * it is in use.
- *
- * @dom: Requested domain number
- *
- * return: domain number on success, HVPCI_DOM_INVALID on failure
- */
-static u16 hv_get_dom_num(u16 dom)
-{
-	unsigned int i;
-
-	if (test_and_set_bit(dom, hvpci_dom_map) == 0)
-		return dom;
-
-	for_each_clear_bit(i, hvpci_dom_map, HVPCI_DOM_MAP_SIZE) {
-		if (test_and_set_bit(i, hvpci_dom_map) == 0)
-			return i;
-	}
-
-	return HVPCI_DOM_INVALID;
-}
-
-/**
- * hv_put_dom_num() - Mark the PCI domain number as free
- * @dom: Domain number to be freed
- */
-static void hv_put_dom_num(u16 dom)
-{
-	clear_bit(dom, hvpci_dom_map);
-}
-
 /**
  * hv_pci_probe() - New VMBus channel probe, for a root PCI bus
  * @hdev:	VMBus's tracking struct for this root PCI bus
@@ -3750,9 +3708,9 @@ static int hv_pci_probe(struct hv_device *hdev,
 {
 	struct pci_host_bridge *bridge;
 	struct hv_pcibus_device *hbus;
-	u16 dom_req, dom;
+	int ret, dom;
+	u16 dom_req;
 	char *name;
-	int ret;
 
 	bridge = devm_pci_alloc_host_bridge(&hdev->device, 0);
 	if (!bridge)
@@ -3779,11 +3737,14 @@ static int hv_pci_probe(struct hv_device *hdev,
 	 * PCI bus (which is actually emulated by the hypervisor) is domain 0.
 	 * (2) There will be no overlap between domains (after fixing possible
 	 * collisions) in the same VM.
+	 *
+	 * Because Gen1 VMs use domain 0, don't allow picking domain 0 here,
+	 * even if bytes 4 and 5 of the instance GUID are both zero. For wider
+	 * userspace compatibility, limit the domain id to a 16-bit value.
 	 */
 	dom_req = hdev->dev_instance.b[5] << 8 | hdev->dev_instance.b[4];
-	dom = hv_get_dom_num(dom_req);
-
-	if (dom == HVPCI_DOM_INVALID) {
+	dom = pci_bus_find_emul_domain_nr(dom_req, 1, U16_MAX);
+	if (dom < 0) {
 		dev_err(&hdev->device,
 			"Unable to use dom# 0x%x or other numbers", dom_req);
 		ret = -EINVAL;
@@ -3917,7 +3878,7 @@ static int hv_pci_probe(struct hv_device *hdev,
 destroy_wq:
 	destroy_workqueue(hbus->wq);
 free_dom:
-	hv_put_dom_num(hbus->bridge->domain_nr);
+	pci_bus_release_emul_domain_nr(hbus->bridge->domain_nr);
 free_bus:
 	kfree(hbus);
 	return ret;
@@ -4042,8 +4003,6 @@ static void hv_pci_remove(struct hv_device *hdev)
 	irq_domain_remove(hbus->irq_domain);
 	irq_domain_free_fwnode(hbus->fwnode);
 
-	hv_put_dom_num(hbus->bridge->domain_nr);
-
 	kfree(hbus);
 }
 
@@ -4217,9 +4176,6 @@ static int __init init_hv_pci_drv(void)
 	if (ret)
 		return ret;
 
-	/* Set the invalid domain number's bit, so it will not be used */
-	set_bit(HVPCI_DOM_INVALID, hvpci_dom_map);
-
 	/* Initialize PCI block r/w interface */
 	hvpci_block_ops.read_block = hv_read_config_block;
 	hvpci_block_ops.write_block = hv_write_config_block;
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index b14dd064006c..64aed8705e91 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -6656,9 +6656,31 @@ static void pci_no_domains(void)
 #endif
 }
 
+#ifdef CONFIG_PCI_DOMAINS
+static DEFINE_IDA(pci_domain_nr_dynamic_ida);
+
+/**
+ * pci_bus_find_emul_domain_nr() - allocate a PCI domain number per constraints
+ * @hint: desired domain, 0 if any id in the range of @min to @max is acceptable
+ * @min: minimum allowable domain
+ * @max: maximum allowable domain, no ids higher than INT_MAX will be returned
+ */
+int pci_bus_find_emul_domain_nr(u32 hint, u32 min, u32 max)
+{
+	return ida_alloc_range(&pci_domain_nr_dynamic_ida, max(hint, min), max,
+			       GFP_KERNEL);
+}
+EXPORT_SYMBOL_GPL(pci_bus_find_emul_domain_nr);
+
+void pci_bus_release_emul_domain_nr(int domain_nr)
+{
+	ida_free(&pci_domain_nr_dynamic_ida, domain_nr);
+}
+EXPORT_SYMBOL_GPL(pci_bus_release_emul_domain_nr);
+#endif
+
 #ifdef CONFIG_PCI_DOMAINS_GENERIC
 static DEFINE_IDA(pci_domain_nr_static_ida);
-static DEFINE_IDA(pci_domain_nr_dynamic_ida);
 
 static void of_pci_reserve_static_domain_nr(void)
 {
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index 0ce98e18b5a8..5e101ced00a5 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -650,6 +650,11 @@ static void pci_release_host_bridge_dev(struct device *dev)
 
 	pci_free_resource_list(&bridge->windows);
 	pci_free_resource_list(&bridge->dma_ranges);
+
+	/* Host bridges only have domain_nr set in the emulation case */
+	if (bridge->domain_nr != PCI_DOMAIN_NR_NOT_SET)
+		pci_bus_release_emul_domain_nr(bridge->domain_nr);
+
 	kfree(bridge);
 }
 
@@ -1130,7 +1135,8 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge)
 	device_del(&bridge->dev);
 free:
 #ifdef CONFIG_PCI_DOMAINS_GENERIC
-	pci_bus_release_domain_nr(parent, bus->domain_nr);
+	if (bridge->domain_nr == PCI_DOMAIN_NR_NOT_SET)
+		pci_bus_release_domain_nr(parent, bus->domain_nr);
 #endif
 	if (bus_registered)
 		put_device(&bus->dev);
-- 
2.51.0

RE: [PATCH v2 1/2] PCI: Enable host bridge emulation for PCI_DOMAINS_GENERIC platforms

[Michael Kelley]

From: Dan Williams <dan.j.williams@intel.com> Sent: Friday, October 24, 2025 3:46 PM
> 
> The ability to emulate a host bridge is useful not only for hardware PCI
> controllers like CONFIG_VMD, or virtual PCI controllers like
> CONFIG_PCI_HYPERV, but also for test and development scenarios like
> CONFIG_SAMPLES_DEVSEC [1].
> 
> One stumbling block for defining CONFIG_SAMPLES_DEVSEC, a sample
> implementation of a platform TSM for PCI Device Security, is the need to
> accommodate PCI_DOMAINS_GENERIC architectures alongside x86 [2].

There's not a [2] tag anywhere below.  Presumably it should be the "Closes:"
link?

> 
> In support of supplementing the existing CONFIG_PCI_BRIDGE_EMUL
> infrastructure for host bridges:
> 
> * Introduce pci_bus_find_emul_domain_nr() as a common way to find a free
>   PCI domain number whether that is to reuse the existing dynamic
>   allocation code in the !ACPI case, or to assign an unused domain above
>   the last ACPI segment.
> 
> * Convert pci-hyperv to the new allocator so that the PCI core can
>   unconditionally assume that bridge->domain_nr != PCI_DOMAIN_NR_NOT_SET
>   is the dynamically allocated case.
> 
> A follow on patch can also convert vmd to the new scheme. Currently vmd
> is limited to CONFIG_PCI_DOMAINS_GENERIC=n (x86) so, unlike pci-hyperv,
> it does not immediately conflict with this new
> pci_bus_find_emul_domain_nr() mechanism.
> 
> Link: https://lore.kernel.org/all/174107249038.1288555.12362100502109498455.stgit@dwillia2-xfh.jf.intel.com [1]
> Reported-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> Closes: https://lore.kernel.org/all/20250311144601.145736-3-suzuki.poulose@arm.com 
> Cc: Lorenzo Pieralisi <lpieralisi@kernel.org>
> Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
> Cc: Rob Herring <robh@kernel.org>
> Cc: Bjorn Helgaas <bhelgaas@google.com>
> Cc: "K. Y. Srinivasan" <kys@microsoft.com>
> Cc: Haiyang Zhang <haiyangz@microsoft.com>
> Cc: Wei Liu <wei.liu@kernel.org>
> Cc: Dexuan Cui <decui@microsoft.com>
> Tested-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> [michael: maintain compatibility with userspace that expects 16-bit ids]

Is the above line spurious?  It doesn't seem to belong here.

> Cc: Michael Kelley <mhklinux@outlook.com>
> Signed-off-by: Dan Williams <dan.j.williams@intel.com>

Tested on x86 and arm64 VMs in the Azure public cloud with
multiple Hyper-V virtual PCI devices in each VM. So covered
both the "CONFIG_PCI_DOMAINS_GENERIC=n" and "=y" cases.
Did manual unbind/rebind of vPCI devices multiple times so the
domain numbers would be freed and reallocated. All was good.
I did not go to the trouble of simulating a domain number
collision.

Reviewed-by: Michael Kelley <mhklinux@outlook.com>
Tested-by: Michael Kelley <mhklinux@outlook.com>

> ---
>  include/linux/pci.h                 |  7 ++++
>  drivers/pci/controller/pci-hyperv.c | 62 +++++------------------------
>  drivers/pci/pci.c                   | 24 ++++++++++-
>  drivers/pci/probe.c                 |  8 +++-
>  4 files changed, 46 insertions(+), 55 deletions(-)
> 
> diff --git a/include/linux/pci.h b/include/linux/pci.h
> index d1fdf81fbe1e..1ef1535802b0 100644
> --- a/include/linux/pci.h
> +++ b/include/linux/pci.h
> @@ -1956,10 +1956,17 @@ DEFINE_GUARD(pci_dev, struct pci_dev *, pci_dev_lock(_T), pci_dev_unlock(_T))
>   */
>  #ifdef CONFIG_PCI_DOMAINS
>  extern int pci_domains_supported;
> +int pci_bus_find_emul_domain_nr(u32 hint, u32 min, u32 max);
> +void pci_bus_release_emul_domain_nr(int domain_nr);
>  #else
>  enum { pci_domains_supported = 0 };
>  static inline int pci_domain_nr(struct pci_bus *bus) { return 0; }
>  static inline int pci_proc_domain(struct pci_bus *bus) { return 0; }
> +static inline int pci_bus_find_emul_domain_nr(u32 hint, u32 min, u32 max)
> +{
> +	return 0;
> +}
> +static inline void pci_bus_release_emul_domain_nr(int domain_nr) { }
>  #endif /* CONFIG_PCI_DOMAINS */
> 
>  /*
> diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c
> index 146b43981b27..64f68efaf547 100644
> --- a/drivers/pci/controller/pci-hyperv.c
> +++ b/drivers/pci/controller/pci-hyperv.c
> @@ -3696,48 +3696,6 @@ static int hv_send_resources_released(struct hv_device *hdev)
>  	return 0;
>  }
> 
> -#define HVPCI_DOM_MAP_SIZE (64 * 1024)
> -static DECLARE_BITMAP(hvpci_dom_map, HVPCI_DOM_MAP_SIZE);
> -
> -/*
> - * PCI domain number 0 is used by emulated devices on Gen1 VMs, so define 0
> - * as invalid for passthrough PCI devices of this driver.
> - */
> -#define HVPCI_DOM_INVALID 0
> -
> -/**
> - * hv_get_dom_num() - Get a valid PCI domain number
> - * Check if the PCI domain number is in use, and return another number if
> - * it is in use.
> - *
> - * @dom: Requested domain number
> - *
> - * return: domain number on success, HVPCI_DOM_INVALID on failure
> - */
> -static u16 hv_get_dom_num(u16 dom)
> -{
> -	unsigned int i;
> -
> -	if (test_and_set_bit(dom, hvpci_dom_map) == 0)
> -		return dom;
> -
> -	for_each_clear_bit(i, hvpci_dom_map, HVPCI_DOM_MAP_SIZE) {
> -		if (test_and_set_bit(i, hvpci_dom_map) == 0)
> -			return i;
> -	}
> -
> -	return HVPCI_DOM_INVALID;
> -}
> -
> -/**
> - * hv_put_dom_num() - Mark the PCI domain number as free
> - * @dom: Domain number to be freed
> - */
> -static void hv_put_dom_num(u16 dom)
> -{
> -	clear_bit(dom, hvpci_dom_map);
> -}
> -
>  /**
>   * hv_pci_probe() - New VMBus channel probe, for a root PCI bus
>   * @hdev:	VMBus's tracking struct for this root PCI bus
> @@ -3750,9 +3708,9 @@ static int hv_pci_probe(struct hv_device *hdev,
>  {
>  	struct pci_host_bridge *bridge;
>  	struct hv_pcibus_device *hbus;
> -	u16 dom_req, dom;
> +	int ret, dom;
> +	u16 dom_req;
>  	char *name;
> -	int ret;
> 
>  	bridge = devm_pci_alloc_host_bridge(&hdev->device, 0);
>  	if (!bridge)
> @@ -3779,11 +3737,14 @@ static int hv_pci_probe(struct hv_device *hdev,
>  	 * PCI bus (which is actually emulated by the hypervisor) is domain 0.
>  	 * (2) There will be no overlap between domains (after fixing possible
>  	 * collisions) in the same VM.
> +	 *
> +	 * Because Gen1 VMs use domain 0, don't allow picking domain 0 here,
> +	 * even if bytes 4 and 5 of the instance GUID are both zero. For wider
> +	 * userspace compatibility, limit the domain id to a 16-bit value.
>  	 */
>  	dom_req = hdev->dev_instance.b[5] << 8 | hdev->dev_instance.b[4];
> -	dom = hv_get_dom_num(dom_req);
> -
> -	if (dom == HVPCI_DOM_INVALID) {
> +	dom = pci_bus_find_emul_domain_nr(dom_req, 1, U16_MAX);
> +	if (dom < 0) {
>  		dev_err(&hdev->device,
>  			"Unable to use dom# 0x%x or other numbers", dom_req);
>  		ret = -EINVAL;
> @@ -3917,7 +3878,7 @@ static int hv_pci_probe(struct hv_device *hdev,
>  destroy_wq:
>  	destroy_workqueue(hbus->wq);
>  free_dom:
> -	hv_put_dom_num(hbus->bridge->domain_nr);
> +	pci_bus_release_emul_domain_nr(hbus->bridge->domain_nr);
>  free_bus:
>  	kfree(hbus);
>  	return ret;
> @@ -4042,8 +4003,6 @@ static void hv_pci_remove(struct hv_device *hdev)
>  	irq_domain_remove(hbus->irq_domain);
>  	irq_domain_free_fwnode(hbus->fwnode);
> 
> -	hv_put_dom_num(hbus->bridge->domain_nr);
> -
>  	kfree(hbus);
>  }
> 
> @@ -4217,9 +4176,6 @@ static int __init init_hv_pci_drv(void)
>  	if (ret)
>  		return ret;
> 
> -	/* Set the invalid domain number's bit, so it will not be used */
> -	set_bit(HVPCI_DOM_INVALID, hvpci_dom_map);
> -
>  	/* Initialize PCI block r/w interface */
>  	hvpci_block_ops.read_block = hv_read_config_block;
>  	hvpci_block_ops.write_block = hv_write_config_block;
> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> index b14dd064006c..64aed8705e91 100644
> --- a/drivers/pci/pci.c
> +++ b/drivers/pci/pci.c
> @@ -6656,9 +6656,31 @@ static void pci_no_domains(void)
>  #endif
>  }
> 
> +#ifdef CONFIG_PCI_DOMAINS
> +static DEFINE_IDA(pci_domain_nr_dynamic_ida);
> +
> +/**
> + * pci_bus_find_emul_domain_nr() - allocate a PCI domain number per constraints
> + * @hint: desired domain, 0 if any id in the range of @min to @max is acceptable
> + * @min: minimum allowable domain
> + * @max: maximum allowable domain, no ids higher than INT_MAX will be returned
> + */
> +int pci_bus_find_emul_domain_nr(u32 hint, u32 min, u32 max)
> +{
> +	return ida_alloc_range(&pci_domain_nr_dynamic_ida, max(hint, min), max,
> +			       GFP_KERNEL);
> +}
> +EXPORT_SYMBOL_GPL(pci_bus_find_emul_domain_nr);
> +
> +void pci_bus_release_emul_domain_nr(int domain_nr)
> +{
> +	ida_free(&pci_domain_nr_dynamic_ida, domain_nr);
> +}
> +EXPORT_SYMBOL_GPL(pci_bus_release_emul_domain_nr);
> +#endif
> +
>  #ifdef CONFIG_PCI_DOMAINS_GENERIC
>  static DEFINE_IDA(pci_domain_nr_static_ida);
> -static DEFINE_IDA(pci_domain_nr_dynamic_ida);
> 
>  static void of_pci_reserve_static_domain_nr(void)
>  {
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index 0ce98e18b5a8..5e101ced00a5 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -650,6 +650,11 @@ static void pci_release_host_bridge_dev(struct device *dev)
> 
>  	pci_free_resource_list(&bridge->windows);
>  	pci_free_resource_list(&bridge->dma_ranges);
> +
> +	/* Host bridges only have domain_nr set in the emulation case */
> +	if (bridge->domain_nr != PCI_DOMAIN_NR_NOT_SET)
> +		pci_bus_release_emul_domain_nr(bridge->domain_nr);
> +
>  	kfree(bridge);
>  }
> 
> @@ -1130,7 +1135,8 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge)
>  	device_del(&bridge->dev);
>  free:
>  #ifdef CONFIG_PCI_DOMAINS_GENERIC
> -	pci_bus_release_domain_nr(parent, bus->domain_nr);
> +	if (bridge->domain_nr == PCI_DOMAIN_NR_NOT_SET)
> +		pci_bus_release_domain_nr(parent, bus->domain_nr);
>  #endif
>  	if (bus_registered)
>  		put_device(&bus->dev);
> --
> 2.51.0

RE: [PATCH v2 1/2] PCI: Enable host bridge emulation for PCI_DOMAINS_GENERIC platforms

[dan.j.williams]

Michael Kelley wrote:
> From: Dan Williams <dan.j.williams@intel.com> Sent: Friday, October 24, 2025 3:46 PM
> > 
> > The ability to emulate a host bridge is useful not only for hardware PCI
> > controllers like CONFIG_VMD, or virtual PCI controllers like
> > CONFIG_PCI_HYPERV, but also for test and development scenarios like
> > CONFIG_SAMPLES_DEVSEC [1].
> > 
> > One stumbling block for defining CONFIG_SAMPLES_DEVSEC, a sample
> > implementation of a platform TSM for PCI Device Security, is the need to
> > accommodate PCI_DOMAINS_GENERIC architectures alongside x86 [2].
> 
> There's not a [2] tag anywhere below.  Presumably it should be the "Closes:"
> link?

Yes, good catch.

> > In support of supplementing the existing CONFIG_PCI_BRIDGE_EMUL
> > infrastructure for host bridges:
> > 
> > * Introduce pci_bus_find_emul_domain_nr() as a common way to find a free
> >   PCI domain number whether that is to reuse the existing dynamic
> >   allocation code in the !ACPI case, or to assign an unused domain above
> >   the last ACPI segment.
> > 
> > * Convert pci-hyperv to the new allocator so that the PCI core can
> >   unconditionally assume that bridge->domain_nr != PCI_DOMAIN_NR_NOT_SET
> >   is the dynamically allocated case.
> > 
> > A follow on patch can also convert vmd to the new scheme. Currently vmd
> > is limited to CONFIG_PCI_DOMAINS_GENERIC=n (x86) so, unlike pci-hyperv,
> > it does not immediately conflict with this new
> > pci_bus_find_emul_domain_nr() mechanism.
> > 
> > Link: https://lore.kernel.org/all/174107249038.1288555.12362100502109498455.stgit@dwillia2-xfh.jf.intel.com [1]
> > Reported-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> > Closes: https://lore.kernel.org/all/20250311144601.145736-3-suzuki.poulose@arm.com 
> > Cc: Lorenzo Pieralisi <lpieralisi@kernel.org>
> > Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
> > Cc: Rob Herring <robh@kernel.org>
> > Cc: Bjorn Helgaas <bhelgaas@google.com>
> > Cc: "K. Y. Srinivasan" <kys@microsoft.com>
> > Cc: Haiyang Zhang <haiyangz@microsoft.com>
> > Cc: Wei Liu <wei.liu@kernel.org>
> > Cc: Dexuan Cui <decui@microsoft.com>
> > Tested-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> > [michael: maintain compatibility with userspace that expects 16-bit ids]
> 
> Is the above line spurious?  It doesn't seem to belong here.

That was documenting the change to make the @max argument to
pci_bus_find_emul_domain_nr() be U16_MAX in the Hyper-V case, but yeah
no need to note that in this case. You did not edit the patch directly.
I am ok if Bjorn deletes that on applying.

If you want a respin Bjorn, let me know.

> > Cc: Michael Kelley <mhklinux@outlook.com>
> > Signed-off-by: Dan Williams <dan.j.williams@intel.com>
> 
> Tested on x86 and arm64 VMs in the Azure public cloud with
> multiple Hyper-V virtual PCI devices in each VM. So covered
> both the "CONFIG_PCI_DOMAINS_GENERIC=n" and "=y" cases.
> Did manual unbind/rebind of vPCI devices multiple times so the
> domain numbers would be freed and reallocated. All was good.
> I did not go to the trouble of simulating a domain number
> collision.
> 
> Reviewed-by: Michael Kelley <mhklinux@outlook.com>
> Tested-by: Michael Kelley <mhklinux@outlook.com>

Thanks, Michael much appreciated!

Re: [PATCH v2 1/2] PCI: Enable host bridge emulation for PCI_DOMAINS_GENERIC platforms

[Dave Jiang]

On 10/24/25 3:46 PM, Dan Williams wrote:
> The ability to emulate a host bridge is useful not only for hardware PCI
> controllers like CONFIG_VMD, or virtual PCI controllers like
> CONFIG_PCI_HYPERV, but also for test and development scenarios like
> CONFIG_SAMPLES_DEVSEC [1].
> 
> One stumbling block for defining CONFIG_SAMPLES_DEVSEC, a sample
> implementation of a platform TSM for PCI Device Security, is the need to
> accommodate PCI_DOMAINS_GENERIC architectures alongside x86 [2].
> 
> In support of supplementing the existing CONFIG_PCI_BRIDGE_EMUL
> infrastructure for host bridges:
> 
> * Introduce pci_bus_find_emul_domain_nr() as a common way to find a free
>   PCI domain number whether that is to reuse the existing dynamic
>   allocation code in the !ACPI case, or to assign an unused domain above
>   the last ACPI segment.
> 
> * Convert pci-hyperv to the new allocator so that the PCI core can
>   unconditionally assume that bridge->domain_nr != PCI_DOMAIN_NR_NOT_SET
>   is the dynamically allocated case.
> 
> A follow on patch can also convert vmd to the new scheme. Currently vmd
> is limited to CONFIG_PCI_DOMAINS_GENERIC=n (x86) so, unlike pci-hyperv,
> it does not immediately conflict with this new
> pci_bus_find_emul_domain_nr() mechanism.
> 
> Link: http://lore.kernel.org/174107249038.1288555.12362100502109498455.stgit@dwillia2-xfh.jf.intel.com [1]
> Reported-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> Closes: http://lore.kernel.org/20250311144601.145736-3-suzuki.poulose@arm.com
> Cc: Lorenzo Pieralisi <lpieralisi@kernel.org>
> Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
> Cc: Rob Herring <robh@kernel.org>
> Cc: Bjorn Helgaas <bhelgaas@google.com>
> Cc: "K. Y. Srinivasan" <kys@microsoft.com>
> Cc: Haiyang Zhang <haiyangz@microsoft.com>
> Cc: Wei Liu <wei.liu@kernel.org>
> Cc: Dexuan Cui <decui@microsoft.com>
> Tested-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> [michael: maintain compatibility with userspace that expects 16-bit ids]
> Cc: Michael Kelley <mhklinux@outlook.com>
> Signed-off-by: Dan Williams <dan.j.williams@intel.com>

Reviewed-by: Dave Jiang <dave.jiang@intel.com>> ---
>  include/linux/pci.h                 |  7 ++++
>  drivers/pci/controller/pci-hyperv.c | 62 +++++------------------------
>  drivers/pci/pci.c                   | 24 ++++++++++-
>  drivers/pci/probe.c                 |  8 +++-
>  4 files changed, 46 insertions(+), 55 deletions(-)
> 
> diff --git a/include/linux/pci.h b/include/linux/pci.h
> index d1fdf81fbe1e..1ef1535802b0 100644
> --- a/include/linux/pci.h
> +++ b/include/linux/pci.h
> @@ -1956,10 +1956,17 @@ DEFINE_GUARD(pci_dev, struct pci_dev *, pci_dev_lock(_T), pci_dev_unlock(_T))
>   */
>  #ifdef CONFIG_PCI_DOMAINS
>  extern int pci_domains_supported;
> +int pci_bus_find_emul_domain_nr(u32 hint, u32 min, u32 max);
> +void pci_bus_release_emul_domain_nr(int domain_nr);
>  #else
>  enum { pci_domains_supported = 0 };
>  static inline int pci_domain_nr(struct pci_bus *bus) { return 0; }
>  static inline int pci_proc_domain(struct pci_bus *bus) { return 0; }
> +static inline int pci_bus_find_emul_domain_nr(u32 hint, u32 min, u32 max)
> +{
> +	return 0;
> +}
> +static inline void pci_bus_release_emul_domain_nr(int domain_nr) { }
>  #endif /* CONFIG_PCI_DOMAINS */
>  
>  /*
> diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c
> index 146b43981b27..64f68efaf547 100644
> --- a/drivers/pci/controller/pci-hyperv.c
> +++ b/drivers/pci/controller/pci-hyperv.c
> @@ -3696,48 +3696,6 @@ static int hv_send_resources_released(struct hv_device *hdev)
>  	return 0;
>  }
>  
> -#define HVPCI_DOM_MAP_SIZE (64 * 1024)
> -static DECLARE_BITMAP(hvpci_dom_map, HVPCI_DOM_MAP_SIZE);
> -
> -/*
> - * PCI domain number 0 is used by emulated devices on Gen1 VMs, so define 0
> - * as invalid for passthrough PCI devices of this driver.
> - */
> -#define HVPCI_DOM_INVALID 0
> -
> -/**
> - * hv_get_dom_num() - Get a valid PCI domain number
> - * Check if the PCI domain number is in use, and return another number if
> - * it is in use.
> - *
> - * @dom: Requested domain number
> - *
> - * return: domain number on success, HVPCI_DOM_INVALID on failure
> - */
> -static u16 hv_get_dom_num(u16 dom)
> -{
> -	unsigned int i;
> -
> -	if (test_and_set_bit(dom, hvpci_dom_map) == 0)
> -		return dom;
> -
> -	for_each_clear_bit(i, hvpci_dom_map, HVPCI_DOM_MAP_SIZE) {
> -		if (test_and_set_bit(i, hvpci_dom_map) == 0)
> -			return i;
> -	}
> -
> -	return HVPCI_DOM_INVALID;
> -}
> -
> -/**
> - * hv_put_dom_num() - Mark the PCI domain number as free
> - * @dom: Domain number to be freed
> - */
> -static void hv_put_dom_num(u16 dom)
> -{
> -	clear_bit(dom, hvpci_dom_map);
> -}
> -
>  /**
>   * hv_pci_probe() - New VMBus channel probe, for a root PCI bus
>   * @hdev:	VMBus's tracking struct for this root PCI bus
> @@ -3750,9 +3708,9 @@ static int hv_pci_probe(struct hv_device *hdev,
>  {
>  	struct pci_host_bridge *bridge;
>  	struct hv_pcibus_device *hbus;
> -	u16 dom_req, dom;
> +	int ret, dom;
> +	u16 dom_req;
>  	char *name;
> -	int ret;
>  
>  	bridge = devm_pci_alloc_host_bridge(&hdev->device, 0);
>  	if (!bridge)
> @@ -3779,11 +3737,14 @@ static int hv_pci_probe(struct hv_device *hdev,
>  	 * PCI bus (which is actually emulated by the hypervisor) is domain 0.
>  	 * (2) There will be no overlap between domains (after fixing possible
>  	 * collisions) in the same VM.
> +	 *
> +	 * Because Gen1 VMs use domain 0, don't allow picking domain 0 here,
> +	 * even if bytes 4 and 5 of the instance GUID are both zero. For wider
> +	 * userspace compatibility, limit the domain id to a 16-bit value.
>  	 */
>  	dom_req = hdev->dev_instance.b[5] << 8 | hdev->dev_instance.b[4];
> -	dom = hv_get_dom_num(dom_req);
> -
> -	if (dom == HVPCI_DOM_INVALID) {
> +	dom = pci_bus_find_emul_domain_nr(dom_req, 1, U16_MAX);
> +	if (dom < 0) {
>  		dev_err(&hdev->device,
>  			"Unable to use dom# 0x%x or other numbers", dom_req);
>  		ret = -EINVAL;
> @@ -3917,7 +3878,7 @@ static int hv_pci_probe(struct hv_device *hdev,
>  destroy_wq:
>  	destroy_workqueue(hbus->wq);
>  free_dom:
> -	hv_put_dom_num(hbus->bridge->domain_nr);
> +	pci_bus_release_emul_domain_nr(hbus->bridge->domain_nr);
>  free_bus:
>  	kfree(hbus);
>  	return ret;
> @@ -4042,8 +4003,6 @@ static void hv_pci_remove(struct hv_device *hdev)
>  	irq_domain_remove(hbus->irq_domain);
>  	irq_domain_free_fwnode(hbus->fwnode);
>  
> -	hv_put_dom_num(hbus->bridge->domain_nr);
> -
>  	kfree(hbus);
>  }
>  
> @@ -4217,9 +4176,6 @@ static int __init init_hv_pci_drv(void)
>  	if (ret)
>  		return ret;
>  
> -	/* Set the invalid domain number's bit, so it will not be used */
> -	set_bit(HVPCI_DOM_INVALID, hvpci_dom_map);
> -
>  	/* Initialize PCI block r/w interface */
>  	hvpci_block_ops.read_block = hv_read_config_block;
>  	hvpci_block_ops.write_block = hv_write_config_block;
> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> index b14dd064006c..64aed8705e91 100644
> --- a/drivers/pci/pci.c
> +++ b/drivers/pci/pci.c
> @@ -6656,9 +6656,31 @@ static void pci_no_domains(void)
>  #endif
>  }
>  
> +#ifdef CONFIG_PCI_DOMAINS
> +static DEFINE_IDA(pci_domain_nr_dynamic_ida);
> +
> +/**
> + * pci_bus_find_emul_domain_nr() - allocate a PCI domain number per constraints
> + * @hint: desired domain, 0 if any id in the range of @min to @max is acceptable
> + * @min: minimum allowable domain
> + * @max: maximum allowable domain, no ids higher than INT_MAX will be returned
> + */
> +int pci_bus_find_emul_domain_nr(u32 hint, u32 min, u32 max)
> +{
> +	return ida_alloc_range(&pci_domain_nr_dynamic_ida, max(hint, min), max,
> +			       GFP_KERNEL);
> +}
> +EXPORT_SYMBOL_GPL(pci_bus_find_emul_domain_nr);
> +
> +void pci_bus_release_emul_domain_nr(int domain_nr)
> +{
> +	ida_free(&pci_domain_nr_dynamic_ida, domain_nr);
> +}
> +EXPORT_SYMBOL_GPL(pci_bus_release_emul_domain_nr);
> +#endif
> +
>  #ifdef CONFIG_PCI_DOMAINS_GENERIC
>  static DEFINE_IDA(pci_domain_nr_static_ida);
> -static DEFINE_IDA(pci_domain_nr_dynamic_ida);
>  
>  static void of_pci_reserve_static_domain_nr(void)
>  {
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index 0ce98e18b5a8..5e101ced00a5 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -650,6 +650,11 @@ static void pci_release_host_bridge_dev(struct device *dev)
>  
>  	pci_free_resource_list(&bridge->windows);
>  	pci_free_resource_list(&bridge->dma_ranges);
> +
> +	/* Host bridges only have domain_nr set in the emulation case */
> +	if (bridge->domain_nr != PCI_DOMAIN_NR_NOT_SET)
> +		pci_bus_release_emul_domain_nr(bridge->domain_nr);
> +
>  	kfree(bridge);
>  }
>  
> @@ -1130,7 +1135,8 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge)
>  	device_del(&bridge->dev);
>  free:
>  #ifdef CONFIG_PCI_DOMAINS_GENERIC
> -	pci_bus_release_domain_nr(parent, bus->domain_nr);
> +	if (bridge->domain_nr == PCI_DOMAIN_NR_NOT_SET)
> +		pci_bus_release_domain_nr(parent, bus->domain_nr);
>  #endif
>  	if (bus_registered)
>  		put_device(&bus->dev);

[PATCH v2 2/2] PCI: vmd: Switch to pci_bus_find_emul_domain_nr()

[Dan Williams]

The new common domain number allocator can replace the custom allocator
in VMD.

Beyond some code reuse benefits it does close a potential race whereby
vmd_find_free_domain() collides with new PCI buses coming online with a
conflicting domain number. Such a race has not been observed in
practice, hence not tagging this change as a fix.

As VMD uses pci_create_root_bus() rather than pci_alloc_host_bridge() +
pci_scan_root_bus_bridge() it has no chance to set ->domain_nr in the
bridge so needs to manage freeing the domain number on its own.

Cc: Szymon Durawa <szymon.durawa@linux.intel.com>
Cc: Nirmal Patel <nirmal.patel@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/pci/controller/vmd.c | 40 +++++++++++++++---------------------
 1 file changed, 17 insertions(+), 23 deletions(-)

diff --git a/drivers/pci/controller/vmd.c b/drivers/pci/controller/vmd.c
index b4b62b9ccc45..03b5920138f0 100644
--- a/drivers/pci/controller/vmd.c
+++ b/drivers/pci/controller/vmd.c
@@ -578,22 +578,6 @@ static void vmd_detach_resources(struct vmd_dev *vmd)
 	vmd->dev->resource[VMD_MEMBAR2].child = NULL;
 }
 
-/*
- * VMD domains start at 0x10000 to not clash with ACPI _SEG domains.
- * Per ACPI r6.0, sec 6.5.6,  _SEG returns an integer, of which the lower
- * 16 bits are the PCI Segment Group (domain) number.  Other bits are
- * currently reserved.
- */
-static int vmd_find_free_domain(void)
-{
-	int domain = 0xffff;
-	struct pci_bus *bus = NULL;
-
-	while ((bus = pci_find_next_bus(bus)) != NULL)
-		domain = max_t(int, domain, pci_domain_nr(bus));
-	return domain + 1;
-}
-
 static int vmd_get_phys_offsets(struct vmd_dev *vmd, bool native_hint,
 				resource_size_t *offset1,
 				resource_size_t *offset2)
@@ -878,13 +862,6 @@ static int vmd_enable_domain(struct vmd_dev *vmd, unsigned long features)
 		.parent = res,
 	};
 
-	sd->vmd_dev = vmd->dev;
-	sd->domain = vmd_find_free_domain();
-	if (sd->domain < 0)
-		return sd->domain;
-
-	sd->node = pcibus_to_node(vmd->dev->bus);
-
 	/*
 	 * Currently MSI remapping must be enabled in guest passthrough mode
 	 * due to some missing interrupt remapping plumbing. This is probably
@@ -910,9 +887,24 @@ static int vmd_enable_domain(struct vmd_dev *vmd, unsigned long features)
 	pci_add_resource_offset(&resources, &vmd->resources[1], offset[0]);
 	pci_add_resource_offset(&resources, &vmd->resources[2], offset[1]);
 
+	sd->vmd_dev = vmd->dev;
+
+	/*
+	 * Emulated domains start at 0x10000 to not clash with ACPI _SEG
+	 * domains.  Per ACPI r6.0, sec 6.5.6,  _SEG returns an integer, of
+	 * which the lower 16 bits are the PCI Segment Group (domain) number.
+	 * Other bits are currently reserved.
+	 */
+	sd->domain = pci_bus_find_emul_domain_nr(0, 0x10000, INT_MAX);
+	if (sd->domain < 0)
+		return sd->domain;
+
+	sd->node = pcibus_to_node(vmd->dev->bus);
+
 	vmd->bus = pci_create_root_bus(&vmd->dev->dev, vmd->busn_start,
 				       &vmd_ops, sd, &resources);
 	if (!vmd->bus) {
+		pci_bus_release_emul_domain_nr(sd->domain);
 		pci_free_resource_list(&resources);
 		vmd_remove_irq_domain(vmd);
 		return -ENODEV;
@@ -1005,6 +997,7 @@ static int vmd_probe(struct pci_dev *dev, const struct pci_device_id *id)
 		return -ENOMEM;
 
 	vmd->dev = dev;
+	vmd->sysdata.domain = PCI_DOMAIN_NR_NOT_SET;
 	vmd->instance = ida_alloc(&vmd_instance_ida, GFP_KERNEL);
 	if (vmd->instance < 0)
 		return vmd->instance;
@@ -1070,6 +1063,7 @@ static void vmd_remove(struct pci_dev *dev)
 	vmd_detach_resources(vmd);
 	vmd_remove_irq_domain(vmd);
 	ida_free(&vmd_instance_ida, vmd->instance);
+	pci_bus_release_emul_domain_nr(vmd->sysdata.domain);
 }
 
 static void vmd_shutdown(struct pci_dev *dev)
-- 
2.51.0

Re: [PATCH v2 2/2] PCI: vmd: Switch to pci_bus_find_emul_domain_nr()

[Dave Jiang]

On 10/24/25 3:46 PM, Dan Williams wrote:
> The new common domain number allocator can replace the custom allocator
> in VMD.
> 
> Beyond some code reuse benefits it does close a potential race whereby
> vmd_find_free_domain() collides with new PCI buses coming online with a
> conflicting domain number. Such a race has not been observed in
> practice, hence not tagging this change as a fix.
> 
> As VMD uses pci_create_root_bus() rather than pci_alloc_host_bridge() +
> pci_scan_root_bus_bridge() it has no chance to set ->domain_nr in the
> bridge so needs to manage freeing the domain number on its own.
> 
> Cc: Szymon Durawa <szymon.durawa@linux.intel.com>
> Cc: Nirmal Patel <nirmal.patel@linux.intel.com>
> Signed-off-by: Dan Williams <dan.j.williams@intel.com>

Reviewed-by: Dave Jiang <dave.jiang@intel.com>> ---
>  drivers/pci/controller/vmd.c | 40 +++++++++++++++---------------------
>  1 file changed, 17 insertions(+), 23 deletions(-)
> 
> diff --git a/drivers/pci/controller/vmd.c b/drivers/pci/controller/vmd.c
> index b4b62b9ccc45..03b5920138f0 100644
> --- a/drivers/pci/controller/vmd.c
> +++ b/drivers/pci/controller/vmd.c
> @@ -578,22 +578,6 @@ static void vmd_detach_resources(struct vmd_dev *vmd)
>  	vmd->dev->resource[VMD_MEMBAR2].child = NULL;
>  }
>  
> -/*
> - * VMD domains start at 0x10000 to not clash with ACPI _SEG domains.
> - * Per ACPI r6.0, sec 6.5.6,  _SEG returns an integer, of which the lower
> - * 16 bits are the PCI Segment Group (domain) number.  Other bits are
> - * currently reserved.
> - */
> -static int vmd_find_free_domain(void)
> -{
> -	int domain = 0xffff;
> -	struct pci_bus *bus = NULL;
> -
> -	while ((bus = pci_find_next_bus(bus)) != NULL)
> -		domain = max_t(int, domain, pci_domain_nr(bus));
> -	return domain + 1;
> -}
> -
>  static int vmd_get_phys_offsets(struct vmd_dev *vmd, bool native_hint,
>  				resource_size_t *offset1,
>  				resource_size_t *offset2)
> @@ -878,13 +862,6 @@ static int vmd_enable_domain(struct vmd_dev *vmd, unsigned long features)
>  		.parent = res,
>  	};
>  
> -	sd->vmd_dev = vmd->dev;
> -	sd->domain = vmd_find_free_domain();
> -	if (sd->domain < 0)
> -		return sd->domain;
> -
> -	sd->node = pcibus_to_node(vmd->dev->bus);
> -
>  	/*
>  	 * Currently MSI remapping must be enabled in guest passthrough mode
>  	 * due to some missing interrupt remapping plumbing. This is probably
> @@ -910,9 +887,24 @@ static int vmd_enable_domain(struct vmd_dev *vmd, unsigned long features)
>  	pci_add_resource_offset(&resources, &vmd->resources[1], offset[0]);
>  	pci_add_resource_offset(&resources, &vmd->resources[2], offset[1]);
>  
> +	sd->vmd_dev = vmd->dev;
> +
> +	/*
> +	 * Emulated domains start at 0x10000 to not clash with ACPI _SEG
> +	 * domains.  Per ACPI r6.0, sec 6.5.6,  _SEG returns an integer, of
> +	 * which the lower 16 bits are the PCI Segment Group (domain) number.
> +	 * Other bits are currently reserved.
> +	 */
> +	sd->domain = pci_bus_find_emul_domain_nr(0, 0x10000, INT_MAX);
> +	if (sd->domain < 0)
> +		return sd->domain;
> +
> +	sd->node = pcibus_to_node(vmd->dev->bus);
> +
>  	vmd->bus = pci_create_root_bus(&vmd->dev->dev, vmd->busn_start,
>  				       &vmd_ops, sd, &resources);
>  	if (!vmd->bus) {
> +		pci_bus_release_emul_domain_nr(sd->domain);
>  		pci_free_resource_list(&resources);
>  		vmd_remove_irq_domain(vmd);
>  		return -ENODEV;
> @@ -1005,6 +997,7 @@ static int vmd_probe(struct pci_dev *dev, const struct pci_device_id *id)
>  		return -ENOMEM;
>  
>  	vmd->dev = dev;
> +	vmd->sysdata.domain = PCI_DOMAIN_NR_NOT_SET;
>  	vmd->instance = ida_alloc(&vmd_instance_ida, GFP_KERNEL);
>  	if (vmd->instance < 0)
>  		return vmd->instance;
> @@ -1070,6 +1063,7 @@ static void vmd_remove(struct pci_dev *dev)
>  	vmd_detach_resources(vmd);
>  	vmd_remove_irq_domain(vmd);
>  	ida_free(&vmd_instance_ida, vmd->instance);
> +	pci_bus_release_emul_domain_nr(vmd->sysdata.domain);
>  }
>  
>  static void vmd_shutdown(struct pci_dev *dev)

Re: [PATCH v2 0/2] PCI: Unify domain emulation

[Bjorn Helgaas]

On Fri, Oct 24, 2025 at 03:46:20PM -0700, Dan Williams wrote:
> Changes since v1 [1]:
> - Rebase on v6.18-rc2
> - Support callers supplying both a hint and a range for the Hyper-V hint
>   + fallback case (Michael)
> - Add comment explaining domain number 0 vs Gen1 VMs, and domain values
>   greater than U16_MAX concerns (Michael)
> - Leave the VMD status quo comment about requesting domain numbers >
>   U16_MAX
> 
> [1]: http://lore.kernel.org/20250716160835.680486-1-dan.j.williams@intel.com
> 
> The PCI/TSM effort created a sample driver to test ABI flows
> (samples/devsec/ [2]). Suzuki observed that it only worked on x86 due to
> its dependency on CONFIG_PCI_DOMAINS_GENERIC=n. I.e. an unfortunate
> restriction for what should be an architecture agnostic test framework.
> 
> Introduce a new pci_bus_find_emul_domain_nr() helper that all "soft"
> host-bridge drivers can share and hide the CONFIG_PCI_DOMAINS_GENERIC
> details behind that helper.
> 
> [2]: https://git.kernel.org/pub/scm/linux/kernel/git/devsec/tsm.git/commit/?id=0e16ce0b9c64
> 
> Dan Williams (2):
>   PCI: Enable host bridge emulation for PCI_DOMAINS_GENERIC platforms
>   PCI: vmd: Switch to pci_bus_find_emul_domain_nr()
> 
>  include/linux/pci.h                 |  7 ++++
>  drivers/pci/controller/pci-hyperv.c | 62 +++++------------------------
>  drivers/pci/controller/vmd.c        | 40 ++++++++-----------
>  drivers/pci/pci.c                   | 24 ++++++++++-
>  drivers/pci/probe.c                 |  8 +++-
>  5 files changed, 63 insertions(+), 78 deletions(-)

Applied to pci/enumeration for v6.19, thanks, Dan!