[PATCH] powernv: Properly mask POWER9 DD1 PVR for different chip types

[PATCH] powernv: Properly mask POWER9 DD1 PVR for different chip types

[Michael Neuling]

Bits 48:51 in the PVR for POWER9 represent different chip types (scale
up vs out and 12 vs 24 core). Current chips have 0 here, but could be
non-zero in the future.

This changes the POWER9 DD1 mask to correctly ignore these bits 48:51.

Signed-off-by: Michael Neuling <mikey@neuling.org>
---
 arch/powerpc/kernel/cputable.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index 9b3e88b1a9..89dcd94237 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -526,8 +526,8 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.machine_check_early	= __machine_check_early_realmode_p8,
 		.platform		= "power8",
 	},
-	{	/* Power9 DD1*/
-		.pvr_mask		= 0xffffff00,
+	{	/* Power9 DD1. Bits 48:51 represent chip type so mask these */
+		.pvr_mask		= 0xffff0f00,
 		.pvr_value		= 0x004e0100,
 		.cpu_name		= "POWER9 (raw)",
 		.cpu_features		= CPU_FTRS_POWER9_DD1,
-- 
2.11.0

Re: [PATCH] powernv: Properly mask POWER9 DD1 PVR for different chip types

[Michael Ellerman]

Michael Neuling <mikey@neuling.org> writes:

> Bits 48:51 in the PVR for POWER9 represent different chip types (scale
> up vs out and 12 vs 24 core). Current chips have 0 here, but could be
> non-zero in the future.
>
> This changes the POWER9 DD1 mask to correctly ignore these bits 48:51.
>
> Signed-off-by: Michael Neuling <mikey@neuling.org>
> ---
>  arch/powerpc/kernel/cputable.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)

Presumably we should backport this?

cheers

> diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
> index 9b3e88b1a9..89dcd94237 100644
> --- a/arch/powerpc/kernel/cputable.c
> +++ b/arch/powerpc/kernel/cputable.c
> @@ -526,8 +526,8 @@ static struct cpu_spec __initdata cpu_specs[] = {
>  		.machine_check_early	= __machine_check_early_realmode_p8,
>  		.platform		= "power8",
>  	},
> -	{	/* Power9 DD1*/
> -		.pvr_mask		= 0xffffff00,
> +	{	/* Power9 DD1. Bits 48:51 represent chip type so mask these */
> +		.pvr_mask		= 0xffff0f00,
>  		.pvr_value		= 0x004e0100,
>  		.cpu_name		= "POWER9 (raw)",
>  		.cpu_features		= CPU_FTRS_POWER9_DD1,
> -- 
> 2.11.0

Re: [PATCH] powernv: Properly mask POWER9 DD1 PVR for different chip types

[Michael Neuling]

On Wed, 2017-06-07 at 17:24 +1000, Michael Ellerman wrote:
> Michael Neuling <mikey@neuling.org> writes:
>=20
> > Bits 48:51 in the PVR for POWER9 represent different chip types (scale
> > up vs out and 12 vs 24 core). Current chips have 0 here, but could be
> > non-zero in the future.
> >=20
> > This changes the POWER9 DD1 mask to correctly ignore these bits 48:51.
> >=20
> > Signed-off-by: Michael Neuling <mikey@neuling.org>
> > ---
> > =C2=A0arch/powerpc/kernel/cputable.c | 4 ++--
> > =C2=A01 file changed, 2 insertions(+), 2 deletions(-)
>=20
> Presumably we should backport this?

Actually, we need to scrap this patch.

Looks like the scale up version will be marked as DD1, but it won't need th=
ese
workarounds. So we need it to match on the other POWER9 entry.

Mikey

>=20
> cheers
>=20
> > diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputa=
ble.c
> > index 9b3e88b1a9..89dcd94237 100644
> > --- a/arch/powerpc/kernel/cputable.c
> > +++ b/arch/powerpc/kernel/cputable.c
> > @@ -526,8 +526,8 @@ static struct cpu_spec __initdata cpu_specs[] =3D {
> > =C2=A0		.machine_check_early	=3D
> > __machine_check_early_realmode_p8,
> > =C2=A0		.platform		=3D "power8",
> > =C2=A0	},
> > -	{	/* Power9 DD1*/
> > -		.pvr_mask		=3D 0xffffff00,
> > +	{	/* Power9 DD1. Bits 48:51 represent chip type so mask
> > these */
> > +		.pvr_mask		=3D 0xffff0f00,
> > =C2=A0		.pvr_value		=3D 0x004e0100,
> > =C2=A0		.cpu_name		=3D "POWER9 (raw)",
> > =C2=A0		.cpu_features		=3D CPU_FTRS_POWER9_DD1,
> > --=C2=A0
> > 2.11.0
>=20
>=20

Re: [PATCH] powernv: Properly mask POWER9 DD1 PVR for different chip types

[Michael Ellerman]

Michael Neuling <mikey@neuling.org> writes:

> On Wed, 2017-06-07 at 17:24 +1000, Michael Ellerman wrote:
>> Michael Neuling <mikey@neuling.org> writes:
>>=20
>> > Bits 48:51 in the PVR for POWER9 represent different chip types (scale
>> > up vs out and 12 vs 24 core). Current chips have 0 here, but could be
>> > non-zero in the future.
>> >=20
>> > This changes the POWER9 DD1 mask to correctly ignore these bits 48:51.
>> >=20
>> > Signed-off-by: Michael Neuling <mikey@neuling.org>
>> > ---
>> > =C2=A0arch/powerpc/kernel/cputable.c | 4 ++--
>> > =C2=A01 file changed, 2 insertions(+), 2 deletions(-)
>>=20
>> Presumably we should backport this?
>
> Actually, we need to scrap this patch.
>
> Looks like the scale up version will be marked as DD1, but it won't need =
these
> workarounds. So we need it to match on the other POWER9 entry.

OK.

/giphy "dumpster fire"

cheers