Date: Sun, 9 Nov 2003 15:16:55 +0300 (MSK) From: Igor Sysoev <is@rambler-co.ru> To: John-Mark Gurney <gurney_j@efn.org> Cc: Alan Cox <alc@cs.rice.edu> Subject: Re: Update: Debox sendfile modifications Message-ID: <Pine.BSF.4.21.0311091515470.18994-100000@is.park.rambler.ru> In-Reply-To: <20031109033026.GF558@funkthat.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 8 Nov 2003, John-Mark Gurney wrote: > Igor Sysoev wrote this message on Wed, Nov 05, 2003 at 12:31 +0300: > > I think it can done in the following way - a socket should have flag > > that says that sendfile() had started the reading a page. > > layer violation... I do not think that it's layer violation. sendfile() works with descriptor so it should know its state. It should know wheather descriptor is non-blocking or has it enough buffer space. > how do you know that the fd is a socket? fp->f_type == DTYPE_SOCKET > > select()/poll()/kevent() should check this flag before the checking > > a socket buffer space. When the page had been read this flag is reset. > > So, what about using sendfile on a pipe? are you going to teach sendfile > how to interact with pipe's too? What about other fd types? > > If you made this a fd transparent operation then I would agree with > it. The current sendfile() implementation works with sockets only. Well, I agree that such sendfile() implementation is a hack. Nowever this implementation is very usefull in the real world - it allows to minimize a data copy in http and ftp servers. I just could not figure to myself where can be usefull the high perfomance sendfile() to a pipe. I think that it's better to leave sendfile() as a sending to a socket only hack. I believe that any sendfile() generalization (e.g. sending data from a socket to a file) is useless. Igor Sysoev http://sysoev.ru/en/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0311091515470.18994-100000>