[PATCH net-next v7 2/2] net: ti: icssg-prueth: Add support for PA Stats

[Jan Kiszka]

On 02.09.24 09:28, Anwar, Md Danish wrote:
> 
> 
> On 9/2/2024 12:52 PM, Jan Kiszka wrote:
>> On 22.08.24 14:26, MD Danish Anwar wrote:
>>> Add support for dumping PA stats registers via ethtool.
>>> Firmware maintained stats are stored at PA Stats registers.
>>> Also modify emac_get_strings() API to use ethtool_puts().
>>>
>>> This commit also maintains consistency between miig_stats and pa_stats by
>>> - renaming the array icssg_all_stats to icssg_all_miig_stats
>>> - renaming the structure icssg_stats to icssg_miig_stats
>>> - renaming ICSSG_STATS() to ICSSG_MIIG_STATS()
>>> - changing order of stats related data structures and arrays so that data
>>>   structures of a certain stats type is clubbed together.
>>>
>>> Signed-off-by: MD Danish Anwar <danishanwar at ti.com>
>>> ---
>>>  drivers/net/ethernet/ti/icssg/icssg_ethtool.c |  19 ++-
>>>  drivers/net/ethernet/ti/icssg/icssg_prueth.c  |   6 +
>>>  drivers/net/ethernet/ti/icssg/icssg_prueth.h  |   9 +-
>>>  drivers/net/ethernet/ti/icssg/icssg_stats.c   |  31 +++-
>>>  drivers/net/ethernet/ti/icssg/icssg_stats.h   | 158 +++++++++++-------
>>>  5 files changed, 140 insertions(+), 83 deletions(-)
>>>
>>
>> ...
>>
>>> diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
>>> index 53a3e44b99a2..f623a0f603fc 100644
>>> --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c
>>> +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
>>> @@ -1182,6 +1182,12 @@ static int prueth_probe(struct platform_device *pdev)
>>>  		return -ENODEV;
>>>  	}
>>>  
>>> +	prueth->pa_stats = syscon_regmap_lookup_by_phandle(np, "ti,pa-stats");
>>> +	if (IS_ERR(prueth->pa_stats)) {
>>> +		dev_err(dev, "couldn't get ti,pa-stats syscon regmap\n");
>>> +		return -ENODEV;
>>
>> I was just beaten for potentially not being backward compatible, but
>> this is definitely not working with existing DTs, just ran into it.
>>
> 
> Jan, I had posted the DT patch needed for this [1]. But the DT patch
> goes to different tree and it was dependent on binding patch of this
> series. My intention was once the binding is in, I will post the DT
> patch in the next window. Till then ICSSG driver will break here but
> once DT gets merged it will be alright. If that's not the best solution,
> I can post DT here, but it will need to get merged via net-next.
> 
> [1] https://lore.kernel.org/all/20240729113226.2905928-4-danishanwar@ti.com/
> 

Why not respect in the driver that the dtbinding says this property is
only optional?

Jan

-- 
Siemens AG, Technology
Linux Expert Center