<yhchuang@xxxxxxxxxxx> wrote: > From: Zong-Zhe Yang <kevin_yang@xxxxxxxxxxx> > > Add a set of logic with corresponding coexistence parameters to > handle the situation under BT inquiry/page. > > We will set PSTDMA while WL-Busy + BT inquiry/page to separate > WL/BT slots. PSTDMA can protect WL data rate and BT performance. > > If WL-Busy + BT inquiry/page and there was BT device paired, > We will set the mechanism to 4Slot PSTDMA. > In 4Slot PSTDMA, the paired devices can perform more smoothly > and prevent some issues trigger from insufficient data. > > And to avoid A2DP glitch or disconnection, we will adjust ACL > data priority higher than inquiry/page. > > In addition, we found sometimes BT inquiry/page still working > last for seconds after BT had notified inquiry/page finished. > It will lead to A2DP glitch cause of ACL data, inquiry/page > priority toggled. To fix the corner, we add a timer to remain > the inquiry/page status. > > And we found WL busy/idle threshold is too sensitive, > it will keep switching in some weak network environment and > coexistence mechanism will switch between TDMA and PSTDMA. > > The very frequently switching may destroyed not only the > handshake with AP, but BT performance. And it will trigger > some unexpected error. > > To prevent the frequently switching, we add a timer to delay > the status change while WL busy switch to idle. > > Signed-off-by: Zong-Zhe Yang <kevin_yang@xxxxxxxxxxx> > Signed-off-by: Yan-Hsuan Chuang <yhchuang@xxxxxxxxxxx> Patch applied to wireless-drivers-next.git, thanks. 001a3c902f6f rtw88: coex: Fix ACL Tx pause during BT inquiry/page. -- https://patchwork.kernel.org/patch/11664065/ https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
