The input file that is created while running monte carlo runs is done
in 3 different places. There was no easy way to collect all of those
lines in one place, so I duplicated the lines in the dryrun function.
So the previous commit broke on the mac. There were missing changes in
TranslationUnitVisitor.cpp. Also there are fixes for clang 3.9 that
were not included yet.
The use of True/False prevents the plotting of booleans, as those values
cannot be parsed as doubles. While it's conceptually more pleasing to
use the actual Python keywords for booleans, numbers will work just as
well and allow plotting. #337
OK, getting all strings that don't end in .so or .a is not good. Lots of
problems. I made a list of file suffixes that include c and c++ source
files now.
There are 2 bugs. The first in ICG.pm did not look for dependencies
that included "src" in the name correctly. The code was inserting the
"src" in the incorrect place. The second bug is in mis_dep.pm. That
bug did not include non ".o" files from the beginning.
Created a new executive job that waits for threads to finish and readies them
for their next frame of execution. Created a new job class system_thread_sync
after the top of frame jobs and before the input processor is run to sync the
threads. Along the way cleaned up instrumentation jobs on the threads to fix#290.
The return code from parsing the monte carlo input was always set to zero and not checked.
Retured the return code from parsing the input. Added an exit if the input was not parsed
correctly.
Tested the name of the header file right when a class is starting processing. If
the class is in a file that is in an ICG_EXCLUDE directory, we quit processing
right there.
refs #251
To support backwards compatibility with measurement units I added the udunit symbol
names that we don't use into 13/15. This includes "rad" instead of "r" and
"degree" instead of "d".
refs #231
Found that when we hex pack numbers out of the random number generator we should zero pad
these numbers to 16 digits. Otherwise if the resulting number is an odd number of digits we
get an error. We already do this for file data.
refs #236
ICG dies on a strdup if the the input file is empty. Added a check for
an empty input file name. We exit with an error message if there is
no input file given.
refs #229
Added a test to set the debugger command to gdb or lldb depending
on availability. Created a new script for lldb to run. Modified
the script for gdb to backtrace all threads.
refs #227
It's been a while since we tested 15.x on the mac. The makefile
for ICG needed updating to what is essentially in 16. Added the
one change required for clang 3.8. Updated the version to 15.2.0
in trick_ver.txt. Finally removed the check for libclang.a. The
more recent clang releases are delivering libclang.dylib and I'm
too lazy to make big changes in the configure script to search
for it. The header file check should be enough.
refs #228
Backporting addition to suspend accepting new variable server connections and
all communications to variable server clients for the duration of reloading
a checkpoint
refs #168
The makefiles need to compile the units directory and they were
not at this time. The units need to be compiled in the native
bit size regardless of whether we are forcing 32 bit or not.
refs #225
