Re: [Patch] shm bug introduced with pagecache in 2.3.11

Manfred Spraul (manfreds@colorfullife.com)
Fri, 19 Nov 1999 21:31:54 +0100

"Benjamin C.R. LaHaise" wrote:
>
> On Fri, 19 Nov 1999, Linus Torvalds wrote:
>
> > Is anybody willing to take a stab at creating a read-write semaphore?
>
> Sure. Does the following sound?
>
> void down_read(struct rw_semaphore *sem);
> void down_write(struct rw_semaphore *sem);
> void up_read(struct rw_semaphore *sem);
> void up_write(struct rw_semaphore *sem);
>
> Looking at existing semaphores and rwlocks, it looks like a quick and
> painless beast to create.
>
There is a subtle problem: a rw semaphore is no real semaphore, it will
not support counts [at least if you base them on the rwlock asm code]:
semaphores allow a call to up() before the corresponding down(), eg
vfork() could do that [on SMP with an extremely long interrupt on the
line with the fork_out-label ;)

I would prefer if you use completely different function names to avoid
any confusion.
eg
acquire_shared()
acquire_exclusive()
release_shared()
release_exclusive()

--
	Manfred


-
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/