mask out underflow in fpe not inexact twice
don't catch underflow floating point exceptions
updated PETSc directory layout to match standard packaging strategiesinclude/petsc finclude,private,mpiunilib/petsc confbin/petsc*
update OPENMPI download filepetscerror_() was wrong for OpenMPI handling of Fortran MPI_Comm objects
manual merge of conflict
Merge branch 'barry/error-print-options'
Merge branch 'barry/fix-valgrind'
when unavailable LU or Cholesky is selected print error messagenow contains pointer to website with list of possible choices
traceback error message should print options databasethis is useful for use to help understand what has gone wrong with user code
valgrind include file should not be public to PETSc users by default
convert some PetscMalloc() to PetscMalloc1()remove unneeded () in first argument to PetscMalloc/Calloc1()fixes for MATLAB examples
Merge branch 'maint'
use xcode for building iOS dynamic library (framework)update opengles for new objective C rules
add newline to error statement
Sys: PetscCheckPointer for PETSC_CHAR checks char*, not char**By dereferencing char**, we access sizeof(void*) bytes of memory insteadof 1. This may be invalid if a short string is placed at the
Sys: PetscCheckPointer for PETSC_CHAR checks char*, not char**By dereferencing char**, we access sizeof(void*) bytes of memory insteadof 1. This may be invalid if a short string is placed at the end of apage and causes warnings from gcc -fsanitize.
show more ...
with with -> with the the -> the
Merge branch 'barry/configure-test-isatty'
Doc PetscPushErrorHandler: make arg order match call order
configure turns off use of isatty() on batch systems since it isatty() appears not to work correctly on cray andon batch systems anyways the output always has to go to files since outpus is not inte
configure turns off use of isatty() on batch systems since it isatty() appears not to work correctly on cray andon batch systems anyways the output always has to go to files since outpus is not interactiveReported-by: Jed Brown <jed@jedbrown.org>
double http:// in messageReported-by: "Jose E. Roman" <jroman@dsic.upv.es>
completed updating of example output for changine in ASCII viewing of option names, types etc
initial arrays that are published to SAWs, otherwise SAWs may access them before PETSc code has put correct values into them
only color the begin and end line for traceback error messages
Shorten default error message printed by PETSc to hopefully get people to read it and email it allEnclosed entire message with ---------- linesColor the begin and end lines and error message :-)
Shorten default error message printed by PETSc to hopefully get people to read it and email it allEnclosed entire message with ---------- linesColor the begin and end lines and error message :-) Jed and Satish will hate this, I am open to removing it or making it optional?
more casting to double into print statements
1...<<11121314151617181920>>...23