diff mbox

[v3,10/16] xen/arm: cpufeature: Provide an helper to check if a capability is supported

Message ID 1465315583-1278-11-git-send-email-julien.grall@arm.com (mailing list archive)
State New, archived
Headers show

Commit Message

Julien Grall June 7, 2016, 4:06 p.m. UTC
The CPU capabilities will be set depending on the value found in the CPU
registers. This patch provides a generic to go through a set of capabilities
and find which one should be enabled.

The parameter "info" is used to display the kind of capability updated (e.g
workaround, feature...).

Signed-off-by: Julien Grall <julien.grall@arm.com>

---
    Changes in v3:
         - Patch added. The code was previously part of "Detect
         silicon...".
---
 xen/arch/arm/cpufeature.c        | 16 ++++++++++++++++
 xen/include/asm-arm/cpufeature.h |  9 +++++++++
 2 files changed, 25 insertions(+)

Comments

Stefano Stabellini June 22, 2016, 9:59 a.m. UTC | #1
On Tue, 7 Jun 2016, Julien Grall wrote:
> The CPU capabilities will be set depending on the value found in the CPU
> registers. This patch provides a generic to go through a set of capabilities
> and find which one should be enabled.
> 
> The parameter "info" is used to display the kind of capability updated (e.g
> workaround, feature...).
> 
> Signed-off-by: Julien Grall <julien.grall@arm.com>
> 
> ---
>     Changes in v3:
>          - Patch added. The code was previously part of "Detect
>          silicon...".
> ---
>  xen/arch/arm/cpufeature.c        | 16 ++++++++++++++++
>  xen/include/asm-arm/cpufeature.h |  9 +++++++++
>  2 files changed, 25 insertions(+)
> 
> diff --git a/xen/arch/arm/cpufeature.c b/xen/arch/arm/cpufeature.c
> index 7a1b56b..088625b 100644
> --- a/xen/arch/arm/cpufeature.c
> +++ b/xen/arch/arm/cpufeature.c
> @@ -24,6 +24,22 @@
>  
>  DECLARE_BITMAP(cpu_hwcaps, ARM_NCAPS);
>  
> +void update_cpu_capabilities(const struct arm_cpu_capabilities *caps,
> +                             const char *info)
> +{
> +    int i;
> +
> +    for ( i = 0; caps[i].matches; i++ )
> +    {
> +        if ( !caps[i].matches(&caps[i]) )
> +            continue;
> +
> +        if ( !cpus_have_cap(caps[i].capability) && caps[i].desc )
> +            printk(XENLOG_INFO "%s: %s\n", info, caps[i].desc);

The patch is OK. I still have the same comment as before: if an info
parameter is passed to this function with a string, then I think the
string should contain some useful information. Not a generic message.
I would just do:

    printk(XENLOG_INFO "enable workaround for: %s\n", caps[i].desc);

If one day we need to change the message depending on the caller, then
that day we'll add a parameter to update_cpu_capabilities.


> +        cpus_set_cap(caps[i].capability);
> +    }
> +}
> +
>  /*
>   * Local variables:
>   * mode: C
> diff --git a/xen/include/asm-arm/cpufeature.h b/xen/include/asm-arm/cpufeature.h
> index 2bebad1..be2414c 100644
> --- a/xen/include/asm-arm/cpufeature.h
> +++ b/xen/include/asm-arm/cpufeature.h
> @@ -62,6 +62,15 @@ static inline void cpus_set_cap(unsigned int num)
>          __set_bit(num, cpu_hwcaps);
>  }
>  
> +struct arm_cpu_capabilities {
> +    const char *desc;
> +    u16 capability;
> +    bool_t (*matches)(const struct arm_cpu_capabilities *);
> +};
> +
> +void update_cpu_capabilities(const struct arm_cpu_capabilities *caps,
> +                             const char *info);
> +
>  #endif /* __ASSEMBLY__ */
>  
>  #endif
> -- 
> 1.9.1
>
Julien Grall June 22, 2016, 10:09 a.m. UTC | #2
Hi Stefano,

On 22/06/16 10:59, Stefano Stabellini wrote:
> On Tue, 7 Jun 2016, Julien Grall wrote:
>> The CPU capabilities will be set depending on the value found in the CPU
>> registers. This patch provides a generic to go through a set of capabilities
>> and find which one should be enabled.
>>
>> The parameter "info" is used to display the kind of capability updated (e.g
>> workaround, feature...).
>>
>> Signed-off-by: Julien Grall <julien.grall@arm.com>
>>
>> ---
>>      Changes in v3:
>>           - Patch added. The code was previously part of "Detect
>>           silicon...".
>> ---
>>   xen/arch/arm/cpufeature.c        | 16 ++++++++++++++++
>>   xen/include/asm-arm/cpufeature.h |  9 +++++++++
>>   2 files changed, 25 insertions(+)
>>
>> diff --git a/xen/arch/arm/cpufeature.c b/xen/arch/arm/cpufeature.c
>> index 7a1b56b..088625b 100644
>> --- a/xen/arch/arm/cpufeature.c
>> +++ b/xen/arch/arm/cpufeature.c
>> @@ -24,6 +24,22 @@
>>
>>   DECLARE_BITMAP(cpu_hwcaps, ARM_NCAPS);
>>
>> +void update_cpu_capabilities(const struct arm_cpu_capabilities *caps,
>> +                             const char *info)
>> +{
>> +    int i;
>> +
>> +    for ( i = 0; caps[i].matches; i++ )
>> +    {
>> +        if ( !caps[i].matches(&caps[i]) )
>> +            continue;
>> +
>> +        if ( !cpus_have_cap(caps[i].capability) && caps[i].desc )
>> +            printk(XENLOG_INFO "%s: %s\n", info, caps[i].desc);
>
> The patch is OK. I still have the same comment as before: if an info
> parameter is passed to this function with a string, then I think the
> string should contain some useful information. Not a generic message.

It is useful. As mentioned in the commit message the parameter "info" is 
used to display the kind of capability updated (e.g workaround, feature, 
...).

> I would just do:
>
>      printk(XENLOG_INFO "enable workaround for: %s\n", caps[i].desc);
>
> If one day we need to change the message depending on the caller, then
> that day we'll add a parameter to update_cpu_capabilities.

We both know that this function will be used for other purpose very 
soon. So I would rather avoid to modify again in the future. That is why 
I described the purpose in the commit message.

Cheers,
Stefano Stabellini June 22, 2016, 10:14 a.m. UTC | #3
On Wed, 22 Jun 2016, Julien Grall wrote:
> Hi Stefano,
> 
> On 22/06/16 10:59, Stefano Stabellini wrote:
> > On Tue, 7 Jun 2016, Julien Grall wrote:
> > > The CPU capabilities will be set depending on the value found in the CPU
> > > registers. This patch provides a generic to go through a set of
> > > capabilities
> > > and find which one should be enabled.
> > > 
> > > The parameter "info" is used to display the kind of capability updated
> > > (e.g
> > > workaround, feature...).
> > > 
> > > Signed-off-by: Julien Grall <julien.grall@arm.com>
> > > 
> > > ---
> > >      Changes in v3:
> > >           - Patch added. The code was previously part of "Detect
> > >           silicon...".
> > > ---
> > >   xen/arch/arm/cpufeature.c        | 16 ++++++++++++++++
> > >   xen/include/asm-arm/cpufeature.h |  9 +++++++++
> > >   2 files changed, 25 insertions(+)
> > > 
> > > diff --git a/xen/arch/arm/cpufeature.c b/xen/arch/arm/cpufeature.c
> > > index 7a1b56b..088625b 100644
> > > --- a/xen/arch/arm/cpufeature.c
> > > +++ b/xen/arch/arm/cpufeature.c
> > > @@ -24,6 +24,22 @@
> > > 
> > >   DECLARE_BITMAP(cpu_hwcaps, ARM_NCAPS);
> > > 
> > > +void update_cpu_capabilities(const struct arm_cpu_capabilities *caps,
> > > +                             const char *info)
> > > +{
> > > +    int i;
> > > +
> > > +    for ( i = 0; caps[i].matches; i++ )
> > > +    {
> > > +        if ( !caps[i].matches(&caps[i]) )
> > > +            continue;
> > > +
> > > +        if ( !cpus_have_cap(caps[i].capability) && caps[i].desc )
> > > +            printk(XENLOG_INFO "%s: %s\n", info, caps[i].desc);
> > 
> > The patch is OK. I still have the same comment as before: if an info
> > parameter is passed to this function with a string, then I think the
> > string should contain some useful information. Not a generic message.
> 
> It is useful. As mentioned in the commit message the parameter "info" is used
> to display the kind of capability updated (e.g workaround, feature, ...).
> 
> > I would just do:
> > 
> >      printk(XENLOG_INFO "enable workaround for: %s\n", caps[i].desc);
> > 
> > If one day we need to change the message depending on the caller, then
> > that day we'll add a parameter to update_cpu_capabilities.
> 
> We both know that this function will be used for other purpose very soon. So I
> would rather avoid to modify again in the future. That is why I described the
> purpose in the commit message.

I am unconvinced, but I'll leave it be.

Acked-by: Stefano Stabellini <sstabellini@kernel.org>
diff mbox

Patch

diff --git a/xen/arch/arm/cpufeature.c b/xen/arch/arm/cpufeature.c
index 7a1b56b..088625b 100644
--- a/xen/arch/arm/cpufeature.c
+++ b/xen/arch/arm/cpufeature.c
@@ -24,6 +24,22 @@ 
 
 DECLARE_BITMAP(cpu_hwcaps, ARM_NCAPS);
 
+void update_cpu_capabilities(const struct arm_cpu_capabilities *caps,
+                             const char *info)
+{
+    int i;
+
+    for ( i = 0; caps[i].matches; i++ )
+    {
+        if ( !caps[i].matches(&caps[i]) )
+            continue;
+
+        if ( !cpus_have_cap(caps[i].capability) && caps[i].desc )
+            printk(XENLOG_INFO "%s: %s\n", info, caps[i].desc);
+        cpus_set_cap(caps[i].capability);
+    }
+}
+
 /*
  * Local variables:
  * mode: C
diff --git a/xen/include/asm-arm/cpufeature.h b/xen/include/asm-arm/cpufeature.h
index 2bebad1..be2414c 100644
--- a/xen/include/asm-arm/cpufeature.h
+++ b/xen/include/asm-arm/cpufeature.h
@@ -62,6 +62,15 @@  static inline void cpus_set_cap(unsigned int num)
         __set_bit(num, cpu_hwcaps);
 }
 
+struct arm_cpu_capabilities {
+    const char *desc;
+    u16 capability;
+    bool_t (*matches)(const struct arm_cpu_capabilities *);
+};
+
+void update_cpu_capabilities(const struct arm_cpu_capabilities *caps,
+                             const char *info);
+
 #endif /* __ASSEMBLY__ */
 
 #endif