> > Is the netlink socket interface the prefered method? I'm just starting the > > design phase and can use netlink or sysfs. > > Netlink is definitely the way to go. > Sysfs was designed for simple one-shot events, like turn on/turn off. > As we saw - simple data reading requires a lot of racy manipulations. > Netlink is a UDP-like protocol, which can be used for > kernelspace<->userspace or user<->user multicast transfers. > Using sequence and acknowledge numbers for request/response design > allows to control the whole dataflow. > Asynchronous events, like new device found or alarm search results, will > be delivered to all "subscribed" applications. Netlink is definitely very nice for this. But (to relate to the other sub-thread about periodic polling), as I understand it, the idea of netlink is to implement only the bare minimum of functionality in kernel space (like hardware initialization and the mechanisms to initiate asynchronous communication from kernel -> user) and leave the rest to the user-space code. Since the 1-wire slave devices cannot send an interrupt and checking for the ALARM state is actually done by polling, I would think that this does not really need to go into the kernel module but can be just as well handled by e.g. owserver. Or am I getting this completely wrong?
Regards, Christoph -- Christoph Scheurer GnuPG key Id: 0x6128C6B6 ------------------------------------------------------- This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642 _______________________________________________ Owfs-developers mailing list Owfs-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/owfs-developers
