On 1 August 2013 11:20, Sudeep KarkadaNagesha
<Sudeep.KarkadaNagesha@xxxxxxx> wrote:
> On 26/07/13 21:25, Markus Mayer wrote:
>> From: Markus Mayer <mmayer@xxxxxxxxxxxx>
>>
>> This patch adds the GPIO driver for the bcm281xx family of chips.
>>
> May be I am missing my mails, but I could not find even in archives.
> The patch says its v3, I could not trace v1/v2 and not change log.
No, you didn't miss any e-mails. We had some internal reviews and the
plan was for me to renumber the patches and start from v1 again for
the public review, but I clearly forgot. Hence v3 is really the first
public version of this patch.
> Why these macros can't be something like this ?
> #define GPIO_GPOR_OFFSET(bank) (0x00000000 + (bank) << 2)
Done.
> Can this be embed in bcm_kona_gpio?
> You can then get rid of all the accesses to global variable
> bcm_kona_gpio.
Still working on this one, but the global variable will go away. The
same is true of the other globals.
>> +static inline u32 bcm_kona_gpio_out_status(int bank)
>> +{
>> + return GPIO_GPOR0_OFFSET + (bank << 2);
>> +}
>> +
>> +static inline u32 bcm_kona_gpio_in_status(int bank)
>> +{
>> + return GPIO_GPIR0_OFFSET + (bank << 2);
>> +}
>> +
>> +static inline u32 bcm_kona_gpio_out_set(int bank)
>> +{
>> + return GPIO_GPORS0_OFFSET + (bank << 2);
>> +}
>> +
>> +static inline u32 bcm_kona_gpio_out_clear(int bank)
>> +{
>> + return GPIO_GPORC0_OFFSET + (bank << 2);
>> +}
>> +
>> +static inline u32 bcm_kona_gpio_int_status(int bank)
>> +{
>> + return GPIO_ISR0_OFFSET + (bank << 2);
>> +}
>> +
>> +static inline u32 bcm_kona_gpio_int_mask(int bank)
>> +{
>> + return GPIO_IMR0_OFFSET + (bank << 2);
>> +}
>> +
>> +static inline u32 bcm_kona_gpio_int_mskclr(int bank)
>> +{
>> + return GPIO_IMRC0_OFFSET + (bank << 2);
>> +}
>> +
>> +static inline u32 bcm_kona_gpio_pwd_status(int bank)
>> +{
>> + return GPIO_GPPLSR0_OFFSET + (bank << 2);
>> +}
>> +
>> +static inline u32 bcm_kona_gpio_control(int gpio)
>> +{
>> + return GPIO_GPCTR0_OFFSET + (gpio << 2);
>> +}
> All these functions can go away if you can have the macros as shown above ?
I took those functions out.
>> +static int bcm_kona_count_irq_resources(struct platform_device *pdev)
>> +{
>> + struct resource *res;
>> + int count = 0;
>> +
>> + for (;;) {
>> + res = platform_get_resource(pdev, IORESOURCE_IRQ, count);
> platform_get_irq instead
Done.
>> + if (!res)
>> + break;
>> + count++;
>> + }
>> + return count;
>> +}
> IMO, you don't require any global data or pointers if probe populates
> all the pdev, irq/gpio_chip data member appropriately
Yes, they'll go away in the next revision of this driver which I am
hoping to post shortly.
Thanks,
-Markus
--
Markus Mayer
Broadcom Landing Team
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
