We have accessing P2SB from a very few places for quite known hardware.
When a new SoC appears in intel-family.h it's not obvious that it needs
to be added to p2sb.c as well. Instead, provide default BDF and refactor
p2sb_get_devfn() to always succeed. If in the future we would need to
exclude something, we may add a list of unsupported IDs.
Without this change the iTCO on Intel Commet Lake SoCs became unavailable:
i801_smbus 0000:00:1f.4: failed to create iTCO device
Fixes: 5c7b9167ddf8 ("i2c: i801: convert to use common P2SB accessor")
Reported-by: Jarkko Nikula <[email protected]>
Signed-off-by: Andy Shevchenko <[email protected]>
---
drivers/platform/x86/p2sb.c | 15 ++++++---------
1 file changed, 6 insertions(+), 9 deletions(-)
diff --git a/drivers/platform/x86/p2sb.c b/drivers/platform/x86/p2sb.c
index 384d0962ae93..1cf2471d54dd 100644
--- a/drivers/platform/x86/p2sb.c
+++ b/drivers/platform/x86/p2sb.c
@@ -19,26 +19,23 @@
#define P2SBC 0xe0
#define P2SBC_HIDE BIT(8)
+#define P2SB_DEVFN_DEFAULT PCI_DEVFN(31, 1)
+
static const struct x86_cpu_id p2sb_cpu_ids[] = {
X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT, PCI_DEVFN(13, 0)),
- X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_D, PCI_DEVFN(31, 1)),
- X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT_D, PCI_DEVFN(31, 1)),
- X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE, PCI_DEVFN(31, 1)),
- X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L, PCI_DEVFN(31, 1)),
- X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE, PCI_DEVFN(31, 1)),
- X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_L, PCI_DEVFN(31, 1)),
{}
};
static int p2sb_get_devfn(unsigned int *devfn)
{
+ unsigned int fn = P2SB_DEVFN_DEFAULT;
const struct x86_cpu_id *id;
id = x86_match_cpu(p2sb_cpu_ids);
- if (!id)
- return -ENODEV;
+ if (id)
+ fn = (unsigned int)id->driver_data;
- *devfn = (unsigned int)id->driver_data;
+ *devfn = fn;
return 0;
}
--
2.35.1
[Public]
> -----Original Message-----
> From: Andy Shevchenko <[email protected]>
> Sent: Friday, November 4, 2022 10:49
> To: Hans de Goede <[email protected]>; Andy Shevchenko
> <[email protected]>; platform-driver-
> [email protected]; [email protected]
> Cc: Mark Gross <[email protected]>; Jarkko Nikula
> <[email protected]>
> Subject: [PATCH v1 1/1] platform/x86: p2sb: Don't fail if unknown CPU is
> found
>
> We have accessing P2SB from a very few places for quite known hardware.
>
> When a new SoC appears in intel-family.h it's not obvious that it needs
> to be added to p2sb.c as well. Instead, provide default BDF and refactor
> p2sb_get_devfn() to always succeed. If in the future we would need to
> exclude something, we may add a list of unsupported IDs.
>
> Without this change the iTCO on Intel Commet Lake SoCs became
Isn't it "Comet Lake"?
> unavailable:
>
> i801_smbus 0000:00:1f.4: failed to create iTCO device
>
> Fixes: 5c7b9167ddf8 ("i2c: i801: convert to use common P2SB accessor")
> Reported-by: Jarkko Nikula <[email protected]>
> Signed-off-by: Andy Shevchenko <[email protected]>
> ---
> drivers/platform/x86/p2sb.c | 15 ++++++---------
> 1 file changed, 6 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/platform/x86/p2sb.c b/drivers/platform/x86/p2sb.c
> index 384d0962ae93..1cf2471d54dd 100644
> --- a/drivers/platform/x86/p2sb.c
> +++ b/drivers/platform/x86/p2sb.c
> @@ -19,26 +19,23 @@
> #define P2SBC 0xe0
> #define P2SBC_HIDE BIT(8)
>
> +#define P2SB_DEVFN_DEFAULT PCI_DEVFN(31, 1)
> +
> static const struct x86_cpu_id p2sb_cpu_ids[] = {
> X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT,
> PCI_DEVFN(13, 0)),
> - X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_D,
> PCI_DEVFN(31, 1)),
> - X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT_D,
> PCI_DEVFN(31, 1)),
> - X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE,
> PCI_DEVFN(31, 1)),
> - X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L,
> PCI_DEVFN(31, 1)),
> - X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE,
> PCI_DEVFN(31, 1)),
> - X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_L,
> PCI_DEVFN(31, 1)),
> {}
> };
>
> static int p2sb_get_devfn(unsigned int *devfn)
> {
> + unsigned int fn = P2SB_DEVFN_DEFAULT;
> const struct x86_cpu_id *id;
>
> id = x86_match_cpu(p2sb_cpu_ids);
> - if (!id)
> - return -ENODEV;
> + if (id)
> + fn = (unsigned int)id->driver_data;
>
> - *devfn = (unsigned int)id->driver_data;
> + *devfn = fn;
> return 0;
> }
>
> --
> 2.35.1
On Fri, Nov 04, 2022 at 03:58:22PM +0000, Limonciello, Mario wrote:
> > -----Original Message-----
> > From: Andy Shevchenko <[email protected]>
> > Sent: Friday, November 4, 2022 10:49
...
> > We have accessing P2SB from a very few places for quite known hardware.
> >
> > When a new SoC appears in intel-family.h it's not obvious that it needs
> > to be added to p2sb.c as well. Instead, provide default BDF and refactor
> > p2sb_get_devfn() to always succeed. If in the future we would need to
> > exclude something, we may add a list of unsupported IDs.
> >
> > Without this change the iTCO on Intel Commet Lake SoCs became
>
> Isn't it "Comet Lake"?
Right, thanks!
In any case I would like to have Jarkko's Tested-by before proceeding with
this. Hence v2 is warranted to appear.
> > unavailable:
> >
> > i801_smbus 0000:00:1f.4: failed to create iTCO device
--
With Best Regards,
Andy Shevchenko
Hi
On 11/4/22 19:22, Andy Shevchenko wrote:
> On Fri, Nov 04, 2022 at 03:58:22PM +0000, Limonciello, Mario wrote:
>> Isn't it "Comet Lake"?
>
> Right, thanks!
>
> In any case I would like to have Jarkko's Tested-by before proceeding with
> this. Hence v2 is warranted to appear.
>
Yes, iTCO registration works again with this patch and is functioning,
i.e. able to reset the machine.
Tested-by: Jarkko Nikula <[email protected]>
Hi,
On 11/4/22 16:49, Andy Shevchenko wrote:
> We have accessing P2SB from a very few places for quite known hardware.
>
> When a new SoC appears in intel-family.h it's not obvious that it needs
> to be added to p2sb.c as well. Instead, provide default BDF and refactor
> p2sb_get_devfn() to always succeed. If in the future we would need to
> exclude something, we may add a list of unsupported IDs.
>
> Without this change the iTCO on Intel Commet Lake SoCs became unavailable:
>
> i801_smbus 0000:00:1f.4: failed to create iTCO device
>
> Fixes: 5c7b9167ddf8 ("i2c: i801: convert to use common P2SB accessor")
> Reported-by: Jarkko Nikula <[email protected]>
> Signed-off-by: Andy Shevchenko <[email protected]>
Thanks. I've merged this now, with the Commet Lake spelling error
fixed and with Jarko's Tested-by added.
I'll include this in my upcoming fixes pull-req to Linus.
Regards,
Hans
> ---
> drivers/platform/x86/p2sb.c | 15 ++++++---------
> 1 file changed, 6 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/platform/x86/p2sb.c b/drivers/platform/x86/p2sb.c
> index 384d0962ae93..1cf2471d54dd 100644
> --- a/drivers/platform/x86/p2sb.c
> +++ b/drivers/platform/x86/p2sb.c
> @@ -19,26 +19,23 @@
> #define P2SBC 0xe0
> #define P2SBC_HIDE BIT(8)
>
> +#define P2SB_DEVFN_DEFAULT PCI_DEVFN(31, 1)
> +
> static const struct x86_cpu_id p2sb_cpu_ids[] = {
> X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT, PCI_DEVFN(13, 0)),
> - X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_D, PCI_DEVFN(31, 1)),
> - X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT_D, PCI_DEVFN(31, 1)),
> - X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE, PCI_DEVFN(31, 1)),
> - X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L, PCI_DEVFN(31, 1)),
> - X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE, PCI_DEVFN(31, 1)),
> - X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_L, PCI_DEVFN(31, 1)),
> {}
> };
>
> static int p2sb_get_devfn(unsigned int *devfn)
> {
> + unsigned int fn = P2SB_DEVFN_DEFAULT;
> const struct x86_cpu_id *id;
>
> id = x86_match_cpu(p2sb_cpu_ids);
> - if (!id)
> - return -ENODEV;
> + if (id)
> + fn = (unsigned int)id->driver_data;
>
> - *devfn = (unsigned int)id->driver_data;
> + *devfn = fn;
> return 0;
> }
>