> We have a little research team which tries various methods to speed > up interprocess communication and this is going to be a part of it, > which I was asked to implement. <snip> > I hope this explains u my position. Not really. > I am trying to implement a memory layer between the kernel and > process, so that process can use it to communicate between each > other. This is going to be a shared memory. It is not clear how you can speed up IPC when all you are doing is using the existing shared memory mechanism. The only difference seems to be when and by whom the shared memory segment is set up. By doing this part in the kernel, you are only avoiding the usual shmget/shmat calls. But how does that improve speed? You still need some way to attach to the segment, and of course the basic IPC mechanism hasn't changed any way. -Ravi. __________________________________________________ Do You Yahoo!? Yahoo! Health - your guide to health and wellness http://health.yahoo.com -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
