Re: [PATCH net-next v6 3/3] net: stmmac: Add driver support for DWMAC5 common safety IRQ

[Serge Semin]

On Mon, Dec 18, 2023 at 03:27:54PM +0530, Suraj Jaiswal wrote:
> 
> Hi Serge,
> Please find commnet inline & let me know if any further action needed
> 
> Thanks
> Suraj
> 
> On 12/14/2023 8:42 PM, Serge Semin wrote:
> > Hi Suraj
> > 
> > On Tue, Dec 12, 2023 at 05:28:41PM +0530, Suraj Jaiswal wrote:
> >> Add support to listen HW safety IRQ like ECC(error
> >> correction code), DPP(data path parity), FSM(finite state
> >> machine) fault in common IRQ line.
> >>
> >> Signed-off-by: Suraj Jaiswal <quic_jsuraj@xxxxxxxxxxx>
> >> ---
> >>  drivers/net/ethernet/stmicro/stmmac/common.h  |  1 +
> >>  drivers/net/ethernet/stmicro/stmmac/stmmac.h  |  3 +++
> >>  .../net/ethernet/stmicro/stmmac/stmmac_main.c | 21 +++++++++++++++++++
> >>  .../ethernet/stmicro/stmmac/stmmac_platform.c |  9 ++++++++
> >>  4 files changed, 34 insertions(+)
> >>
> >> diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
> >> index 721c1f8e892f..b9233b09b80f 100644
> >> --- a/drivers/net/ethernet/stmicro/stmmac/common.h
> >> +++ b/drivers/net/ethernet/stmicro/stmmac/common.h
> >> @@ -344,6 +344,7 @@ enum request_irq_err {
> >>  	REQ_IRQ_ERR_ALL,
> >>  	REQ_IRQ_ERR_TX,
> >>  	REQ_IRQ_ERR_RX,
> >> +	REQ_IRQ_ERR_SFTY,
> >>  	REQ_IRQ_ERR_SFTY_UE,
> >>  	REQ_IRQ_ERR_SFTY_CE,
> >>  	REQ_IRQ_ERR_LPI,
> >> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
> >> index 9f89acf31050..ca3d93851bed 100644
> >> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
> >> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
> >> @@ -31,6 +31,7 @@ struct stmmac_resources {
> >>  	int wol_irq;
> >>  	int lpi_irq;
> >>  	int irq;
> >> +	int sfty_irq;
> >>  	int sfty_ce_irq;
> >>  	int sfty_ue_irq;
> >>  	int rx_irq[MTL_MAX_RX_QUEUES];
> >> @@ -297,6 +298,7 @@ struct stmmac_priv {
> >>  	void __iomem *ptpaddr;
> >>  	void __iomem *estaddr;
> >>  	unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
> >> +	int sfty_irq;
> >>  	int sfty_ce_irq;
> >>  	int sfty_ue_irq;
> >>  	int rx_irq[MTL_MAX_RX_QUEUES];
> >> @@ -305,6 +307,7 @@ struct stmmac_priv {
> >>  	char int_name_mac[IFNAMSIZ + 9];
> >>  	char int_name_wol[IFNAMSIZ + 9];
> >>  	char int_name_lpi[IFNAMSIZ + 9];
> >> +	char int_name_sfty[IFNAMSIZ + 10];
> >>  	char int_name_sfty_ce[IFNAMSIZ + 10];
> >>  	char int_name_sfty_ue[IFNAMSIZ + 10];
> >>  	char int_name_rx_irq[MTL_MAX_TX_QUEUES][IFNAMSIZ + 14];
> >> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> >> index 47de466e432c..6cf289f192a7 100644
> >> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> >> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> >> @@ -3592,6 +3592,10 @@ static void stmmac_free_irq(struct net_device *dev,
> >>  		if (priv->wol_irq > 0 && priv->wol_irq != dev->irq)
> >>  			free_irq(priv->wol_irq, dev);
> >>  		fallthrough;
> >> +	case REQ_IRQ_ERR_SFTY:
> >> +		if (priv->sfty_irq > 0 && priv->sfty_irq != dev->irq)
> >> +			free_irq(priv->sfty_irq, dev);
> >> +		fallthrough;
> >>  	case REQ_IRQ_ERR_WOL:
> >>  		free_irq(dev->irq, dev);
> >>  		fallthrough;
> >> @@ -3759,6 +3763,7 @@ static int stmmac_request_irq_single(struct net_device *dev)
> >>  	struct stmmac_priv *priv = netdev_priv(dev);
> >>  	enum request_irq_err irq_err;
> >>  	int ret;
> > 
> >> +	char *int_name;
> > 
> > See my comment below.
> > 
> >>  
> >>  	ret = request_irq(dev->irq, stmmac_interrupt,
> >>  			  IRQF_SHARED, dev->name, dev);
> >> @@ -3798,6 +3803,20 @@ static int stmmac_request_irq_single(struct net_device *dev)
> >>  		}
> >>  	}
> >>  
> > 
> >> +	if (priv->sfty_irq > 0 && priv->sfty_irq != dev->irq) {
> >> +		int_name = priv->int_name_sfty;
> >> +		sprintf(int_name, "%s:%s", dev->name, "safety");
> >> +		ret = request_irq(priv->sfty_irq, stmmac_safety_interrupt,
> >> +				  0, int_name, dev);
> >> +		if (unlikely(ret < 0)) {
> >> +			netdev_err(priv->dev,
> >> +				   "%s: alloc safety failed %d (error: %d)\n",
> >> +				   __func__, priv->sfty_irq, ret);
> >> +			irq_err = REQ_IRQ_ERR_SFTY;
> >> +			goto irq_error;
> >> +		}
> >> +	}
> >> +
> > 

> > Omg, I thought this change belonged to stmmac_request_irq_multi_msi().
> > My bad, sorry. Please move the code above to
> > stmmac_request_irq_multi_msi() and get back the part in
> > stmmac_request_irq_single() as it was in v5,

Please note my comment regarding the common safety IRQ being supported
in both stmmac_request_irq_single() and stmmac_request_irq_multi_msi()
methods.

> but instead of specifying
> > "safety" IRQ name use "dev->name" as the rest of similar code snippets
> > in here have:
> > 
> > +	if (priv->sfty_irq > 0 && priv->sfty_irq != dev->irq) {
> > +		ret = request_irq(priv->sfty_irq, stmmac_safety_interrupt,
> > +				  0, dev->name, dev);
> > +		if (unlikely(ret < 0)) {
> > +			netdev_err(priv->dev,
> > +				   "%s: alloc safety failed %d (error: %d)\n",
> > +				   __func__, priv->sfty_irq, ret);
> > +			irq_err = REQ_IRQ_ERR_SFTY;
> > +			goto irq_error;
> > +		}
> > +	}
> 

> <Suraj> We can not use "dev->name" as this is name already used by "stmmac_interrupt" @ https://elixir.bootlin.com/linux/v6.1.68/source/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c#L3655.

It's not that much of the problem. The main idea is to convert your
solution to following the _local_ coding convention. See, the rest of
the IRQs in stmmac_request_irq_single() are requested with "dev->name"
being specified as the IRQ name (irrespective to having such solution
being not that correct). That's what I was talking about. If you want
the safety IRQ to have an unique name, then please submit this patch
as I suggested above and _then_, on top of it, add a new patch which
would convert the entire stmmac_request_irq_single() method to
creating all IRQ names as it's, for instance, done in
stmmac_request_irq_multi_msi().

> <
> ret = request_irq(dev->irq, stmmac_interrupt,
> 			  IRQF_SHARED, dev->name, dev);
> >

> if we are using same "dev->name" while requesting safety IRQ as well then  "/proc/interrupt" will show same name eth0/eth1 for  both "stmmac_interrupt" & "safety interrupt" and by looking at "/proc/interrupt" output we can not say which IRQ is for safety and which is for stmmac_interrupt.

Thanks. I am perfectly aware of that. Please see my comment above.

> > 
> > I guess at some point afterwards we'll need to refactor the IRQs
> > request part of this driver: replace stmmac_request_irq_single() body
> > with the upper part of the stmmac_request_irq_multi_msi() method and
> > then just make the former method being called from the later one...
> > 
> >>  	return 0;
> >>  
> >>  irq_error:
> >> @@ -7462,8 +7481,10 @@ int stmmac_dvr_probe(struct device *device,
> >>  	priv->dev->irq = res->irq;
> >>  	priv->wol_irq = res->wol_irq;
> >>  	priv->lpi_irq = res->lpi_irq;
> >> +	priv->sfty_irq = res->sfty_irq;
> >>  	priv->sfty_ce_irq = res->sfty_ce_irq;
> >>  	priv->sfty_ue_irq = res->sfty_ue_irq;
> > 
> >> +
> > 
> > Please drop this change. The code below is attached to the code above
> > because it basically does the same but in the loop.

> <Suraj> below loop code "for (i = 0; i < MTL_MAX_RX_QUEUES; i++) priv->rx_irq[i] = res->rx_irq[i];" is not for rx_irq array and will not help for safety irq.
> Let me know if I got your commnet properly .

Sorry, you didn't. My comment concerned the _empty_ line you placed
between the code above and below. You shouldn't have done that.

> > 
> >>  	for (i = 0; i < MTL_MAX_RX_QUEUES; i++)
> >>  		priv->rx_irq[i] = res->rx_irq[i];
> >>  	for (i = 0; i < MTL_MAX_TX_QUEUES; i++)
> >> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
> >> index 1ffde555da47..3808a3225a7d 100644
> >> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
> >> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
> >> @@ -726,6 +726,15 @@ int stmmac_get_platform_resources(struct platform_device *pdev,
> >>  		dev_info(&pdev->dev, "IRQ eth_lpi not found\n");
> >>  	}
> >>  
> >> +	stmmac_res->sfty_irq =
> >> +		platform_get_irq_byname_optional(pdev, "sfty");
> > 
> >> +
> > 
> > Please drop this change too. It's normal to have a method call
> > attached to the error check statement especially seeing the rest of
> > the similar code snippets are designed that way in this function.

> <Suraj> Do you means to remove all below code where we are printing the dev_info() message ?

No. I was referring to the _empty_ line between the method above and
the error check code below. It's pointless and at the very least
breaks the local coding convention.

-Serge(y)

> We added this code similar to LPM code.
> > 
> > -Serge(y)
> > 
> >> +	if (stmmac_res->sfty_irq < 0) {
> >> +		if (stmmac_res->sfty_irq == -EPROBE_DEFER)
> >> +			return -EPROBE_DEFER;
> >> +		dev_info(&pdev->dev, "IRQ safety IRQ not found\n");
> >> +	}
> >> +
> >>  	stmmac_res->addr = devm_platform_ioremap_resource(pdev, 0);
> >>  
> >>  	return PTR_ERR_OR_ZERO(stmmac_res->addr);
> >> -- 
> >> 2.25.1
> >>
> >>