On Mon, Nov 18, 2019 at 7:29 PM Andrew Donnellan <ajd(a)linux.ibm.com> wrote:
On 19/11/19 1:48 pm, Alastair D'Silva wrote:
> On Tue, 2019-11-19 at 10:47 +1100, Andrew Donnellan wrote:
>> On 15/11/19 3:35 am, Dan Williams wrote:
>>>> Have you discussed with the directory owner if it's ok to split
>>>> driver over several files?
>>> My thought is to establish drivers/opencapi/ and move this and the
>>> existing drivers/misc/ocxl/ bits there.
>> Is there any other justification for this we can think of apart from
>> wanting to put this driver in the nvdimm directory? OpenCAPI drivers
>> aren't really a category of driver unto themselves.
> There is a precedent for bus-based dirs, eg. drivers/(ide|w1|spi) all
> contain drivers for both controllers & connected devices.
> Fred, how do you feel about moving the generic OpenCAPI driver out of
Instinctively I don't like the idea of creating a whole opencapi
directory, as OpenCAPI is a generic bus which is not tightly coupled to
any particular application area, and drivers for other OpenCAPI devices
are already spread throughout the tree (e.g. cxlflash in drivers/scsi).
I'm not suggesting all opencapi drivers go there, nor the entirety of
this driver, just common infrastructure. That said, it's hard to talk
about specifics given the current state of the patch set. I have not
even taken a deeper look past the changelog as this 3K lines-of-code
submission needs to be broken up into smaller pieces before we settle
on what pieces belong where.
Just looking at the diffstat, at a minimum it's not appropriate for
them to live in drivers/nvdimm/ directly, drivers/nvdimm/oxcl/ would
be an acceptable starting point.