[U-Boot-Users] Please pull u-boot-83xx.git

[Timur Tabi <timur@freescale.com>]

Stefan Roese wrote:
> Hi Kim,
> 
> On Saturday 04 November 2006 03:11, Kim Phillips wrote:
>> Please pull from 'master' branch of:
>>
>>   http://opensource.freescale.com/pub/scm/u-boot-83xx.git
>>
>> to receive the following updates (essentially MPC8349mITX and MPC8360EMDS
>> support):
> 
> I did a quick check. The overall impression is quite good. Thanks a lot
> for your contribution (and you colleagues too of course).

We're happy to make it available.  Once you pull this code into your tree, 
there's a lot more coming.

Kim Phillips is out today, so I'm going to address your concerns in his stead. 
I can make the changes, but they won't be posted until after Kim reviews them.

> Here a few notes:
> 
> - Please don't add changelog comments in the files (like in
>   cpu/mpc83xx/i2c.c or spd_sdram.c). Instead please remove all the
>   file internal changelog comments, since we decided to only use
>   git for this history.

Okay.

> - Please add the missing entries to the MAINTAINERS files (MPC8349EMDS &
>   MPC8360EMDS).

Okay.

> Another idea (this would also affect the other Freescale board ports):
> What do you think of moving all your Freescale board ports into a
> Freescale board directory (as done for AMCC or esd for example).
> So we would get something like:

Yes, we can do that.  I agree that the current code base is a little messy 
because of so many board-specific files.  We already have plans to address that, 
but again, we'll start work once the current code has been pulled.

> We could start with the mpc83xx boards and contine soon with the
> 85xx and 86xx eval boards (and so on...).

That's how I was planning on doing it.

>> +ulong get_ddr_clk(ulong dummy)
>> +{
>> +	return gd->ddr_clk;
>> +}
> 
> Hmmm. Is this function really needed?

I'll check.

> 
>> Author: Timur Tabi <timur@freescale.com>  2006-10-26 01:45:23
>> Committer: Kim Phillips <kim.phillips@freescale.com>  2006-11-04 02:42:19
>> Parent: 31068b7c4abeefcb2c8fd4fbeccc8ec6c6d0475a (mpc83xx: Add support for variable flash memory sizes on 83xx systems)
>> Child:  bed85caf872714ebf53013967a695c9d63acfc68 (mpc83xx: Add support for Errata DDR6 on MPC 834x systems)
>> Branches: master, 83xx
>> Follows: U-Boot-1_1_6
>> Precedes: 
>>
>>     mpc83xx: fix TQM build by defining a CFG_FLASH_SIZE for it
>>
>> -------------------------- include/configs/TQM834x.h --------------------------
>> index b1f033d..f0e4900 100644
>> @@ -95,6 +95,7 @@
>>  #define CFG_FLASH_CFI_DRIVER			/* use the CFI driver */
>>  #undef CFG_FLASH_CHECKSUM
>>  #define CFG_FLASH_BASE		0x80000000	/* start of FLASH   */
>> +#define CFG_FLASH_SIZE		8		/* FLASH size in MB */
> 
> Do you only support sizes bigger or equal to 1 MByte? What if a board
> only has 512kBytes?

As I mentioned in my other post, this code is for the LBLAWAR1 code in start.S. 
  That code assumes that CFG_FLASH_SIZE is at least 1.  I'm not aware of any 
83xx board that has less than that much flash, but you're right - it is 
possible.  If you like, I can provide an enhanced version of this code to 
support flash sizes less than 1MB, but I would prefer to do that after you pull 
our current code.

> I also do get some warning upon compiling for MPC8349ITX:
> mpc8349itx.c: In function 'misc_init_r':
> mpc8349itx.c:398: warning: pointer targets in passing argument 4 of 'i2c_read' differ in signedness
> mpc8349itx.c:443: warning: pointer targets in passing argument 4 of 'i2c_write' differ in signedness
> 
> And for MPC8360EMDS:
> uccf.c: In function 'ucc_set_clk_src':
> uccf.c:121: warning: 'reg_num' is used uninitialized in this function
> uccf.c:105: warning: 'shift' may be used uninitialized in this function
> uccf.c:103: warning: 'p_cmxucr' may be used uninitialized in this function
> 
> Could you please clean this up too?

I've never seen these, but I'll look into it.

-- 
Timur Tabi
Linux Kernel Developer @ Freescale