| Commit message (Collapse) | Author | Age |
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4076 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
| |
This closes PR Cactus/1945: "CCTK_ActiveTimeLevels does not check its arguments".
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4075 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
| |
(unsigned int) instead of %d (signed int) to prevent a (theoretical)
buffer overflow on 32 bit machines.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4073 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4071 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
groups.
Whether the synchronisation was done by CCTK_CallFunction or not was
stored in the "synchronised" flag of a t_sched_data object. This
object exists only once for the simulation, and not once for each
recursive call of the scheduler when it traverses groups. Thus the
information was correct only for the innermost schedule item, namely
functions, and incorrect for schedule groups.
Because PUGH does not overload CCTK_CallFunction, and thus this flag
always stays false, I assume that this error does not occur with PUGH
simulations. It does occur with Carpet simulations.
As a quick solution, I moved the "synchronised" flag from the
t_sched_data structure into the t_attribute structure. This does not
work for recursive calls.
It may be that the t_attribute structure is meant for unchanging
information. In that case, it is necessary to construct a stack of
t_sched_data objects.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4060 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
|
|
| |
CCTK_TraverseString if a name is a valid group name, but the group has
zero variables. Instead, do nothing in this case, which is correct,
because the callback function should be called for each variable in
the group.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4057 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
|
| |
Use 'memcpy (&aq, &ap, sizeof (ap));' if it isn't available.
This fixes a linker problem with older C compilers (eg. PGI 5.1).
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4053 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
| |
rescaling the arguments. This is a standard algorithm.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4050 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4047 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
| |
http://cactus.cct.lsu.edu/pipermail/developers/2005-May/000920.html
http://cactus.cct.lsu.edu/pipermail/developers/2005-May/000938.html
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4044 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4028 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
| |
and potentially uninitialized variables.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4021 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4018 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4017 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
scheduling error neither aborts Cactus run nor gives useful error message
so scheduling errors now abort the Cactus run.
In particular, this commit changes the error-handling strategy of
CCTKi_DoScheduleSortAllGroups()
ScheduleSortGroup()
from "print a message to stderr, then return error code"
to CCTK_VWarn(CCTK_WARN_ABORT, ...), i.e.
"print a message to stderr, then abort the Cactus run".
The CCTKi_DoScheduleSortAllGroups() message is the same one as before.
The ScheduleSortGroup() message is changed to include the group name.
(N.b. the other part of Cactus/1908, that the schedule-error
message is uninformative, is *not* fixed by this commit.)
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4016 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4014 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4013 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
| |
since there are now fairly extensive latex docs
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4012 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
| |
will grok them (keep hex values in comments since they're convenient
for C bitwise-oring)
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4011 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
| |
(all inside #ifdef UTIL_TABLE_TEST)
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4010 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
|
|
|
| |
also add another error check in insert_table_entry()
to catch the case where the type code is invalid and hence
CCTK_VarTypeSize() returns -1
--> this gives a new error return from this function, and hence
from all the Util_TableSet* functions
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4008 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
|
| |
* add a bunch of debug printing to the standalone test drivers
(which still core-dumps on my laptop after corrupting the malloc heap;
I'm in the process of tracking down what's wrong...)
git-svn-id: http://svn.cactuscode.org/flesh/trunk@4006 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3995 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3994 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
| |
another comment.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3991 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
|
| |
cctk_Interp.h with CCODE ifdefs.
This allows Fortran code to use the same #def'd error codes as the C code.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3989 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3980 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3979 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
|
|
|
| |
Turn CCTK_STRING from a #define into a typedef.
Do not declare CCTK_CHAR in Fortran, as it does not make sense.
Add some comments.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3975 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3974 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
|
|
|
|
| |
variables, and CCTK_CHAR (like CCTK_STRING) only for other purposes,
e.g. tables.
* Introduce unique integer constants for them
* Add missing conversion functions from and to strings
* Add table functions for CCTK_BYTE.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3972 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3971 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3970 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
|
| |
them. Technically, Fortran has no character type, but only strings,
hence the interface to that routine needs to be different from C.
Util_TableSetString is still there, used, and works.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3967 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3966 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3962 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
| |
compiler warnings about the restrict qualifier.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3952 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
| |
also use int for lengths and not size_t.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3950 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3949 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
| |
Move #include statements below include guards.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3948 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
| |
-- it's Util_StrSep(), not CCTK_StrSep()
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3947 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3946 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
| |
Sorry.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3945 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
| |
same name as the one from <math.h> (which is #included) but has a
different return type, which is an error for gcc 4.0.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3944 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3941 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3940 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3939 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
| |
which turns on/off highlighting of CCTK warning messages using ANSI escape
sequences. Default is on.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3937 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
| |
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3936 17b73243-c579-4c4c-a9d2-2d5706c11dac
|
|
|
|
|
|
|
|
|
| |
source code.
They have integrated a fix to correctly convert floating point numbers
into their string representation.
git-svn-id: http://svn.cactuscode.org/flesh/trunk@3935 17b73243-c579-4c4c-a9d2-2d5706c11dac
|