Re: [PATCH V2 2/3] iommu/rockchip: add multi irqs support

[Heiko Stuebner <heiko@sntech.de>]

Am Freitag, 21. Juli 2017, 15:54:40 CEST schrieb xxm:
> Hi Heiko,
> 
> 
> On 07/21/2017 03:07 PM, Heiko Stuebner wrote:
> > Am Freitag, 21. Juli 2017, 14:27:09 CEST schrieb Simon Xue:
> >> From: Simon <xxm@rock-chips.com>
> >>
> >> RK3368 vpu mmu have two irqs, this patch support multi irqs
> >>
> >> Signed-off-by: Simon <xxm@rock-chips.com>
> >> ---
> >> changes since V1:
> >>   - use devm_kcalloc instead of devm_kzalloc when alloc irq array
> >>
> >>   drivers/iommu/rockchip-iommu.c | 34 ++++++++++++++++++++++++----------
> >>   1 file changed, 24 insertions(+), 10 deletions(-)
> >>
> >> diff --git a/drivers/iommu/rockchip-iommu.c b/drivers/iommu/rockchip-iommu.c
> >> index 4ba48a2..3c462c0 100644
> >> --- a/drivers/iommu/rockchip-iommu.c
> >> +++ b/drivers/iommu/rockchip-iommu.c
> >> @@ -90,7 +90,8 @@ struct rk_iommu {
> >>   	struct device *dev;
> >>   	void __iomem **bases;
> >>   	int num_mmu;
> >> -	int irq;
> >> +	int *irq;
> >> +	int num_irq;
> >>   	struct iommu_device iommu;
> >>   	struct list_head node; /* entry in rk_iommu_domain.iommus */
> >>   	struct iommu_domain *domain; /* domain to which iommu is attached */
> >> @@ -825,10 +826,12 @@ static int rk_iommu_attach_device(struct iommu_domain *domain,
> >>   
> >>   	iommu->domain = domain;
> >>   
> >> -	ret = devm_request_irq(iommu->dev, iommu->irq, rk_iommu_irq,
> >> -			       IRQF_SHARED, dev_name(dev), iommu);
> >> -	if (ret)
> >> -		return ret;
> >> +	for (i = 0; i < iommu->num_irq; i++) {
> >> +		ret = devm_request_irq(iommu->dev, iommu->irq[i], rk_iommu_irq,
> >> +				       IRQF_SHARED, dev_name(dev), iommu);
> >> +		if (ret)
> >> +			return ret;
> >> +	}
> >>   
> >>   	for (i = 0; i < iommu->num_mmu; i++) {
> >>   		rk_iommu_write(iommu->bases[i], RK_MMU_DTE_ADDR,
> >> @@ -878,7 +881,8 @@ static void rk_iommu_detach_device(struct iommu_domain *domain,
> >>   	}
> >>   	rk_iommu_disable_stall(iommu);
> >>   
> >> -	devm_free_irq(iommu->dev, iommu->irq, iommu);
> >> +	for (i = 0; i < iommu->num_irq; i++)
> >> +		devm_free_irq(iommu->dev, iommu->irq[i], iommu);
> >>   
> >>   	iommu->domain = NULL;
> >>   
> >> @@ -1157,10 +1161,20 @@ static int rk_iommu_probe(struct platform_device *pdev)
> >>   	if (iommu->num_mmu == 0)
> >>   		return PTR_ERR(iommu->bases[0]);
> >>   
> >> -	iommu->irq = platform_get_irq(pdev, 0);
> >> -	if (iommu->irq < 0) {
> >> -		dev_err(dev, "Failed to get IRQ, %d\n", iommu->irq);
> >> -		return -ENXIO;
> >> +	while (platform_get_irq(pdev, iommu->num_irq) >= 0)
> >> +		iommu->num_irq++;
> > Hmm, this could also result in a iommu having 0 irqs if wrongly
> > configured and probe would still suceed. This sounds somehow
> > wrong to me.
> >
> > But I'm not sure if there is precedent on how to handle a variable
> > number of interrupts correctly somewhere.
> 
> How about add a judgement for iommu->num_irq ? like this:
> if (!iommu->num_irq)
> 	return -ENOXIO;

platform devices already have a function that gets you the number of irqs.
Re-using that is way better than open-coding it, so

iommu->num_irq = platform_irq_count(pdev);
if (iommu->num_irq < 0)
	return iommu->num_irq;
if (iommu->num_irq == 0)
	return -ENXIO;


Heiko