Signed-off-by: Prashant Shah <[email protected]>
---
drivers/net/ethernet/8390/wd.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/8390/wd.c b/drivers/net/ethernet/8390/wd.c
index 03eb3ee..b43a63f 100644
--- a/drivers/net/ethernet/8390/wd.c
+++ b/drivers/net/ethernet/8390/wd.c
@@ -308,7 +308,7 @@ static int __init wd_probe1(struct net_device *dev, int ioaddr)
/* Snarf the interrupt now. There's no point in waiting since we cannot
share and the board will usually be enabled. */
- i = request_irq(dev->irq, ei_interrupt, 0, DRV_NAME, dev);
+ i = request_irq(dev->irq, ei_interrupt, 0, dev->name, dev);
if (i) {
printk (" unable to get IRQ %d.\n", dev->irq);
return i;
--
1.7.9.5
On 24/07/13 16:09, Prashant Shah wrote:
> Signed-off-by: Prashant Shah <[email protected]>
> ---
> drivers/net/ethernet/8390/wd.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/8390/wd.c b/drivers/net/ethernet/8390/wd.c
> index 03eb3ee..b43a63f 100644
> --- a/drivers/net/ethernet/8390/wd.c
> +++ b/drivers/net/ethernet/8390/wd.c
> @@ -308,7 +308,7 @@ static int __init wd_probe1(struct net_device *dev, int ioaddr)
>
> /* Snarf the interrupt now. There's no point in waiting since we cannot
> share and the board will usually be enabled. */
> - i = request_irq(dev->irq, ei_interrupt, 0, DRV_NAME, dev);
> + i = request_irq(dev->irq, ei_interrupt, 0, dev->name, dev);
You should also remove the definition of DRV_NAME, since it is no longer
used. The changelog should probably mention that this will change the
interrupt name (which appears in /proc/interrupts for example) from "wd"
to "eth%d".
~Ryan
Hi,
> You should also remove the definition of DRV_NAME, since it is no longer
> used. The changelog should probably mention that this will change the
> interrupt name (which appears in /proc/interrupts for example) from "wd"
> to "eth%d".
>
I will resend the patch.
Regards.
Hi,
On Thu, Jul 25, 2013 at 10:49 AM, Prashant Shah <[email protected]> wrote:
> Hi,
>
>> You should also remove the definition of DRV_NAME, since it is no longer
>> used. The changelog should probably mention that this will change the
>> interrupt name (which appears in /proc/interrupts for example) from "wd"
>> to "eth%d".
>>
I was going through the code. The apne driver apne.c in the same
folder is using DRV_NAME in request_region()
> if (!request_region(IOBASE, 0x20, DRV_NAME)) {
I can change the wd.c request_region() code to use DRV_NAME. Currently
it is using a string constant.
> r = request_region(base_addr, WD_IO_EXTENT, "wd-probe");
This will make it more consistent. Please suggest which change is more
preferable.
@Matthew, following drivers are not using dev->name in request_irq()
axnet_cs.c
mac8390.c
hydra.c
ne-h8300.c
apne.c
stnic.c
zorro8390.c
Regards.
On Thu, Jul 25, 2013 at 8:38 AM, Prashant Shah <[email protected]> wrote:
> On Thu, Jul 25, 2013 at 10:49 AM, Prashant Shah <[email protected]> wrote:
>>> You should also remove the definition of DRV_NAME, since it is no longer
>>> used. The changelog should probably mention that this will change the
>>> interrupt name (which appears in /proc/interrupts for example) from "wd"
>>> to "eth%d".
>
> I was going through the code. The apne driver apne.c in the same
> folder is using DRV_NAME in request_region()
>
>> if (!request_region(IOBASE, 0x20, DRV_NAME)) {
>
> I can change the wd.c request_region() code to use DRV_NAME. Currently
> it is using a string constant.
>
>> r = request_region(base_addr, WD_IO_EXTENT, "wd-probe");
>
> This will make it more consistent. Please suggest which change is more
> preferable.
If you convert each driver to use devres API, you may avoid a lot of
useless work.
--
With Best Regards,
Andy Shevchenko
On Thu, Jul 25, 2013 at 6:20 PM, tedheadster <[email protected]> wrote:
> Andy,
> I can't find a single example of a devres_* call in
> drivers/net/ethernet/*. Does any networking code exist that we can look at
> as an example for conversion to the devres API?
devres API usually means managed version of the functions that used
mostly at probe() stage.
For example, devm_*() or pcim_*() calls.
--
With Best Regards,
Andy Shevchenko
Hello.
On 07/25/2013 10:31 PM, Andy Shevchenko wrote:
>> Andy,
>> I can't find a single example of a devres_* call in
>> drivers/net/ethernet/*. Does any networking code exist that we can look at
>> as an example for conversion to the devres API?
> devres API usually means managed version of the functions that used
> mostly at probe() stage.
> For example, devm_*() or pcim_*() calls.
It's also called managed device API. In fact, I've never heard it named
devres API.
WBR, Sergei
Hi,
On Thu, Jul 25, 2013 at 10:33:24PM +0400, Sergei Shtylyov wrote:
> It's also called managed device API. In fact, I've never heard it
> named devres API.
Never read the documentation. :-)
$ head -1 Documentation/driver-model/devres.txt
Devres - Managed Device Resource
A.
Hi,
On Fri, Jul 26, 2013 at 12:08 AM, Aaro Koskinen <[email protected]> wrote:
> Hi,
>
> On Thu, Jul 25, 2013 at 10:33:24PM +0400, Sergei Shtylyov wrote:
>> It's also called managed device API. In fact, I've never heard it
>> named devres API.
If I understood it correctly it has to just calls
devm_request_region() with the struct device pointer and there are no
deallocation functions to call ? It manages deallocation on it own.
Regards.
On 27-07-2013 12:58, Prashant Shah wrote:
>>> It's also called managed device API. In fact, I've never heard it
>>> named devres API.
> If I understood it correctly it has to just calls
> devm_request_region() with the struct device pointer and there are no
> deallocation functions to call ? It manages deallocation on it own.
Yes.
> Regards.
WBR, Sergei