Thread: Use macros for calculating LWLock offset

Hi,

In the lwlock.c, InitializeLWLocks() calculate the LWLock offset by 
itself (c319991bcad),
however, there are macros defined in lwlock.h, I think, we can use the 
macros.

diff --git a/src/backend/storage/lmgr/lwlock.c 
b/src/backend/storage/lmgr/lwlock.c
index 2fa90cc095..108e652179 100644
--- a/src/backend/storage/lmgr/lwlock.c
+++ b/src/backend/storage/lmgr/lwlock.c
@@ -525,18 +525,17 @@ InitializeLWLocks(void)
                 LWLockInitialize(&lock->lock, id);

         /* Initialize buffer mapping LWLocks in main array */
-       lock = MainLWLockArray + NUM_INDIVIDUAL_LWLOCKS;
+       lock = MainLWLockArray + BUFFER_MAPPING_LWLOCK_OFFSET;
         for (id = 0; id < NUM_BUFFER_PARTITIONS; id++, lock++)
                 LWLockInitialize(&lock->lock, LWTRANCHE_BUFFER_MAPPING);

         /* Initialize lmgrs' LWLocks in main array */
-       lock = MainLWLockArray + NUM_INDIVIDUAL_LWLOCKS + 
NUM_BUFFER_PARTITIONS;
+       lock = MainLWLockArray + LOCK_MANAGER_LWLOCK_OFFSET;
         for (id = 0; id < NUM_LOCK_PARTITIONS; id++, lock++)
                 LWLockInitialize(&lock->lock, LWTRANCHE_LOCK_MANAGER);

         /* Initialize predicate lmgrs' LWLocks in main array */
-       lock = MainLWLockArray + NUM_INDIVIDUAL_LWLOCKS +
-               NUM_BUFFER_PARTITIONS + NUM_LOCK_PARTITIONS;
+       lock = MainLWLockArray + PREDICATELOCK_MANAGER_LWLOCK_OFFSET;
         for (id = 0; id < NUM_PREDICATELOCK_PARTITIONS; id++, lock++)
                 LWLockInitialize(&lock->lock, 
LWTRANCHE_PREDICATE_LOCK_MANAGER);


--
Best regards
Japin Li

On Thu, Nov 19, 2020 at 03:39:51PM +0800, japin wrote:
> In the lwlock.c, InitializeLWLocks() calculate the LWLock offset by itself
> (c319991bcad),
> however, there are macros defined in lwlock.h, I think, we can use the
> macros.

I agree that this makes this code a bit cleaner, so let's use those
macros.  Others may have some comments here, so let's wait a bit
first.
--
Michael

On Fri, Nov 20, 2020 at 03:25:50PM +0900, Michael Paquier wrote:
> I agree that this makes this code a bit cleaner, so let's use those
> macros.  Others may have some comments here, so let's wait a bit
> first.

Got this one committed as of d03d754.
--
Michael