Re: [RFC] ipw2100 ipw2200 ps3_gelic rtl8712 --- Are we ready for wext cleanup?

[Kalle Valo]

Philipp Hortmann <philipp.g.hortmann@xxxxxxxxx> writes:

> origin of this question was the following patch series from Arnd Bergmann
> [PATCH 00/10] Remove obsolete and orphaned wifi drivers
> https://lore.kernel.org/linux-staging/20231023131953.2876682-1-arnd@xxxxxxxxxx/
>
> Here the remaining files that use iw_handler_def:
> drivers/net/ethernet/toshiba/ps3_gelic_wireless.c:static const struct
> iw_handler_def gelic_wl_wext_handler_def = {
> drivers/net/wireless/intel/ipw2x00/ipw2100.c:static const struct
> iw_handler_def ipw2100_wx_handler_def;
> drivers/net/wireless/intel/ipw2x00/ipw2100.c:static const struct
> iw_handler_def ipw2100_wx_handler_def = {
> drivers/net/wireless/intel/ipw2x00/ipw2200.c:static const struct
> iw_handler_def ipw_wx_handler_def = {
> drivers/staging/rtl8712/os_intfs.c:     pnetdev->wireless_handlers =
> (struct iw_handler_def *)
> drivers/staging/rtl8712/rtl871x_ioctl.h:extern struct iw_handler_def
> r871x_handlers_def;
> drivers/staging/rtl8712/rtl871x_ioctl_linux.c:struct iw_handler_def
> r871x_handlers_def = {
>
>
> In this Email Greg writes over rtl8192e:
> https://lore.kernel.org/linux-staging/2024100810-payback-suds-8c15@gregkh/
> "...
> No staging driver should ever get in the way of api changes elsewhere in
> the kernel, that's one of the rules of this part of the tree.  So from
> my opinion, it's fine to delete it now.  It can always come back in a
> new way later on.
> ..."
>
> So it should not be an issue to remove rtl8712.
>
> Stefan Lippers-Hollmann was one year ago still using the ipw2200.
> https://lore.kernel.org/linux-staging/20231024014302.0a0b79b0@mir/
>
> Here my opinion why I think we should reconsider this:
>
> I really like to use old hardware. One of my computers is from trash
> and the other one is bought for 50€ three years ago. But non of my
> hardware is from before 2012. Do we as a community really need to
> support hardware from 2003 in kernel 6.13 for WLAN that evolved so
> rapidly? I do not think so.
>
> People around me are complaining that the 2,4GHz WLAN is difficult to
> use because so many devices are using it. Such slow devices consume a
> lot of time to send and receive the data and block therefore other
> devices.
>
> The longterm kernels will still support this hardware for years.
>
> Please explain to our very high value resources (Maintainers,
> Developers with wext and mac80211 expierience) that you cannot find
> any other solution that is within technical possibility and budget
> (USB WLAN Stick or exchange of WLAN module) and that they need to
> invest their time for maintenance.
> Here the example of invested time from Johannes Berg:
> https://lore.kernel.org/all/20241007213525.8b2d52b60531.I6a27aaf30bded9a0977f07f47fba2bd31a3b3330@changeid/
>
> I cannot ask the Linux kernel community to support my test hardware
> just because I bought it some time ago. Rather, I have to show that I
> use it for private or business purposes on a regular basis and that I
> cannot easily change.
>
> Using this hardware is security wise not state of the art as WPA3 is
> not supported. We put so much effort into security. Why not here?

I didn't quite get what you are saying here, are you proposing that we
should remove ancient drivers faster?

-- 
https://patchwork.kernel.org/project/linux-wireless/list/

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