blob-cvs-commit

RE: Blob boot from Compact Flash (ext2fs)

[Christopher H. <ch...@mu...>]

> Also, I'm looking at the state of blob wrt there being no 
> maintainer. I have patches, too, that need to be integrated.n

Send the patches (in clean, concise pieces, i.e., standard patch submission
technique) to blo...@li....

One of us with write access to cvs will apply them, if they are acceptable.

-ch

Re: Blob boot from Compact Flash (ext2fs)

[Marc S. <el...@bu...>]

I think that 'clean' is part of the problem.  I'm working with the
Lineo patches for the Sharp ARM SoC's.  Have you seen their work?  I'm
making some changes so that their code works properly with another
implementation of the Sharp CPU.

I'll post something when I've got it working. 

On Thu, Oct 02, 2003 at 10:06:31AM -0700, Christopher Hoover wrote:
> 
> > Also, I'm looking at the state of blob wrt there being no 
> > maintainer. I have patches, too, that need to be integrated.n
> 
> Send the patches (in clean, concise pieces, i.e., standard patch submission
> technique) to blo...@li....
> 
> One of us with write access to cvs will apply them, if they are acceptable.
> 
> -ch

Re: Blob boot from Compact Flash (ext2fs)

[Brad P. <br...@he...>]

Hi

I made some changes to the blob+lineo code for my A400 board and got
disgusted and did some refactoring of their code (mostly to look the
4,000,000 #ifdef's they put everywhere).

Then, while working on an xscale project I noticed the latest blob cvs
seems to have done a lot of this refactoring already and integrated some
new cpu's cleanly.

So I add "put your changes into the newest blob cvs" to my to-do list and
promptly started ignoring it  :-)

My biggest beef with the lineo code is the use of "LH7A400" when they
really ment KEV7A400 and "LH79520" when they really ment KEV79520.
i.e. mixing the notion of the cpu with the board instantiation as if there
would only ever be one A400 board in the world.

So if some/any/all of the lineo code gets put in, please define the 
Sharp eval boards as KEV7A400 and KEV79520 like this snipped below
(from configure)

	...
	cat >> confdefs.h <<\EOF
#define LH79520 1
EOF

	BLOB_PLATFORM_OBJ="kev79520.o"
	BLOB_FLASH_OBJS="sharp_lh28f320.o"
	use_cpu="lh79520"
	...

that way we can have multiple boards which use the same cpu and not all
pull our hair out.

I sure you know all this, and sorry for the rant - I had some pent up 
frustration I needed to release :-)

[and having said that, I'm happy to build and test blob for the KEV79520
and KEV7A400 since I have one of each right now, as well as my own A400
board]

-brad

Marc Singer wrote:
>I think that 'clean' is part of the problem.  I'm working with the
>Lineo patches for the Sharp ARM SoC's.  Have you seen their work?  I'm
>making some changes so that their code works properly with another
>implementation of the Sharp CPU.
>
>I'll post something when I've got it working. 
>
>On Thu, Oct 02, 2003 at 10:06:31AM -0700, Christopher Hoover wrote:
>> 
>> > Also, I'm looking at the state of blob wrt there being no 
>> > maintainer. I have patches, too, that need to be integrated.n
>> 
>> Send the patches (in clean, concise pieces, i.e., standard patch submission
>> technique) to blo...@li....
>> 
>> One of us with write access to cvs will apply them, if they are acceptable.
>> 
>> -ch
>--
>To unsubscribe from this list: send the line "unsubscribe lart" in
>the body of a message to maj...@la...
>Please read the LART FAQ at http://www.lart.tudelft.nl/faq.php3
>

Re: Blob boot from Compact Flash (ext2fs)

[Marc S. <el...@bu...>]

So, are you saying that the code for the Sharp boards got integrated
already?  If so, I'd really appreciate a snapshot from CVS.

On Thu, Oct 02, 2003 at 03:16:21PM -0400, Brad Parker wrote:
> 
> Hi
> 
> I made some changes to the blob+lineo code for my A400 board and got
> disgusted and did some refactoring of their code (mostly to look the
> 4,000,000 #ifdef's they put everywhere).
> 
> Then, while working on an xscale project I noticed the latest blob cvs
> seems to have done a lot of this refactoring already and integrated some
> new cpu's cleanly.
> 
> So I add "put your changes into the newest blob cvs" to my to-do list and
> promptly started ignoring it  :-)
> 
> My biggest beef with the lineo code is the use of "LH7A400" when they
> really ment KEV7A400 and "LH79520" when they really ment KEV79520.
> i.e. mixing the notion of the cpu with the board instantiation as if there
> would only ever be one A400 board in the world.

A common mistake.  The uboot code is much much worse.

> 
> So if some/any/all of the lineo code gets put in, please define the 
> Sharp eval boards as KEV7A400 and KEV79520 like this snipped below
> (from configure)
> 
> 	...
> 	cat >> confdefs.h <<\EOF
> #define LH79520 1
> EOF
> 
> 	BLOB_PLATFORM_OBJ="kev79520.o"
> 	BLOB_FLASH_OBJS="sharp_lh28f320.o"
> 	use_cpu="lh79520"
> 	...
> 
> that way we can have multiple boards which use the same cpu and not all
> pull our hair out.
> 
> I sure you know all this, and sorry for the rant - I had some pent up 
> frustration I needed to release :-)

I've been very restrained about criticizing those who came before me.
Their code has always been welcome as it lightens my burden.  Yet, I
too have found the Lineo work wanting

> 
> [and having said that, I'm happy to build and test blob for the KEV79520
> and KEV7A400 since I have one of each right now, as well as my own A400
> board]

Good to know.  I rewrote the memory initialization code once already.
I'll do so again as soon as I can get something running on the
KEV7A400.