On Thu, Oct 17, 2002 at 05:59:10PM +0530, Nagaraj wrote: > > > > Well I dont mean to intrude, but IMHO, this looks like a multithreading > > requirement in userspace. > > Probably yes. I didnt try that. I bet no. > >writes are generally blocking calls and if the > > application intends to fill in another buffer during the time the write > call is > > being executed, let him use multithreading in userland and that will solve > his > > problem.. Or am I missing something ? Yes... > Yeh, if u can somehow arrange to > prepare the second buffer and send it immediately > when blocking write returns, and setup a new dma for that, > it should also solve the problem. > > any volunteers to try that out ? You are both missing the poll system call and the O_NONBLOCK fcntl flags (note that only blocking call gives scheduler an oportunity to switch threads, so once you use O_NONBLOCK, threads can help nothing more). However, you did not post the code with both read and write yet... (or did I mis it?) ------------------------------------------------------------------------------- Jan 'Bulb' Hudec <bulb@ucw.cz> -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
