]> granicus.if.org Git - postgresql/log
postgresql
23 years agoDisable query cancel during HandleDeadLock().
Hiroshi Inoue [Tue, 9 Jan 2001 09:38:57 +0000 (09:38 +0000)]
Disable query cancel during HandleDeadLock().

23 years ago1. Checkpoint.undo may be after checkpoint itself:
Vadim B. Mikheev [Tue, 9 Jan 2001 06:24:33 +0000 (06:24 +0000)]
1. Checkpoint.undo may be after checkpoint itself:
   - no more elog(STOP) in StartupXLOG();
   - both checkpoint' undo & redo are used to define
     oldest on-line log file.
2. Ability to pre-allocate a few log files at checkpoint time
   (wal_files option). Off by default.

23 years agoUpdate TODO list.
Bruce Momjian [Tue, 9 Jan 2001 06:16:23 +0000 (06:16 +0000)]
Update TODO list.

23 years agoREADME.mb has been unified into SGML documents.
Tatsuo Ishii [Tue, 9 Jan 2001 04:40:31 +0000 (04:40 +0000)]
README.mb has been unified into SGML documents.

23 years agoFix oversight in planning of GROUP queries: when an expression is used
Tom Lane [Tue, 9 Jan 2001 03:48:51 +0000 (03:48 +0000)]
Fix oversight in planning of GROUP queries: when an expression is used
as both a GROUP BY item and an output expression, the top-level Group
node should just copy up the evaluated expression value from its input,
rather than re-evaluating the expression.  Aside from any performance
benefit this might offer, this avoids a crash when there is a sub-SELECT
in said expression.

23 years agoFix small but critical typo ...
Tom Lane [Tue, 9 Jan 2001 02:15:16 +0000 (02:15 +0000)]
Fix small but critical typo ...

23 years agoPrevent vacuumdb from trying to vacuum template0.
Tom Lane [Mon, 8 Jan 2001 23:02:36 +0000 (23:02 +0000)]
Prevent vacuumdb from trying to vacuum template0.

23 years agoDocument the system attributes ctid and tableoid, which for some reason
Tom Lane [Mon, 8 Jan 2001 22:07:47 +0000 (22:07 +0000)]
Document the system attributes ctid and tableoid, which for some reason
were never yet mentioned anywhere in our documentation.  Improve
explanations of the other system attributes, too.

23 years agoDocument that we don't support ORDER BY with general expressions on
Tom Lane [Mon, 8 Jan 2001 21:30:37 +0000 (21:30 +0000)]
Document that we don't support ORDER BY with general expressions on
the output of UNION/INTERSECT/EXCEPT.

23 years agoAdd rudimentary section about controlling kernel's file and process limits.
Peter Eisentraut [Mon, 8 Jan 2001 21:01:54 +0000 (21:01 +0000)]
Add rudimentary section about controlling kernel's file and process limits.

23 years agoRemove compiler warning about uninitialized warnings.
Bruce Momjian [Mon, 8 Jan 2001 20:54:24 +0000 (20:54 +0000)]
Remove compiler warning about uninitialized warnings.

23 years agocheck for failure after vacuuming each DB, not only the last one.
Tom Lane [Mon, 8 Jan 2001 20:32:27 +0000 (20:32 +0000)]
check for failure after vacuuming each DB, not only the last one.

23 years agoAdd some debugging support code (ifdef'd out in normal use).
Tom Lane [Mon, 8 Jan 2001 18:34:44 +0000 (18:34 +0000)]
Add some debugging support code (ifdef'd out in normal use).

23 years agoLockBuffer should not elog while holding buffer's cntx_lock.
Tom Lane [Mon, 8 Jan 2001 18:31:49 +0000 (18:31 +0000)]
LockBuffer should not elog while holding buffer's cntx_lock.

23 years agoKeep relations open until they are no longer needed.
Hiroshi Inoue [Mon, 8 Jan 2001 03:14:58 +0000 (03:14 +0000)]
Keep relations open until they are no longer needed.

23 years agoMake outfuncs/readfuncs treat OIDs properly as unsigned values. Clean up
Tom Lane [Mon, 8 Jan 2001 00:31:43 +0000 (00:31 +0000)]
Make outfuncs/readfuncs treat OIDs properly as unsigned values.  Clean up
inconsistent coding practices for handling Index values and booleans,
too.

23 years agoCorrect nasty error in heap_update: it was releasing the buffer refcount
Tom Lane [Sun, 7 Jan 2001 22:14:31 +0000 (22:14 +0000)]
Correct nasty error in heap_update: it was releasing the buffer refcount
before calling RelationInvalidateHeapTuple(), which is bad because the
latter needs to look at the tuple data, which is in the shared disk
buffer.  If another backend manages to recycle the buffer while this
is going on, we will compute the wrong hashindex for the tuple or
maybe even crash outright.  Must hold buffer refcount until afterwards.
(This bug is not in 7.0.*; seems to be have introduced during WAL changes.)

23 years agoClear QueryCancel and ProcDiePending at start of proc_exit, to ensure
Tom Lane [Sun, 7 Jan 2001 04:30:41 +0000 (04:30 +0000)]
Clear QueryCancel and ProcDiePending at start of proc_exit, to ensure
that leftover cancel/die requests cannot interfere with exit activities.

23 years agoFix recent breakage of query-cancel logic, see my pghackers message
Tom Lane [Sun, 7 Jan 2001 04:17:29 +0000 (04:17 +0000)]
Fix recent breakage of query-cancel logic, see my pghackers message
of 6 Jan 2001 21:55.

23 years agotag configure as beta2 .. REL7_1_BETA2
PostgreSQL Daemon [Sun, 7 Jan 2001 02:23:25 +0000 (02:23 +0000)]
tag configure as beta2 ..

23 years agoResultmap updates for OpenBSD, per report from bpalmer@crimelabs.net.
Tom Lane [Sun, 7 Jan 2001 01:14:35 +0000 (01:14 +0000)]
Resultmap updates for OpenBSD, per report from bpalmer@crimelabs.net.

23 years agoModify readfuncs so that recursive use of stringToNode will not crash
Tom Lane [Sun, 7 Jan 2001 01:08:48 +0000 (01:08 +0000)]
Modify readfuncs so that recursive use of stringToNode will not crash
and burn.  Just for added luck, change reading of CONST nodes so that
we do not need to consult pg_type rows while reading them; this means
that no database access occurs during stringToNode.  This requires
changing the order in which const-node fields are written, which means
an initdb is forced.

23 years agoClean up checking of relkind for ALTER TABLE and LOCK TABLE commands.
Tom Lane [Sun, 7 Jan 2001 00:05:22 +0000 (00:05 +0000)]
Clean up checking of relkind for ALTER TABLE and LOCK TABLE commands.
Disallow cases like adding constraints to sequences :-(, and eliminate
now-unnecessary search of pg_rewrite to decide if a relation is a view.

23 years agoLog memory context stats to stderr when reporting a 'Memory exhausted'
Tom Lane [Sat, 6 Jan 2001 21:59:39 +0000 (21:59 +0000)]
Log memory context stats to stderr when reporting a 'Memory exhausted'
error, so as to provide a starting point for debugging.

23 years agoFix memory leak in relcache handling of rules: allocate rule parsetrees
Tom Lane [Sat, 6 Jan 2001 21:53:18 +0000 (21:53 +0000)]
Fix memory leak in relcache handling of rules: allocate rule parsetrees
in per-entry sub-memory-context, where they were supposed to go, rather
than in CacheMemoryContext where the code was putting them.  Must've
suffered a severe brain fade when I wrote this :-(

23 years agoUse more portable syntax for 'find'.
Peter Eisentraut [Sat, 6 Jan 2001 21:24:01 +0000 (21:24 +0000)]
Use more portable syntax for 'find'.

23 years agoPolish help output. Allow --help to work with BSD getopts.
Peter Eisentraut [Sat, 6 Jan 2001 20:57:26 +0000 (20:57 +0000)]
Polish help output.  Allow --help to work with BSD getopts.

23 years agoNo need for screen_size to be static.
Tom Lane [Sat, 6 Jan 2001 17:43:01 +0000 (17:43 +0000)]
No need for screen_size to be static.

23 years agoSimplify rules to build man pages so they run a lot faster and create less
Peter Eisentraut [Sat, 6 Jan 2001 16:54:16 +0000 (16:54 +0000)]
Simplify rules to build man pages so they run a lot faster and create less
noise.

23 years agoUpdate multibyte Japanese doc for 7.1.
Tatsuo Ishii [Sat, 6 Jan 2001 12:38:28 +0000 (12:38 +0000)]
Update multibyte Japanese doc for 7.1.

23 years agoEXECUTE documentation, from "Robert B. Easter" <reaster@comptechnews.com>.
Peter Eisentraut [Sat, 6 Jan 2001 12:26:08 +0000 (12:26 +0000)]
EXECUTE documentation, from "Robert B. Easter" <reaster@comptechnews.com>.
I threw in spell check run over the whole file.

23 years agoUpdate section on SQL syntax. (Still a lot to be done though.) Add
Peter Eisentraut [Sat, 6 Jan 2001 11:58:56 +0000 (11:58 +0000)]
Update section on SQL syntax.  (Still a lot to be done though.)  Add
appendix with comprehensive list of key words.

23 years agoSimplify the rules that explicitly allowed TYPE as a type name (which is
Peter Eisentraut [Sat, 6 Jan 2001 10:50:02 +0000 (10:50 +0000)]
Simplify the rules that explicitly allowed TYPE as a type name (which is
no longer the case).  Add AND and TRAILING to ColLabel.  All key words
except AS are now at least ColLabel's.

23 years agoBring CREATE RULE reference page into some semblance of agreement with
Tom Lane [Sat, 6 Jan 2001 04:14:35 +0000 (04:14 +0000)]
Bring CREATE RULE reference page into some semblance of agreement with
what's actually implemented.

23 years agoFix copy to make it more robust against unexpected character
Tatsuo Ishii [Sat, 6 Jan 2001 03:33:17 +0000 (03:33 +0000)]
Fix copy to make it more robust against unexpected character
sequences. This is done by disabling multi-byte awareness when it's
not necessary. This is kind of a workaround, not a perfect solution.
However, there is no ideal way to parse broken multi-byte character
sequences. So I guess this is the best way what we could do right
now...

23 years agoinit_irels() is changed to be called in RelationCacheInitializePhase2()
Hiroshi Inoue [Sat, 6 Jan 2001 01:48:59 +0000 (01:48 +0000)]
init_irels() is changed to be called in RelationCacheInitializePhase2()
so that transactional control could guarantee the consistency.

23 years agoFix NOT NULL option for plpgsql variables (doesn't look like it
Tom Lane [Sat, 6 Jan 2001 01:43:01 +0000 (01:43 +0000)]
Fix NOT NULL option for plpgsql variables (doesn't look like it
could ever have worked...)

23 years agoFix misplaced strdup(), which could lead to error messages referencing
Tom Lane [Sat, 6 Jan 2001 01:39:01 +0000 (01:39 +0000)]
Fix misplaced strdup(), which could lead to error messages referencing
deallocated memory later on.

23 years agoRename and document some invalidation routines to make it clearer that
Tom Lane [Fri, 5 Jan 2001 22:54:37 +0000 (22:54 +0000)]
Rename and document some invalidation routines to make it clearer that
they don't themselves flush any cache entries, only add to to-do lists
that will be processed later.

23 years agoRemove not-really-standard implementation of CREATE TABLE's UNDER clause,
Tom Lane [Fri, 5 Jan 2001 06:34:23 +0000 (06:34 +0000)]
Remove not-really-standard implementation of CREATE TABLE's UNDER clause,
and revert documentation to describe the existing INHERITS clause
instead, per recent discussion in pghackers.  Also fix implementation
of SQL_inheritance SET variable: it is not cool to look at this var
during the initial parsing phase, only during parse_analyze().  See
recent bug report concerning misinterpretation of date constants just
after a SET TIMEZONE command.  gram.y really has to be an invariant
transformation of the query string to a raw parsetree; anything that
can vary with time must be done during parse analysis.

23 years agoDisallow creation of a child table by a user who does not own the parent
Tom Lane [Fri, 5 Jan 2001 02:58:16 +0000 (02:58 +0000)]
Disallow creation of a child table by a user who does not own the parent
table, per pghackers discussion around 22-Dec-00.

23 years agoUpdate TODO list.
Bruce Momjian [Thu, 4 Jan 2001 18:51:18 +0000 (18:51 +0000)]
Update TODO list.

23 years agoAllow NetBSD's libedit to be used instead of GNU Readline. (This simply
Peter Eisentraut [Thu, 4 Jan 2001 17:58:48 +0000 (17:58 +0000)]
Allow NetBSD's libedit to be used instead of GNU Readline.  (This simply
amounts to checking for -ledit instead of -lreadline.)

23 years agoCorrect path where to check for password file existance.
Peter Eisentraut [Thu, 4 Jan 2001 17:25:09 +0000 (17:25 +0000)]
Correct path where to check for password file existance.

23 years agoUpdate TODO list.
Bruce Momjian [Thu, 4 Jan 2001 08:24:59 +0000 (08:24 +0000)]
Update TODO list.

23 years agoUpdate TODO list.
Bruce Momjian [Thu, 4 Jan 2001 08:15:19 +0000 (08:15 +0000)]
Update TODO list.

23 years agoUpdate TODO list.
Bruce Momjian [Thu, 4 Jan 2001 08:13:26 +0000 (08:13 +0000)]
Update TODO list.

23 years agoClean up some unnecessary fragility in EXECUTE command.
Tom Lane [Thu, 4 Jan 2001 02:38:02 +0000 (02:38 +0000)]
Clean up some unnecessary fragility in EXECUTE command.

23 years agoRepair guaranteed core dump in SPI_exec(). Guess this routine wasn't
Tom Lane [Thu, 4 Jan 2001 02:36:52 +0000 (02:36 +0000)]
Repair guaranteed core dump in SPI_exec().  Guess this routine wasn't
used before ...

23 years agoI neglected to remove a debug message,sorry.
Hiroshi Inoue [Thu, 4 Jan 2001 02:24:22 +0000 (02:24 +0000)]
I neglected to remove a debug message,sorry.

23 years agopg_dump failed to handle backslashes embedded in function definitions
Tom Lane [Thu, 4 Jan 2001 01:23:47 +0000 (01:23 +0000)]
pg_dump failed to handle backslashes embedded in function definitions
(and most other places where it needed to output a string literal, too,
except for data INSERT statements).  Per bug report from Easter, 12/1/00.

23 years agoFix breakage of rules using NOTIFY actions, per bug report and patch
Tom Lane [Wed, 3 Jan 2001 22:01:05 +0000 (22:01 +0000)]
Fix breakage of rules using NOTIFY actions, per bug report and patch
from sergiop@sinectis.com.ar.

23 years agoNew file format for COPY BINARY, in accordance with pghackers discussions
Tom Lane [Wed, 3 Jan 2001 20:04:10 +0000 (20:04 +0000)]
New file format for COPY BINARY, in accordance with pghackers discussions
of early December 2000.  COPY BINARY is now TOAST-safe.

23 years agoMakeRetrieveViewRuleName was scribbling on memory that didn't belong
Tom Lane [Wed, 3 Jan 2001 18:43:09 +0000 (18:43 +0000)]
MakeRetrieveViewRuleName was scribbling on memory that didn't belong
to it.  Bad dog.

23 years agoRepair always-broken date_part('quarter',timestamp).
Thomas G. Lockhart [Wed, 3 Jan 2001 16:48:02 +0000 (16:48 +0000)]
Repair always-broken date_part('quarter',timestamp).
Previous result did not have correct month boundaries so anything near edge
 cases was suspect (e.g. April was in Q1 and July, August were lumped into
 Q2).
Thanks to Denis Osadchy <osadchy@turbo.nsk.su> for the report.

23 years agoUpdate TODO list.
Bruce Momjian [Wed, 3 Jan 2001 06:56:52 +0000 (06:56 +0000)]
Update TODO list.

Index: TODO
===================================================================
RCS file: /home/projects/pgsql/cvsroot/pgsql/doc/TODO,v
retrieving revision 1.311
diff -r1.311 TODO
3c3
< Last updated: Tue Jan  2 21:23:11 EST 2001
---
> Last updated: Wed Jan  3 01:56:41 EST 2001
10c10
< A dash(-) marks changes that will appear in the upcoming 7.1 release.
---
> A dash (-) marks changes that will appear in the upcoming 7.1 release.
23c23
< * -Creating index of TIMESTAMP & RELTIME fails, or rename to DATETIME(Thomas)
---
> * -Creating index of TIMESTAMP & RELTIME fails, or rename to DATETIME (Thomas)
35c35
< * -Views containing aggregates sometimes fail(Jan)
---
> * -Views containing aggregates sometimes fail (Jan)
37,38c37,38
<   duplicates(inherit)
< * -Disallow LOCK on view(Mark Hollomon)
---
>   duplicates (inherit)
> * -Disallow LOCK on view (Mark Hollomon)
58,60c58,60
< * -Add OUTER joins, left and right[outer](Tom, Thomas)
< * -Allow long tuples by chaining or auto-storing outside db (TOAST)(Jan)
< * Fix memory leak for expressions[memory](Tom?)
---
> * -Add OUTER joins, left and right[outer] (Tom, Thomas)
> * -Allow long tuples by chaining or auto-storing outside db (TOAST) (Jan)
> * Fix memory leak for expressions[memory] (Tom?)
70c70
< * Allow row re-use without vacuum(Vadim)
---
> * Allow row re-use without vacuum (Vadim)
74c74
< * More access control over who can create tables and use locks(Karel)
---
> * More access control over who can create tables and use locks (Karel)
92c92
<  o -Fix large object mapping scheme, own typeid or reltype(Peter)
---
>  o -Fix large object mapping scheme, own typeid or reltype (Peter)
97c97
< * Fix improper masking of some inet/cidr types [cidr]
---
> * -Fix improper masking of some inet/cidr types [cidr]
113c113
< * Get BIT type working
---
> * -Get BIT type working
119c119
< * Update macaddr manufacturer numbers, or remove the function macaddr_manuf()
---
> * -Update macaddr manufacturer numbers, or remove the function macaddr_manuf()
147c147
< * -Add ALTER TABLE ALTER COLUMN feature(Peter E)
---
> * -Add ALTER TABLE ALTER COLUMN feature (Peter E)
157c157
< * Auto-destroy sequence on DROP of table with SERIAL(Ryan)
---
> * Auto-destroy sequence on DROP of table with SERIAL (Ryan)
166a167
> * Allow INSERT INTO tab (col1, ..) VALUES (val1, ..), (val2, ..)
172d172
< * Allow INSERT INTO tab (col1, ..) VALUES (val1, ..), (val2, ..)
176c176
< * Make NULL's come out at the beginning or end depending on the
---
> * Make NULL's come out at the beginning or end depending on the
220c220
< * Certain indexes will not shrink, i.e. oid indexes with many inserts(Vadim)
---
> * Certain indexes will not shrink, i.e. oid indexes with many inserts (Vadim)
226c226
<   with after-row images(Vadim) [logging](Vadim)
---
>   with after-row images (Vadim) [logging]
232c232
< * -Redesign the function call interface to handle NULLs better[function](Tom)
---
> * -Redesign the function call interface to handle NULLs better[function] (Tom)
239c239
< * Encrpyt passwords in pg_shadow table using MD5(Vince)
---
> * Encrpyt passwords in pg_shadow table using MD5 (Vince)
298a299,300
> * Create a transaction processor to aid in persistent connections and
>   connection pooling
307c309
< * Remove SET KSQO option now that OR processing is improved(Tom)
---
> * Remove SET KSQO option now that OR processing is improved (Tom)

23 years agoUpdate TODO list.
Bruce Momjian [Wed, 3 Jan 2001 02:23:22 +0000 (02:23 +0000)]
Update TODO list.

Index: TODO
===================================================================
RCS file: /home/projects/pgsql/cvsroot/pgsql/doc/TODO,v
retrieving revision 1.310
diff -r1.310 TODO
3c3
< Last updated: Wed Dec 27 00:44:44 EST 2000
---
> Last updated: Tue Jan  2 21:23:11 EST 2001
310a311
> * Allow libedit to be used in place of libreadline

23 years agoOnly update stamp-h if config.status actually looks at config.h.
Peter Eisentraut [Tue, 2 Jan 2001 22:13:05 +0000 (22:13 +0000)]
Only update stamp-h if config.status actually looks at config.h.

23 years agoI've found a memory leak in libecpg of PostgreSQL 7.0.3.
Bruce Momjian [Tue, 2 Jan 2001 22:03:02 +0000 (22:03 +0000)]
I've found a memory leak in libecpg of PostgreSQL 7.0.3.
The leak is caused by the memory allocation in
src/interfaces/ecpg/lib/execute.c in line 669 which is never freed.
Adding a "free(array_query);" after PQexec in line 671 seems to fix the
leak.

Thorsten Knabe

23 years agoDocument tuple ordering differences as a possible cause of
Tom Lane [Tue, 2 Jan 2001 05:56:02 +0000 (05:56 +0000)]
Document tuple ordering differences as a possible cause of
regression test 'failures'.

23 years agoClean up non-reentrant interface for hash_seq/HashTableWalk, so that
Tom Lane [Tue, 2 Jan 2001 04:33:24 +0000 (04:33 +0000)]
Clean up non-reentrant interface for hash_seq/HashTableWalk, so that
starting a new hashtable search no longer clobbers any other search
active anywhere in the system.  Fix RelationCacheInvalidate() so that
it will not crash or go into an infinite loop if invoked recursively,
as for example by a second SI Reset message arriving while we are still
processing a prior one.

23 years agoTweak temporary-installation setup so that it doesn't break when the
Tom Lane [Tue, 2 Jan 2001 02:13:48 +0000 (02:13 +0000)]
Tweak temporary-installation setup so that it doesn't break when the
configured install --prefix begins with /data/...

23 years agoUpdate geometry-alpha-precision.out per reports from Brent Verner and
Tom Lane [Mon, 1 Jan 2001 23:18:18 +0000 (23:18 +0000)]
Update geometry-alpha-precision.out per reports from Brent Verner and
Adriaan Joubert.

23 years agoDefine HAVE_LIBZ only if we detect <zlib.h> as well as libz.a/.so.
Tom Lane [Mon, 1 Jan 2001 23:10:09 +0000 (23:10 +0000)]
Define HAVE_LIBZ only if we detect <zlib.h> as well as libz.a/.so.
Otherwise, build falls over on a machine with a non-devel RPM of libz.

23 years agoCLUSTER forgot to create a TOAST table for the clustered relation.
Tom Lane [Mon, 1 Jan 2001 21:35:00 +0000 (21:35 +0000)]
CLUSTER forgot to create a TOAST table for the clustered relation.

23 years agoEnsure attcacheoff is written out as -1 when writing pg_attribute
Tom Lane [Mon, 1 Jan 2001 21:33:31 +0000 (21:33 +0000)]
Ensure attcacheoff is written out as -1 when writing pg_attribute
tuples for a relation.  Needed to prevent Assert failure in CLUSTER.

23 years agoUpdate comment.
Tom Lane [Mon, 1 Jan 2001 21:22:54 +0000 (21:22 +0000)]
Update comment.

23 years agoMark geometric 'overlaps' operators (&&) as self-commutative.
Tom Lane [Sun, 31 Dec 2000 22:34:04 +0000 (22:34 +0000)]
Mark geometric 'overlaps' operators (&&) as self-commutative.

23 years agoFix typo in error message.
Tom Lane [Sun, 31 Dec 2000 22:24:14 +0000 (22:24 +0000)]
Fix typo in error message.

23 years agoDon't say 'export PGHOST' or 'export PGPORT' unless we actually define
Tom Lane [Sun, 31 Dec 2000 18:38:44 +0000 (18:38 +0000)]
Don't say 'export PGHOST' or 'export PGPORT' unless we actually define
those variables.  Some shells will invent an empty-string definition in
this case, which is not what we want.

23 years agoOn further thought, we need a defense against empty PGPORT here too.
Tom Lane [Sun, 31 Dec 2000 18:23:21 +0000 (18:23 +0000)]
On further thought, we need a defense against empty PGPORT here too.

23 years agoIgnore PGPORT environment variable if it is an empty string.
Tom Lane [Sun, 31 Dec 2000 18:15:58 +0000 (18:15 +0000)]
Ignore PGPORT environment variable if it is an empty string.

23 years agoReverse #if test to be defined(__osf__) rather than not-any-of-a-lot-
Tom Lane [Sun, 31 Dec 2000 18:04:35 +0000 (18:04 +0000)]
Reverse #if test to be defined(__osf__) rather than not-any-of-a-lot-
of-others.

23 years agoRemove incorrect use of rl_special_prefixes until further evaluation.
Peter Eisentraut [Sun, 31 Dec 2000 11:57:58 +0000 (11:57 +0000)]
Remove incorrect use of rl_special_prefixes until further evaluation.

23 years agoNetBSD/Alpha porting fixes from tom@minnesota.com.
Tom Lane [Sun, 31 Dec 2000 03:34:01 +0000 (03:34 +0000)]
NetBSD/Alpha porting fixes from tom@minnesota.com.

23 years agoBe more careful about the difference between signed and unsigned ints.
Tom Lane [Sat, 30 Dec 2000 19:17:47 +0000 (19:17 +0000)]
Be more careful about the difference between signed and unsigned ints.
Bug is revealed by OID regress test on 64-bit platforms.

23 years agoCorrect UNDER syntax.
Peter Eisentraut [Sat, 30 Dec 2000 19:11:45 +0000 (19:11 +0000)]
Correct UNDER syntax.

23 years agoCorrect UNDER syntax.
Peter Eisentraut [Sat, 30 Dec 2000 19:00:11 +0000 (19:00 +0000)]
Correct UNDER syntax.

23 years agoRefinements
Peter Eisentraut [Sat, 30 Dec 2000 17:11:32 +0000 (17:11 +0000)]
Refinements

23 years agoRemove C++ comment.
Peter Eisentraut [Sat, 30 Dec 2000 16:48:49 +0000 (16:48 +0000)]
Remove C++ comment.

23 years ago1. WAL needs in zero-ed content of newly initialized page.
Vadim B. Mikheev [Sat, 30 Dec 2000 15:19:57 +0000 (15:19 +0000)]
1. WAL needs in zero-ed content of newly initialized page.
2. Log record for PageRepaireFragmentation now keeps array
   of !LP_USED offnums to redo cleanup properly.

23 years agoAdd mention of sysctl(8) for IPC tuning on Linux.
Peter Eisentraut [Sat, 30 Dec 2000 15:03:09 +0000 (15:03 +0000)]
Add mention of sysctl(8) for IPC tuning on Linux.

23 years agoRemove incorrect assert.
Peter Eisentraut [Sat, 30 Dec 2000 14:47:06 +0000 (14:47 +0000)]
Remove incorrect assert.

23 years agoFixed misprint in heap update WALoging.
Vadim B. Mikheev [Sat, 30 Dec 2000 06:52:34 +0000 (06:52 +0000)]
Fixed misprint in heap update WALoging.

23 years agoImporve messages.
Tatsuo Ishii [Sat, 30 Dec 2000 06:10:43 +0000 (06:10 +0000)]
Imporve messages.

23 years agoApparently, special float8 comparison file for Alpha is only needed
Tom Lane [Sat, 30 Dec 2000 02:48:04 +0000 (02:48 +0000)]
Apparently, special float8 comparison file for Alpha is only needed
when using vendor cc, not gcc.

23 years agoClean up spinlock assembly code slightly (just cosmetic improvements)
Tom Lane [Sat, 30 Dec 2000 02:34:56 +0000 (02:34 +0000)]
Clean up spinlock assembly code slightly (just cosmetic improvements)
for Alpha gcc case.  For Alpha non-gcc case, replace use of
__INTERLOCKED_TESTBITSS_QUAD builtin with __LOCK_LONG_RETRY and
__UNLOCK_LONG.  The former does not execute an MB instruction and
therefore was guaranteed not to work on multiprocessor machines.
The LOCK_LONG builtins produce code that is the same in all essential
details as the gcc assembler code.

23 years agoParanoia about possible values of errno after a shmget/semget failure.
Tom Lane [Sat, 30 Dec 2000 01:20:55 +0000 (01:20 +0000)]
Paranoia about possible values of errno after a shmget/semget failure.
In theory we should always get EEXIST if there's a key collision, but
if the kernel code tests error conditions in a weird order, perhaps
EACCES or EIDRM could occur too.

23 years agoRemove obsolete and unportable enable_plpgsql script. createlang has
Tom Lane [Sat, 30 Dec 2000 00:50:44 +0000 (00:50 +0000)]
Remove obsolete and unportable enable_plpgsql script.  createlang has
been the supported and documented way to do this for a long time...

23 years agoFix unportable use of '!' in shell commands.
Peter Eisentraut [Sat, 30 Dec 2000 00:24:09 +0000 (00:24 +0000)]
Fix unportable use of '!' in shell commands.

23 years agocolumn and tuple numbers should be int not size_t.
Tom Lane [Fri, 29 Dec 2000 22:46:37 +0000 (22:46 +0000)]
column and tuple numbers should be int not size_t.

23 years agoFix failure in CreateCheckPoint on some Alpha boxes --- it's not OK to
Tom Lane [Fri, 29 Dec 2000 21:31:21 +0000 (21:31 +0000)]
Fix failure in CreateCheckPoint on some Alpha boxes --- it's not OK to
assume that TAS() will always succeed the first time, even if the lock
is known to be free.  Also, make sure that code will eventually time out
and report a stuck spinlock, rather than looping forever.  Small cleanups
in s_lock.h, too.

23 years agoMUST update (in-memory) data page BEFORE XLogInsert to log
Vadim B. Mikheev [Fri, 29 Dec 2000 20:47:17 +0000 (20:47 +0000)]
MUST update (in-memory) data page BEFORE XLogInsert to log
NEW page content if WAL will decide to backup page.

23 years agostamp-h needs to be made by config.status, not elsewhere, per recipe in
Tom Lane [Fri, 29 Dec 2000 20:39:09 +0000 (20:39 +0000)]
stamp-h needs to be made by config.status, not elsewhere, per recipe in
Autoconf manual.  In particular, touching it before creating config.status
is guaranteed to lose.

23 years agonbtree_xlog_newroot: set meta flag in meta page opaque.
Vadim B. Mikheev [Fri, 29 Dec 2000 08:08:59 +0000 (08:08 +0000)]
nbtree_xlog_newroot: set meta flag in meta page opaque.

23 years agoAttached are patches for two fixes to reduce memory usage by the JDBC
Bruce Momjian [Thu, 28 Dec 2000 23:56:46 +0000 (23:56 +0000)]
Attached are patches for two fixes to reduce memory usage by the JDBC
drivers.

The first fix fixes the PreparedStatement object to not allocate
unnecessary objects when converting native types to Stings.  The old
code used the following format:
        (new Integer(x)).toString()
whereas this can more efficiently be occompilshed by:
        Integer.toString(x);
avoiding the unnecessary object creation.

The second fix is to release some resources on the close() of a
ResultSet.  Currently the close() method on ResultSet is a noop.  The
purpose of the close() method is to release resources when the ResultSet
is no longer needed.  The fix is to free the tuples cached by the
ResultSet when it is closed (by clearing out the Vector object that
stores the tuples).  This is important for my application, as I have a
cache of Statement objects that I reuse.  Since the Statement object
maintains a reference to the ResultSet and the ResultSet kept references
to the old tuples, my cache was holding on to a lot of memory.

Barry Lind

23 years agoQualify %.in rule to avoid triggering on configure.in, repair unportable
Peter Eisentraut [Thu, 28 Dec 2000 17:34:52 +0000 (17:34 +0000)]
Qualify %.in rule to avoid triggering on configure.in, repair unportable
attempt to install more than one file per 'install' invocation, clean up
some other oddities.

23 years agoNew WAL version - CRC and data blocks backup.
Vadim B. Mikheev [Thu, 28 Dec 2000 13:00:29 +0000 (13:00 +0000)]
New WAL version - CRC and data blocks backup.

23 years agoLet's try this again on accepting the correct range of Oid input values
Tom Lane [Thu, 28 Dec 2000 01:51:15 +0000 (01:51 +0000)]
Let's try this again on accepting the correct range of Oid input values
for 64-bit platforms ...

23 years agoCorrect erroneous documentation of PQsetnonblocking().
Tom Lane [Thu, 28 Dec 2000 00:16:11 +0000 (00:16 +0000)]
Correct erroneous documentation of PQsetnonblocking().