[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[Yuan Yao]

I2C1 can work on ls102xa rev2.0 SD boot, so add
ID EEPROM for SD boot.

Signed-off-by: Yuan Yao <yao.yuan@freescale.com>
---
 include/configs/ls1021atwr.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/include/configs/ls1021atwr.h b/include/configs/ls1021atwr.h
index 7dcb719..59e704e 100644
--- a/include/configs/ls1021atwr.h
+++ b/include/configs/ls1021atwr.h
@@ -278,7 +278,6 @@
 #define CONFIG_SYS_I2C_MXC_I2C3		/* enable I2C bus 3 */
 
 /* EEPROM */
-#ifndef CONFIG_SD_BOOT
 #define CONFIG_ID_EEPROM
 #define CONFIG_SYS_I2C_EEPROM_NXID
 #define CONFIG_SYS_EEPROM_BUS_NUM		1
@@ -286,7 +285,6 @@
 #define CONFIG_SYS_I2C_EEPROM_ADDR_LEN		1
 #define CONFIG_SYS_EEPROM_PAGE_WRITE_BITS	3
 #define CONFIG_SYS_EEPROM_PAGE_WRITE_DELAY_MS	5
-#endif
 
 /*
  * MMC
-- 
2.1.0.27.g96db324

[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[York Sun]

On 09/23/2015 12:48 AM, Yuan Yao wrote:
> I2C1 can work on ls102xa rev2.0 SD boot, so add
> ID EEPROM for SD boot.
> 
> Signed-off-by: Yuan Yao <yao.yuan@freescale.com>
> ---
>  include/configs/ls1021atwr.h | 2 --
>  1 file changed, 2 deletions(-)
> 
> diff --git a/include/configs/ls1021atwr.h b/include/configs/ls1021atwr.h
> index 7dcb719..59e704e 100644
> --- a/include/configs/ls1021atwr.h
> +++ b/include/configs/ls1021atwr.h
> @@ -278,7 +278,6 @@
>  #define CONFIG_SYS_I2C_MXC_I2C3		/* enable I2C bus 3 */
>  
>  /* EEPROM */
> -#ifndef CONFIG_SD_BOOT
>  #define CONFIG_ID_EEPROM
>  #define CONFIG_SYS_I2C_EEPROM_NXID
>  #define CONFIG_SYS_EEPROM_BUS_NUM		1
> @@ -286,7 +285,6 @@
>  #define CONFIG_SYS_I2C_EEPROM_ADDR_LEN		1
>  #define CONFIG_SYS_EEPROM_PAGE_WRITE_BITS	3
>  #define CONFIG_SYS_EEPROM_PAGE_WRITE_DELAY_MS	5
> -#endif
>  
>  /*
>   * MMC
> 

What would happen if you run the new u-boot image on earlier SoC? Do you need to
check SVR in function mac_read_from_eeprom()?

York

[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[Yao Yuan]

Hi york,

The earlier SoC is just LS1021a rev1.0, but rev1.0 haven't delivery to the customer.
Also the rev1.0 has since gone out of production.
So we don't have necessary to support rev1.0 because no one will or possibly to use rev1.0.

Best Regards,
Yuan Yao

> -----Original Message-----
> From: York Sun [mailto:yorksun at freescale.com]
> Sent: Thursday, October 15, 2015 5:22 AM
> To: Yuan Yao-B46683 <yao.yuan@freescale.com>; Wang Huan-B18965
> <alison.wang@freescale.com>
> Cc: u-boot at lists.denx.de
> Subject: Re: [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot
> 
> On 09/23/2015 12:48 AM, Yuan Yao wrote:
> > I2C1 can work on ls102xa rev2.0 SD boot, so add ID EEPROM for SD boot.
> >
> > Signed-off-by: Yuan Yao <yao.yuan@freescale.com>
> > ---
> >  include/configs/ls1021atwr.h | 2 --
> >  1 file changed, 2 deletions(-)
> >
> > diff --git a/include/configs/ls1021atwr.h
> > b/include/configs/ls1021atwr.h index 7dcb719..59e704e 100644
> > --- a/include/configs/ls1021atwr.h
> > +++ b/include/configs/ls1021atwr.h
> > @@ -278,7 +278,6 @@
> >  #define CONFIG_SYS_I2C_MXC_I2C3		/* enable I2C bus 3 */
> >
> >  /* EEPROM */
> > -#ifndef CONFIG_SD_BOOT
> >  #define CONFIG_ID_EEPROM
> >  #define CONFIG_SYS_I2C_EEPROM_NXID
> >  #define CONFIG_SYS_EEPROM_BUS_NUM		1
> > @@ -286,7 +285,6 @@
> >  #define CONFIG_SYS_I2C_EEPROM_ADDR_LEN		1
> >  #define CONFIG_SYS_EEPROM_PAGE_WRITE_BITS	3
> >  #define CONFIG_SYS_EEPROM_PAGE_WRITE_DELAY_MS	5
> > -#endif
> >
> >  /*
> >   * MMC
> >
> 
> What would happen if you run the new u-boot image on earlier SoC? Do you
> need to check SVR in function mac_read_from_eeprom()?
> 
> York

[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[York Sun]

On 10/15/2015 12:55 AM, Yuan Yao-B46683 wrote:
> Hi york,
> 
> The earlier SoC is just LS1021a rev1.0, but rev1.0 haven't delivery to the customer.
> Also the rev1.0 has since gone out of production.
> So we don't have necessary to support rev1.0 because no one will or possibly to use rev1.0.
> 

OK.

York

[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[Sinan Akman]

   Hi Yuan

On 15/10/15 12:40 PM, York Sun wrote:
>
>
> On 10/15/2015 12:55 AM, Yuan Yao-B46683 wrote:
>> Hi york,
>>
>> The earlier SoC is just LS1021a rev1.0, but rev1.0 haven't delivery to the customer.
>> Also the rev1.0 has since gone out of production.
>> So we don't have necessary to support rev1.0 because no one will or possibly to use rev1.0.

   Just to make this clear, the boards we bought all show :

CPU:   Freescale LayerScape LS1021E, Version: 1.0, (0x87081110)

   When you mention "rev1.0 haven't delivery to the customer" are
you not referring to Rev 1.0 silicon ?

   Regards
   Sinan Akman

>>
>
> OK.
>
> York
> _______________________________________________
> U-Boot mailing list
> U-Boot at lists.denx.de
> http://lists.denx.de/mailman/listinfo/u-boot
>

[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[Yao Yuan]

Hi Sinan Akman,

Yes, I mean the Rev 1.0 silicon.
Sorry, I can't guarantee that there aren't any boards with Rev1.0 silicon are in user's hands.  
Because we have also delivery very little board with Rev1.0 silicon to  customer  or developer for developing, assessing and verifying in the early stages. 
Also we haven't support Rev1.0 in our SDK now.

Thanks.

Best Regards,
Yuan Yao

> -----Original Message-----
> From: Sinan Akman [mailto:sinan at writeme.com]
> Sent: Friday, October 16, 2015 3:37 AM
> To: Sun York-R58495 <yorksun@freescale.com>; Yuan Yao-B46683
> <yao.yuan@freescale.com>; Wang Huan-B18965 <alison.wang@freescale.com>
> Cc: u-boot at lists.denx.de
> Subject: Re: [U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD
> boot
> 
> 
>    Hi Yuan
> 
> On 15/10/15 12:40 PM, York Sun wrote:
> >
> >
> > On 10/15/2015 12:55 AM, Yuan Yao-B46683 wrote:
> >> Hi york,
> >>
> >> The earlier SoC is just LS1021a rev1.0, but rev1.0 haven't delivery to the
> customer.
> >> Also the rev1.0 has since gone out of production.
> >> So we don't have necessary to support rev1.0 because no one will or
> possibly to use rev1.0.
> 
>    Just to make this clear, the boards we bought all show :
> 
> CPU:   Freescale LayerScape LS1021E, Version: 1.0, (0x87081110)
> 
>    When you mention "rev1.0 haven't delivery to the customer" are you not
> referring to Rev 1.0 silicon ?
> 
>    Regards
>    Sinan Akman
> 
> >>
> >
> > OK.
> >
> > York
> > _______________________________________________
> > U-Boot mailing list
> > U-Boot at lists.denx.de
> > http://lists.denx.de/mailman/listinfo/u-boot
> >

[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[Sinan Akman]

   Hi Yuan

On 19/10/15 05:21 AM, Yao Yuan wrote:
> Hi Sinan Akman,
>
> Yes, I mean the Rev 1.0 silicon.
> Sorry, I can't guarantee that there aren't any boards with Rev1.0 silicon are in user's hands.
> Because we have also delivery very little board with Rev1.0 silicon to  customer  or developer for developing, assessing and verifying in the early stages.

   Thanks for the follow up on this, but I don't think that there were only
very few rev1.0 boards delivered. These were the boards bought from
distributor and during the time span of many weeks.

   I don't know the details and the impact of this patch exactly but
if the patch will break any rev1.0 board I don't think this would be
acceptable. There are definitely enough number of rev1.0 boards
made generally available (not specifically for early adapters etc)
that we should not break u-boot running on them.

> Also we haven't support Rev1.0 in our SDK now.

   Sorry, but this is a bit irrelevant here. I'd expect u-boot mainline
does not necessarily follow what your SDK stops supporting. If
u-boot supported rev1.0 boards and if there are still many of
  them out there I see no reason to break this.

   Regards
   Sinan Akman

>
> Thanks.
>
> Best Regards,
> Yuan Yao
>
>> -----Original Message-----
>> From: Sinan Akman [mailto:sinan at writeme.com]
>> Sent: Friday, October 16, 2015 3:37 AM
>> To: Sun York-R58495 <yorksun@freescale.com>; Yuan Yao-B46683
>> <yao.yuan@freescale.com>; Wang Huan-B18965 <alison.wang@freescale.com>
>> Cc: u-boot at lists.denx.de
>> Subject: Re: [U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD
>> boot
>>
>>
>>     Hi Yuan
>>
>> On 15/10/15 12:40 PM, York Sun wrote:
>>>
>>> On 10/15/2015 12:55 AM, Yuan Yao-B46683 wrote:
>>>> Hi york,
>>>>
>>>> The earlier SoC is just LS1021a rev1.0, but rev1.0 haven't delivery to the
>> customer.
>>>> Also the rev1.0 has since gone out of production.
>>>> So we don't have necessary to support rev1.0 because no one will or
>> possibly to use rev1.0.
>>
>>     Just to make this clear, the boards we bought all show :
>>
>> CPU:   Freescale LayerScape LS1021E, Version: 1.0, (0x87081110)
>>
>>     When you mention "rev1.0 haven't delivery to the customer" are you not
>> referring to Rev 1.0 silicon ?
>>
>>     Regards
>>     Sinan Akman
>>
>>> OK.
>>>
>>> York
>>> _______________________________________________
>>> U-Boot mailing list
>>> U-Boot at lists.denx.de
>>> http://lists.denx.de/mailman/listinfo/u-boot
>>>
> _______________________________________________
> U-Boot mailing list
> U-Boot at lists.denx.de
> http://lists.denx.de/mailman/listinfo/u-boot
>

[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[York Sun]

On 10/19/2015 04:14 AM, Sinan Akman wrote:
> 
>    Hi Yuan
> 
> On 19/10/15 05:21 AM, Yao Yuan wrote:
>> Hi Sinan Akman,
>>
>> Yes, I mean the Rev 1.0 silicon.
>> Sorry, I can't guarantee that there aren't any boards with Rev1.0 silicon are in user's hands.
>> Because we have also delivery very little board with Rev1.0 silicon to  customer  or developer for developing, assessing and verifying in the early stages.
> 
>    Thanks for the follow up on this, but I don't think that there were only
> very few rev1.0 boards delivered. These were the boards bought from
> distributor and during the time span of many weeks.
> 
>    I don't know the details and the impact of this patch exactly but
> if the patch will break any rev1.0 board I don't think this would be
> acceptable. There are definitely enough number of rev1.0 boards
> made generally available (not specifically for early adapters etc)
> that we should not break u-boot running on them.
> 
>> Also we haven't support Rev1.0 in our SDK now.
> 
>    Sorry, but this is a bit irrelevant here. I'd expect u-boot mainline
> does not necessarily follow what your SDK stops supporting. If
> u-boot supported rev1.0 boards and if there are still many of
>   them out there I see no reason to break this.
> 

Yuan,

Please test the patch on rev 1.0 SoC and make changes if needed. This feature
has been disabled for rev 1.0.

York

[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[Yao Yuan]

Hi york,

I have tested the patch on Rev1.0.

It seems there is no effect for Rev1.0.
Just the ID EEPROM can't work and will print some error info for I2C1.
There isn't any effect for others.
Is it acceptable?

Best Regards,
Yuan Yao

> -----Original Message-----
> From: York Sun [mailto:yorksun at freescale.com]
> Sent: Monday, October 19, 2015 11:02 PM
> To: sinan at writeme.com; Yuan Yao-B46683 <yao.yuan@freescale.com>; Wang
> Huan-B18965 <alison.wang@freescale.com>
> Cc: u-boot at lists.denx.de
> Subject: Re: [U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD
> boot
> 
> 
> 
> On 10/19/2015 04:14 AM, Sinan Akman wrote:
> >
> >    Hi Yuan
> >
> > On 19/10/15 05:21 AM, Yao Yuan wrote:
> >> Hi Sinan Akman,
> >>
> >> Yes, I mean the Rev 1.0 silicon.
> >> Sorry, I can't guarantee that there aren't any boards with Rev1.0 silicon are
> in user's hands.
> >> Because we have also delivery very little board with Rev1.0 silicon to
> customer  or developer for developing, assessing and verifying in the early
> stages.
> >
> >    Thanks for the follow up on this, but I don't think that there were
> > only very few rev1.0 boards delivered. These were the boards bought
> > from distributor and during the time span of many weeks.
> >
> >    I don't know the details and the impact of this patch exactly but
> > if the patch will break any rev1.0 board I don't think this would be
> > acceptable. There are definitely enough number of rev1.0 boards made
> > generally available (not specifically for early adapters etc) that we
> > should not break u-boot running on them.
> >
> >> Also we haven't support Rev1.0 in our SDK now.
> >
> >    Sorry, but this is a bit irrelevant here. I'd expect u-boot
> > mainline does not necessarily follow what your SDK stops supporting.
> > If u-boot supported rev1.0 boards and if there are still many of
> >   them out there I see no reason to break this.
> >
> 
> Yuan,
> 
> Please test the patch on rev 1.0 SoC and make changes if needed. This feature
> has been disabled for rev 1.0.
> 
> York

[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[York Sun]

That's good. Thanks.

York


-------- Original message --------
From: Yuan Yao-B46683
Date:10/20/2015 8:59 PM (GMT-08:00)
To: Sun York-R58495 , sinan at writeme.com, Wang Huan-B18965
Cc: u-boot at lists.denx.de
Subject: RE: [U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

Hi york,

I have tested the patch on Rev1.0.

It seems there is no effect for Rev1.0.
Just the ID EEPROM can't work and will print some error info for I2C1.
There isn't any effect for others.
Is it acceptable?

Best Regards,
Yuan Yao

> -----Original Message-----
> From: York Sun [mailto:yorksun at freescale.com]
> Sent: Monday, October 19, 2015 11:02 PM
> To: sinan at writeme.com; Yuan Yao-B46683 <yao.yuan@freescale.com>; Wang
> Huan-B18965 <alison.wang@freescale.com>
> Cc: u-boot at lists.denx.de
> Subject: Re: [U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD
> boot
>
>
>
> On 10/19/2015 04:14 AM, Sinan Akman wrote:
> >
> >    Hi Yuan
> >
> > On 19/10/15 05:21 AM, Yao Yuan wrote:
> >> Hi Sinan Akman,
> >>
> >> Yes, I mean the Rev 1.0 silicon.
> >> Sorry, I can't guarantee that there aren't any boards with Rev1.0 silicon are
> in user's hands.
> >> Because we have also delivery very little board with Rev1.0 silicon to
> customer  or developer for developing, assessing and verifying in the early
> stages.
> >
> >    Thanks for the follow up on this, but I don't think that there were
> > only very few rev1.0 boards delivered. These were the boards bought
> > from distributor and during the time span of many weeks.
> >
> >    I don't know the details and the impact of this patch exactly but
> > if the patch will break any rev1.0 board I don't think this would be
> > acceptable. There are definitely enough number of rev1.0 boards made
> > generally available (not specifically for early adapters etc) that we
> > should not break u-boot running on them.
> >
> >> Also we haven't support Rev1.0 in our SDK now.
> >
> >    Sorry, but this is a bit irrelevant here. I'd expect u-boot
> > mainline does not necessarily follow what your SDK stops supporting.
> > If u-boot supported rev1.0 boards and if there are still many of
> >   them out there I see no reason to break this.
> >
>
> Yuan,
>
> Please test the patch on rev 1.0 SoC and make changes if needed. This feature
> has been disabled for rev 1.0.
>
> York

[U-Boot] [PATCH] configs: ls1021atwr: Enable ID EEPROM for SD boot

[York Sun]

On 09/23/2015 12:48 AM, Yuan Yao wrote:
> I2C1 can work on ls102xa rev2.0 SD boot, so add
> ID EEPROM for SD boot.
> 
> Signed-off-by: Yuan Yao <yao.yuan@freescale.com>
> ---


Applied to u-boot-fsl-qoriq. Awaiting upstream. Thanks.

York