Eric Wong <e...@80x24.org> wrote: > This bug is noticeable on a amd64 FreeBSD 9.2 VM, and possible under > Linux, too. This happens as a zero-copy sendfile implementation means > pages queued for transmission by the sendfile system call should not be > modified at any point after the sendfile syscall is made.
Btw, this gotcha is now noted in the Linux sendfile(2) manpage: http://git.kernel.org/cgit/docs/man-pages/man-pages.git/commit/man2/sendfile.2?id=7b6a3299776b5c1c4f169a591434a855d50c68b4