On Tue, Jul 13, 2021 at 1:36 AM Bjorn Helgaas [off-list ref] wrote:

On Mon, Jul 12, 2021 at 02:48:12PM +0300, Andy Shevchenko wrote:

quoted

On Thu, Jul 08, 2021 at 04:47:06PM -0500, Bjorn Helgaas wrote:

quoted

On Thu, Apr 02, 2020 at 11:23:27PM +0300, Andy Shevchenko wrote:

...

quoted

Datasheets are publicly available (at least one may google and find some
information about those PCH chips). I have in possession the hardware for
gpio-pch. I can easily test that part at least.

If you have a URL for those datasheets, can you share it?  I spent
some time looking but all I found was 1-2 page marketing brochures.

It's a part of the so called EG20T PCH. It's part of in particular
Intel Galileo (Quark SoC) and Intel Minnowboard (v1) (Atom E6xx SoC).
Hence the easily found links:

http://minnowboard.outof.biz/MinnowBoard.html
https://www.elinux.org/Minnowboard:Minnow_Original
https://en.wikipedia.org/wiki/List_of_Intel_Atom_microprocessors

https://www.intel.com/content/www/us/en/embedded/products/quark/x1000/documentation.html?grouping=EMT_Content%20Type&sort=title:asc
(Chapter 19)

https://ark.intel.com/content/www/us/en/ark/products/codename/37567/products-formerly-tunnel-creek.html

Hmm... Funny, the document #324211 can't be downloaded
https://download.intel.com/embedded/chipsets/datasheet/324211.pdf

I guess you may ping Intel and tell them that they should play nice
when talking about "open hardware" (MinnowBoard initiative).
Nevertheless, the (Old? #457798 is a specification update under NDA.
Okay, it refers to rev 8, while Mouser, see below, provides rev 9)
copy is available on other sites, such as

https://www.mouser.tw/pdfdocs/Intel_Platform_Controller_Hub_EG20T_datasheet.pdf
(Chapter 16)

quoted

quoted

that's not a trivial task, and I don't think that burden should fall
on anyone who wants to make any improvements to these drivers.

quoted

Another alternative would be to remove legacy PCI PM usage
(ioh_gpio_suspend() and ioh_gpio_resume()) from gpio-ml-ioh.  That
would mean gpio-ml-ioh wouldn't support power management at all, which
isn't a good thing, but maybe it would be even more motivation to
unify it with gpio-pch (which has already been converted by
226e6b866d74 ("gpio: pch: Convert to dev_pm_ops"))?

With regard to (1) probably we may exceptionally accept the fix to
gpio-ml-ioh, but I really prefer to do the much more _useful_ job on
it by unifying the two.

Should Vaibhav re-post this patch, or do you want to pull it from the
archives?  I just checked and it still applies cleanly to v5.14-rc1.

Here it is for reference:
  https://lore.kernel.org/lkml/20200402155057.30667-1-vaibhavgupta40@gmail.com/ (local)

I'll take from the archives.

I'll post a couple small patches toward unifying them.  They don't do
the whole job but they're baby steps.

Thanks! I look forward to seeing them soon!

-- 
With Best Regards,
Andy Shevchenko