Bug 206567

Summary:	Patch for Samsung Galaxy book Atheros QCA 6174
Product:	Networking	Reporter:	Aurélien C (aurelien.k.ernel)
Component:	Wireless	Assignee:	networking_wireless (networking_wireless)
Status:	NEEDINFO ---
Severity:	high	CC:	aurelien.k.ernel, briannorris, hartrumpf
Priority:	P1
Hardware:	All
OS:	Linux
Kernel Version:	5.5.4	Subsystem:
Regression:	No	Bisected commit-id:
Attachments:	regd_common.h patch

Description Aurélien C 2020-02-17 12:03:55 UTC

Hi, 

The Samsung Galaxy book wireless card Atheros QCA 6174 is not usable on standard kernel since the driver doesn't know the 0x5F reg domain.
A patch is available since 2 years and solve the problem (see patched attached to this bug). More information can be found here:  https://bugzilla.redhat.com/show_bug.cgi?id=1615436

Would it be possible to integrate that patch to the kernel source in order to get this card working on samsung devices?
Thank you in advance.

Comment 1 Aurélien C 2020-02-17 12:04:26 UTC

Created attachment 287437 [details]
regd_common.h patch

Comment 2 hartrumpf 2021-02-06 16:38:55 UTC

I know several owners of these laptops that are frustrated by this bug for a long time, esp. because the tiny bug fix is known for a long time and reported as fine.

Comment 3 Brian Norris 2021-02-08 19:12:49 UTC

bugzilla is not really the place for submitting patches, nor expecting responses from maintainers. The documentation for submitting patches for drivers/net/wireless/ is here:

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches

Bugzilla can be OK for documenting the problem, so you have a good link when sending a formal patch, but unfortunately, bugzilla does not get regular attention.

And even if you can't go through submitting the patch yourself, can we get a proper attribution and Signed-off-by from the patch author? With that, an enterprising individual could submit such a patch on your behalf.

Also, for the patch:

> -       {CTRY_KOREA_ROC, APL9_WORLD, "KR"},
> +    {CTRY_KOREA_ROC, APL10_WORLD, "KR"},

This part feels wrong to me, but I'm not that familiar with ath regd handling. It seems like you might need a new entry, rather than replacing one. But I'd leave that up to code review on the linux-wireless mailing list, I suppose.

Marking NEEDINFO, because the patch can't go anywhere without providing the appropriate Signed-off-by and From lines, at a minimum.