parport starts using device-model and we now have parport under
/sys/bus. As the ports are discovered they are added as device under
/sys/bus/parport. As and when other drivers register new device,
they will be registered as a subdevice under the relevant parport.
Signed-off-by: Sudip Mukherjee
-
As suggested by Greg, new functions were introduced in the parport
subsystem which will use the device-model and we convert the drivers
one by one. So accordingly we now have parport_register_drv(),
parport_register_dev() and attach_ret() which are using the device-model.
And we have a flag to ind
modified the functions to use the new device-model of parport.
Signed-off-by: Sudip Mukherjee
---
drivers/i2c/busses/i2c-parport.c | 19 ---
1 file changed, 12 insertions(+), 7 deletions(-)
diff --git a/drivers/i2c/busses/i2c-parport.c b/drivers/i2c/busses/i2c-parport.c
index a1
as parport starts using the device-model, update the documentation
to show the newly added functions and update TODO with some other
planned modifications.
Signed-off-by: Sudip Mukherjee
---
Documentation/parport-lowlevel.txt | 49 ++
drivers/parport/TODO-parp
modified the required functions to start using the new parport
device-model
Signed-off-by: Sudip Mukherjee
---
drivers/staging/panel/panel.c | 20 +++-
1 file changed, 11 insertions(+), 9 deletions(-)
diff --git a/drivers/staging/panel/panel.c b/drivers/staging/panel/panel.c
ind
Sorry, I still haven't done a proper review.
On Wed, Apr 15, 2015 at 01:18:41PM +0530, Sudip Mukherjee wrote:
> +struct pardevice *
> +parport_register_dev(struct parport *port, const char *name,
> + int (*pf)(void *), void (*kf)(void *),
> + void (*irq_func)(void
On Wed, Apr 15, 2015 at 01:18:41PM +0530, Sudip Mukherjee wrote:
> @@ -575,6 +647,7 @@ parport_register_device(struct parport *port, const char
> *name,
> tmp->irq_func = irq_func;
> tmp->waiting = 0;
> tmp->timeout = 5 * HZ;
> + tmp->devmodel = false;
>
> /* Chain th
On Wed, Apr 15, 2015 at 11:27:46AM +0300, Dan Carpenter wrote:
> Sorry, I still haven't done a proper review.
for almost all your points: it came as i copied the parport_register_dev
from parport_register_device and just added some part leaving everything
else same. I will fix these points in v2 o
On Wed, Apr 15, 2015 at 11:33:59AM +0300, Dan Carpenter wrote:
> On Wed, Apr 15, 2015 at 01:18:41PM +0530, Sudip Mukherjee wrote:
>
>
> The difference between parport_register_device() and
> parport_register_dev() isn't clear from the name.
i kept the name similar deliberately as I thought that a
On Wed, Apr 15, 2015 at 02:50:55PM +0530, Sudip Mukherjee wrote:
> this PARPORT_DEVPROC_REGISTERED flag is cleared in parport_unregister_device()
> and is set in parport_register_dev[ice], so when we call
> parport_register_device() or parport_register_dev() it will be not set
> and the condition w
On Wed, Apr 15, 2015 at 02:50:55PM +0530, Sudip Mukherjee wrote:
>
> > > + tmp->name = name;
> >
> > I wonder who frees this name variable. My concern is that it gets
> > freed before we are done using it or something. (I have not looked at
> > the details).
> it will be done in free_port() the
On Wed, Apr 15, 2015 at 12:45:00PM +0300, Dan Carpenter wrote:
> On Wed, Apr 15, 2015 at 02:50:55PM +0530, Sudip Mukherjee wrote:
> >
> > > > + tmp->name = name;
> > >
> > > I wonder who frees this name variable. My concern is that it gets
> > > freed before we are done using it or somethi
On Wed, Apr 15, 2015 at 01:18:41PM +0530, Sudip Mukherjee wrote:
> parport starts using device-model and we now have parport under
> /sys/bus. As the ports are discovered they are added as device under
> /sys/bus/parport. As and when other drivers register new device,
> they will be registered as a
On Wed, Apr 15, 2015 at 01:18:41PM +0530, Sudip Mukherjee wrote:
> --- a/drivers/parport/share.c
> +++ b/drivers/parport/share.c
> @@ -10,6 +10,8 @@
> * based on work by Grant Guenther
> * and Philip Blundell
> *
> + * Added Device-Model - Sudip Mukherjee
Changelog handles this, n
On Wed, Apr 15, 2015 at 01:18:41PM +0530, Sudip Mukherjee wrote:
> @@ -29,6 +31,7 @@
> #include
> #include
> #include
> +#include
>
> #include
> #include
> @@ -100,6 +103,11 @@ static struct parport_operations dead_ops = {
> .owner = NULL,
> };
>
> +struct bus_type pa
On Wed, Apr 15, 2015 at 03:31:15PM +0200, Greg Kroah-Hartman wrote:
> On Wed, Apr 15, 2015 at 01:18:41PM +0530, Sudip Mukherjee wrote:
> > @@ -29,6 +31,7 @@
> > +struct bus_type parport_bus_type = {
> > + .name = "parport",
> > +};
> > +EXPORT_SYMBOL(parport_bus_type);
>
> They bus ty
On 14.04.2015 10:00, Dan Carpenter wrote:
> BAReq->addr1 is part of struct rtllib_hdr_2addr. It's 4 bytes into
> the struct so it's fine.
You're right - should I explicitly add __aligned() in such places or just
leave comment in that case (for future commiters)?
> I sort of like the ether_addr_c
On 14.04.2015 10:07, Dan Carpenter wrote:
> On Mon, Apr 13, 2015 at 11:47:36PM +0200, Mateusz Kulikowski wrote:
>> Fix most of remaining PREFER_PR_LEVEL warnings in rtllib.
>> Replace printk() with pr_* where possible.
(...)
>> #define RTLLIB_DEBUG_DATA(level, data, datalen) \
>> do {
On Wed, 2015-04-15 at 23:08 +0200, Mateusz Kulikowski wrote:
> - Discuss and try to submit ether_addr_copy_unaligned on netdev list
I've no issue with a patch that does that.
https://lkml.org/lkml/2014/1/20/338
___
devel mailing list
de...@linuxdriver
On 14.04.2015 10:40, Dan Carpenter wrote:
> On Mon, Apr 13, 2015 at 11:47:38PM +0200, Mateusz Kulikowski wrote:
>> +if (*rfa_pti_r > 4) {
>> +(*rfa_pti_r)--;
>
>
> Honestly, I thought that patch 14 was too hard to review and did too
> many things, b
20 matches
Mail list logo
