Re: PCI/kernel msi code vs GIC ITS driver conflict?

[John Garry <john.garry@huawei.com>]

On 03/09/2019 17:16, Marc Zyngier wrote:
> Hi John,
>
> On 03/09/2019 15:09, John Garry wrote:
>> Hi Marc, Bjorn, Thomas,

Hi Marc,

>>
>> We've come across a conflict with the kernel/pci msi code and GIC ITS
>> driver on our arm64 system, whereby we can't unbind and re-bind a PCI
>> device driver under special conditions. I'll explain...
>>
>> Our PCI device support 32 MSIs. The driver attempts to allocate msi
>> vectors with min msi=17, max msi = 32, and affd.pre vectors = 16. For
>> our test we make nr_cpus = 1 (just anything less than 16).
>
> Just to confirm: this PCI device is requiring Multi-MSI, right? As
> opposed to MSI-X?

Right, Multi-MSI.

>
>> We find that the pci/kernel msi code gives us 17 vectors, but the GIC
>> ITS code reserves 32 lpi maps in its_irq_domain_alloc(). The problem
>> then occurs when unbinding the driver in its_irq_domain_free() call,
>> where we only clear bits for 17 vectors. So if we unbind the driver and
>> then attempt to bind again, it fails.
>
> Is this device, by any chance, sharing its requested-id with another
> device? By being behind a bridge of some sort?There is some code to
> deal with it, but I'm not sure it has ever been verified in anger...

It's a RC iEP and there should be no requested-id sharing:

root@ubuntu:/home/john#  lspci -s 74:02.0 -v
74:02.0 Serial Attached SCSI controller: Huawei Technologies Co., Ltd. 
HiSilicon SAS 3.0 HBA (rev 20)
Flags: bus master, fast devsel, latency 0, IRQ 23, NUMA node 0
Memory at a2000000 (32-bit, non-prefetchable) [size=32K]
Capabilities: [40] Express Root Complex Integrated Endpoint, MSI 00
Capabilities: [80] MSI: Enable+ Count=32/32 Maskable+ 64bit+
Capabilities: [b0] Power Management version 3
Kernel driver in use: hisi_sas_v3_hw

>
>> Where the fault lies, I can't say. Maybe the kernel msi code should
>> always give power of 2 vectors - as I understand, the PCI spec mandates
>> this. Or maybe the GIC ITS driver has a problem in the free path, as
>> above. Or maybe the PCI driver should not be allowed to request !power
>> of 2 min/max vectors.
>>
>> Opinion?
>
> My hunch is that it is an ITS driver bug: the PCI layer is allowed to
> give any number of MSIs to an endpoint driver, as long as they match the
> requirements of the allocation for Multi-MSI.

I would tend to say that, but isn't the requirement to allocate power of 
2 msi vectors, which doesn't seem to be enforced in the kernel msi layer?

  That's the responsibility
> of the ITS driver. If unbind/bind fails, it means that somehow we've
> missed the freeing of the LPIs, which isn't good.
>
> Is the device common enough that I can try and reproduce the issue?

No, it's integrated into the hi1620 SoC found in the D06 dev board only, 
but I don't think that there is anything special about this HW.

If
> there's a Linux driver somewhere, I can always hack something in
> emulation and find out...

Ok, the interrupt allocation for this particular driver in this test is 
in 
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c#n2393

Cheers,
John

>
> Thanks,
>
> 	M.
>