Hi! Iain,
From: Iain Hibbert <plu...@rya-online.net> Date: Mon, 18 Jan 2010 13:43:09 +0000 (GMT) > On Mon, 18 Jan 2010, KIYOHARA Takashi wrote: > > > I am supporting NOKIA DTL-1/4 now. > > (SOCKET Bluetooth CF 8510-00159F and older version) > > Linux supported it by drivers/bluetooth/dtl1_cs.c. > > > > This obtains BDADDR now and does inquiry. > > Hm, does it work with ACL data yet? I think that your > btuart_dtl_output_acl() function strips the ACLDATA header.. I don't > understand how does the device know the connection handle? Only type (1byte) of acldata adjust to me. Perhaps, the problem is not in this. > > However, this should attach/ detach the 4byte header at btuart. > > Is this the only difference, that there is an extra wrapper around every > packet in and out? I looked like that. > > Should we support ioctl for this? Or, do you have a better idea? > > As you may recall, I don't want the driver to be poking inside the netbt > stack structures. The way you have that has catastrophy pending :) > > I think that because the btuart driver is fairly simple, it would be best > to just clone it and make a btdtl.c that handles the appropriate IO > directly.. that way, btattach can attach it with a different name and no > ioctl is needed to activate the difference..? I apologizes for insufficient explanation. I want to call btuartioctl() for NOKIA DTL. Therefore, it is necessary to return btuart.h removed before. (btuart.h v1.4) > I think there is some specialist 'nokia' control packets possible with > this protocol, do we need to use them during the operation of the device? > (in setup, the btattach module can send/recv what it likes..) NOKIA DTL always need btuart_dtl_header for all packet. However it not need process for setup. It detect clock speed automatically perhaps. I could operate 9600 and 115200. # btattach -f /dev/tty01 (9600|115200) Moreover, I understood there was a method where it reset it by using the MODEM control register (com(4)'s MCR). Thanks, -- kiyohara
