Re: [PATCH]IPC locks breaking down with RCU

[Hugh Dickins]

On Tue, 22 Oct 2002, Dipankar Sarma wrote:
> 
> I took a quick look at the original ipc code and I don't understand
> something - it seems to me the ipc_ids structs are protected by the semaphore
> inside for all operations, so why do we need the spinlock in the
> first place ? Am I missing something here ?

I made that mistake too at first, Mingming set me straight.
Many of the entry points down() the ipc_ids.sem semaphore, but the
most significant ones do not.  ipc/sem.c is probably the best example
(if confusing, since it involves quite different meanings of semaphore):
sys_semop() is the frequent, fast entry point, uses sem_lock without down.

Hugh