Messages in this thread |  | | | Date | Tue, 15 Dec 1998 00:10:40 +1100 | | From | Richard Gooch <> | | Subject | Re: PATCH: Raw device IO for 2.1.131 |
| |
Jes Sorensen writes: > >>>>> "Richard" == Richard Gooch <rgooch@atnf.csiro.au> writes: > > Richard> Jes Sorensen writes: > >> SGI managed to get zero copy right for write() on a socket and it > >> works great, I'd love having Linux do the same. > > Richard> To get you what you want without resorting to sendfile(), > Richard> we'd have to be able to be able to pin down user pages and > Richard> then initiate DMA. Linus has said he doesn't like that idea > Richard> and has also pointed out he feels that the copy operation > Richard> would not be a significant overhead. Other have mentioned > Richard> other applications (video capture and processing) where they > Richard> feel an extra copy *is* significant. But is this the case > Richard> with your application? It seems to me that if you have a > Richard> bunch of user pages you want to DMA out, you have already > Richard> spent a considerable amount of time generating the data, so > Richard> an extra copy is not significant. Is that so? > > No most of our applications do read(filefd), write(sockfd) and as > such it could probably be done with sendfile, the main issue is that > it requires rewriting the code as a special case for Linux.
How about mmap(filefd); write (sockfd, mmap_addr); ?
Regards,
Richard.... - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
|  |