On Fri, Dec 13, 2024 at 07:56:28PM +0200, Jussi Laakkonen wrote: > This RFC patchset contains changes to address the issue with WLAN > drivers, which may not stop the initiated scheduled scans, or to provide > results on them. This has been noticed to happen with sprd-wlan driver on > MT6762 chipset. Unfortunately I'm not very familiar with how various drivers and vendors have implemented scheduled scanning and the related nl80211 commands/events. Has this issue been discussed with the particular vendor referenced here? If not, it would be good to do so and try to see whether the driver could be fixed to follow what the nl80211 interface is documented to do for scheduled scans. > In the internal testing this has shown to solve the issue, but without > the detailed knowledge whether this is a correct approach or not, the > changes should receive more comments on how to do this properly. On well > behaving drivers this does not seem to be an issue, so might these > changes be better serving their purpose as an option configurable on > build time instead of being like presented here. It is interested that the option of fixing the driver is not even considered here.. That should be the first goal. If that can be clearly shown to be impossible, some workarounds could be considered, but I do not think extra complexity in user space to avoid issues with misbehaving drivers is a good approach and that should be avoiding. -- Jouni Malinen PGP id EFC895FA _______________________________________________ Hostap mailing list Hostap@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/hostap
