[PATCH v2] powerpc/pseries: Fix MSI-X allocation failure when quota is exceeded

[Madhavan Srinivasan]

On 1/7/26 7:55 PM, Nilay Shroff wrote:
>
> On 1/7/26 3:32 PM, Nam Cao wrote:
>> Nilay reported that since commit daaa574aba6f ("powerpc/pseries/msi: Switch
>> to msi_create_parent_irq_domain()"), the NVMe driver cannot enable MSI-X
>> when the device's MSI-X table size is larger than the firmware's MSI quota
>> for the device.
>>
>> This is because the commit changes how rtas_prepare_msi_irqs() is called:
>>
>>    - Before, it is called when interrupts are allocated at the global
>>      interrupt domain with nvec_in being the number of allocated interrupts.
>>      rtas_prepare_msi_irqs() can return a positive number and the allocation
>>      will be retried.
>>
>>    - Now, it is called at the creation of per-device interrupt domain with
>>      nvec_in being the number of interrupts that the device supports. If
>>      rtas_prepare_msi_irqs() returns positive, domain creation just fails.
>>
>> For Nilay's NVMe driver case, rtas_prepare_msi_irqs() returns a positive
>> number (the quota). This causes per-device interrupt domain creation to
>> fail and thus the NVMe driver cannot enable MSI-X.
>>
>> Rework to make this scenario works again:
>>
>>    - pseries_msi_ops_prepare() only prepares as many interrupts as the quota
>>      permit.
>>
>>    - pseries_irq_domain_alloc() fails if the device's quota is exceeded.
>>
>> Now, if the quota is exceeded, pseries_msi_ops_prepare() will only prepare
>> as allowed by the quota. If device drivers attempt to allocate more
>> interrupts than the quota permits, pseries_irq_domain_alloc() will return
>> an error code and msi_handle_pci_fail() will allow device drivers a retry.
>>
>> Reported-by: Nilay Shroff <nilay at linux.ibm.com>
>> Closes: https://lore.kernel.org/linuxppc-dev/6af2c4c2-97f6-4758-be33-256638ef39e5@linux.ibm.com/
>> Fixes: daaa574aba6f ("powerpc/pseries/msi: Switch to msi_create_parent_irq_domain()")
>> Signed-off-by: Nam Cao <namcao at linutronix.de>
>> Acked-by: Nilay Shroff <nilay at linux.ibm.com>
>> Cc: stable at vger.kernel.org
>> ---
>> v2:
>>    - change pseries_msi_ops_prepare()'s allocation logic to match the
>>      original logic in __pci_enable_msix_range()
>>
>>    - fix up Nilay's email address
>> ---
>>   arch/powerpc/platforms/pseries/msi.c | 44 ++++++++++++++++++++++++++--
>>   1 file changed, 41 insertions(+), 3 deletions(-)
>>
>> diff --git a/arch/powerpc/platforms/pseries/msi.c b/arch/powerpc/platforms/pseries/msi.c
>> index a82aaa786e9e..edc30cda5dbc 100644
>> --- a/arch/powerpc/platforms/pseries/msi.c
>> +++ b/arch/powerpc/platforms/pseries/msi.c
>> @@ -19,6 +19,11 @@
>>   
>>   #include "pseries.h"
>>   
>> +struct pseries_msi_device {
>> +	unsigned int msi_quota;
>> +	unsigned int msi_used;
>> +};
>> +
>>   static int query_token, change_token;
>>   
>>   #define RTAS_QUERY_FN		0
>> @@ -433,8 +438,28 @@ static int pseries_msi_ops_prepare(struct irq_domain *domain, struct device *dev
>>   	struct msi_domain_info *info = domain->host_data;
>>   	struct pci_dev *pdev = to_pci_dev(dev);
>>   	int type = (info->flags & MSI_FLAG_PCI_MSIX) ? PCI_CAP_ID_MSIX : PCI_CAP_ID_MSI;
>> +	int ret;
>> +
>> +	struct pseries_msi_device *pseries_dev __free(kfree)
>> +		= kmalloc(sizeof(*pseries_dev), GFP_KERNEL);
>> +	if (!pseries_dev)
>> +		return -ENOMEM;
>> +
>> +	while (1) {
>> +		ret = rtas_prepare_msi_irqs(pdev, nvec, type, arg);
>> +		if (!ret)
>> +			break;
>> +		else if (ret > 0)
>> +			nvec = ret;
>> +		else
>> +			return ret;
>> +	}
>>   
>> -	return rtas_prepare_msi_irqs(pdev, nvec, type, arg);
>> +	pseries_dev->msi_quota = nvec;
>> +	pseries_dev->msi_used = 0;
>> +
>> +	arg->scratchpad[0].ptr = no_free_ptr(pseries_dev);
>> +	return 0;
>>   }
>>   
>>   /*
>> @@ -443,9 +468,13 @@ static int pseries_msi_ops_prepare(struct irq_domain *domain, struct device *dev
>>    */
>>   static void pseries_msi_ops_teardown(struct irq_domain *domain, msi_alloc_info_t *arg)
>>   {
>> +	struct pseries_msi_device *pseries_dev = arg->scratchpad[0].ptr;
>>   	struct pci_dev *pdev = to_pci_dev(domain->dev);
>>   
>>   	rtas_disable_msi(pdev);
>> +
>> +	WARN_ON(pseries_dev->msi_used);
>> +	kfree(pseries_dev);
>>   }
>>   
>>   static void pseries_msi_shutdown(struct irq_data *d)
>> @@ -546,12 +575,18 @@ static int pseries_irq_domain_alloc(struct irq_domain *domain, unsigned int virq
>>   				    unsigned int nr_irqs, void *arg)
>>   {
>>   	struct pci_controller *phb = domain->host_data;
>> +	struct pseries_msi_device *pseries_dev;
>>   	msi_alloc_info_t *info = arg;
>>   	struct msi_desc *desc = info->desc;
>>   	struct pci_dev *pdev = msi_desc_to_pci_dev(desc);
>>   	int hwirq;
>>   	int i, ret;
>>   
>> +	pseries_dev = info->scratchpad[0].ptr;
>> +
>> +	if (pseries_dev->msi_used + nr_irqs > pseries_dev->msi_quota)
>> +		return -ENOSPC;
>> +
>>   	hwirq = rtas_query_irq_number(pci_get_pdn(pdev), desc->msi_index);
>>   	if (hwirq < 0) {
>>   		dev_err(&pdev->dev, "Failed to query HW IRQ: %d\n", hwirq);
>> @@ -567,9 +602,10 @@ static int pseries_irq_domain_alloc(struct irq_domain *domain, unsigned int virq
>>   			goto out;
>>   
>>   		irq_domain_set_hwirq_and_chip(domain, virq + i, hwirq + i,
>> -					      &pseries_msi_irq_chip, domain->host_data);
>> +					      &pseries_msi_irq_chip, pseries_dev);
>>   	}
>>   
>> +	pseries_dev->msi_used++;
>>   	return 0;
>>   
>>   out:
>> @@ -582,9 +618,11 @@ static void pseries_irq_domain_free(struct irq_domain *domain, unsigned int virq
>>   				    unsigned int nr_irqs)
>>   {
>>   	struct irq_data *d = irq_domain_get_irq_data(domain, virq);
>> -	struct pci_controller *phb = irq_data_get_irq_chip_data(d);
>> +	struct pseries_msi_device *pseries_dev = irq_data_get_irq_chip_data(d);
>> +	struct pci_controller *phb = domain->host_data;
>>   
>>   	pr_debug("%s bridge %pOF %d #%d\n", __func__, phb->dn, virq, nr_irqs);
>> +	pseries_dev->msi_used -= nr_irqs;
>>   	irq_domain_free_irqs_parent(domain, virq, nr_irqs);
>>   }
>>   
> I just tested this change on my system using the latest mainline kernel and it works
> well for me. So with that, please fell free to add,
>
> Tested-by: Nilay Shroff <nilay at linux.ibm.com>

Thanks will pull this in.
Maddy

>