[PATCHES] Fix argument of SetBufferCommitInfoNeedsSave at _bt_check_unique

2005-10-12 Thread Atsushi Ogawa 
The argument of SetBufferCommitInfoNeedsSave at _bt_check_unique
might be not correct. When the next page is checked, I think that
the argument is not buf but nbuf.

regards,

--- Atsushi Ogawa
*** ./src/backend/access/nbtree/nbtinsert.c.orig	2005-10-13 00:59:55.599839064 +0900
--- ./src/backend/access/nbtree/nbtinsert.c	2005-10-13 01:02:43.335339392 +0900
***
*** 261,267 
   hbuffer) == HEAPTUPLE_DEAD)
  	{
  		curitemid-lp_flags |= LP_DELETE;
! 		SetBufferCommitInfoNeedsSave(buf);
  	}
  	LockBuffer(hbuffer, BUFFER_LOCK_UNLOCK);
  }
--- 261,272 
   hbuffer) == HEAPTUPLE_DEAD)
  	{
  		curitemid-lp_flags |= LP_DELETE;
! 
! 		/* Make a buffer of curitemid dirty. */ 
! 		if (nbuf == InvalidBuffer)
! 			SetBufferCommitInfoNeedsSave(buf);
! 		else
! 			SetBufferCommitInfoNeedsSave(nbuf);
  	}
  	LockBuffer(hbuffer, BUFFER_LOCK_UNLOCK);
  }

---(end of broadcast)---
TIP 1: if posting/reading through Usenet, please send an appropriate
   subscribe-nomail command to [EMAIL PROTECTED] so that your
   message can get through to the mailing list cleanly

Re: [PATCHES] Fix argument of SetBufferCommitInfoNeedsSave at _bt_check_unique

2005-10-12 Thread Tom Lane 
Atsushi Ogawa [EMAIL PROTECTED] writes:
 The argument of SetBufferCommitInfoNeedsSave at _bt_check_unique
 might be not correct. When the next page is checked, I think that
 the argument is not buf but nbuf.

You are right --- thanks for finding this!

regards, tom lane

---(end of broadcast)---
TIP 5: don't forget to increase your free space map settings