Hi all, I am getting unimplemented instruction warnings followed by an infinite loop. After some research I found out that x87 instructions are unimplemented and I should pass some flags to compiler ( http://www.mail-archive.com/gem5-users@gem5.org/msg04837.html). I compiled my binary to use SSE instructions as suggested with the following flags.
-mfpmath=sse -march=native -mfpmath=sse I still got stuck in infinite loop. When I look at gcc documentation, ( http://gcc.gnu.org/onlinedocs/gcc/i386-and-x86_002d64-Options.html) *For the i386 compiler, you must use -march=cpu-type, -msse or -msse2switches to enable SSE extensions and make this option effective. For the x86-64 compiler, these extensions are enabled by default.* I am using 64bit gcc and so those flags shouldn't matter. *What ISAs should I specify to compiler to get rid of these warnings? I have cheked maillist but there are no solutions related to this issue. How do you guys workaround this??* For side note; I am trying to run NAS parallel benchmarks on gem5. The code is written in fortran, c (uses openmp) and I compile it using gfortran, gcc respectively. Instruction below are unimplemented. warn: instruction 'wbinvd' unimplemented warn: instruction 'fldcw_Mw' unimplemented warn: instruction 'fwait' unimplemented Thanks Mehmet
_______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
