On 24 Jun 2010, at 13:35, Naveen Krishna Ch <naveenkrishna.ch@xxxxxxxxx> wrote:
> Hi Mark,
>
> On 24 June 2010 17:02, Mark Brown <broonie@xxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
>>
>> On Wed, Jun 23, 2010 at 11:15:55AM +0530, Naveen Krishna Ch wrote:
>>
>>> Where should i be looking,
>>
>> I would suggest looking at the driver and comparing it with the
>> datasheet.
>>
>>> Will changing the spin counts work?
>>
>> I'd expect this to be unlikely. Note that if we run out of spins we
>> still go for the 1ms sleep that was being used originally so it's more
>> likely that there's a bug in the chip when that register is being read
>> (though this would be surprising) or some other change in your kernel is
>> causing problems.
>
> --- a/drivers/i2c/busses/i2c-s3c2410.c
> +++ b/drivers/i2c/busses/i2c-s3c2410.c
> @@ -527,7 +527,7 @@ static int s3c24xx_i2c_doxfer(struct s3c24xx_i2c *i2c,
> /* first, try busy waiting briefly */
> do {
> iicstat = readl(i2c->regs + S3C2410_IICSTAT);
> - } while ((iicstat & S3C2410_IICSTAT_START) && --spins);
> + } while (--spins && (iicstat & S3C2410_IICSTAT_START));
>
> /* if that timed out sleep */
> if (!spins) {
>
> Does the trick..
That smells rather like a compiler bug to me - the two versions
ought to be equivalent apart from the ordering of the comparisons.--
To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
