Re: parallel writes to the same file, 2.3.7

Linus Torvalds (torvalds@transmeta.com)
Tue, 22 Jun 1999 10:54:20 -0700 (PDT)

On Tue, 22 Jun 1999, Ingo Molnar wrote:
>
> hm, i thought we want to protect preallocated blocks and other 'on the
> fly' nonpersistent inode-metadata with the inode semaphore, so we can
> later on remove the big kernel lock _and_ the per-filesystem superblock
> lock without worrying too much.

I thought so, yes. However, I decided that it's too large a price to pay
for too little gain (the price being the deadlock on the inode semaphore
in the case we want to page out).

NOTE! The low-level filesystem may choose to use the inode semaphore for
that still - it's just that it's not mandated by the kernel any more, and
if the low-level FS does that it had better guarantee that while the inode
is locked there is no memory allocation activity that can cause a
deadlock..

Linus

-
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/