This small series of patches aims to add devicetree bindings support for
the Virtual Machine Generation ID (vmgenid).
Virtual Machine Generation ID was introduced in commit af6b54e2b5ba
("virt: vmgenid: notify RNG of VM fork and supply generation ID") as an
ACPI only device.
VMGenID specification http://go.microsoft.com/fwlink/?LinkId=260709 defines
a mechanism for the BIOS/hypervisors to communicate to the virtual machine
that it is executed with a different configuration (e.g. snapshot execution
or creation from a template).
The guest operating system can use the notification for various purposes
such as re-initializing its random number generator etc.
More references to vmgenid specs:
- https://www.qemu.org/docs/master/specs/vmgenid.html
- https://learn.microsoft.com/en-us/windows/win32/hyperv_v2/
virtual-machine-generation-identifier
*Reason for this change*:
Chosing ACPI or devicetree is an intrinsic part of an hypervisor design.
Without going into details of why a hypervisor would choose DT over ACPI,
we would like to highlight that the hypervisors that have chosen devicetree
and now want to make use of the vmgenid functionality cannot do so today
because vmgenid is an ACPI only device.
This forces these hypervisors to change their design which could have
undesirable impacts on their use-cases, test-scenarios etc.
vmgenid exposes to the guest a 16-byte cryptographically random number,
the value of which changes every time it starts executing from a new
configuration (snapshot, backup, etc.). During initialization, the device
exposes to the guest the address of the generation ID and
an interrupt number, which the device will use to notify the guest when
the generation ID changes.
These attributes can be trivially communicated via device tree bindings.
We believe that adding a devicetree binding for vmgenid is a simpler
alternative way to expose the device to the guest than forcing the
hypervisors to implement ACPI.
Addtional notes:
While adding the devicetree support we considered re-using existing
structures/code to avoid duplicating code and reduce maintenance; so,
we used the same driver to be configured either by ACPI or by DT.
This also meant reimplementing the existing vmgenid ACPI bus driver as a
platform driver and making it discoverable using `driver.of_match_table`
and `driver.acpi_match_table`.
There is no user impact or change in vmgenid functionality when used
with ACPI. We verified ACPI support of these patches on X86 and DT
support on ARM using Firecracker hypervisor
https://github.com/firecracker-microvm/firecracker.
To check schema and syntax errors, the bindings file is verified with:
```
make dt_binding_check \
DT_SCHEMA_FILES=\
Documentation/devicetree/bindings/rng/microsoft,vmgenid.yaml
```
and the patches were verified with:
`scripts/checkpatch.pl --strict v6-000*`.
Changelog with respect to version 5:
- Added myself as an author and in SoB of the patches. Kept Sudan
as a co-author.
Changelog with respect to version 4:
- Removed __maybe_unused attribute from vmgenid_of_irq_handler since it
is always compiled in (used by vmgenid_add_of).
Changelog with respect to version 3:
- Changed the compatible string from "virtual,vmgenctr" to
"microsoft,vmgenid" as per review comments.
- Renamed vmgenid.yaml to follow DT file naming convention.
- Updated the description of properties and example in vmgenid yaml file.
- Addressed the review comments to remove all ifdefs in vmgenid.c with one
exception which still needs to be under CONFIG_ACPI.
- reformated the code with clang-format.
- Tested code with W=1, Sparse, Smatch and Coccinelle tools.
Changelog with respect to version 2:
- As per review comments, used platform apis instead of "of_*" APIs,
removed unnecessary #include and used IF_ENABLED instead of ifdef.
- Added more info for vmgenid buffer address and corrected the formatting.
- Replaced the compatible string from "linux,*" to "virtual,*" because,
the device does not have a vendor.
Changelog with respect to version 1:
- Moved vmgenid.yaml bindings to the more related "rng" folder.
- Removed `vmgenid_remove` to since it is unrelated to the
current goal of the patch.
- Updated the cover letter and bindings commit
"[PATCH v2 3/4] dt-bindings: rng: Add vmgenid support" to
provide more information on vmgenid.
- Compiled with and without CONFIG_OF/CONFIG_ACPI and fixed
compilers errors/warnings.
Babis Chalios (5):
virt: vmgenid: re-arrange code to make review easier
virt: vmgenid: change implementation to use a platform driver
virt: vmgenid: enable driver regardless of ACPI config
dt-bindings: rng: Add vmgenid support
virt: vmgenid: add support for devicetree bindings
.../bindings/rng/microsoft,vmgenid.yaml | 49 +++++
MAINTAINERS | 1 +
drivers/virt/Kconfig | 1 -
drivers/virt/vmgenid.c | 168 ++++++++++++++----
4 files changed, 180 insertions(+), 39 deletions(-)
create mode 100644 Documentation/devicetree/bindings/rng/microsoft,vmgenid.yaml
Reviewed-by: Alexander Graf <[email protected]>
--
2.40.1
Re-arrange the functions of vmgenid to make the next commit,
which re-implements vmgenid as a platform driver, easier to review.
Co-authored-by: Sudan Landge <[email protected]>
Signed-off-by: Babis Chalios <[email protected]>
Reviewed-by: Alexander Graf <[email protected]>
---
drivers/virt/vmgenid.c | 30 +++++++++++++++---------------
1 file changed, 15 insertions(+), 15 deletions(-)
diff --git a/drivers/virt/vmgenid.c b/drivers/virt/vmgenid.c
index b67a28da4702..a167ccfad299 100644
--- a/drivers/virt/vmgenid.c
+++ b/drivers/virt/vmgenid.c
@@ -21,6 +21,20 @@ struct vmgenid_state {
u8 this_id[VMGENID_SIZE];
};
+static void vmgenid_notify(struct acpi_device *device, u32 event)
+{
+ struct vmgenid_state *state = acpi_driver_data(device);
+ char *envp[] = { "NEW_VMGENID=1", NULL };
+ u8 old_id[VMGENID_SIZE];
+
+ memcpy(old_id, state->this_id, sizeof(old_id));
+ memcpy(state->this_id, state->next_id, sizeof(state->this_id));
+ if (!memcmp(old_id, state->this_id, sizeof(old_id)))
+ return;
+ add_vmfork_randomness(state->this_id, sizeof(state->this_id));
+ kobject_uevent_env(&device->dev.kobj, KOBJ_CHANGE, envp);
+}
+
static int vmgenid_add(struct acpi_device *device)
{
struct acpi_buffer parsed = { ACPI_ALLOCATE_BUFFER };
@@ -65,25 +79,12 @@ static int vmgenid_add(struct acpi_device *device)
return ret;
}
-static void vmgenid_notify(struct acpi_device *device, u32 event)
-{
- struct vmgenid_state *state = acpi_driver_data(device);
- char *envp[] = { "NEW_VMGENID=1", NULL };
- u8 old_id[VMGENID_SIZE];
-
- memcpy(old_id, state->this_id, sizeof(old_id));
- memcpy(state->this_id, state->next_id, sizeof(state->this_id));
- if (!memcmp(old_id, state->this_id, sizeof(old_id)))
- return;
- add_vmfork_randomness(state->this_id, sizeof(state->this_id));
- kobject_uevent_env(&device->dev.kobj, KOBJ_CHANGE, envp);
-}
-
static const struct acpi_device_id vmgenid_ids[] = {
{ "VMGENCTR", 0 },
{ "VM_GEN_COUNTER", 0 },
{ }
};
+MODULE_DEVICE_TABLE(acpi, vmgenid_ids);
static struct acpi_driver vmgenid_driver = {
.name = "vmgenid",
@@ -97,7 +98,6 @@ static struct acpi_driver vmgenid_driver = {
module_acpi_driver(vmgenid_driver);
-MODULE_DEVICE_TABLE(acpi, vmgenid_ids);
MODULE_DESCRIPTION("Virtual Machine Generation ID");
MODULE_LICENSE("GPL v2");
MODULE_AUTHOR("Jason A. Donenfeld <[email protected]>");
--
2.40.1
Since with next commits vmgenid driver will support both ACPI and
devicetree, and since either one of CONFIG_ACPI or CONFIG_OF will
always be enabled, there is no need for the driver compilation to
depend on ACPI/OF. So, remove ACPI dependency for compiling the driver.
Co-authored-by: Sudan Landge <[email protected]>
Signed-off-by: Babis Chalios <[email protected]>
Reviewed-by: Alexander Graf <[email protected]>
---
drivers/virt/Kconfig | 1 -
drivers/virt/vmgenid.c | 16 +++++++++++-----
2 files changed, 11 insertions(+), 6 deletions(-)
diff --git a/drivers/virt/Kconfig b/drivers/virt/Kconfig
index 40129b6f0eca..d8c848cf09a6 100644
--- a/drivers/virt/Kconfig
+++ b/drivers/virt/Kconfig
@@ -16,7 +16,6 @@ if VIRT_DRIVERS
config VMGENID
tristate "Virtual Machine Generation ID driver"
default y
- depends on ACPI
help
Say Y here to use the hypervisor-provided Virtual Machine Generation ID
to reseed the RNG when the VM is cloned. This is highly recommended if
diff --git a/drivers/virt/vmgenid.c b/drivers/virt/vmgenid.c
index c028e2064fdd..3d93e3fb94c4 100644
--- a/drivers/virt/vmgenid.c
+++ b/drivers/virt/vmgenid.c
@@ -36,13 +36,15 @@ static void vmgenid_notify(struct device *device)
kobject_uevent_env(&device->kobj, KOBJ_CHANGE, envp);
}
-static void vmgenid_acpi_handler(acpi_handle __always_unused handle,
- u32 __always_unused event, void *dev)
+static void __maybe_unused
+vmgenid_acpi_handler(acpi_handle __always_unused handle,
+ u32 __always_unused event, void *dev)
{
vmgenid_notify(dev);
}
-static int setup_vmgenid_state(struct vmgenid_state *state, u8 *next_id)
+static int __maybe_unused
+setup_vmgenid_state(struct vmgenid_state *state, u8 *next_id)
{
if (IS_ERR(next_id))
return PTR_ERR(next_id);
@@ -53,9 +55,10 @@ static int setup_vmgenid_state(struct vmgenid_state *state, u8 *next_id)
return 0;
}
-static int vmgenid_add_acpi(struct device *dev,
- struct vmgenid_state *state)
+static int vmgenid_add_acpi(struct device __maybe_unused *dev,
+ struct vmgenid_state __maybe_unused *state)
{
+#if IS_ENABLED(CONFIG_ACPI)
struct acpi_device *device = ACPI_COMPANION(dev);
struct acpi_buffer parsed = { ACPI_ALLOCATE_BUFFER };
union acpi_object *obj;
@@ -98,6 +101,9 @@ static int vmgenid_add_acpi(struct device *dev,
out:
ACPI_FREE(parsed.pointer);
return ret;
+#else
+ return -EINVAL;
+#endif
}
static int vmgenid_add(struct platform_device *pdev)
--
2.40.1
Virtual Machine Generation ID driver was introduced in commit af6b54e2b5ba
("virt: vmgenid: notify RNG of VM fork and supply generation ID"), as an
ACPI only device.
VMGenID specification http://go.microsoft.com/fwlink/?LinkId=260709 defines
a mechanism for the BIOS/hypervisors to communicate to the virtual machine
that it is executed with a different configuration (e.g. snapshot execution
or creation from a template).
The guest operating system can use the notification for various purposes
such as re-initializing its random number generator etc.
As per the specs, hypervisor should provide a globally unique identified,
or GUID via ACPI.
This patch tries to mimic the mechanism to provide the same functionality
which is for a hypervisor/BIOS to notify the virtual machine when it is
executed with a different configuration.
As part of this support the devicetree bindings requires the hypervisors or
BIOS to provide a memory address which holds the GUID and an IRQ which is
used to notify when there is a change in the GUID.
The memory exposed in the DT should follow the rules defined in the
vmgenid spec mentioned above.
*Reason for this change*:
Chosing ACPI or devicetree is an intrinsic part of an hypervisor design.
Without going into details of why a hypervisor would chose DT over ACPI,
we would like to highlight that the hypervisors that have chose devicetree
and now want to make use of the vmgenid functionality cannot do so today
because vmgenid is an ACPI only device.
This forces these hypervisors to change their design which could have
undesirable impacts on their use-cases, test-scenarios etc.
The point of vmgenid is to provide a mechanism to discover a GUID when
the execution state of a virtual machine changes and the simplest
way to do it is pass a memory location and an interrupt via devicetree.
It would complicate things unnecessarily if instead of using devicetree,
we try to implement a new protocol or modify other protocols to somehow
provide the same functionility.
We believe that adding a devicetree binding for vmgenid is a simpler,
better alternative to provide the same functionality and will allow
such hypervisors as mentioned above to continue using devicetree.
More references to vmgenid specs:
- https://www.qemu.org/docs/master/specs/vmgenid.html
- https://learn.microsoft.com/en-us/windows/win32/hyperv_v2/virtual-
machine-generation-identifier
Co-authored-by: Sudan Landge <[email protected]>
Signed-off-by: Babis Chalios <[email protected]>
Reviewed-by: Rob Herring <[email protected]>
Reviewed-by: Alexander Graf <[email protected]>
---
.../bindings/rng/microsoft,vmgenid.yaml | 49 +++++++++++++++++++
MAINTAINERS | 1 +
2 files changed, 50 insertions(+)
create mode 100644 Documentation/devicetree/bindings/rng/microsoft,vmgenid.yaml
diff --git a/Documentation/devicetree/bindings/rng/microsoft,vmgenid.yaml b/Documentation/devicetree/bindings/rng/microsoft,vmgenid.yaml
new file mode 100644
index 000000000000..8f20dee93e7e
--- /dev/null
+++ b/Documentation/devicetree/bindings/rng/microsoft,vmgenid.yaml
@@ -0,0 +1,49 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/rng/microsoft,vmgenid.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Virtual Machine Generation ID
+
+maintainers:
+ - Jason A. Donenfeld <[email protected]>
+
+description:
+ Firmwares or hypervisors can use this devicetree to describe an
+ interrupt and a shared resource to inject a Virtual Machine Generation ID.
+ Virtual Machine Generation ID is a globally unique identifier (GUID) and
+ the devicetree binding follows VMGenID specification defined in
+ http://go.microsoft.com/fwlink/?LinkId=260709.
+
+properties:
+ compatible:
+ const: microsoft,vmgenid
+
+ reg:
+ description:
+ Specifies a 16-byte VMGenID in endianness-agnostic hexadecimal format.
+ maxItems: 1
+
+ interrupts:
+ description:
+ Interrupt used to notify that a new VMGenID is available.
+ maxItems: 1
+
+required:
+ - compatible
+ - reg
+ - interrupts
+
+additionalProperties: false
+
+examples:
+ - |
+ #include <dt-bindings/interrupt-controller/arm-gic.h>
+ rng@80000000 {
+ compatible = "microsoft,vmgenid";
+ reg = <0x80000000 0x1000>;
+ interrupts = <GIC_SPI 35 IRQ_TYPE_EDGE_RISING>;
+ };
+
+...
diff --git a/MAINTAINERS b/MAINTAINERS
index c23fda1aa1f0..efd8de759d95 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -18469,6 +18469,7 @@ M: "Theodore Ts'o" <[email protected]>
M: Jason A. Donenfeld <[email protected]>
S: Maintained
T: git https://git.kernel.org/pub/scm/linux/kernel/git/crng/random.git
+F: Documentation/devicetree/bindings/rng/microsoft,vmgenid.yaml
F: drivers/char/random.c
F: drivers/virt/vmgenid.c
--
2.40.1
Extend the vmgenid platform driver to support devicetree bindings.
With this support, hypervisors can send vmgenid notifications to
the virtual machine without the need to enable ACPI.
The bindings are located at:
Documentation/devicetree/bindings/rng/microsoft,vmgenid.yaml
Co-authored-by: Sudan Landge <[email protected]>
Signed-off-by: Babis Chalios <[email protected]>
Reviewed-by: Alexander Graf <[email protected]>
---
drivers/virt/vmgenid.c | 53 ++++++++++++++++++++++++++++++++++++++++--
1 file changed, 51 insertions(+), 2 deletions(-)
diff --git a/drivers/virt/vmgenid.c b/drivers/virt/vmgenid.c
index 3d93e3fb94c4..9a3c12e5dd6e 100644
--- a/drivers/virt/vmgenid.c
+++ b/drivers/virt/vmgenid.c
@@ -2,12 +2,13 @@
/*
* Copyright (C) 2022 Jason A. Donenfeld <[email protected]>. All Rights Reserved.
*
- * The "Virtual Machine Generation ID" is exposed via ACPI and changes when a
+ * The "Virtual Machine Generation ID" is exposed via ACPI or DT and changes when a
* virtual machine forks or is cloned. This driver exists for shepherding that
* information to random.c.
*/
#include <linux/acpi.h>
+#include <linux/interrupt.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/platform_device.h>
@@ -20,6 +21,7 @@ enum { VMGENID_SIZE = 16 };
struct vmgenid_state {
u8 *next_id;
u8 this_id[VMGENID_SIZE];
+ int irq;
};
static void vmgenid_notify(struct device *device)
@@ -43,6 +45,14 @@ vmgenid_acpi_handler(acpi_handle __always_unused handle,
vmgenid_notify(dev);
}
+static irqreturn_t
+vmgenid_of_irq_handler(int __always_unused irq, void *dev)
+{
+ vmgenid_notify(dev);
+
+ return IRQ_HANDLED;
+}
+
static int __maybe_unused
setup_vmgenid_state(struct vmgenid_state *state, u8 *next_id)
{
@@ -106,6 +116,35 @@ static int vmgenid_add_acpi(struct device __maybe_unused *dev,
#endif
}
+static int vmgenid_add_of(struct platform_device *pdev,
+ struct vmgenid_state *state)
+{
+ u8 *virt_addr;
+ int ret = 0;
+
+ virt_addr = (u8 *)devm_platform_get_and_ioremap_resource(pdev, 0, NULL);
+ if (IS_ERR(virt_addr))
+ return PTR_ERR(virt_addr);
+
+ ret = setup_vmgenid_state(state, virt_addr);
+ if (ret)
+ return ret;
+
+ ret = platform_get_irq(pdev, 0);
+ if (ret < 0)
+ return ret;
+
+ state->irq = ret;
+ pdev->dev.driver_data = state;
+
+ ret = devm_request_irq(&pdev->dev, state->irq, vmgenid_of_irq_handler,
+ IRQF_SHARED, "vmgenid", &pdev->dev);
+ if (ret)
+ pdev->dev.driver_data = NULL;
+
+ return ret;
+}
+
static int vmgenid_add(struct platform_device *pdev)
{
struct vmgenid_state *state;
@@ -116,7 +155,10 @@ static int vmgenid_add(struct platform_device *pdev)
if (!state)
return -ENOMEM;
- ret = vmgenid_add_acpi(dev, state);
+ if (dev->of_node)
+ ret = vmgenid_add_of(pdev, state);
+ else
+ ret = vmgenid_add_acpi(dev, state);
if (ret)
devm_kfree(dev, state);
@@ -124,6 +166,12 @@ static int vmgenid_add(struct platform_device *pdev)
return ret;
}
+static const struct of_device_id vmgenid_of_ids[] = {
+ { .compatible = "microsoft,vmgenid", },
+ { },
+};
+MODULE_DEVICE_TABLE(of, vmgenid_of_ids);
+
static const struct acpi_device_id vmgenid_acpi_ids[] = {
{ "VMGENCTR", 0 },
{ "VM_GEN_COUNTER", 0 },
@@ -136,6 +184,7 @@ static struct platform_driver vmgenid_plaform_driver = {
.driver = {
.name = "vmgenid",
.acpi_match_table = vmgenid_acpi_ids,
+ .of_match_table = vmgenid_of_ids,
},
};
--
2.40.1
On 4/17/24 14:09, Krzysztof Kozlowski wrote:
>
> On 17/04/2024 12:40, Babis Chalios wrote:
>> Virtual Machine Generation ID driver was introduced in commit af6b54e2b5ba
>> ("virt: vmgenid: notify RNG of VM fork and supply generation ID"), as an
>> ACPI only device.
>>
>> VMGenID specification http://go.microsoft.com/fwlink/?LinkId=260709 defines
>> a mechanism for the BIOS/hypervisors to communicate to the virtual machine
>> that it is executed with a different configuration (e.g. snapshot execution
>> or creation from a template).
>> The guest operating system can use the notification for various purposes
>> such as re-initializing its random number generator etc.
>>
>> As per the specs, hypervisor should provide a globally unique identified,
>> or GUID via ACPI.
>>
>> This patch tries to mimic the mechanism to provide the same functionality
>> which is for a hypervisor/BIOS to notify the virtual machine when it is
>> executed with a different configuration.
>>
>> As part of this support the devicetree bindings requires the hypervisors or
>> BIOS to provide a memory address which holds the GUID and an IRQ which is
>> used to notify when there is a change in the GUID.
>> The memory exposed in the DT should follow the rules defined in the
>> vmgenid spec mentioned above.
>>
>> *Reason for this change*:
>> Chosing ACPI or devicetree is an intrinsic part of an hypervisor design.
>> Without going into details of why a hypervisor would chose DT over ACPI,
>> we would like to highlight that the hypervisors that have chose devicetree
>> and now want to make use of the vmgenid functionality cannot do so today
>> because vmgenid is an ACPI only device.
>> This forces these hypervisors to change their design which could have
>> undesirable impacts on their use-cases, test-scenarios etc.
>>
>> The point of vmgenid is to provide a mechanism to discover a GUID when
>> the execution state of a virtual machine changes and the simplest
>> way to do it is pass a memory location and an interrupt via devicetree.
>> It would complicate things unnecessarily if instead of using devicetree,
>> we try to implement a new protocol or modify other protocols to somehow
>> provide the same functionility.
>>
>> We believe that adding a devicetree binding for vmgenid is a simpler,
>> better alternative to provide the same functionality and will allow
>> such hypervisors as mentioned above to continue using devicetree.
>>
>> More references to vmgenid specs:
>> - https://www.qemu.org/docs/master/specs/vmgenid.html
>> - https://learn.microsoft.com/en-us/windows/win32/hyperv_v2/virtual-
>> machine-generation-identifier
>>
>> Co-authored-by: Sudan Landge <[email protected]>
>> Signed-off-by: Babis Chalios <[email protected]>
> What happened here?
>
> NAK
>
> You are no the author of this patch. You changed here nothing and you
> took authorship?
>
> Read carefully submitting patches, this is not acceptable.
Hi Krzysztof,
Thanks for your review and your comments (both here and at the thread in
the previous version). I will read again the documentation for DCO on
submitting
patches.
In the meantime, I followed the suggestion of Alex in the discussion of the
previous thread and had already sent v6 before receiving yours and Jason's
responses. In my defense, I didn't want to plagiarize Sudan here. It
just seemed
from the discussion in the previous version that this is the standard
practice
when someone is taking over the submission of a patch-set.
I will re-create the patches with correct authorship, my SoB and the
Reviewed-by
tags I had received in previous versions and send a v7.
Cheers,
Babis
>
> Best regards,
> Krzysztof
>
On Wed, Apr 17, 2024 at 5:19 PM Babis Chalios <[email protected]> wrote:
> I will re-create the patches with correct authorship, my SoB and the
> Reviewed-by
> tags I had received in previous versions and send a v7.
Please don't bother.
Jason
Hi Krzysztof,
On Wed, Apr 17, 2024 at 6:02 PM Krzysztof Kozlowski <[email protected]> wrote:
> Usually we do not recommend hiding code behind !CONFIG_OF because this
> limits possible usage on ACPI systems via PRP0001. Not sure if it is
> applicable here, because there is already ACPI matching.
>
> I would suggest choose whatever makes code simpler... Or just mark some
> pieces with __maybe_unused if they are really not used? That would avoid
> ifdeffery.
Interesting about PRP0001. Alright. It looks like I can't quite do
without ifdefs because the code dereferences ->handle in a acpi_device
struct, but I can minimize it all to just a single ifdef.
Jason