Re: [PATCH] PCI: Add pci reset quirk for Huawei Intelligent NIC virtual function

[Alex Williamson <alex.williamson@redhat.com>]

On Thu, 3 Dec 2020 19:29:17 +0800
Chiqijun <chiqijun@huawei.com> wrote:

> On 2020/12/3 1:46, Alex Williamson wrote:
> > On Wed, 2 Dec 2020 17:18:12 +0800
> > Chiqijun <chiqijun@huawei.com> wrote:
> >   
> >> On 2020/11/30 23:46, Alex Williamson wrote:  
> >>> On Sat, 28 Nov 2020 17:29:19 -0600
> >>> Bjorn Helgaas <helgaas@kernel.org> wrote:
> >>>      
> >>>> [+cc Alex]
> >>>>
> >>>> On Sat, Nov 28, 2020 at 02:18:25PM +0800, Chiqijun wrote:  
> >>>>> When multiple VFs do FLR at the same time, the firmware is
> >>>>> processed serially, resulting in some VF FLRs being delayed more
> >>>>> than 100ms, when the virtual machine restarts and the device
> >>>>> driver is loaded, the firmware is doing the corresponding VF
> >>>>> FLR, causing the driver to fail to load.
> >>>>>
> >>>>> To solve this problem, add host and firmware status synchronization
> >>>>> during FLR.  
> >>>>
> >>>> Is this because the Huawei Intelligent NIC isn't following the spec,
> >>>> or is it because Linux isn't correctly waiting for the FLR to
> >>>> complete?  
> >>>
> >>> Seems like a spec compliance issue, I don't recall anything in the spec
> >>> about coordinating FLR between VFs.  
> >>
> >> The spec stipulates that the FLR time of a single VF does not exceed
> >> 100ms, but when multiple VMs are reset concurrently in Linux, there will
> >> be multiple VF parallel FLRs, VF of Huawei Intelligent NIC
> >>    FLR will exceed 100ms in this case.
> >>  
> >>>        
> >>>> If this is a Huawei Intelligent NIC defect, is there documentation
> >>>> somewhere (errata) that you can reference?  Will it be fixed in future
> >>>> designs, so we don't have to add future Device IDs to the quirk?
> >>>>     
> >>>>> Signed-off-by: Chiqijun <chiqijun@huawei.com>
> >>>>> ---
> >>>>>    drivers/pci/quirks.c | 67 ++++++++++++++++++++++++++++++++++++++++++++
> >>>>>    1 file changed, 67 insertions(+)
> >>>>>
> >>>>> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> >>>>> index f70692ac79c5..bd6236ea9064 100644
> >>>>> --- a/drivers/pci/quirks.c
> >>>>> +++ b/drivers/pci/quirks.c
> >>>>> @@ -3912,6 +3912,71 @@ static int delay_250ms_after_flr(struct pci_dev *dev, int probe)
> >>>>>    	return 0;
> >>>>>    }
> >>>>>    
> >>>>> +#define PCI_DEVICE_ID_HINIC_VF  0x375E
> >>>>> +#define HINIC_VF_FLR_TYPE       0x1000
> >>>>> +#define HINIC_VF_OP             0xE80
> >>>>> +#define HINIC_OPERATION_TIMEOUT 15000
> >>>>> +
> >>>>> +/* Device-specific reset method for Huawei Intelligent NIC virtual functions */
> >>>>> +static int reset_hinic_vf_dev(struct pci_dev *pdev, int probe)
> >>>>> +{
> >>>>> +	unsigned long timeout;
> >>>>> +	void __iomem *bar;
> >>>>> +	u16 old_command;
> >>>>> +	u32 val;
> >>>>> +
> >>>>> +	if (probe)
> >>>>> +		return 0;
> >>>>> +
> >>>>> +	bar = pci_iomap(pdev, 0, 0);
> >>>>> +	if (!bar)
> >>>>> +		return -ENOTTY;
> >>>>> +
> >>>>> +	pci_read_config_word(pdev, PCI_COMMAND, &old_command);
> >>>>> +
> >>>>> +	/*
> >>>>> +	 * FLR cap bit bit30, FLR ACK bit: bit18, to avoid big-endian conversion
> >>>>> +	 * the big-endian bit6, bit10 is directly operated here
> >>>>> +	 */
> >>>>> +	val = readl(bar + HINIC_VF_FLR_TYPE);
> >>>>> +	if (!(val & (1UL << 6))) {
> >>>>> +		pci_iounmap(pdev, bar);
> >>>>> +		return -ENOTTY;
> >>>>> +	}  
> >>>
> >>>
> >>> I don't know exactly what this is testing, but it seems like a
> >>> feature/capability test that can fail, why is it not done as part of
> >>> the probe?  Can we define bit 6 with a macro?  Same for bit 10 in the
> >>> VF op register below.  
> >>
> >> The firmware of Huawei Intelligent NIC does not support this feature in
> >> the old version. here is the reading ability to determine whether the
> >> firmware supports it.
> >> In the next patch, I will add a comment here and replace bit 6 and bit
> >> 10 with macro definitions.  
> > 
> > 
> > The question remains why this is not done as part of the probe.  If the
> > device firmware doesn't support it, isn't it better to try a regular
> > FLR and have it return error if the time is exceeded rather than claim
> > we have a functional device specific reset quirk that will always fail
> > without ever attempting to FLR the VF?  Thanks,
> > 
> > Alex
> >   
> 
> The firmware has always supported regular FLR. The regular FLR process 
> waits for 100ms after the FLR is triggered and the FLR is considered to 
> be completed, but the Huawei Intelligent NIC will exceed 100ms when the 
> VF FLR is parallel, so we now need to increase the host to confirm that 
> the firmware completes the FLR processing operation.
> So in the probe stage, we return to support FLR, but there is no place 
> to return whether the firmware supports FLR completion ack capability. 
> We need to add checks during FLR, If the firmware does not support FLR 
> completion ack capability, then return -ENOTTY, the kernel will still 
> execute the regular FLR process.

I see, so we implicitly know the device supports FLR and even though
it's the device specific reset that essentially acks support for a
function level reset, we can still fall through to the base FLR reset
when we're called in the non-probe case.  A bit inconsistent, but OK.
Thanks,

Alex

 
> >>>>> +
> >>>>> +	val = readl(bar + HINIC_VF_OP);
> >>>>> +	val = val | (1UL << 10);
> >>>>> +	writel(val, bar + HINIC_VF_OP);
> >>>>> +
> >>>>> +	/* Perform the actual device function reset */
> >>>>> +	pcie_flr(pdev);
> >>>>> +
> >>>>> +	pci_write_config_word(pdev, PCI_COMMAND,
> >>>>> +			      old_command | PCI_COMMAND_MEMORY);
> >>>>> +
> >>>>> +	/* Waiting for device reset complete */
> >>>>> +	timeout = jiffies + msecs_to_jiffies(HINIC_OPERATION_TIMEOUT);  
> >>>
> >>> Yikes, 15s timeout!  
> >>
> >> Huawei Intelligent NIC supports a maximum of 496 VFs, so the total
> >> timeout period is set to 15s, which will not reach the timeout time
> >> under normal circumstances.
> >>  
> >>>      
> >>>>> +	do {
> >>>>> +		val = readl(bar + HINIC_VF_OP);
> >>>>> +		if (!(val & (1UL << 10)))
> >>>>> +			goto reset_complete;
> >>>>> +		msleep(20);
> >>>>> +	} while (time_before(jiffies, timeout));
> >>>>> +
> >>>>> +	val = readl(bar + HINIC_VF_OP);
> >>>>> +	if (!(val & (1UL << 10)))
> >>>>> +		goto reset_complete;
> >>>>> +
> >>>>> +	pci_warn(pdev, "Reset dev timeout, flr ack reg: %x\n",
> >>>>> +		 be32_to_cpu(val));
> >>>>> +
> >>>>> +reset_complete:
> >>>>> +	pci_write_config_word(pdev, PCI_COMMAND, old_command);
> >>>>> +	pci_iounmap(pdev, bar);
> >>>>> +
> >>>>> +	return 0;
> >>>>> +}
> >>>>> +
> >>>>>    static const struct pci_dev_reset_methods pci_dev_reset_methods[] = {
> >>>>>    	{ PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82599_SFP_VF,
> >>>>>    		 reset_intel_82599_sfp_virtfn },
> >>>>> @@ -3923,6 +3988,8 @@ static const struct pci_dev_reset_methods pci_dev_reset_methods[] = {
> >>>>>    	{ PCI_VENDOR_ID_INTEL, 0x0953, delay_250ms_after_flr },
> >>>>>    	{ PCI_VENDOR_ID_CHELSIO, PCI_ANY_ID,
> >>>>>    		reset_chelsio_generic_dev },
> >>>>> +	{ PCI_VENDOR_ID_HUAWEI, PCI_DEVICE_ID_HINIC_VF,
> >>>>> +		reset_hinic_vf_dev },
> >>>>>    	{ 0 }
> >>>>>    };
> >>>>>    
> >>>>> -- 
> >>>>> 2.17.1
> >>>>>         
> >>>>     
> >>>
> >>> .
> >>>      
> >>  
> > 
> > .
> >   
>