It would also be very nice to be able to enable zero copy for mmapped files
by a config option.
Many applications use mapped memory to serve lots of small and
medium sized files (4 - 1024 KB) or even a few big files
(think at web servers, i.e. Apache 2, etc.); this is done to better
serve multiple / parallel downloads being done on the same files.
Using or not using mmap() is a userland / application choice that depends
on lots of factors / strategies that are outside kernel scope.
Besides this there seems to be some work in progress to speed up
mmap() calls, thus why not doing all the work right ?
-- Nick Name: A.D.F. E-Mail: adefacc@tin.it E-Mail-Font: Courier New (plain text, no html)-- - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/