"Jeff V. Merkey" wrote:
>
> Alexander Viro wrote:
> >
>
> > Erm... OK, I could see the point if you only took that lock for
> > directories that have ->i_sem or ->i_zombie taken, but then.. what's the
> > point of taking it at all?
I guess the real question here is how much serialization does the VFS
provide above you. I discovered last year in my initial forays with
Linux that parallelism is fully exposed to the file system. The locks
are there to prevent folks from stomping on FAT chains. I can forgoe
the locking here unless a FAT chain is being monkeyed with. The real
tradeoff is that I don't have to ever lock the FAT tables provided I
follow some simple rules about how fat chains get changed. i.e. cluster
allocation from the fat table is atomic, and fat chains are locked if
they are read, written, or truncated.
In the case of directories, the locks protect the block count. If the
VFS above provides serialization for changes to directory data, this
lock can be dropped.
Jeff
-
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/
This archive was generated by hypermail 2b29 : Mon May 15 2000 - 21:00:21 EST