[PATCH] DaVinci: correct MDSTAT_STATE_MASK

[PATCH] DaVinci: correct MDSTAT_STATE_MASK

[Sergei Shtylyov]

MDSTAT.STATE occupies bits 0..5 according to all available documentation, so fix
the #define MDSTAT_STATE_MASK at last. Using the wrong value seems to have been
harmless though...

This was noticed by me back in 2009 but I didn't follow up with the patch back
then... :-/

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>

---
The patch is against the recent DaVinci tree.

 arch/arm/mach-davinci/include/mach/psc.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Index: linux-davinci/arch/arm/mach-davinci/include/mach/psc.h
===================================================================
--- linux-davinci.orig/arch/arm/mach-davinci/include/mach/psc.h
+++ linux-davinci/arch/arm/mach-davinci/include/mach/psc.h
@@ -243,7 +243,7 @@
 #define PSC_STATE_DISABLE	2
 #define PSC_STATE_ENABLE	3
 
-#define MDSTAT_STATE_MASK	0x1f
+#define MDSTAT_STATE_MASK	0x3f
 #define MDCTL_FORCE		BIT(31)
 
 #ifndef __ASSEMBLER__

[PATCH] DaVinci: correct MDSTAT_STATE_MASK

[Sergei Shtylyov]

Hello.

I wrote:

> MDSTAT.STATE occupies bits 0..5 according to all available documentation, so fix
> the #define MDSTAT_STATE_MASK at last. Using the wrong value seems to have been
> harmless though...

    Except maybe sleep.S -- davinci_ddr_psc_config() could exit too early when 
disabling DDR clock. The same is true with davinci_psc_config() in general but I 
don't think that anything depended on its timely return in that case.

> This was noticed by me back in 2009 but I didn't follow up with the patch back
> then... :-/

    Shame on me. :-)

> Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>

WBR, Sergei

[PATCH] DaVinci: correct MDSTAT_STATE_MASK

[Nori, Sekhar]

On Fri, Jul 08, 2011 at 20:54:57, Sergei Shtylyov wrote:
> MDSTAT.STATE occupies bits 0..5 according to all available documentation, so fix
> the #define MDSTAT_STATE_MASK at last. Using the wrong value seems to have been
> harmless though...
> 
> This was noticed by me back in 2009 but I didn't follow up with the patch back
> then... :-/
> 
> Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>

I applied this with two changes.

Changed the headline to "ARM: davinci: correct MDSTAT_STATE_MASK"
to get the standardization requested by Arnd.

Also dropped the second paragraph of the commit text since it
doesn't really belong to the permanent history ;)

Thanks,
Sekhar

[PATCH] DaVinci: correct MDSTAT_STATE_MASK

[Sergei Shtylyov]

Hello.

On 07/27/2011 08:18 PM, Nori, Sekhar wrote:

>> MDSTAT.STATE occupies bits 0..5 according to all available documentation, so fix
>> the #define MDSTAT_STATE_MASK at last. Using the wrong value seems to have been
>> harmless though...

>> This was noticed by me back in 2009 but I didn't follow up with the patch back
>> then... :-/

>> Signed-off-by: Sergei Shtylyov<sshtylyov@ru.mvista.com>

> I applied this with two changes.

> Changed the headline to "ARM: davinci: correct MDSTAT_STATE_MASK"
> to get the standardization requested by Arnd.

> Also dropped the second paragraph of the commit text since it
> doesn't really belong to the permanent history ;)

    You have queued it for 3.2, right?

> Thanks,
> Sekhar

WBR, Sergei

[PATCH] DaVinci: correct MDSTAT_STATE_MASK

[Nori, Sekhar]

Hi Sergei,

On Tue, Sep 06, 2011 at 19:20:20, Sergei Shtylyov wrote:
> Hello.
> 
> On 07/27/2011 08:18 PM, Nori, Sekhar wrote:
> 
> >> MDSTAT.STATE occupies bits 0..5 according to all available documentation, so fix
> >> the #define MDSTAT_STATE_MASK at last. Using the wrong value seems to have been
> >> harmless though...
> 
> >> This was noticed by me back in 2009 but I didn't follow up with the patch back
> >> then... :-/
> 
> >> Signed-off-by: Sergei Shtylyov<sshtylyov@ru.mvista.com>
> 
> > I applied this with two changes.
> 
> > Changed the headline to "ARM: davinci: correct MDSTAT_STATE_MASK"
> > to get the standardization requested by Arnd.
> 
> > Also dropped the second paragraph of the commit text since it
> > doesn't really belong to the permanent history ;)
> 
>     You have queued it for 3.2, right?

No, actually planning to send for v3.1. Have been delayed a bit
due to my recent travel.

Thanks,
Sekhar