Thanks everyone. Pushed upstream today. Sorry about the delay in pushing it.

Thanks everyone for accepting the changes.

@vmaffione, @gallatin and @yuripv,

Addressed comments on V2 patch

Addressed review comments on v1 patch

@vmaffione, I have used igor and mandoc to ensure no warnings before submitting the patch. I have informed the same in Test plan section

@vmaffione and @gallatin

Added PGP public key along with other previous changes

@vmaffione, Do you have any comments?

Ok. I will create another review. Thanks for your time and immediate response.

Thanks @mav for reviewing and pushing the patch upstream. I verified it to work properly. I have some more minor changes to this Driver (basically to add support for another PCI Device ID). Shall I submit the additional patch here itself? or should I open another review for the same?

Is there anything else needed here? or can this be taken upstream?

@mav, Did you get a chance to look at this?

@mav: Sorry to disturb you on this again. But, please have a look at this when you get a chance and let me know if there any comments to address here.

@mav, Any more comments on this patch?

Made necessary changes as per the previous discussion. Attached V6 patch.

@mav: I checked about that alignment requirements. Seems, the alignment should be on-par to the bar size (and not a fixed value) and the granularity range what I mentioned earlier for LUT based addressing (which is not implemented here). So, I need to change the alignment equal to the bar size. Do you have any comments on the previous discussed changes?

No new patch attached, but looking for few clarification on your previous comments.

Addressed comment from @mav regarding the alignment and explained the reason for those checks for xlat and limit registers.

Any more comments on this patch?

Submitting V4 patch, addressing comment from @imp

In D18774#403351, @imp wrote:

I just made a pass through for anything that caught my eye, and found a couple. I have not reviewed it to see if it matches the hardware spec for what it's trying to control.

Addressed commented from @imp and @mav and submitted V3 patch. Also, added amd_ntb_mw_set_wc routine.

V2 patch addressing the previous comments from cem.

Hi cem,

Thanks Macy.

Hi Macy,

Hi Macy/Warner,

make tinderbox reported some failure with arm architecture. Fixed that.

Hi Oleksandr,

In D16865#371516, @gonzo wrote:

Sorry for delays with answers

To be honest, the current code freeze seems quite... flexible. As in, there's certainly some new stuff coming in, and from my understanding adding a new driver (as opposed to modifying something that's already there) is rather unlikely to break something. Maybe just send an email to re@ and see what they say?

Yes, and IMHO this flexibility is not a very good thing (speaking from personal experience as a part-time RE at $WORK). I don't want to submit more new feature changes to re@ that is strictly necessary and so far I can't find pro arguments for this driver. As far as I understand this driver is going to be used as a building block for other components that are going to use it either as a control method for peripherals or as an interrupt source. If I am wrong and there are readily available immediate use cases, for instance, AMD-based embedded platforms where GPIO can be used from the userland or AMD-based laptops where gpioctl can be used directly to enable certain functionality that is otherwise unavailable I'll bring the patch up to re@ for the consideration. Otherwise, I'd rather wait for the code thaw and submit it to -HEAD with further MFCs to -12 and -11, it will be available in 12.1 in a year or so.

Rajesh, if you're familiar with the solutions available on the market could you share such use cases so I could argue for the inclusion of the driver?

Thanks

Hi Oleksandr,

Hi Oleksandr,

Hi Olexsandr,

1. Added the static keywords to those mentioned structures in H file, as the intent is code readability
2. Cleared the unwanted whitespaces/tabs and followed style(9) format for indentation in both C and H file.

Hi Oleksandr,

Thanks for taking time to review the first patch and clarifying on the release process.

Hi Gonzo,

1. Removed static keywords from pin grouping definitions in H file.
2. Moved static struct "resource_spec" to C file.
3. Removed unwanted static declaration of probe, attach and detach in H file.
4. Removed redundant file in C file

Hi Gonzo,

Thank you Gonzo.

Hi Gonzo, Did you get a chance to look at this?