Robert Haas <robertmh...@gmail.com> writes:
> This discussion seems to miss the fact that there are two levels of
> reordering that can happen. First, the compiler can move things
> around. Second, the CPU can move things around.
Right, I think that's exactly the problem with the previous wording of
that comment; it doesn't address the two logical levels involved.
I've rewritten it, see what you think.
* Another caution for users of these macros is that it is the caller's
* responsibility to ensure that the compiler doesn't re-order accesses
* to shared memory to precede the actual lock acquisition, or follow the
* lock release. Typically we handle this by using volatile-qualified
* pointers to refer to both the spinlock itself and the shared data
* structure being accessed within the spinlocked critical section.
* That fixes it because compilers are not allowed to re-order accesses
* to volatile objects relative to other such accesses.
*
* On platforms with weak memory ordering, the TAS(), TAS_SPIN(), and
* S_UNLOCK() macros must further include hardware-level memory fence
* instructions to prevent similar re-ordering at the hardware level.
* TAS() and TAS_SPIN() must guarantee that loads and stores issued after
* the macro are not executed until the lock has been obtained.
Conversely,
* S_UNLOCK() must guarantee that loads and stores issued before the macro
* have been executed before the lock is released.
regards, tom lane
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
