[U-Boot] i2c_set_bus_num freeze

[Ran Shalit]

On Fri, Jan 27, 2017 at 9:08 AM, Ran Shalit <ranshalit at gmail.com> wrote:

>
>
> On Fri, Jan 27, 2017 at 8:46 AM, Heiko Schocher <hs at denx.de> wrote:
>
>> Hello Ran,
>>
>> Am 26.01.2017 um 18:42 schrieb Ran Shalit:
>>
>>> Hello,
>>>
>>> I have some issue in u-boot trying to access bus #1.
>>>
>>
>> On which hardware? Is the board in mainline?
>
>
> Yes, it's TI's dm8148 (ti81xx, ti814x)
>
>>
>>
>> I can access bus #0, #2 with i2c dev 0/2
>>>
>>> I can also access bus #0,1,2 from linux.
>>>
>>> But on trying to access i2c #1 in u-boot in hangs:
>>>
>>> The function i2c_set_bus_num get stuck .
>>>
>>>
>>> ret = i2c_set_bus_num(1);
>>>
>>> <<------- never gets here .........
>>>   if (ret)
>>>       printf("Failure changing bus number (%d)\n", ret);
>>>
>>> It hangs in i2c_set_bus_num->i2c_init->
>>>
>>> ......
>>>
>>>      printf("i2c_init 8\n");
>>>
>>
>> You see this printf?
>
>
> Yes, it seems to get stuck here:
>
> if (readw (I2C_CON) & I2C_CON_EN) {
> <<--------- we never gets here !!!!!!!!!!!!!!!!!!!!!!!!!!!
> writew (0, I2C_CON);
>
>
>
>>
>>
>> bus_initialized[current_bus] = 1;
>>> if (readw (I2C_CON) & I2C_CON_EN) {
>>> writew (0, I2C_CON);
>>> udelay (50000);
>>> }
>>>
>>> writew(psc, I2C_PSC);
>>> writew(scll, I2C_SCLL);
>>> writew(sclh, I2C_SCLH);
>>>
>>> /* own address */
>>> writew (slaveadd, I2C_OA);
>>> writew (I2C_CON_EN, I2C_CON);
>>>
>>>      printf("i2c_init 9\n");  <<------ never get printed
>>> ..................
>>>
>>> I have seen a similar question, yet I am doing this access after u-boot
>>> complete, and command prompt.
>>>
>>> Is anyone familiar with this ?
>>>
>>
>> Hmm.. no reason, why the last printf should not come ...
>>
>> As I do not know, on which hw ... I just can speculate ...
>>
>> Is the clock for the i2c subsystem 1 in the SoC enabled?
>>
>
> TI's DM8148.
> I remember that I once used that code and it works, so I quite amazed that
> it stopped working now.
> The board had some changes from that time when I tried the code, and the
> sw too some minor changes (i2c eerom added in bus #0),  but I really don't
> understand why it doesn't work any more :(
>
> It might be clock issue, I'll check.
>
>

I still don't understand what reason for it to get stuck in:
 if (readw (I2C_CON) & I2C_CON_EN)
It's a simple register reading, Right ?
So why should it hangs after calling ?

Regards,
Ran


>> bye,
>> Heiko
>>
>>
>>> Regards,
>>>
>>> Ran
>>> _______________________________________________
>>> U-Boot mailing list
>>> U-Boot at lists.denx.de
>>> http://lists.denx.de/mailman/listinfo/u-boot
>>>
>>>
>> --
>> DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
>> HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
>>
>
>