MPILIB: disable usage of floating point registers on parisc
authorHelge Deller <deller@gmx.de>
Sun, 5 May 2013 22:08:22 +0000 (22:08 +0000)
committerHelge Deller <deller@gmx.de>
Fri, 24 May 2013 20:30:11 +0000 (22:30 +0200)
commit70ef5578dd8011eeafa999a18194d9b548a4d889
tree9d6c403711e784ccca4a38761435424e44c27306
parent17fdfd0851617b6c18c0913364caf2a54171ce85
MPILIB: disable usage of floating point registers on parisc

The umul_ppmm() macro for parisc uses the xmpyu assembler statement
which does calculation via a floating point register.

But usage of floating point registers inside the Linux kernel are not
allowed and gcc will stop compilation due to the -mdisable-fpregs
compiler option.

Fix this by disabling the umul_ppmm() and udiv_qrnnd() macros. The
mpilib will then use the generic built-in implementations instead.

Signed-off-by: Helge Deller <deller@gmx.de>
lib/mpi/longlong.h