http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54151
Bug #: 54151
Summary: Comparing if float is equal to decimal constant on
Microblaze with mhard-float results in internal
compiler error.
Classification: Unclassified
Product: gcc
Version: 4.7.1
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c++
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: cyberwizzard+...@gmail.com
Created attachment 27920
--> http://gcc.gnu.org/bugzilla/attachment.cgi?id=27920
Test case triggering the internal compiler error
When determining if a floating point variable is equal to a decimal constant
(without float typing) the Microblaze GCC with 'mhard-float' throws an internal
error: test.cc:1:5: internal compiler error: in prepare_float_lib_cmp, at
optabs.c:4272
Host sytem used to build GCC 4.7.1 for Microblaze: Ubuntu 12.04 64-bit.
The example compiles fine on x86_64 and on Microblaze without 'mhard-float'.
Example triggering the internal error:
int main(){
float a;
if(a==1.0) { return 1; }
return 0;
}
Output of g++ for Microblaze:
mb-g++ -v -save-temps -mhard-float 'test.cc'
Using built-in specs.
COLLECT_GCC=/home/cyberwizzard/Documents/gcc/install/bin/mb-g++
COLLECT_LTO_WRAPPER=/home/cyberwizzard/Documents/gcc/install/libexec/gcc/microblaze-xilinx-elf/4.7.1/lto-wrapper
Target: microblaze-xilinx-elf
Configured with: ../configure --target=microblaze-xilinx-elf
--program-prefix=mb- --prefix=/home/cyberwizzard/Documents/gcc/install
--enable-languages=c --disable-nls --without-headers --disable-libssp
--with-newlib : (reconfigured) ../configure --target=microblaze-xilinx-elf
--program-prefix=mb- --prefix=/home/cyberwizzard/Documents/gcc/install
--enable-languages=c,c++ --disable-nls --without-headers --disable-libssp
--with-newlib
Thread model: single
gcc version 4.7.1 (GCC)
COLLECT_GCC_OPTIONS='-v' '-save-temps' '-mhard-float'
/home/cyberwizzard/Documents/gcc/install/libexec/gcc/microblaze-xilinx-elf/4.7.1/cc1plus
-E -quiet -v test.cc -mhard-float -fpch-preprocess -o test.ii
ignoring nonexistent directory
"/home/cyberwizzard/Documents/gcc/install/lib/gcc/microblaze-xilinx-elf/4.7.1/../../../../microblaze-xilinx-elf/sys-include"
#include "..." search starts here:
#include <...> search starts here:
/home/cyberwizzard/Documents/gcc/install/lib/gcc/microblaze-xilinx-elf/4.7.1/../../../../microblaze-xilinx-elf/include/c++/4.7.1
/home/cyberwizzard/Documents/gcc/install/lib/gcc/microblaze-xilinx-elf/4.7.1/../../../../microblaze-xilinx-elf/include/c++/4.7.1/microblaze-xilinx-elf
/home/cyberwizzard/Documents/gcc/install/lib/gcc/microblaze-xilinx-elf/4.7.1/../../../../microblaze-xilinx-elf/include/c++/4.7.1/backward
/home/cyberwizzard/Documents/gcc/install/lib/gcc/microblaze-xilinx-elf/4.7.1/include
/home/cyberwizzard/Documents/gcc/install/lib/gcc/microblaze-xilinx-elf/4.7.1/include-fixed
/home/cyberwizzard/Documents/gcc/install/lib/gcc/microblaze-xilinx-elf/4.7.1/../../../../microblaze-xilinx-elf/include
End of search list.
COLLECT_GCC_OPTIONS='-v' '-save-temps' '-mhard-float'
/home/cyberwizzard/Documents/gcc/install/libexec/gcc/microblaze-xilinx-elf/4.7.1/cc1plus
-fpreprocessed test.ii -quiet -dumpbase test.cc -mhard-float -auxbase test
-version -o test.s
GNU C++ (GCC) version 4.7.1 (microblaze-xilinx-elf)
compiled by GNU C version 4.6.3, GMP version 4.3.2, MPFR version 2.4.2,
MPC version 0.8.1
GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072
GNU C++ (GCC) version 4.7.1 (microblaze-xilinx-elf)
compiled by GNU C version 4.6.3, GMP version 4.3.2, MPFR version 2.4.2,
MPC version 0.8.1
GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072
Compiler executable checksum: 4290bd7345f2e2cf3d0e24567cded0ba
test.cc: In function 'int main()':
test.cc:1:5: internal compiler error: in prepare_float_lib_cmp, at
optabs.c:4272
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
