Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-30 Thread Laxman Dewangan
Hi Mark/Samuel, There is one more change for doing cleanups in regulator parsing where I moved the regulator parsing from mfd to regulator driver: [PATCH] mfd: tps6586x: move regulator dt parsing to regulator driver Although this series and above change is independent but I like to go these

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-30 Thread Laxman Dewangan
Hi Mark/Samuel, There is one more change for doing cleanups in regulator parsing where I moved the regulator parsing from mfd to regulator driver: [PATCH] mfd: tps6586x: move regulator dt parsing to regulator driver Although this series and above change is independent but I like to go these

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-30 Thread Laxman Dewangan
Hi Mark/Samuel, There is one more change for doing cleanups in regulator parsing where I moved the regulator parsing from mfd to regulator driver: [PATCH] mfd: tps6586x: move regulator dt parsing to regulator driver Although this series and above change is independent but I like to go these

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-30 Thread Laxman Dewangan
Hi Mark/Samuel, There is one more change for doing cleanups in regulator parsing where I moved the regulator parsing from mfd to regulator driver: [PATCH] mfd: tps6586x: move regulator dt parsing to regulator driver Although this series and above change is independent but I like to go these

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-09 Thread Mark Brown
On Tue, Oct 09, 2012 at 11:34:47AM +0530, Laxman Dewangan wrote: > The bit definitions are also different in status and mask register > and yes, this will be again problem as we do in isr thread > data->status_buf[i] &= ~data->mask_buf[i]; > So given the constraint, generic will not work here.

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-09 Thread Laxman Dewangan
On Tuesday 09 October 2012 11:53 AM, Mark Brown wrote: On Tue, Oct 09, 2012 at 11:18:24AM +0530, Laxman Dewangan wrote: On Tuesday 09 October 2012 11:36 AM, Mark Brown wrote: but can you convert to regmap_irq? Yes, the motivation was this when I started this cleanups but found that there is 4

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-09 Thread Mark Brown
On Tue, Oct 09, 2012 at 11:18:24AM +0530, Laxman Dewangan wrote: > On Tuesday 09 October 2012 11:36 AM, Mark Brown wrote: > >but can you convert to regmap_irq? > Yes, the motivation was this when I started this cleanups but found > that there is 4 interrupt status and 5 interrupt mask register.

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-09 Thread Laxman Dewangan
On Tuesday 09 October 2012 11:36 AM, Mark Brown wrote: On Mon, Oct 08, 2012 at 09:43:23PM +0530, Laxman Dewangan wrote: Allocate the irq base if it base is not porvided i.e. in case of device tree invocation of this driver. Convert the tps6586x driver to irq domain, using a legacy IRQ mapping

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-09 Thread Mark Brown
On Mon, Oct 08, 2012 at 09:43:23PM +0530, Laxman Dewangan wrote: > Allocate the irq base if it base is not porvided i.e. > in case of device tree invocation of this driver. > Convert the tps6586x driver to irq domain, using a > legacy IRQ mapping if an irq_base is specified in > platform data or

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-09 Thread Mark Brown
On Mon, Oct 08, 2012 at 09:43:23PM +0530, Laxman Dewangan wrote: Allocate the irq base if it base is not porvided i.e. in case of device tree invocation of this driver. Convert the tps6586x driver to irq domain, using a legacy IRQ mapping if an irq_base is specified in platform data or

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-09 Thread Laxman Dewangan
On Tuesday 09 October 2012 11:36 AM, Mark Brown wrote: On Mon, Oct 08, 2012 at 09:43:23PM +0530, Laxman Dewangan wrote: Allocate the irq base if it base is not porvided i.e. in case of device tree invocation of this driver. Convert the tps6586x driver to irq domain, using a legacy IRQ mapping

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-09 Thread Mark Brown
On Tue, Oct 09, 2012 at 11:18:24AM +0530, Laxman Dewangan wrote: On Tuesday 09 October 2012 11:36 AM, Mark Brown wrote: but can you convert to regmap_irq? Yes, the motivation was this when I started this cleanups but found that there is 4 interrupt status and 5 interrupt mask register.

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-09 Thread Laxman Dewangan
On Tuesday 09 October 2012 11:53 AM, Mark Brown wrote: On Tue, Oct 09, 2012 at 11:18:24AM +0530, Laxman Dewangan wrote: On Tuesday 09 October 2012 11:36 AM, Mark Brown wrote: but can you convert to regmap_irq? Yes, the motivation was this when I started this cleanups but found that there is 4

Re: [PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-09 Thread Mark Brown
On Tue, Oct 09, 2012 at 11:34:47AM +0530, Laxman Dewangan wrote: The bit definitions are also different in status and mask register and yes, this will be again problem as we do in isr thread data-status_buf[i] = ~data-mask_buf[i]; So given the constraint, generic will not work here. OK,

[PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-08 Thread Laxman Dewangan
Allocate the irq base if it base is not porvided i.e. in case of device tree invocation of this driver. Convert the tps6586x driver to irq domain, using a legacy IRQ mapping if an irq_base is specified in platform data or dynamically allocated and otherwise using a linear mapping. Signed-off-by:

[PATCH 1/3] mfd: Convert tps6586x to irq_domain

2012-10-08 Thread Laxman Dewangan
Allocate the irq base if it base is not porvided i.e. in case of device tree invocation of this driver. Convert the tps6586x driver to irq domain, using a legacy IRQ mapping if an irq_base is specified in platform data or dynamically allocated and otherwise using a linear mapping. Signed-off-by: