buffer manager with 'pg_clog', a specialized access method modeled
on pg_xlog. This simplifies startup (don't need to play games to
open pg_log; among other things, OverrideTransactionSystem goes away),
should improve performance a little, and opens the door to recycling
commit log space by removing no-longer-needed segments of the commit
log. Actual recycling is not there yet, but I felt I should commit
this part separately since it'd still be useful if we chose not to
do transaction ID wraparound.
pgsql-hackers. pg_opclass now has a row for each opclass supported by each
index AM, not a row for each opclass name. This allows pg_opclass to show
directly whether an AM supports an opclass, and furthermore makes it possible
to store additional information about an opclass that might be AM-dependent.
pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we
previously expected the user to remember to provide in CREATE INDEX commands.
Lossiness is no longer an index-level property, but is associated with the
use of a particular operator in a particular index opclass.
Along the way, IndexSupportInitialize now uses the syscaches to retrieve
pg_amop and pg_amproc entries. I find this reduces backend launch time by
about ten percent, at the cost of a couple more special cases in catcache.c's
IndexScanOK.
Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane.
initdb forced.
has an alias SERIAL4 and a sister SERIAL8. SERIAL8 is just the same
except the created column is type int8 not int4.
initdb forced. Note this also breaks any chance of pg_upgrade from 7.1,
unless we hack up pg_upgrade to drop and recreate sequences. (Which is
not out of the question, but I don't wanna do it.)
Allow pg_shadow to be MD5 encrypted.
Add ENCRYPTED/UNENCRYPTED option to CREATE/ALTER user.
Add password_encryption postgresql.conf option.
Update wire protocol version to 2.1.
for speed reasons; its result type also changes to int8. avg() on these
datatypes now accumulates the running sum in int8 for speed; but we still
deliver the final result as numeric, so that fractional accuracy is
preserved.
count() now counts and returns in int8, not int4. I am a little nervous
about this possibly breaking users' code, but there didn't seem to be
a strong sentiment for avoiding the problem. If we get complaints during
beta, we can change count back to int4 and add a "count8" aggregate.
For that matter, users can do it for themselves with a simple CREATE
AGGREGATE command; the int4inc function is still present, so no C hacking
is needed.
Also added max() and min() aggregates for OID that do proper unsigned
comparison, instead of piggybacking on int4 aggregates.
initdb forced.
a tad sloppy about generating the targetlist for some nodes, by generating
a tlist entry that claimed to be a constant when the value wasn't actually
constant. This caused setrefs.c to do the wrong thing later on.
syntax for language names (instead of 'string').
createlang now handles the case where a second language uses the same call
handler as an already installed language (e.g., plperl/plperlu).
droplang now handles the reverse case, i.e., dropping a language where
the call handler is still used by another language. Moreover, droplang
can now be used to drop any user-defined language, not just the supplied
ones.
default, but OIDS are removed from many system catalogs that don't need them.
Some interesting side effects: TOAST pointers are 20 bytes not 32 now;
pg_description has a three-column key instead of one.
Bugs fixed in passing: BINARY cursors work again; pg_class.relhaspkey
has some usefulness; pg_dump dumps comments on indexes, rules, and
triggers in a valid order.
initdb forced.
(as proposed in http://fts.postgresql.org/db/mw/msg.html?mid=1028327)
2. support for 'pass-by-value' arguments - to test this
we used special opclass for int4 with values in range [0-2^15]
More testing will be done after resolving problem with
index_formtuple and implementation of B-tree using GiST
3. small patch to contrib modules (seg,cube,rtree_gist,intarray) -
mark functions as 'isstrict' where needed.
Oleg Bartunov
clauses are equal(), before trying to match them up using btree opclass
inference rules. This allows it to recognize many simple cases involving
non-btree operations, for example 'x IS NULL'. Clean up code a little.
Make sure it exits immediately when collector process dies --- in old code,
buffer process would hang around and compete with the new buffer process
for packets. Make sure it doesn't block on writing the pipe when the
collector falls more than a pipeload behind. Avoid leaking pgstats FDs
into every backend.
cvs.
The Debian bug report says, "The upstream source makes use of NOFILE
unconditionalized. As the Hurd doesn't have an arbitrary limit on the
number of open files, this is not defined. But _SC_OPEN_MAX works fine
and returns 1024 (applications can increase this as they want), so I
suggest the below diff. Please forward this upstream, too."
Oliver Elphick
platforms system(2) gets confused unless the signal handler is set to
SIG_DFL, not SIG_IGN. pgstats.c now uses pqsignal() as it should,
not signal(). Also, arrange for the stats collector process to show
a reasonable ID in 'ps', rather than looking like a postmaster.
that call is not needed to prepare for SO_PEERCRED. Also, simplify code
so that #ifdef SO_PEERCRED appears in only one place, to make it easier
to support other platforms with variants of this capability.
points out how silly it is to use Autoconf to test for a preprocessor
symbol, when one can equally easily #ifdef on the symbol itself.
Accordingly, revert configure to prior state and do it that way.
system supports SO_PEERCRED requests for Unix sockets. This is an
amalgamation of patches submitted by Helge Bahmann and Oliver Elphick,
with some editorializing by yours truly.
number in the data structure so that we can give at least a minimally
useful idea of where the mistake is when we issue syntax error messages.
Move the ClientAuthentication() call to where it should have been in
the first place, so that postmaster memory releasing can happen in a
reasonable place also. Update obsolete comments, correct one real bug
(auth_argument was not picked up correctly).
more care with resjunk tlist entries than it was doing. The original
coding ignored resjunk entries entirely, but a resjunk entry that is
in either the distinctClause or sortClause lists indicates that DISTINCT
ON was used. It's important for ruleutils.c to get this right, else we
may dump views using DISTINCT ON incorrectly.
has a DISTINCT ON clause, per bug report from Anthony Wood. While at it,
improve the DISTINCT-ON-clause recognizer routine to not be fooled by out-
of-order DISTINCT lists.
rather than deleting them only to have to create more. Steady state
is 2*CHECKPOINT_SEGMENTS + WAL_FILES + 1 segment files, which will
simply be renamed rather than constantly deleted and recreated.
To make this safe, added current XLOG file/offset number to page
header of XLOG pages, so that an un-overwritten page from an old
incarnation of a logfile can be reliably told from a valid page.
This change means that if you try to restart postmaster in a CVS-tip
database after installing the change, you'll get a complaint about
bad XLOG page magic number. If you don't want to initdb, run
contrib/pg_resetxlog (and be sure you shut down the old postmaster
cleanly).
in GetSnapshotData, GetNewTransactionId, CommitTransaction, AbortTransaction,
etc. Correct race condition in transaction status testing in
HeapTupleSatisfiesVacuum --- this wasn't important for old VACUUM with
exclusive lock on its table, but it sure is important now. All per
pghackers discussion 7/11/01 and 7/12/01.
Note: I didn't force an initdb, figuring that one today was enough.
However, there is a new function in pg_proc.h, and pg_dump won't be
able to dump partial indexes until you add that function.
per previous discussion on pghackers. Most of the duplicate code in
different AMs' ambuild routines has been moved out to a common routine
in index.c; this means that all index types now do the right things about
inserting recently-dead tuples, etc. (I also removed support for EXTEND
INDEX in the ambuild routines, since that's about to go away anyway, and
it cluttered the code a lot.) The retail indextuple deletion routines have
been replaced by a "bulk delete" routine in which the indexscan is inside
the access method. I haven't pushed this change as far as it should go yet,
but it should allow considerable simplification of the internal bookkeeping
for deletions. Also, add flag columns to pg_am to eliminate various
hardcoded tests on AM OIDs, and remove unused pg_am columns.
Fix rtree and gist index types to not attempt to store NULLs; before this,
gist usually crashed, while rtree managed not to crash but computed wacko
bounding boxes for NULL entries (which might have had something to do with
the performance problems we've heard about occasionally).
Add AtEOXact routines to hash, rtree, and gist, all of which have static
state that needs to be reset after an error. We discovered this need long
ago for btree, but missed the other guys.
Oh, one more thing: concurrent VACUUM is now the default.
validity checking rules for VACUUM. Make some other rearrangements of the
VACUUM code to allow more code to be shared between full and lazy VACUUM.
Minor code cleanups and added comments for TransactionId manipulations.
a lie on many Unixen), invoke listen() with MIN(MaxBackends*2, 10000).
The clamp value 10000 is configurable in config.h.in, if that proves
to be necessary --- hopefully it won't.
This makes VACUUM work properly with partial indexes, and avoids memory
leakage with functional indexes. Also, suppress complaint about fewer
index tuples than heap tuples when the index is a partial index.
From Martijn van Oosterhout.
USER and ALTER USER to appear in any order, not only the fixed order
they used to be required to appear in.
Also, some changes from Tom Lane to create a FULL option for VACUUM;
it doesn't do anything yet, but I needed to change many of the same
files to make that happen, so now seemed like a good time.
in cases of qualified rules as well as unqualified ones. Tweak rules
test to avoid cluttering output with dummy SELECT results. Update
documentation to match code.
constraint. This case (a) is useless, (b) violates SQL92, and
(c) is certain to cause a failure downstream when we try to create
an index with duplicated column names. So give an appropriate error
message instead of letting the index failure occur. Per report from
Colin Strickland. NOTE: currently, CREATE INDEX fooi ON foo(f1,f1)
still fails with 'cannot insert duplicate key' error. Should we
change that too? What about functional indexes?
useful as yet, since its primary source of information is (full) VACUUM,
which makes a concerted effort to get rid of free space before telling
the map about it ... next stop is concurrent VACUUM ...
immediately, we will fork a child even if the database state does not
permit connections to be accepted (eg, we are in recovery mode).
The child process will correctly reject the connection and exit as
soon as it's finished collecting the connection request message.
However, this means that reaper() must be prepared to see child
process exit signals even while it's waiting for startup or shutdown
process to finish. As was, a connection request arriving during a
database recovery or shutdown would cause postmaster abort.
stub) into the rest of the system. Adopt a cleaner approach to preventing
deadlock in concurrent heap_updates: allow RelationGetBufferForTuple to
select any page of the rel, and put the onus on it to lock both buffers
in a consistent order. Remove no-longer-needed isExtend hack from
API of ReleaseAndReadBuffer.
have any newly-dead tuples on them. This is a longstanding deficiency
that prevents VACUUM from compacting a file as much as one would expect.
Change requires fixing repair_frag to not assume that fraged_pages is
a subset of vacuum_pages.
Also make some further cleanups of places that assumed page numbers fit
in int and tuple counts fit in uint32.
do anything yet, but it has the necessary connections to initialization
and so forth. Make some gestures towards allowing number of blocks in
a relation to be BlockNumber, ie, unsigned int, rather than signed int.
(I doubt I got all the places that are sloppy about it, yet.) On the
way, replace the hardwired NLOCKS_PER_XACT fudge factor with a GUC
variable.
IS TRUE, etc, with some degree of verisimilitude. Split out
selectivity support functions from builtins.h into a new header
file selfuncs.h, so as to reduce the number of header files builtins.h
must depend on. Fix a few missing inclusions exposed thereby.
From Joe Conway, with some kibitzing from Tom Lane.
> > secure_ctx changes too. it will be PGC_BACKEND after '-p'.
>
> Oh, okay, I missed that part. Could we see the total state of the
> patch --- ie, a diff against current CVS, not a bunch of deltas?
> I've gotten confused about what's in and what's out.
Ok, here it is. Cleared the ctx comment too - after -p
it will be PGC_BACKEND in any case.
Marko Kreen
a new postmaster child process. This should eliminate problems with
authentication blocking (e.g., ident, SSL init) and also reduce problems
with the accept queue filling up under heavy load.
The option to send elog output to a different file per backend (postgres -o)
has been disabled for now because the initialization would have to happen
in a different order and it's not clear we want to keep this anyway.
tests to return the correct results per SQL9x when given NULL inputs.
Reimplement these tests as well as IS [NOT] NULL to have their own
expression node types, instead of depending on special functions.
From Joe Conway, with a little help from Tom Lane.
SI messages now include the relevant database OID, so that operations
in one database do not cause useless cache flushes in backends attached
to other databases. Declare SI messages properly using a union, to
eliminate the former assumption that Oid is the same size as int or Index.
Rewrite the nearly-unreadable code in inval.c, and document it better.
Arrange for catcache flushes at end of command/transaction to happen before
relcache flushes do --- this avoids loading a new tuple into the catcache
while setting up new relcache entry, only to have it be flushed again
immediately.
Here is Tomified version of my 2 pending patches.
Dropped the set_.._real change as it is not needed.
Desc would be:
* use GUC for settings from cmdline
Marko Kreen
Tom Lane
Bruce Momjian
Peter Eisentraut
Hiroshi Inoue
Tatsuo Ishii
Jan Wieck