- changed status to open
- removed comment
Support CCTK_REAL4 in AHFinderDirect
Issue #846
closed
Index: src/include/config.h =================================================================== --- src/include/config.h (revision 1563) +++ src/include/config.h (working copy) @@ -54,8 +54,15 @@ * FIXME: this assumes fp == C 'double' * CCTK_REAL_PRECISION_{4,8,16} are helpful, but not quite enough) */ +#include <cctk.h> #undef FP_IS_FLOAT +#undef FP_IS_DOUBLE +#ifdef CCTK_REAL_PRECISION_4 +#define FP_IS_FLOAT +#endif +#ifdef CCTK_REAL_PRECISION_8 #define FP_IS_DOUBLE +#endif
#if defined(FP_IS_FLOAT)
#define FP_SCANF_FORMAT "%f"
Keyword:
Comments (7)
-
reporter
-
- removed comment
Wouldn't this patch invalidate the comment directly above it? Otherwise it looks ok to apply, but I didn't test. I don't have a configuration using CCTK_REAL4 for testing right now.
-
reporter
- removed comment
The code still doesn't test whether a CCTK_REAL8 is really a double, hence the comment remains valid. Of course, this is only of theoretical interest, since it would be very strange to have a machine where a double does not use 8 bytes.
-
reporter
- changed milestone to ET_2012_05
- removed comment
-
- changed status to open
- removed comment
Please apply
-
reporter
- changed status to resolved
- removed comment
Applied.
-
- edited description
- changed status to closed
- Log in to comment
- Assignee
- –
- Type
- Priority
- Status
- closed
- Component
- EinsteinToolkit thorn
- Milestone
- ET_2012_05
- Version
- –
- Votes
- 0
- Watchers
- 0
