[U-Boot-Users] [PATCH 2/4 V2] PPC: Change lib_ppc/bat_rw.c to use high bats

[U-Boot-Users] [PATCH 2/4 V2] PPC: Change lib_ppc/bat_rw.c to use high bats

[Becky.bruce at freescale.com]

From: Becky Bruce <becky.bruce@freescale.com>

Currently, this code only deals with BATs 0-3, which makes
it useless on systems that support BATs 4-7.  Add the
support for these registers.

Signed-off-by: Becky Bruce <Becky.bruce@freescale.com>
---
 include/asm-ppc/mmu.h |    4 ++
 lib_ppc/bat_rw.c      |  134 ++++++++++++++++++++++++++++++++++---------------
 2 files changed, 97 insertions(+), 41 deletions(-)

diff --git a/include/asm-ppc/mmu.h b/include/asm-ppc/mmu.h
index 49d6860..f836270 100644
--- a/include/asm-ppc/mmu.h
+++ b/include/asm-ppc/mmu.h
@@ -141,6 +141,10 @@ extern void _tlbia(void);		/* invalidate all TLB entries */
 typedef enum {
 	IBAT0 = 0, IBAT1, IBAT2, IBAT3,
 	DBAT0, DBAT1, DBAT2, DBAT3
+#ifdef CONFIG_HIGH_BATS
+	, IBAT4, IBAT5, IBAT6, IBAT7,
+	DBAT4, DBAT5, DBAT6, DBAT7
+#endif
 } ppc_bat_t;
 
 extern int read_bat(ppc_bat_t bat, unsigned long *upper, unsigned long *lower);
diff --git a/lib_ppc/bat_rw.c b/lib_ppc/bat_rw.c
index 912efa7..6288164 100644
--- a/lib_ppc/bat_rw.c
+++ b/lib_ppc/bat_rw.c
@@ -28,47 +28,73 @@
 
 int write_bat (ppc_bat_t bat, unsigned long upper, unsigned long lower)
 {
-	switch (bat) {
+	switch(bat) {
+	case DBAT0:
+		mtspr (DBAT0L, lower);
+		mtspr (DBAT0U, upper);
+		break;
 	case IBAT0:
 		mtspr (IBAT0L, lower);
 		mtspr (IBAT0U, upper);
 		break;
-
+	case DBAT1:
+		mtspr (DBAT1L, lower);
+		mtspr (DBAT1U, upper);
+		break;
 	case IBAT1:
 		mtspr (IBAT1L, lower);
 		mtspr (IBAT1U, upper);
 		break;
-
+	case DBAT2:
+		mtspr (DBAT2L, lower);
+		mtspr (DBAT2U, upper);
+		break;
 	case IBAT2:
 		mtspr (IBAT2L, lower);
 		mtspr (IBAT2U, upper);
 		break;
-
+	case DBAT3:
+		mtspr (DBAT3L, lower);
+		mtspr (DBAT3U, upper);
+		break;
 	case IBAT3:
 		mtspr (IBAT3L, lower);
 		mtspr (IBAT3U, upper);
 		break;
-
-	case DBAT0:
-		mtspr (DBAT0L, lower);
-		mtspr (DBAT0U, upper);
+#ifdef CONFIG_HIGH_BATS
+	case DBAT4:
+		mtspr (DBAT4L, lower);
+		mtspr (DBAT3U, upper);
 		break;
-
-	case DBAT1:
-		mtspr (DBAT1L, lower);
-		mtspr (DBAT1U, upper);
+	case IBAT4:
+		mtspr (IBAT4L, lower);
+		mtspr (IBAT4U, upper);
 		break;
-
-	case DBAT2:
-		mtspr (DBAT2L, lower);
-		mtspr (DBAT2U, upper);
+	case DBAT5:
+		mtspr (DBAT5L, lower);
+		mtspr (DBAT5U, upper);
 		break;
-
-	case DBAT3:
-		mtspr (DBAT3L, lower);
-		mtspr (DBAT3U, upper);
+	case IBAT5:
+		mtspr (IBAT5L, lower);
+		mtspr (IBAT5U, upper);
 		break;
-
+	case DBAT6:
+		mtspr (DBAT6L, lower);
+		mtspr (DBAT6U, upper);
+		break;
+	case IBAT6:
+		mtspr (IBAT6L, lower);
+		mtspr (IBAT6U, upper);
+		break;
+	case DBAT7:
+		mtspr (DBAT7L, lower);
+		mtspr (DBAT7U, upper);
+		break;
+	case IBAT7:
+		mtspr (IBAT7L, lower);
+		mtspr (IBAT7U, upper);
+		break;
+#endif
 	default:
 		return (-1);
 	}
@@ -82,46 +108,72 @@ int read_bat (ppc_bat_t bat, unsigned long *upper, unsigned long *lower)
 	unsigned long register l;
 
 	switch (bat) {
+	case DBAT0:
+		l = mfspr (DBAT0L);
+		u = mfspr (DBAT0U);
+		break;
 	case IBAT0:
 		l = mfspr (IBAT0L);
 		u = mfspr (IBAT0U);
 		break;
-
+	case DBAT1:
+		l = mfspr (DBAT1L);
+		u = mfspr (DBAT1U);
+		break;
 	case IBAT1:
 		l = mfspr (IBAT1L);
 		u = mfspr (IBAT1U);
 		break;
-
+	case DBAT2:
+		l = mfspr (DBAT2L);
+		u = mfspr (DBAT2U);
+		break;
 	case IBAT2:
 		l = mfspr (IBAT2L);
 		u = mfspr (IBAT2U);
 		break;
-
+	case DBAT3:
+		l = mfspr (DBAT3L);
+		u = mfspr (DBAT3U);
+		break;
 	case IBAT3:
 		l = mfspr (IBAT3L);
 		u = mfspr (IBAT3U);
 		break;
-
-	case DBAT0:
-		l = mfspr (DBAT0L);
-		u = mfspr (DBAT0U);
+#ifdef CONFIG_HIGH_BATS
+	case DBAT4:
+		l = mfspr (DBAT4L);
+		u = mfspr (DBAT4U);
 		break;
-
-	case DBAT1:
-		l = mfspr (DBAT1L);
-		u = mfspr (DBAT1U);
+	case IBAT4:
+		l = mfspr (IBAT4L);
+		u = mfspr (IBAT4U);
 		break;
-
-	case DBAT2:
-		l = mfspr (DBAT2L);
-		u = mfspr (DBAT2U);
+	case DBAT5:
+		l = mfspr (DBAT5L);
+		u = mfspr (DBAT5U);
 		break;
-
-	case DBAT3:
-		l = mfspr (DBAT3L);
-		u = mfspr (DBAT3U);
+	case IBAT5:
+		l = mfspr (IBAT5L);
+		u = mfspr (IBAT5U);
 		break;
-
+	case DBAT6:
+		l = mfspr (DBAT6L);
+		u = mfspr (DBAT6U);
+		break;
+	case IBAT6:
+		l = mfspr (IBAT6L);
+		u = mfspr (IBAT6U);
+		break;
+	case DBAT7:
+		l = mfspr (DBAT7L);
+		u = mfspr (DBAT7U);
+		break;
+	case IBAT7:
+		l = mfspr (IBAT7L);
+		u = mfspr (IBAT7U);
+		break;
+#endif
 	default:
 		return (-1);
 	}
-- 
1.5.4.1

[U-Boot-Users] [PATCH 2/4 V2] PPC: Change lib_ppc/bat_rw.c to use high bats

[Scott Wood]

On Thu, May 08, 2008 at 07:02:23PM -0500, Becky.bruce at freescale.com wrote:
> diff --git a/include/asm-ppc/mmu.h b/include/asm-ppc/mmu.h
> index 49d6860..f836270 100644
> --- a/include/asm-ppc/mmu.h
> +++ b/include/asm-ppc/mmu.h
> @@ -141,6 +141,10 @@ extern void _tlbia(void);		/* invalidate all TLB entries */
>  typedef enum {
>  	IBAT0 = 0, IBAT1, IBAT2, IBAT3,
>  	DBAT0, DBAT1, DBAT2, DBAT3
> +#ifdef CONFIG_HIGH_BATS
> +	, IBAT4, IBAT5, IBAT6, IBAT7,
> +	DBAT4, DBAT5, DBAT6, DBAT7
> +#endif

C allows trailing commas on such lists; no need to move it inside the
ifdef.

-Scott

[U-Boot-Users] [PATCH 2/4 V2] PPC: Change lib_ppc/bat_rw.c to use high bats

[Becky Bruce]

On May 9, 2008, at 10:28 AM, Scott Wood wrote:
> On Thu, May 08, 2008 at 07:02:23PM -0500, Becky.bruce at freescale.com  
> wrote:
>> diff --git a/include/asm-ppc/mmu.h b/include/asm-ppc/mmu.h
>> index 49d6860..f836270 100644
>> --- a/include/asm-ppc/mmu.h
>> +++ b/include/asm-ppc/mmu.h
>> @@ -141,6 +141,10 @@ extern void _tlbia(void);		/* invalidate all  
>> TLB entries */
>> typedef enum {
>> 	IBAT0 = 0, IBAT1, IBAT2, IBAT3,
>> 	DBAT0, DBAT1, DBAT2, DBAT3
>> +#ifdef CONFIG_HIGH_BATS
>> +	, IBAT4, IBAT5, IBAT6, IBAT7,
>> +	DBAT4, DBAT5, DBAT6, DBAT7
>> +#endif
>
> C allows trailing commas on such lists; no need to move it inside the
> ifdef.

Dangling commas bother me in the extreme.  If this is the concensus,  
I'll move it, but I prefer it where it is.

-B

[U-Boot-Users] [PATCH 2/4 V2] PPC: Change lib_ppc/bat_rw.c to use high bats

[Jerry Van Baren]

Becky Bruce wrote:
> On May 9, 2008, at 10:28 AM, Scott Wood wrote:
>> On Thu, May 08, 2008 at 07:02:23PM -0500, Becky.bruce at freescale.com  
>> wrote:
>>> diff --git a/include/asm-ppc/mmu.h b/include/asm-ppc/mmu.h
>>> index 49d6860..f836270 100644
>>> --- a/include/asm-ppc/mmu.h
>>> +++ b/include/asm-ppc/mmu.h
>>> @@ -141,6 +141,10 @@ extern void _tlbia(void);		/* invalidate all  
>>> TLB entries */
>>> typedef enum {
>>> 	IBAT0 = 0, IBAT1, IBAT2, IBAT3,
>>> 	DBAT0, DBAT1, DBAT2, DBAT3
>>> +#ifdef CONFIG_HIGH_BATS
>>> +	, IBAT4, IBAT5, IBAT6, IBAT7,
>>> +	DBAT4, DBAT5, DBAT6, DBAT7
>>> +#endif
>> C allows trailing commas on such lists; no need to move it inside the
>> ifdef.
> 
> Dangling commas bother me in the extreme.  If this is the concensus,  
> I'll move it, but I prefer it where it is.
> 
> -B

+1 for moving.  Dangling commas is the second best thing in C. ;-)

Best regards,
gvb

[U-Boot-Users] [PATCH 2/4 V2] PPC: Change lib_ppc/bat_rw.c to use high bats

[Jean-Christophe PLAGNIOL-VILLARD]

On 14:45 Fri 09 May     , Jerry Van Baren wrote:
> Becky Bruce wrote:
> > On May 9, 2008, at 10:28 AM, Scott Wood wrote:
> >> On Thu, May 08, 2008 at 07:02:23PM -0500, Becky.bruce at freescale.com  
> >> wrote:
> >>> diff --git a/include/asm-ppc/mmu.h b/include/asm-ppc/mmu.h
> >>> index 49d6860..f836270 100644
> >>> --- a/include/asm-ppc/mmu.h
> >>> +++ b/include/asm-ppc/mmu.h
> >>> @@ -141,6 +141,10 @@ extern void _tlbia(void);		/* invalidate all  
> >>> TLB entries */
> >>> typedef enum {
> >>> 	IBAT0 = 0, IBAT1, IBAT2, IBAT3,
> >>> 	DBAT0, DBAT1, DBAT2, DBAT3
> >>> +#ifdef CONFIG_HIGH_BATS
> >>> +	, IBAT4, IBAT5, IBAT6, IBAT7,
> >>> +	DBAT4, DBAT5, DBAT6, DBAT7
> >>> +#endif
> >> C allows trailing commas on such lists; no need to move it inside the
> >> ifdef.
> > 
> > Dangling commas bother me in the extreme.  If this is the concensus,  
> > I'll move it, but I prefer it where it is.
> > 
> > -B
> 
> +1 for moving.  Dangling commas is the second best thing in C. ;-)
+1 for moving.  Dangling commas is the second best thing in C. ;-)

Best Regards,
J.

[U-Boot-Users] [PATCH 2/4 V2] PPC: Change lib_ppc/bat_rw.c to use high bats

[Stefan Roese]

On Friday 09 May 2008, Jerry Van Baren wrote:
> >> C allows trailing commas on such lists; no need to move it inside the
> >> ifdef.
> >
> > Dangling commas bother me in the extreme.  If this is the concensus,
> > I'll move it, but I prefer it where it is.
> >
> > -B
>
> +1 for moving.  Dangling commas is the second best thing in C. ;-)

Now I'm curious. :) And what's supposed to be "the best thing in C"?

Best regards,
Stefan

=====================================================================
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142-66989-80  Email: office at denx.de
=====================================================================

[U-Boot-Users] [PATCH 2/4 V2] PPC: Change lib_ppc/bat_rw.c to use high bats

[Becky Bruce]

On May 9, 2008, at 2:14 PM, Jean-Christophe PLAGNIOL-VILLARD wrote:
> On 14:45 Fri 09 May     , Jerry Van Baren wrote:
>> Becky Bruce wrote:
>>> On May 9, 2008, at 10:28 AM, Scott Wood wrote:
>>>> On Thu, May 08, 2008 at 07:02:23PM -0500, Becky.bruce at freescale.com
>>>> wrote:
>>>>> diff --git a/include/asm-ppc/mmu.h b/include/asm-ppc/mmu.h
>>>>> index 49d6860..f836270 100644
>>>>> --- a/include/asm-ppc/mmu.h
>>>>> +++ b/include/asm-ppc/mmu.h
>>>>> @@ -141,6 +141,10 @@ extern void _tlbia(void);		/* invalidate all
>>>>> TLB entries */
>>>>> typedef enum {
>>>>> 	IBAT0 = 0, IBAT1, IBAT2, IBAT3,
>>>>> 	DBAT0, DBAT1, DBAT2, DBAT3
>>>>> +#ifdef CONFIG_HIGH_BATS
>>>>> +	, IBAT4, IBAT5, IBAT6, IBAT7,
>>>>> +	DBAT4, DBAT5, DBAT6, DBAT7
>>>>> +#endif
>>>> C allows trailing commas on such lists; no need to move it inside  
>>>> the
>>>> ifdef.
>>>
>>> Dangling commas bother me in the extreme.  If this is the concensus,
>>> I'll move it, but I prefer it where it is.
>>>
>>> -B
>>
>> +1 for moving.  Dangling commas is the second best thing in C. ;-)
> +1 for moving.  Dangling commas is the second best thing in C. ;-)
>

I also want to know what's the first best thing!  I have to guess that  
the dangling comma is really the fifteenth-best thing, or  
something..... but I don't want this thread to degenerate into a holy  
war.

Ayway, point taken, and I'll move it.  But I don't have to like it :)

Cheers,
B

[U-Boot-Users] [PATCH 2/4 V2] PPC: Change lib_ppc/bat_rw.c to use high bats

[Jerry Van Baren]

Stefan Roese wrote:
> On Friday 09 May 2008, Jerry Van Baren wrote:
>>>> C allows trailing commas on such lists; no need to move it inside the
>>>> ifdef.
>>> Dangling commas bother me in the extreme.  If this is the concensus,
>>> I'll move it, but I prefer it where it is.
>>>
>>> -B
>> +1 for moving.  Dangling commas is the second best thing in C. ;-)
> 
> Now I'm curious. :) And what's supposed to be "the best thing in C"?
> 
> Best regards,
> Stefan

I figured everybody has a favorite feature of C, so by claiming *second 
best*, I wouldn't have to argue with everybody that it was the best. :-D 
  Great conversation starter too, I use it in the bar all the time.  ;-)

As Calvin says, "I try to make everyone's day a little more surreal."
The Essential Calvin and Hobbes, p152-3
   <http://www.calvin-and-hobbes.org/information/quotes>

gvb

[U-Boot-Users] [PATCH 2/4 V2] PPC: Change lib_ppc/bat_rw.c to use high bats

[Wolfgang Denk]

In message <6514CE95-5C25-4C36-88DA-63E03DE87024@freescale.com> you wrote:
> 
> >> TLB entries */
> >> typedef enum {
> >> 	IBAT0 = 0, IBAT1, IBAT2, IBAT3,
> >> 	DBAT0, DBAT1, DBAT2, DBAT3
> >> +#ifdef CONFIG_HIGH_BATS
> >> +	, IBAT4, IBAT5, IBAT6, IBAT7,
> >> +	DBAT4, DBAT5, DBAT6, DBAT7
> >> +#endif
> >
> > C allows trailing commas on such lists; no need to move it inside the
> > ifdef.
> 
> Dangling commas bother me in the extreme.  If this is the concensus,  
> I'll move it, but I prefer it where it is.

Please move it. As is it looks irregular and ugly to me.

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Q: What do you get when you cross an ethernet with an income statement?
A: A local area networth.