* [PATCH] net: sfp: Unique GPIO interrupt names
@ 2020-07-06 19:38 Chris Healy
2020-07-06 20:07 ` Andrew Lunn
2020-07-06 20:42 ` Russell King - ARM Linux admin
0 siblings, 2 replies; 5+ messages in thread
From: Chris Healy @ 2020-07-06 19:38 UTC (permalink / raw)
To: Russell King - ARM Linux, Andrew Lunn, Florian Fainelli,
Heiner Kallweit, David S. Miller, kuba, netdev, linux-kernel
Dynamically generate a unique GPIO interrupt name, based on the
device name and the GPIO name. For example:
103: 0 sx1503q 12 Edge sff2-los
104: 0 sx1503q 13 Edge sff3-los
The sffX indicates the SFP the loss of signal GPIO is associated with.
Signed-off-by: Chris Healy <cphealy@gmail.com>
---
drivers/net/phy/sfp.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index 73c2969f11a4..9b03c7229320 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -220,6 +220,7 @@ struct sfp {
struct phy_device *mod_phy;
const struct sff_data *type;
u32 max_power_mW;
+ char sfp_irq_name[32];
unsigned int (*get_state)(struct sfp *);
void (*set_state)(struct sfp *, unsigned int);
@@ -2349,12 +2350,15 @@ static int sfp_probe(struct platform_device *pdev)
continue;
}
+ snprintf(sfp->sfp_irq_name, sizeof(sfp->sfp_irq_name),
+ "%s-%s", dev_name(sfp->dev), gpio_of_names[i]);
+
err = devm_request_threaded_irq(sfp->dev, sfp->gpio_irq[i],
NULL, sfp_irq,
IRQF_ONESHOT |
IRQF_TRIGGER_RISING |
IRQF_TRIGGER_FALLING,
- dev_name(sfp->dev), sfp);
+ sfp->sfp_irq_name, sfp);
if (err) {
sfp->gpio_irq[i] = 0;
sfp->need_poll = true;
--
2.21.3
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] net: sfp: Unique GPIO interrupt names
2020-07-06 19:38 [PATCH] net: sfp: Unique GPIO interrupt names Chris Healy
@ 2020-07-06 20:07 ` Andrew Lunn
2020-07-06 23:03 ` Chris Healy
2020-07-06 20:42 ` Russell King - ARM Linux admin
1 sibling, 1 reply; 5+ messages in thread
From: Andrew Lunn @ 2020-07-06 20:07 UTC (permalink / raw)
To: Chris Healy
Cc: Russell King - ARM Linux, Florian Fainelli, Heiner Kallweit,
David S. Miller, kuba, netdev, linux-kernel
On Mon, Jul 06, 2020 at 12:38:37PM -0700, Chris Healy wrote:
> Dynamically generate a unique GPIO interrupt name, based on the
> device name and the GPIO name. For example:
>
> 103: 0 sx1503q 12 Edge sff2-los
> 104: 0 sx1503q 13 Edge sff3-los
>
> The sffX indicates the SFP the loss of signal GPIO is associated with.
Hi Chris
For netdev, please put inside the [PATCH] part of the subject, which
tree this is for, i.e. net-next.
> Signed-off-by: Chris Healy <cphealy@gmail.com>
> ---
> drivers/net/phy/sfp.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
> index 73c2969f11a4..9b03c7229320 100644
> --- a/drivers/net/phy/sfp.c
> +++ b/drivers/net/phy/sfp.c
> @@ -220,6 +220,7 @@ struct sfp {
> struct phy_device *mod_phy;
> const struct sff_data *type;
> u32 max_power_mW;
> + char sfp_irq_name[32];
>
> unsigned int (*get_state)(struct sfp *);
> void (*set_state)(struct sfp *, unsigned int);
> @@ -2349,12 +2350,15 @@ static int sfp_probe(struct platform_device *pdev)
> continue;
> }
>
> + snprintf(sfp->sfp_irq_name, sizeof(sfp->sfp_irq_name),
> + "%s-%s", dev_name(sfp->dev), gpio_of_names[i]);
> +
This is perfectly O.K, but you could consider using
devm_kasprintf(). That will allocate as much memory as needed for the
string, and hence avoid truncation issues, which we have seen before
with other interrupt names.
Andrew
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] net: sfp: Unique GPIO interrupt names
2020-07-06 19:38 [PATCH] net: sfp: Unique GPIO interrupt names Chris Healy
2020-07-06 20:07 ` Andrew Lunn
@ 2020-07-06 20:42 ` Russell King - ARM Linux admin
2020-07-06 23:04 ` Chris Healy
1 sibling, 1 reply; 5+ messages in thread
From: Russell King - ARM Linux admin @ 2020-07-06 20:42 UTC (permalink / raw)
To: Chris Healy
Cc: Andrew Lunn, Florian Fainelli, Heiner Kallweit, David S. Miller,
kuba, netdev, linux-kernel
On Mon, Jul 06, 2020 at 12:38:37PM -0700, Chris Healy wrote:
> Dynamically generate a unique GPIO interrupt name, based on the
> device name and the GPIO name. For example:
>
> 103: 0 sx1503q 12 Edge sff2-los
> 104: 0 sx1503q 13 Edge sff3-los
>
> The sffX indicates the SFP the loss of signal GPIO is associated with.
>
> Signed-off-by: Chris Healy <cphealy@gmail.com>
This doesn't work in all cases.
> ---
> drivers/net/phy/sfp.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
> index 73c2969f11a4..9b03c7229320 100644
> --- a/drivers/net/phy/sfp.c
> +++ b/drivers/net/phy/sfp.c
> @@ -220,6 +220,7 @@ struct sfp {
> struct phy_device *mod_phy;
> const struct sff_data *type;
> u32 max_power_mW;
> + char sfp_irq_name[32];
>
> unsigned int (*get_state)(struct sfp *);
> void (*set_state)(struct sfp *, unsigned int);
> @@ -2349,12 +2350,15 @@ static int sfp_probe(struct platform_device *pdev)
> continue;
> }
>
> + snprintf(sfp->sfp_irq_name, sizeof(sfp->sfp_irq_name),
> + "%s-%s", dev_name(sfp->dev), gpio_of_names[i]);
sfp_irq_name will be overwritten for each GPIO IRQ claimed, which means
all IRQs for a particular cage will end up with the same name.
sfp_irq_name[] therefore needs to be an array of names, one per input.
Thanks.
--
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] net: sfp: Unique GPIO interrupt names
2020-07-06 20:07 ` Andrew Lunn
@ 2020-07-06 23:03 ` Chris Healy
0 siblings, 0 replies; 5+ messages in thread
From: Chris Healy @ 2020-07-06 23:03 UTC (permalink / raw)
To: Andrew Lunn
Cc: Russell King - ARM Linux, Florian Fainelli, Heiner Kallweit,
David S. Miller, kuba, netdev, linux-kernel
On Mon, Jul 6, 2020 at 1:07 PM Andrew Lunn <andrew@lunn.ch> wrote:
>
> On Mon, Jul 06, 2020 at 12:38:37PM -0700, Chris Healy wrote:
> > Dynamically generate a unique GPIO interrupt name, based on the
> > device name and the GPIO name. For example:
> >
> > 103: 0 sx1503q 12 Edge sff2-los
> > 104: 0 sx1503q 13 Edge sff3-los
> >
> > The sffX indicates the SFP the loss of signal GPIO is associated with.
>
> Hi Chris
>
> For netdev, please put inside the [PATCH] part of the subject, which
> tree this is for, i.e. net-next.
Will do.
>
> > Signed-off-by: Chris Healy <cphealy@gmail.com>
> > ---
> > drivers/net/phy/sfp.c | 6 +++++-
> > 1 file changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
> > index 73c2969f11a4..9b03c7229320 100644
> > --- a/drivers/net/phy/sfp.c
> > +++ b/drivers/net/phy/sfp.c
> > @@ -220,6 +220,7 @@ struct sfp {
> > struct phy_device *mod_phy;
> > const struct sff_data *type;
> > u32 max_power_mW;
> > + char sfp_irq_name[32];
> >
> > unsigned int (*get_state)(struct sfp *);
> > void (*set_state)(struct sfp *, unsigned int);
> > @@ -2349,12 +2350,15 @@ static int sfp_probe(struct platform_device *pdev)
> > continue;
> > }
> >
> > + snprintf(sfp->sfp_irq_name, sizeof(sfp->sfp_irq_name),
> > + "%s-%s", dev_name(sfp->dev), gpio_of_names[i]);
> > +
>
> This is perfectly O.K, but you could consider using
> devm_kasprintf(). That will allocate as much memory as needed for the
> string, and hence avoid truncation issues, which we have seen before
> with other interrupt names.
I'll give this a try for the next version of this patch.
>
> Andrew
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] net: sfp: Unique GPIO interrupt names
2020-07-06 20:42 ` Russell King - ARM Linux admin
@ 2020-07-06 23:04 ` Chris Healy
0 siblings, 0 replies; 5+ messages in thread
From: Chris Healy @ 2020-07-06 23:04 UTC (permalink / raw)
To: Russell King - ARM Linux admin
Cc: Andrew Lunn, Florian Fainelli, Heiner Kallweit, David S. Miller,
kuba, netdev, linux-kernel
On Mon, Jul 6, 2020 at 1:42 PM Russell King - ARM Linux admin
<linux@armlinux.org.uk> wrote:
>
> On Mon, Jul 06, 2020 at 12:38:37PM -0700, Chris Healy wrote:
> > Dynamically generate a unique GPIO interrupt name, based on the
> > device name and the GPIO name. For example:
> >
> > 103: 0 sx1503q 12 Edge sff2-los
> > 104: 0 sx1503q 13 Edge sff3-los
> >
> > The sffX indicates the SFP the loss of signal GPIO is associated with.
> >
> > Signed-off-by: Chris Healy <cphealy@gmail.com>
>
> This doesn't work in all cases.
>
> > ---
> > drivers/net/phy/sfp.c | 6 +++++-
> > 1 file changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
> > index 73c2969f11a4..9b03c7229320 100644
> > --- a/drivers/net/phy/sfp.c
> > +++ b/drivers/net/phy/sfp.c
> > @@ -220,6 +220,7 @@ struct sfp {
> > struct phy_device *mod_phy;
> > const struct sff_data *type;
> > u32 max_power_mW;
> > + char sfp_irq_name[32];
> >
> > unsigned int (*get_state)(struct sfp *);
> > void (*set_state)(struct sfp *, unsigned int);
> > @@ -2349,12 +2350,15 @@ static int sfp_probe(struct platform_device *pdev)
> > continue;
> > }
> >
> > + snprintf(sfp->sfp_irq_name, sizeof(sfp->sfp_irq_name),
> > + "%s-%s", dev_name(sfp->dev), gpio_of_names[i]);
>
> sfp_irq_name will be overwritten for each GPIO IRQ claimed, which means
> all IRQs for a particular cage will end up with the same name.
> sfp_irq_name[] therefore needs to be an array of names, one per input.
Good point. I'll add the necessary support to deal with this case and
test on my side with a hacked up devicetree file providing some
additional GPIOs and include this in the next version of the patch.
>
> Thanks.
>
> --
> RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
> FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2020-07-06 23:05 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-07-06 19:38 [PATCH] net: sfp: Unique GPIO interrupt names Chris Healy
2020-07-06 20:07 ` Andrew Lunn
2020-07-06 23:03 ` Chris Healy
2020-07-06 20:42 ` Russell King - ARM Linux admin
2020-07-06 23:04 ` Chris Healy
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).