Hi Marcel, Unfortunately, I don't have a possibility to test 5.5 kernel. As far as I can see, the most recent version available to the public via git.kernel.org is 5.3-rc3 and the problematic part in /net/bluetooth/l2cap_core.c file is the same for 4.20 and 5.3-rc3. Hence I suppose it will behave the same way as 4.20. Best regards, Konstantin On Wed, 2020-01-29 at 04:56 +0100, Marcel Holtmann wrote: > Hi Konstantin, > > > It seems that there's a bug in Bluetooth stack of 4.20 kernel. I > > discovered it during Bluetooth qualification tests. > > L2CAP/COS/CFD/BV- > > 12-C [Unknown Option Response] test fails because instead of > > "Configuration Response" with error code 0x0003 "unknown option" > > the > > Bluetooth stack generates "Command Reject". > > > > I think, this happens because the tester (Bluetooth SIG PTS) uses > > hints > > as unknown options, and the Bluetooth stack has special handling > > for > > hints. > > > > I made a small patch in order to overcome this problem. Please > > confirm, > > that the patch is feasible. If not, please suggest how to fix the > > "Command Reject" problem. > > is the limited to 4.20 kernel or does it also happen with the latest > 5.5 kernel? > > Regards > > Marcel > > > > >
