diff mbox

[U-Boot] arm: socfpga: reset: FIX address of tstscratch register

Message ID 1447348990-2948-1-git-send-email-ilu@linutronix.de
State Accepted
Delegated to: Marek Vasut
Headers show

Commit Message

Philipp Rosenberger Nov. 12, 2015, 5:23 p.m. UTC
The Cyclone V Hard Processor System Technical Reference Manual in the
chapter about the Reset Manager Module Address Map stats that the offset
of the tstscratch register ist 0x54 not 0x24.

Cyclone V Hard Processor System Technical Reference Manual cv_5v4 2015.11.02
page 3-17 Reset Manager Module Address Map

Signed-off-by: Philipp Rosenberger <ilu@linutronix.de>
---
 arch/arm/mach-socfpga/include/mach/reset_manager.h | 1 +
 1 file changed, 1 insertion(+)

Comments

Marek Vasut Nov. 12, 2015, 6:12 p.m. UTC | #1
On Thursday, November 12, 2015 at 06:23:10 PM, Philipp Rosenberger wrote:
> The Cyclone V Hard Processor System Technical Reference Manual in the
> chapter about the Reset Manager Module Address Map stats that the offset
> of the tstscratch register ist 0x54 not 0x24.
> 
> Cyclone V Hard Processor System Technical Reference Manual cv_5v4
> 2015.11.02 page 3-17 Reset Manager Module Address Map
> 
> Signed-off-by: Philipp Rosenberger <ilu@linutronix.de>

Acked-by: Marek Vasut <marex@denx.de>

Thanks for spotting this!

> ---
>  arch/arm/mach-socfpga/include/mach/reset_manager.h | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/arch/arm/mach-socfpga/include/mach/reset_manager.h
> b/arch/arm/mach-socfpga/include/mach/reset_manager.h index
> 8e59578..6eb6011 100644
> --- a/arch/arm/mach-socfpga/include/mach/reset_manager.h
> +++ b/arch/arm/mach-socfpga/include/mach/reset_manager.h
> @@ -25,6 +25,7 @@ struct socfpga_reset_manager {
>  	u32	per2_mod_reset;
>  	u32	brg_mod_reset;
>  	u32	misc_mod_reset;
> +	u32	padding2[12];
>  	u32	tstscratch;
>  };

Best regards,
Marek Vasut
Marek Vasut Nov. 12, 2015, 9:58 p.m. UTC | #2
On Thursday, November 12, 2015 at 06:23:10 PM, Philipp Rosenberger wrote:
> The Cyclone V Hard Processor System Technical Reference Manual in the
> chapter about the Reset Manager Module Address Map stats that the offset
> of the tstscratch register ist 0x54 not 0x24.
> 
> Cyclone V Hard Processor System Technical Reference Manual cv_5v4
> 2015.11.02 page 3-17 Reset Manager Module Address Map
> 
> Signed-off-by: Philipp Rosenberger <ilu@linutronix.de>

Applied, thanks!

Best regards,
Marek Vasut
Stefan Roese Nov. 13, 2015, 6:11 a.m. UTC | #3
Hi Philipp,

On 12.11.2015 18:23, Philipp Rosenberger wrote:
> The Cyclone V Hard Processor System Technical Reference Manual in the
> chapter about the Reset Manager Module Address Map stats that the offset
> of the tstscratch register ist 0x54 not 0x24.
>
> Cyclone V Hard Processor System Technical Reference Manual cv_5v4 2015.11.02
> page 3-17 Reset Manager Module Address Map
>
> Signed-off-by: Philipp Rosenberger <ilu@linutronix.de>
> ---
>   arch/arm/mach-socfpga/include/mach/reset_manager.h | 1 +
>   1 file changed, 1 insertion(+)
>
> diff --git a/arch/arm/mach-socfpga/include/mach/reset_manager.h b/arch/arm/mach-socfpga/include/mach/reset_manager.h
> index 8e59578..6eb6011 100644
> --- a/arch/arm/mach-socfpga/include/mach/reset_manager.h
> +++ b/arch/arm/mach-socfpga/include/mach/reset_manager.h
> @@ -25,6 +25,7 @@ struct socfpga_reset_manager {
>   	u32	per2_mod_reset;
>   	u32	brg_mod_reset;
>   	u32	misc_mod_reset;
> +	u32	padding2[12];
>   	u32	tstscratch;
>   };

Thanks. But usually such padding things are added as "u8" (1 byte)
variables. This makes it easier to calculate the offsets. In this
case:

+	u8	padding2[0x30];

which I would prefer.

Thanks,
Stefan
Marek Vasut Nov. 13, 2015, 7:13 a.m. UTC | #4
On Friday, November 13, 2015 at 07:11:18 AM, Stefan Roese wrote:
> Hi Philipp,
> 
> On 12.11.2015 18:23, Philipp Rosenberger wrote:
> > The Cyclone V Hard Processor System Technical Reference Manual in the
> > chapter about the Reset Manager Module Address Map stats that the offset
> > of the tstscratch register ist 0x54 not 0x24.
> > 
> > Cyclone V Hard Processor System Technical Reference Manual cv_5v4
> > 2015.11.02 page 3-17 Reset Manager Module Address Map
> > 
> > Signed-off-by: Philipp Rosenberger <ilu@linutronix.de>
> > ---
> > 
> >   arch/arm/mach-socfpga/include/mach/reset_manager.h | 1 +
> >   1 file changed, 1 insertion(+)
> > 
> > diff --git a/arch/arm/mach-socfpga/include/mach/reset_manager.h
> > b/arch/arm/mach-socfpga/include/mach/reset_manager.h index
> > 8e59578..6eb6011 100644
> > --- a/arch/arm/mach-socfpga/include/mach/reset_manager.h
> > +++ b/arch/arm/mach-socfpga/include/mach/reset_manager.h
> > @@ -25,6 +25,7 @@ struct socfpga_reset_manager {
> > 
> >   	u32	per2_mod_reset;
> >   	u32	brg_mod_reset;
> >   	u32	misc_mod_reset;
> > 
> > +	u32	padding2[12];
> > 
> >   	u32	tstscratch;
> >   
> >   };
> 
> Thanks. But usually such padding things are added as "u8" (1 byte)
> variables. This makes it easier to calculate the offsets. In this
> case:
> 
> +	u8	padding2[0x30];
> 
> which I would prefer.

I don't mind either way, I can amend the patch (if you don't mind),
so let's hear Dinh's final word.

Best regards,
Marek Vasut
Philipp Rosenberger Nov. 13, 2015, 8:02 a.m. UTC | #5
Hi Marek,

On 13.11.2015 08:13, Marek Vasut wrote:
> On Friday, November 13, 2015 at 07:11:18 AM, Stefan Roese wrote:
>> Hi Philipp,
>>
>> On 12.11.2015 18:23, Philipp Rosenberger wrote:
>>> The Cyclone V Hard Processor System Technical Reference Manual in the
>>> chapter about the Reset Manager Module Address Map stats that the offset
>>> of the tstscratch register ist 0x54 not 0x24.
>>>
>>> Cyclone V Hard Processor System Technical Reference Manual cv_5v4
>>> 2015.11.02 page 3-17 Reset Manager Module Address Map
>>>
>>> Signed-off-by: Philipp Rosenberger <ilu@linutronix.de>
>>> ---
>>>
>>>   arch/arm/mach-socfpga/include/mach/reset_manager.h | 1 +
>>>   1 file changed, 1 insertion(+)
>>>
>>> diff --git a/arch/arm/mach-socfpga/include/mach/reset_manager.h
>>> b/arch/arm/mach-socfpga/include/mach/reset_manager.h index
>>> 8e59578..6eb6011 100644
>>> --- a/arch/arm/mach-socfpga/include/mach/reset_manager.h
>>> +++ b/arch/arm/mach-socfpga/include/mach/reset_manager.h
>>> @@ -25,6 +25,7 @@ struct socfpga_reset_manager {
>>>
>>>   	u32	per2_mod_reset;
>>>   	u32	brg_mod_reset;
>>>   	u32	misc_mod_reset;
>>>
>>> +	u32	padding2[12];
>>>
>>>   	u32	tstscratch;
>>>   
>>>   };
>>
>> Thanks. But usually such padding things are added as "u8" (1 byte)
>> variables. This makes it easier to calculate the offsets. In this
>> case:
>>
>> +	u8	padding2[0x30];
>>
>> which I would prefer.
> 
> I don't mind either way, I can amend the patch (if you don't mind),
> so let's hear Dinh's final word.

If we use u8 in the place we should change the padding1 in the same
struct as well.

Best Regards
Philipp Rosenberger
Marek Vasut Nov. 13, 2015, 8:04 a.m. UTC | #6
On Friday, November 13, 2015 at 09:02:43 AM, Philipp Rosenberger wrote:
> Hi Marek,
> 
> On 13.11.2015 08:13, Marek Vasut wrote:
> > On Friday, November 13, 2015 at 07:11:18 AM, Stefan Roese wrote:
> >> Hi Philipp,
> >> 
> >> On 12.11.2015 18:23, Philipp Rosenberger wrote:
> >>> The Cyclone V Hard Processor System Technical Reference Manual in the
> >>> chapter about the Reset Manager Module Address Map stats that the
> >>> offset of the tstscratch register ist 0x54 not 0x24.
> >>> 
> >>> Cyclone V Hard Processor System Technical Reference Manual cv_5v4
> >>> 2015.11.02 page 3-17 Reset Manager Module Address Map
> >>> 
> >>> Signed-off-by: Philipp Rosenberger <ilu@linutronix.de>
> >>> ---
> >>> 
> >>>   arch/arm/mach-socfpga/include/mach/reset_manager.h | 1 +
> >>>   1 file changed, 1 insertion(+)
> >>> 
> >>> diff --git a/arch/arm/mach-socfpga/include/mach/reset_manager.h
> >>> b/arch/arm/mach-socfpga/include/mach/reset_manager.h index
> >>> 8e59578..6eb6011 100644
> >>> --- a/arch/arm/mach-socfpga/include/mach/reset_manager.h
> >>> +++ b/arch/arm/mach-socfpga/include/mach/reset_manager.h
> >>> @@ -25,6 +25,7 @@ struct socfpga_reset_manager {
> >>> 
> >>>   	u32	per2_mod_reset;
> >>>   	u32	brg_mod_reset;
> >>>   	u32	misc_mod_reset;
> >>> 
> >>> +	u32	padding2[12];
> >>> 
> >>>   	u32	tstscratch;
> >>>   
> >>>   };
> >> 
> >> Thanks. But usually such padding things are added as "u8" (1 byte)
> >> variables. This makes it easier to calculate the offsets. In this
> >> case:
> >> 
> >> +	u8	padding2[0x30];
> >> 
> >> which I would prefer.
> > 
> > I don't mind either way, I can amend the patch (if you don't mind),
> > so let's hear Dinh's final word.
> 
> If we use u8 in the place we should change the padding1 in the same
> struct as well.

In that case, we keep it as-is . Subsequent patch is welcome if anyone cares.

Best regards,
Marek Vasut
diff mbox

Patch

diff --git a/arch/arm/mach-socfpga/include/mach/reset_manager.h b/arch/arm/mach-socfpga/include/mach/reset_manager.h
index 8e59578..6eb6011 100644
--- a/arch/arm/mach-socfpga/include/mach/reset_manager.h
+++ b/arch/arm/mach-socfpga/include/mach/reset_manager.h
@@ -25,6 +25,7 @@  struct socfpga_reset_manager {
 	u32	per2_mod_reset;
 	u32	brg_mod_reset;
 	u32	misc_mod_reset;
+	u32	padding2[12];
 	u32	tstscratch;
 };