[PATCH] OMAP3: pmu: make CPU_HAS_PMU dependent on OMAP3_EMU

[PATCH] OMAP3: pmu: make CPU_HAS_PMU dependent on OMAP3_EMU

[Will Deacon]

CPU performance event counters on v7 cores will only operate
if either the NIDEN or DBGEN signals are driven high.

For the OMAP3 platform, these signals are driven low by default [i.e.
when a JTAG debugger is not connected] but DBGEN can be asserted in
software by selecting the OMAP3_EMU Kconfig option, which enables the
virtual clock for hardware debugging peripherals.

This patch makes CPU_HAS_PMU depend on OMAP3_EMU for OMAP3 platforms.

Cc: Tony Lindgren <tony@atomide.com>
Cc: Jean Pihet <jpihet@mvista.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
---

Since I don't have an OMAP3 board to test this on I would appreciate
it if somebody with some hardware could (a) confirm that this problem
exists and (b) validate that this patch solves the issue. Thanks!

 arch/arm/Kconfig |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 1f254bd..98922f7 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -955,7 +955,8 @@ config XSCALE_PMU
 	default y
 
 config CPU_HAS_PMU
-	depends on CPU_V6 || CPU_V7 || XSCALE_PMU
+	depends on (CPU_V6 || CPU_V7 || XSCALE_PMU) && \
+		   (!ARCH_OMAP3 || OMAP3_EMU)
 	default y
 	bool
 
-- 
1.6.3.3

[PATCH] OMAP3: pmu: make CPU_HAS_PMU dependent on OMAP3_EMU

[Jean Pihet]

Hi Will,

Sorry for the late reply. Here below are the test results.
Re-sending in plain text only, sorry for the inconvenience.

On Thu, May 27, 2010 at 17:00, Will Deacon <will.deacon@arm.com> wrote:
> CPU performance event counters on v7 cores will only operate
> if either the NIDEN or DBGEN signals are driven high.
>
> For the OMAP3 platform, these signals are driven low by default [i.e.
> when a JTAG debugger is not connected] but DBGEN can be asserted in
> software by selecting the OMAP3_EMU Kconfig option, which enables the
> virtual clock for hardware debugging peripherals.
>
> This patch makes CPU_HAS_PMU depend on OMAP3_EMU for OMAP3 platforms.
>
> Cc: Tony Lindgren <tony@atomide.com>
> Cc: Jean Pihet <jpihet@mvista.com>
> Signed-off-by: Will Deacon <will.deacon@arm.com>
> ---
>
> Since I don't have an OMAP3 board to test this on I would appreciate
> it if somebody with some hardware could (a) confirm that this problem
> exists and (b) validate that this patch solves the issue. Thanks!
>
>  arch/arm/Kconfig |    3 ++-
>  1 files changed, 2 insertions(+), 1 deletions(-)
>
> diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
> index 1f254bd..98922f7 100644
> --- a/arch/arm/Kconfig
> +++ b/arch/arm/Kconfig
> @@ -955,7 +955,8 @@ config XSCALE_PMU
>        default y
>
>  config CPU_HAS_PMU
> -       depends on CPU_V6 || CPU_V7 || XSCALE_PMU
> +       depends on (CPU_V6 || CPU_V7 || XSCALE_PMU) && \
> +                  (!ARCH_OMAP3 || OMAP3_EMU)
>        default y
>        bool
>
> --
> 1.6.3.3
>
>

The patch nicely fixes the issue. Tested with Oprofile on OMAP3530
EVM. RETention and OFF power modes succesfully tested.

Acked-by: Jean Pihet <jpihet@mvista.com>

Regards,
Jean

[PATCH] OMAP3: pmu: make CPU_HAS_PMU dependent on OMAP3_EMU

[Will Deacon]

Jean,

> > diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
> > index 1f254bd..98922f7 100644
> > --- a/arch/arm/Kconfig
> > +++ b/arch/arm/Kconfig
> > @@ -955,7 +955,8 @@ config XSCALE_PMU
> >        default y
> >
> >  config CPU_HAS_PMU
> > -       depends on CPU_V6 || CPU_V7 || XSCALE_PMU
> > +       depends on (CPU_V6 || CPU_V7 || XSCALE_PMU) && \
> > +                  (!ARCH_OMAP3 || OMAP3_EMU)
> >        default y
> >        bool
> >
> > --
> > 1.6.3.3
> >
> >
> 
> The patch nicely fixes the issue. Tested with Oprofile on OMAP3530
> EVM. RETention and OFF power modes succesfully tested.
> 
> Acked-by: Jean Pihet <jpihet@mvista.com>

Brilliant - thanks for testing this.
Tony: would you mind picking this patch up please or would
you prefer that I posted it to Russell's patch system?

Thanks,

Will

[PATCH] OMAP3: pmu: make CPU_HAS_PMU dependent on OMAP3_EMU

[Will Deacon]

Hello,

* I wrote: 

> Tony: would you mind picking this patch up please or would
> you prefer that I posted it to Russell's patch system?

Since this only modifies arch/arm/KConfig and Jean has acked it,
I've submitted it to Russell's patch system as #6195/1.

Will

[PATCH] OMAP3: pmu: make CPU_HAS_PMU dependent on OMAP3_EMU

[Tony Lindgren]

* Will Deacon <will.deacon@arm.com> [100621 17:29]:
> Hello,
> 
> * I wrote: 
> 
> > Tony: would you mind picking this patch up please or would
> > you prefer that I posted it to Russell's patch system?
> 
> Since this only modifies arch/arm/KConfig and Jean has acked it,
> I've submitted it to Russell's patch system as #6195/1.

Yeah good, thanks.

Tony

[PATCH] OMAP3: pmu: make CPU_HAS_PMU dependent on OMAP3_EMU

[karl leplat]

Patch:

CPU performance event counters on v7 cores will only operate
if either the NIDEN or DBGEN signals are driven high.

For the OMAP3 platform, these signals are driven low by default [i.e.
when a JTAG debugger is not connected] but DBGEN can be asserted in
software by selecting the OMAP3_EMU Kconfig option, which enables the
virtual clock for hardware debugging peripherals.

This patch makes CPU_HAS_PMU depend on OMAP3_EMU for OMAP3 platforms.

If i selected OMAP3_EMU option, i didn't see changes, DBGEN is already low. How can i activated the debug mode ?

Thanks !