Dear Timur Tabi,
In message <[EMAIL PROTECTED]> you wrote:
>
> I don't know much about 8xx, but on 8xxx, all the information needed for I2C
> configuration is in the board configuration files. Couldn't we just move
> i2c_init earlier?
I don;t remember all the details, sorry. One problem might h
Dear Timur,
In message <[EMAIL PROTECTED]> you wrote:
>
> > Very early. We have a board with the environment in EEPROM,
> > and EEPROM at I2C.
>
> Which board is that? Do you use i2c_reg_read() to read EEPROM, or i2c_read()?
Well, just grep where i2c_reg_read() is actually used - it's a LOT o
Dear Gary,
In message <[EMAIL PROTECTED]> you wrote:
>
> If you look at lib_ppc/board.c you will see that
> a) i2c_init() is called _while still running out of flash_ because it's
> in init_sequence[] which is used in board_init_f().
> b) i2c_init() is also called in devices_init() which is invok
On Wed, 22 Oct 2008 17:53:00 -0500
Timur Tabi <[EMAIL PROTECTED]> wrote:
> Detlef Vollmann wrote:
>
> > Very early. We have a board with the environment in EEPROM,
> > and EEPROM at I2C.
>
> Which board is that? Do you use i2c_reg_read() to read EEPROM, or i2c_read()?
>
If you look at lib_pp
Wolfgang Denk wrote:
>> would it be okay to make a best-guess attempt at modifying the code to call
>> i2c_init() earlier for those boards that need it (e.g. call it from
>> scc_init()
>> on the lwmon board), and hope someone else notices any bugs in other 8xx
>> boards?
>
> That would mean to
Hello Wolfgang,
> We need a formal Signed-off-by: line for you for this patch.
Something like this?
Signed-off-by: Pieter Voorthuijsen <[EMAIL PROTECTED]>
I'm sure the code will be adapted by someone as it is only for 1.2.0, will they
also add their signed-off lines?
> I think that such a li
Evrika Group - cursuri de perfectionare :
- contabilitate costul cursului este de 300 ron cu incepere din 10 noiembrie
2008 .
- expert fiscal costul cursului este de 1000 ron cu incepere in 01 noiembrie
2008.
- inspector protectia muncii costul cursului este de 300 ron cu incepere din
Dear Timur,
In message <[EMAIL PROTECTED]> you wrote:
>
> The only platform that has this problem is 8xx. Can't we do something like
> this:
>
> init_fnc_t *init_sequence[] = {
>
> +#ifdef CONFIG_8xx
> + init_func_i2c(),
> +#endif
>
> #if defined(CONFIG_BOARD_EARLY_INIT_F)
> boar
On Thu, Oct 23, 2008 at 2:10 PM, Wolfgang Denk <[EMAIL PROTECTED]> wrote:
> We can - but I have no way of testing if this breaks some boards.
Ok, so this goes back to a question I asked earlier. Would you accept
a patch like this, knowing that it might break some boards that we
can't test, with
