Messages in this thread | | | Subject | Re: mmap() is slower than read() on SCSI/IDE on 2.0 and 2.1 | Date | Mon, 14 Dec 1998 05:05:34 -0800 | From | "Jay Nordwick" <> |
| |
>read() does more prefetching than mmap() faults do, at least this was >the story last time I was playing around in this area... > >Later, >David S. Miller >davem@dm.cobaltmicro.com >
How about cluing in the VM to prefetch with madvise() ? How would this effect performance?
As side questions: - why would read() do more prefetching than mmap()? - If read() caused prefetching of blocks, the when a page faulted in from an mmap() mapping, why wouldn't this cause prefetching? - When a binary executable image is mmap()ed to be run, wouldn't you want to prefetch blocks of code and data -- what would be the difference between this and the user mmap()ing in a file?
Just getting new to Linux kernel sources (coming from spare time FreeBSD hacking) and would appreciate any useful source people can point me to.
Thanks, -jay
- 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/
| |