| 2010-03-14 |
Stefan Manegold |
fixes for HP-UX:
Nov2009
|
| 2010-03-14 |
Martin Kersten |
The garbage collection administration has been changed. Every variable record
|
| 2010-03-14 |
Martin Kersten |
The garbage collector optimizer should be the last in the pipelines.
|
| 2010-03-14 |
Martin Kersten |
The garbage collection administration has been changed. Every variable record
|
| 2010-03-14 |
Martin Kersten |
The garbage collection administration has been changed. Every variable record
|
| 2010-03-14 |
Martin Kersten |
The MAL debugger list command has been extended with
|
| 2010-03-14 |
Martin Kersten |
The MAL debugger list command has been extended with
|
| 2010-03-14 |
Martin Kersten |
The dataflow scheduler has been revamped to allow for
|
| 2010-03-14 |
Martin Kersten |
The garbage collection administration has been changed. Every variable record
|
| 2010-03-13 |
Niels Nes |
LLONG_MAX should be defined before compiling BUNlast (needs this because of
|
| 2010-03-12 |
Fabian Groffen |
optimisations step two (a little less trivial): push MAX/MIN down into the MAP program
|
| 2010-03-12 |
Fabian Groffen |
optimisations step one (still trivial): push ORDER BY [DESC], LIMIT and/or OFFSET down into the MAP program
|
| 2010-03-12 |
Martin Kersten |
Move history to the attic
|
| 2010-03-11 |
Niels Nes |
isspace needs int on hpux
|
| 2010-03-11 |
Sjoerd Mullender |
There is no test called "compress00".
|
| 2010-03-11 |
Fabian Groffen |
do what mat.pack and mergetable optimiser were supposed to do, sorting and other operations now seem to work correctly
|
| 2010-03-10 |
Niels Nes |
calib requires getpagesize, HPUX doesn't export it when
|
| 2010-03-10 |
Niels Nes |
Fix format string for PTRDIFF type on HPUX.
|
| 2010-03-10 |
Niels Nes |
find pthread libray requirements on HP UX
|
| 2010-03-10 |
Niels Nes |
add test for bug in in with or processing
Feb2010
|
| 2010-03-10 |
Niels Nes |
fix bug in IN handling
Feb2010
|
| 2010-03-10 |
Sjoerd Mullender |
Layout: I like it when opening braces of functions appear at the start
Feb2010
|
| 2010-03-10 |
Sjoerd Mullender |
propagated changes of Wednesday Mar 10 2010
M5XQ
|
| 2010-03-10 |
Sjoerd Mullender |
propagated changes of Wednesday Mar 10 2010
XQFT
|
| 2010-03-10 |
Sjoerd Mullender |
propagated changes of Tuesday Mar 09 2010 - Wednesday Mar 10 2010
|
| 2010-03-10 |
Sjoerd Mullender |
propagated changes of Wednesday Mar 10 2010
|
| 2010-03-10 |
Fabian Groffen |
do a two-phase approach were we first detect all sql.binds, such that we reliably get all columns detected; it seems that some operations are done inbetween the column binds
|
| 2010-03-10 |
Sjoerd Mullender |
propagated changes of Wednesday Mar 03 2010 - Wednesday Mar 10 2010
|
| 2010-03-10 |
Sjoerd Mullender |
propagated changes of Friday Mar 05 2010 - Wednesday Mar 10 2010
|
| 2010-03-10 |
Sjoerd Mullender |
propagated changes of Monday Mar 01 2010 - Wednesday Mar 10 2010
|
| 2010-03-10 |
Sjoerd Mullender |
propagated changes of Tuesday Mar 09 2010 - Wednesday Mar 10 2010
|
| 2010-03-10 |
Sjoerd Mullender |
propagated changes of Friday Mar 05 2010 - Wednesday Mar 10 2010
|
| 2010-03-10 |
Sjoerd Mullender |
Remove duplicate file.
Feb2010
|
| 2010-03-10 |
Fabian Groffen |
mapreduce optimizer v2.0: switch to pessimistic approach, allowing to execute any query, fetching only the data in the MAP phase. Mergetable optimiser is the one here that messes up any operation like sorting. From here on, the optimiser should try to find operations to push into the MAP phase.
|
| 2010-03-10 |
Fabian Groffen |
Add mvcRef, sort
|
| 2010-03-09 |
Jan Rittinger |
-- Performance fix in set property inference.
Feb2010
|
| 2010-03-09 |
Fabian Groffen |
add minimal_pipe definition, so one can easily switch to basic plans
|
| 2010-03-09 |
Fabian Groffen |
(hopefully) Fix problem found by Arjen in compiling the java sources. Set JAVA_HOME to the compiler we found with configure, such that ant will use that one too.
Feb2010
|
| 2010-03-09 |
Fabian Groffen |
Derive JAVA_HOME, which is essential in the Java module to get Ant to do what one would expect
Feb2010
|
| 2010-03-07 |
Martin Kersten |
Approve output
|
| 2010-03-07 |
Martin Kersten |
Also support compile time execution of barrier statement by turning them
|
| 2010-03-07 |
Martin Kersten |
Extra debugging for chasing possible race conditions, which may be
|
| 2010-03-06 |
Stefan Manegold |
approved new output:
|
| 2010-03-06 |
Stefan Manegold |
approved new output of `stethoscope --help`
|
| 2010-03-06 |
Stefan Manegold |
for the records:
|
| 2010-03-06 |
Martin Kersten |
Stethoscope has been moved to clients repository
|
| 2010-03-06 |
Martin Kersten |
Stethoscope help information test should be re-approved.
|
| 2010-03-06 |
Niels Nes |
bison/yacc rule may depend on direct sources now (ie x.y), not only on
|
| 2010-03-05 |
Fabian Groffen |
It's better to allocate some memory to write to. Change signature of generateSalt, since it doesn't allocate memory as I mistakenly thought it did.
Feb2010
|
| 2010-03-05 |
Sjoerd Mullender |
CRLF -> LF.
Feb2010
|
| 2010-03-05 |
Fabian Groffen |
Fix bug #2964247, thanks Sjoerd for debugging.
Feb2010
|
| 2010-03-05 |
Niels Nes |
we need mutils too
|
| 2010-03-05 |
Fabian Groffen |
check return of fopen() in more places, as pointed out by Mark Bucciarelli
Feb2010
|
| 2010-03-05 |
Fabian Groffen |
revert back to deprecated signal style, since windows doesn't seem to like sigaction at all
|
| 2010-03-05 |
Sjoerd Mullender |
Fix of previous fix. Deal with s == sub correctly (that implies s[-1]
Feb2010
|
| 2010-03-05 |
Sjoerd Mullender |
Avoid dereferencing a pointer before the start of the buffer.
Feb2010
|
| 2010-03-05 |
Sjoerd Mullender |
Let configure try linking with -lcrypto when testing -lssl.
Feb2010
|
| 2010-03-04 |
Fabian Groffen |
fix unstoppable behaviour with a stupid workaround: even if we have to monitor a single server, do it in a separate thread
|
| 2010-03-04 |
Fabian Groffen |
swith to new-style signal handlers, don't help to solve lockup up problems when running single threaded
|
| 2010-03-04 |
Fabian Groffen |
stethoscope was moved
|
