Re: [PATCH v2 4/6] fpga: dfl: add generic support for MSIX interrupts

[matthew.gerlach@linux.intel.com]

[-- Attachment #1: Type: text/plain, Size: 3887 bytes --]



On Fri, 23 Sep 2022, Ilpo Järvinen wrote:

> On Fri, 23 Sep 2022, matthew.gerlach@linux.intel.com wrote:
>
>> From: Matthew Gerlach <matthew.gerlach@linux.intel.com>
>>
>> Define and use a DFHv1 parameter to add generic support for MSIX
>> interrupts for DFL devices.
>>
>> Signed-off-by: Matthew Gerlach <matthew.gerlach@linux.intel.com>
>> ---
>> v2: fix kernel doc
>>     clarify use of DFH_VERSION field
>> ---
>>  drivers/fpga/dfl.c  | 60 +++++++++++++++++++++++++++++++++++++++++----
>>  include/linux/dfl.h | 14 +++++++++++
>>  2 files changed, 69 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/fpga/dfl.c b/drivers/fpga/dfl.c
>> index 1132f3c10440..dfd3f563c92d 100644
>> --- a/drivers/fpga/dfl.c
>> +++ b/drivers/fpga/dfl.c
>> @@ -941,23 +941,22 @@ static int parse_feature_irqs(struct build_feature_devs_info *binfo,
>>  	void __iomem *base = binfo->ioaddr + ofst;
>>  	unsigned int i, ibase, inr = 0;
>>  	enum dfl_id_type type;
>> -	int virq;
>> +	int virq, off;
>>  	u64 v;
>>
>>  	type = feature_dev_id_type(binfo->feature_dev);
>>
>>  	/*
>>  	 * Ideally DFL framework should only read info from DFL header, but
>> -	 * current version DFL only provides mmio resources information for
>> +	 * current version, DFHv0, only provides mmio resources information for
>>  	 * each feature in DFL Header, no field for interrupt resources.
>>  	 * Interrupt resource information is provided by specific mmio
>>  	 * registers of each private feature which supports interrupt. So in
>>  	 * order to parse and assign irq resources, DFL framework has to look
>>  	 * into specific capability registers of these private features.
>>  	 *
>> -	 * Once future DFL version supports generic interrupt resource
>> -	 * information in common DFL headers, the generic interrupt parsing
>> -	 * code will be added. But in order to be compatible to old version
>> +	 * DFHv1 supports generic interrupt resource information in DFHv1
>> +	 * parameter blocks. But in order to be compatible to old version
>>  	 * DFL, the driver may still fall back to these quirks.
>>  	 */
>>  	if (type == PORT_ID) {
>> @@ -981,6 +980,36 @@ static int parse_feature_irqs(struct build_feature_devs_info *binfo,
>>  		}
>>  	}
>>
>> +	if (fid != FEATURE_ID_AFU && fid != PORT_FEATURE_ID_ERROR &&
>> +	    fid != PORT_FEATURE_ID_UINT && fid != FME_FEATURE_ID_GLOBAL_ERR) {
>> +
>> +		v = FIELD_GET(DFH_VERSION, readq(base));
>
> I'd call this variable version (or ver) if you want to store it but it
> would also fit to switch () line so that no extra variable is needed.

I will change the v to dfh_ver to be clearer.  I want to store the 
value because it is used in the default case in the error message.  The 
error message helps to debug broken FPGA images.

>
>> +		switch (v) {
>> +		case 0:
>> +			break;
>> +
>> +		case 1:
>> +			v =  readq(base + DFHv1_CSR_SIZE_GRP);
>
> Extra space.

I will remove extra space.

>
>> +			if (FIELD_GET(DFHv1_CSR_SIZE_GRP_HAS_PARAMS, v)) {
>> +				off = dfl_find_param(base + DFHv1_PARAM_HDR, ofst,
>> +						     DFHv1_PARAM_ID_MSIX);
>> +				if (off >= 0) {
>
> I'd reverse these 2 conditions and break when there's nothing to do.

I'm not sure what you mean by reversing these conditions because a DFHv1 
may or may not have parameters (the first condition), and a DFHv1 may have
parameters but may not have a MSI-X parameter (the second condition).

>
>> +					ibase = readl(base + DFHv1_PARAM_HDR +
>> +						      off + DFHv1_PARAM_MSIX_STARTV);
>> +					inr = readl(base + DFHv1_PARAM_HDR +
>> +						    off + DFHv1_PARAM_MSIX_NUMV);
>> +					dev_dbg(binfo->dev, "start %d num %d fid 0x%x\n",
>> +						ibase, inr, fid);
>> +				}
>> +			}
>> +			break;
>> +
>> +		default:
>> +			dev_warn(binfo->dev, "unexpected DFH version %lld\n", v);
>> +			break;
>> +		}
>> +	}
>> +
>>  	if (!inr) {
>>  		*irq_base = 0;
>>  		*nr_irqs = 0;
>
> -- 
> i.
>
>