Re: [HACKERS] Re: starting to review the Extend NOT NULL representation to pg_constraint patch

2011-06-25 Thread Dean Rasheed
On 25 June 2011 01:59, Alvaro Herrera wrote: Excerpts from Robert Haas's message of vie jun 24 19:01:49 -0400 2011: I would tend to think of the not-null-ness that is required by the primary constraint as a separate constraint, not an intrinsic part of the primary

Re: [HACKERS] WIP: Fast GiST index build

2011-06-25 Thread Jesper Krogh
On 2011-06-06 09:42, Heikki Linnakangas wrote: took about 15 hours without the patch, and 2 hours with it. That's quite dramatic. With the precense of robust consumer-class SSD-drives that can be found in sizes where they actually can fit many database usage scenarios. A PostgreSQL version is

Re: [HACKERS] WIP: Fast GiST index build

2011-06-25 Thread Alexander Korotkov
On Sat, Jun 25, 2011 at 11:03 AM, Jesper Krogh wrote: * Wouldn't it be natural to measure the performance benefits of disc-bound tests in an SSD setup? Sure, it would be great to run performance tests on SSD drives too. Unfortunately, I don't have corresponding test platform

Re: [HACKERS] debugging tools inside postgres

2011-06-25 Thread Martijn van Oosterhout
On Fri, Jun 24, 2011 at 02:35:08PM +0800, HuangQi wrote: Hi, I'm trying to debug a modification for the query planner. But I found it seems the data structure of my planned query is incorrect. I was trying to print out the data structure by use the p command in gdb which is quite

Re: [HACKERS] heap_hot_search_buffer refactoring

2011-06-25 Thread Jeff Davis
On Fri, 2011-06-24 at 15:32 -0400, Robert Haas wrote: On Sun, Jun 19, 2011 at 2:16 PM, Robert Haas wrote: New patch attached, with that one-line change. Jeff, are you planning to review this further? Do you think it's OK to commit? 1. Patch does not apply to master

Re: [HACKERS] Repeated PredicateLockRelation calls during seqscan

2011-06-25 Thread Kevin Grittner
Heikki Linnakangas wrote: BTW, isn't bitgetpage() in nodeBitmapHeapscan.c missing PredicateLockTuple() and CheckForSerializableConflictOut() calls in the codepath for a lossy bitmap? In the non-lossy case, heap_hot_search_buffer() takes care of it, but not in the lossy case. I think the

Re: [HACKERS] possible connection leak in dblink?

2011-06-25 Thread Joe Conway
On 06/14/2011 07:41 PM, Fujii Masao wrote: On Wed, Jun 15, 2011 at 5:34 AM, Peter Eisentraut wrote: Otherwise the connection might not get freed. Could someone verify that? ISTM that the root problem is that dblink_send_query calls DBLINK_GET_CONN though it doesn't accept

Re: [HACKERS] possible connection leak in dblink?

2011-06-25 Thread Peter Eisentraut
On lör, 2011-06-25 at 13:36 -0700, Joe Conway wrote: However, since this is really just a case of unused variables and not a leaked connection, I'm inclined to just fix git master -- comments on that? Please put it into 9.1 as well, so we can get a clean compile with gcc 4.6 there. -- Sent

[HACKERS] Range Types, constructors, and the type system

2011-06-25 Thread Jeff Davis
Different ranges over the same subtype make sense when using different total orders for the subtype. This is most apparent with text collation, but makes sense (at least mathematically, if not practically) for any subtype. For instance: [a, Z) is a valid range in en_US, but not in C, so it makes

Re: [HACKERS] spinlock contention

2011-06-25 Thread Greg Stark
On Thu, Jun 23, 2011 at 4:42 PM, Robert Haas wrote: ProcArrayLock looks like a tougher nut to crack - there's simply no way, with the system we have right now, that you can take a snapshot without locking the list of running processes.  I'm not sure what to do about that,

[HACKERS] Word-smithing doc changes

2011-06-25 Thread Greg Stark
I think this commit was ill-advised:;a=commitdiff;h=a03feb9354bda5084f19cc952bc52ba7be89f372 In a concurrent index build, the index is actually entered into the system catalogs in one transaction, then the two table scans occur in a -