On Tue, 2023-08-29 at 09:21 -0700, Jeff Johnson wrote:
> Just want to remind that there were two proposals given to allow setting
> this from upstream:
> <https://lore.kernel.org/all/[email protected]/>
>
> In the thread you asked for additional information:
> <https://lore.kernel.org/all/[email protected]/>
Yes, I know. I figured it was long enough ago :P
> So to see if there is any update on that request Cc: Xinyue Ling
> <[email protected]>
Sure.
> Would hate to see this ripped out just to need to replace it later.
I actually wouldn't mind that much. I have a gut feeling that at least
the mac80211 part will actually not be of any use to Qualcomm since most
of the things I see these days are for non-upstream drivers anyway.
Therefore, it would actually make sense to understand how this could be
done properly _without_ supporting it in mac80211, because it's clear
that just adding support for it as proposed in those patches will break
in mac80211 pretty badly.
If we even decide to go that route, because honestly, I'm getting pretty
upset about the whole Qualcomm adding various APIs that we never see any
users for, then not maintaining it, _and_ complaining that we don't fix
things upstream quickly enough.
johannes
On 8/29/2023 9:24 AM, Johannes Berg wrote:
> If we even decide to go that route, because honestly, I'm getting pretty
> upset about the whole Qualcomm adding various APIs that we never see any
> users for, then not maintaining it, _and_ complaining that we don't fix
> things upstream quickly enough.
Part of my new role within Qualcomm is to address this.