hi all, i read the O_DIRECT january thread: http://kerneltrap.org/mailarchive/linux-kernel/2007/1/11/44365
i am interested in using mmap as a quick way of getting data to and from disk. i also want to handle errors (like bad blocks) as well as is possible using usual read/write calls. would i be right in saying that a) mmap+madvise+msync provides almost the same function as b) read/write(2)+posix_fadvise+fsync i have read that fault-tolerance or error-handling is 'better' using (b). what does this mean exactly? i note that write(2) may return EIO - what happens if the same underlying error occurs when accessing a mmap'ed file? i notice that in the implementation of msync, fsync is called so errno might be set to EIO... ta, jack (please cc: me) -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
