Re: [PATCH v2] cxl/port: Fix endpoint RAS error unmasking

[Dave Jiang <dave.jiang@intel.com>]

On 11/3/26 11:35 AM, Cheatham, Benjamin wrote:
> On 11/3/2026 12:34 AM, Richard Cheng wrote:
>> Moving endpoint RAS handling from cxl_pci to cxl_port left
>> cxl_ras_unmask() checking whether port->uport_dev is a PCI device.
>> Endpoint ports always use a cxl_memdev as uport_dev, so this check
>> returns early and leaves the CXL RAS error masks unchanged.
>>
>> Resolve the PCI device directly through the memdev's parent. Rename
>> helper to cxl_endpoint_ras_unmask() and its param to endpoint to make
>> its endpoint-only scope explicit.
>>
>> Fixes: dab7162d0ae7 ("cxl/port: Move endpoint component register management to cxl_port")
>> Signed-off-by: Richard Cheng <icheng@nvidia.com>
>> ---
> 
> ...
> 
>> -static int cxl_ras_unmask(struct cxl_port *port)
>> +static int cxl_endpoint_ras_unmask(struct cxl_port *endpoint)
>>  {
>> +	struct cxl_memdev *cxlmd = to_cxl_memdev(endpoint->uport_dev);
>> +	struct device *parent = cxlmd->dev.parent;
> 
> Two small nits:
> 
> 1. Maybe just use endpoint->uport_dev.parent directly and drop the cxlmd part? It's less
> readable, but it's future proofing for when/if CXL.cache devices come along.
> 
> 2. I don't think you need to rename the port parameter if you change the name to endpoint. It's
> self-explanatory at that point and cuts down on the diff.

I asked for that change. Since we are making changes on that line already, may as well make it explicit for readability. Or at least satisfies my OCD :)

DJ

> 
> Both are really up to you on whether you think they're worthwhile, and the change makes sense as-is,
> so regardless of whether you implement the above:
> 
> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com>
> 
> Thanks,
> Ben