On Mon, Jul 28, 2008 at 5:23 PM, Ben Dooks <[EMAIL PROTECTED]> wrote:
> On Sun, Jul 27, 2008 at 02:41:54PM +0800, Bryan Wu wrote:
>> From: Michael Hennerich <[EMAIL PROTECTED]>
>>
>> PM_SUSPEND_MEM: Blackfin does not maintain register state through
>> Hibernate. Save and restore peripheral base initialization during
>> PM transitions.
>>
>> Signed-off-by: Michael Hennerich <[EMAIL PROTECTED]>
>> Signed-off-by: Bryan Wu <[EMAIL PROTECTED]>
>> ---
>> drivers/i2c/busses/i2c-bfin-twi.c | 35 ++++++++++++++++++++++++-----------
>> 1 files changed, 24 insertions(+), 11 deletions(-)
>>
>> diff --git a/drivers/i2c/busses/i2c-bfin-twi.c
>> b/drivers/i2c/busses/i2c-bfin-twi.c
>> index 48d084b..3c855ff 100644
>> --- a/drivers/i2c/busses/i2c-bfin-twi.c
>> +++ b/drivers/i2c/busses/i2c-bfin-twi.c
>> @@ -49,6 +49,8 @@ struct bfin_twi_iface {
>> struct i2c_msg *pmsg;
>> int msg_num;
>> int cur_msg;
>> + u16 saved_clkdiv;
>> + u16 saved_control;
>> void __iomem *regs_base;
>> };
>>
>> @@ -565,32 +567,43 @@ static u32 bfin_twi_functionality(struct i2c_adapter
>> *adap)
>> I2C_FUNC_I2C;
>> }
>>
>> -
>> static struct i2c_algorithm bfin_twi_algorithm = {
>> .master_xfer = bfin_twi_master_xfer,
>> .smbus_xfer = bfin_twi_smbus_xfer,
>> .functionality = bfin_twi_functionality,
>> };
>>
>> -
>> -static int i2c_bfin_twi_suspend(struct platform_device *dev, pm_message_t
>> state)
>> +static int i2c_bfin_twi_suspend(struct platform_device *pdev, pm_message_t
>> state)
>> {
>> - struct bfin_twi_iface *iface = platform_get_drvdata(dev);
>> + struct bfin_twi_iface *iface = platform_get_drvdata(pdev);
>> +
>> + iface->saved_clkdiv = read_CLKDIV(iface);
>> + iface->saved_control = read_CONTROL(iface);
>> +
>> + free_irq(iface->irq, iface);
>>
>> /* Disable TWI */
>> - write_CONTROL(iface, read_CONTROL(iface) & ~TWI_ENA);
>> - SSYNC();
>> + write_CONTROL(iface, iface->saved_control & ~TWI_ENA);
>
> I assume removing the SSYNC() call is valid?
>
Yes, I think it is OK for that. We tested on our platform.
Actually, keeping the SSYNC() is not a big deal for this driver.
How do you think Michael?
>> return 0;
>> }
>>
>> -static int i2c_bfin_twi_resume(struct platform_device *dev)
>> +static int i2c_bfin_twi_resume(struct platform_device *pdev)
>> {
>> - struct bfin_twi_iface *iface = platform_get_drvdata(dev);
>> + struct bfin_twi_iface *iface = platform_get_drvdata(pdev);
>>
>> - /* Enable TWI */
>> - write_CONTROL(iface, read_CONTROL(iface) | TWI_ENA);
>> - SSYNC();
>> + int rc = request_irq(iface->irq, bfin_twi_interrupt_entry,
>> + IRQF_DISABLED, pdev->name, iface);
>> + if (rc) {
>> + dev_err(&pdev->dev, "Can't get IRQ %d !\n", iface->irq);
>> + return -ENODEV;
>> + }
>> +
>> + /* Resume TWI interface clock as specified */
>> + write_CLKDIV(iface, iface->saved_clkdiv);
>> +
>> + /* Resume TWI */
>> + write_CONTROL(iface, iface->saved_control);
>>
>> return 0;
>> }
>> --
>> 1.5.6
>
> if removing the SSYNC() call is ok, then I'll queue this into
> my tree which I will be requesting a pull asap.
>
Thanks lot, I'd like to see that.
-Bryan
_______________________________________________
i2c mailing list
[email protected]
http://lists.lm-sensors.org/mailman/listinfo/i2c
