[llvm-commits] [llvm] r169626 - in /llvm/trunk: include/llvm/MC/MCAssembler.h lib/MC/MCAssembler.cpp lib/MC/MCELFStreamer.cpp lib/MC/MCMachOStreamer.cpp lib/MC/MCObjectStreamer.cpp lib/MC/MCPureStreamer.cpp lib/MC/WinCOFFStreamer.cpp

[Rafael Espíndola]

> Thanks for doing the measurement. It's good to know that practice
> sometimes aligns with theory ;-)
> Can you share your measurement methodology, by the way? You ran
> llvm-mc on textual assembly produced by Clang? How many times did you
> run it / average / median? I'm asking because >1% differences in
> performance tend to be notoriously unreliable.

Sure. What I did was

* Close every other application. I still had X and deamons running,
but that was it.
* Disable address space randomization:
   sudo sysctl kernel.randomize_va_space=0
* produce the assembly:
  clang -S  -fno-verbose-asm gcc.c -w
* run the benchmark:
  sudo nice -n -20  perf  stat -r 10 llvm-mc -filetype=obj gcc.s -o
gcc.o -mc-relax-all

> Eli

Cheers,
Rafael