Messages in this thread |  | | | From | "Anthony Barbachan" <> | | Subject | Re: Question on FFS support | | Date | Sat, 9 Oct 1999 04:31:55 -0400 |
| |
> > included in the Linux kernel. I would have thought that with the need for a > > journaling filesystem FFS would have been adopted, especially with the BSD > > code available. Is there some issue that has prevented this? > > a) FFS is not journaling.
I stand corrected then. While inquiring about FreeBSD large file support and the availability of a journaling filesystem because of the possible need for some of my Samba/NFS servers to serve > 2GB files as well as the definate need to alleviate the long fsck delay, one of the responses claimed that journaling support was there and his FreeBSD systems fscked large volumes very quicky. > b) FFS is supported on Linux.
After rechecking the kernel menuconfig I found the option combination to activate it. Originally I had looked for FFS which appeared and said it was for the Amiga not FreeBSD, I didn't expect to find it under a different name UFS which I've always assumed was Solaris.
> c) FFS under FreeBSD uses Kirk's softupdates. That's what you ar thinking > about.
And softupdates is what gave me responder his quick fsck time? Ok I see. Thanks.
> d) s-u is not fs-specific, at least between the FFS/UFS/EXT2 > e) IIRC EXT2 driver under -CURRENT also uses s-u. > f) Kirk's code seriously depends on the BSD page/buffer cache (Linux is > moving to the similar variant) _and_ on the BSD VFS. > g) Linux VFS differs. Big way. > h) The old pre-alpha code around s-u for Linux that I had in Spring is (a) > buggy and (b) depends on the old buffer cache. >
- 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/
|  |