Commit 954c1813 authored by Tom Lane's avatar Tom Lane

Remove an unnecessary HOLD_INTERRUPTS/RESUME_INTERRUPTS pair.

This was required back when RESUME_INTERRUPTS could actually
execute ProcessInterrupts, but that hasn't been true since 2001...
parent bd95c749
...@@ -8,7 +8,7 @@ ...@@ -8,7 +8,7 @@
* *
* *
* IDENTIFICATION * IDENTIFICATION
* $PostgreSQL: pgsql/src/backend/storage/buffer/bufmgr.c,v 1.212 2006/10/04 00:29:57 momjian Exp $ * $PostgreSQL: pgsql/src/backend/storage/buffer/bufmgr.c,v 1.213 2006/10/22 20:34:54 tgl Exp $
* *
*------------------------------------------------------------------------- *-------------------------------------------------------------------------
*/ */
...@@ -694,21 +694,14 @@ retry: ...@@ -694,21 +694,14 @@ retry:
BufTableDelete(&oldTag, oldHash); BufTableDelete(&oldTag, oldHash);
/* /*
* Avoid accepting a cancel interrupt when we release the mapping lock; * Done with mapping lock.
* that would leave the buffer free but not on the freelist. (Which would
* not be fatal, since it'd get picked up again by the clock scanning
* code, but we'd rather be sure it gets to the freelist.)
*/ */
HOLD_INTERRUPTS();
LWLockRelease(oldPartitionLock); LWLockRelease(oldPartitionLock);
/* /*
* Insert the buffer at the head of the list of free buffers. * Insert the buffer at the head of the list of free buffers.
*/ */
StrategyFreeBuffer(buf, true); StrategyFreeBuffer(buf, true);
RESUME_INTERRUPTS();
} }
/* /*
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment