diff mbox series

[RFC,03/21] pci: Define TEE-IO bit in PCIe device capabilities

Message ID 20240823132137.336874-4-aik@amd.com (mailing list archive)
State New
Headers show
Series Secure VFIO, TDISP, SEV TIO | expand

Commit Message

Alexey Kardashevskiy Aug. 23, 2024, 1:21 p.m. UTC
A new bit #30 from the PCI Express Device Capabilities Register is defined
in PCIe 6.1 as "TEE Device Interface Security Protocol (TDISP)".

Define the macro.

Signed-off-by: Alexey Kardashevskiy <aik@amd.com>
---
 include/uapi/linux/pci_regs.h | 1 +
 1 file changed, 1 insertion(+)

Comments

Bjorn Helgaas Aug. 23, 2024, 10:19 p.m. UTC | #1
On Fri, Aug 23, 2024 at 11:21:17PM +1000, Alexey Kardashevskiy wrote:
> A new bit #30 from the PCI Express Device Capabilities Register is defined
> in PCIe 6.1 as "TEE Device Interface Security Protocol (TDISP)".

Include spec section number.
Jonathan Cameron Aug. 28, 2024, 1:54 p.m. UTC | #2
On Fri, 23 Aug 2024 23:21:17 +1000
Alexey Kardashevskiy <aik@amd.com> wrote:

> A new bit #30 from the PCI Express Device Capabilities Register is defined
> in PCIe 6.1 as "TEE Device Interface Security Protocol (TDISP)".
No it isn't.

TEE-IO supported - When Set, this bit indicates the Function implements the TEE-IO
functionality as described by ....

So it is defined as TEE-IO not TDISP even though that definition is in the
TDISP section fo the spec.

As Bjorn said, spec reference.

Jonathan

> 
> Define the macro.
> 
> Signed-off-by: Alexey Kardashevskiy <aik@amd.com>
> ---
>  include/uapi/linux/pci_regs.h | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/include/uapi/linux/pci_regs.h b/include/uapi/linux/pci_regs.h
> index 94c00996e633..0011a301b8c5 100644
> --- a/include/uapi/linux/pci_regs.h
> +++ b/include/uapi/linux/pci_regs.h
> @@ -498,6 +498,7 @@
>  #define  PCI_EXP_DEVCAP_PWR_VAL	0x03fc0000 /* Slot Power Limit Value */
>  #define  PCI_EXP_DEVCAP_PWR_SCL	0x0c000000 /* Slot Power Limit Scale */
>  #define  PCI_EXP_DEVCAP_FLR     0x10000000 /* Function Level Reset */
> +#define  PCI_EXP_DEVCAP_TEE_IO  0x40000000 /* TEE-IO Supported (TDISP) */
>  #define PCI_EXP_DEVCTL		0x08	/* Device Control */
>  #define  PCI_EXP_DEVCTL_CERE	0x0001	/* Correctable Error Reporting En. */
>  #define  PCI_EXP_DEVCTL_NFERE	0x0002	/* Non-Fatal Error Reporting Enable */
Dan Williams Aug. 30, 2024, 2:21 a.m. UTC | #3
Alexey Kardashevskiy wrote:
> A new bit #30 from the PCI Express Device Capabilities Register is defined
> in PCIe 6.1 as "TEE Device Interface Security Protocol (TDISP)".
> 
> Define the macro.
> 
> Signed-off-by: Alexey Kardashevskiy <aik@amd.com>
> ---
>  include/uapi/linux/pci_regs.h | 1 +
>  1 file changed, 1 insertion(+)

Not sure this is justified as a standalone patch, lets fold it in with
its user.
Alexey Kardashevskiy Aug. 30, 2024, 4:04 a.m. UTC | #4
On 30/8/24 12:21, Dan Williams wrote:
> Alexey Kardashevskiy wrote:
>> A new bit #30 from the PCI Express Device Capabilities Register is defined
>> in PCIe 6.1 as "TEE Device Interface Security Protocol (TDISP)".
>>
>> Define the macro.
>>
>> Signed-off-by: Alexey Kardashevskiy <aik@amd.com>
>> ---
>>   include/uapi/linux/pci_regs.h | 1 +
>>   1 file changed, 1 insertion(+)
> 
> Not sure this is justified as a standalone patch, lets fold it in with
> its user.

not sure either but this one is already defined in the PCIe spec for 
some time and lspci knows it but it is going quite some time before "its 
user" makes it to the upstream linux.
Dan Williams Aug. 30, 2024, 9:37 p.m. UTC | #5
Alexey Kardashevskiy wrote:
> On 30/8/24 12:21, Dan Williams wrote:
> > Alexey Kardashevskiy wrote:
> >> A new bit #30 from the PCI Express Device Capabilities Register is defined
> >> in PCIe 6.1 as "TEE Device Interface Security Protocol (TDISP)".
> >>
> >> Define the macro.
> >>
> >> Signed-off-by: Alexey Kardashevskiy <aik@amd.com>
> >> ---
> >>   include/uapi/linux/pci_regs.h | 1 +
> >>   1 file changed, 1 insertion(+)
> > 
> > Not sure this is justified as a standalone patch, lets fold it in with
> > its user.
> 
> not sure either but this one is already defined in the PCIe spec for 
> some time and lspci knows it but it is going quite some time before "its 
> user" makes it to the upstream linux.

So, wait. I.e. if the answer to the question "what does Linux lose by
not merging a patch?" is "nothing", then there is no urgency to merge
it.
diff mbox series

Patch

diff --git a/include/uapi/linux/pci_regs.h b/include/uapi/linux/pci_regs.h
index 94c00996e633..0011a301b8c5 100644
--- a/include/uapi/linux/pci_regs.h
+++ b/include/uapi/linux/pci_regs.h
@@ -498,6 +498,7 @@ 
 #define  PCI_EXP_DEVCAP_PWR_VAL	0x03fc0000 /* Slot Power Limit Value */
 #define  PCI_EXP_DEVCAP_PWR_SCL	0x0c000000 /* Slot Power Limit Scale */
 #define  PCI_EXP_DEVCAP_FLR     0x10000000 /* Function Level Reset */
+#define  PCI_EXP_DEVCAP_TEE_IO  0x40000000 /* TEE-IO Supported (TDISP) */
 #define PCI_EXP_DEVCTL		0x08	/* Device Control */
 #define  PCI_EXP_DEVCTL_CERE	0x0001	/* Correctable Error Reporting En. */
 #define  PCI_EXP_DEVCTL_NFERE	0x0002	/* Non-Fatal Error Reporting Enable */