Messages in this thread | | | From | "Stephen C. Tweedie" <> | Date | Tue, 20 Jul 1999 14:11:32 +0100 (BST) | Subject | Re: [RFC] - Some notions that I would like comments on |
| |
Hi,
On Mon, 19 Jul 1999 14:37:27 -0700 (PDT), Jeremy Fitzhardinge <jeremy@goop.org> said:
> Initially I hoped that the filesystem would lay out blocks linearly on > disk in the order which I wrote them,
It should do so, pretty much, although you may end up with a lot of inode indirection information polluting the first few blocks of the executable if you do it this way.
However, any filesystem reorganisation will not help. The trouble is, the kernel assumes that binaries are allocated reasonably sequentially, and so when doing clustered paging, it naturally reads in a 64K contiguous lump of the file. If you have reordered the file so that each 64k chunk of data is occupying non-contiguous blocks on disk, then you just end up making the kernel pager much less efficient.
--Stephen
- 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/
| |