search for: getelemenptr

...nvolving at least two multiplications and one addition) hence it actualy should have been moved out of the inner loop. At the very least inside the loop it should have been calculated once , and not twice. Anyway this is just a syptom of a general phenomenon in which it seems my LLVM assumes that getelemenptrs are atomic and it does not try to do any optimization on them. Can someone tell me if there is a known solution /existing code transformation that does the optimization ?

...elementptr instructions > is quite expensive (involving at least two multiplications and one > addition) hence it actualy should have been moved out of the inner loop. Right. > and not twice. Anyway this is just a syptom of a general > phenomenon in which it seems my LLVM assumes that getelemenptrs are > atomic and it does not try to do any optimization on > them. LLVM assumes that the code generator will handle this optimization when the getelementptr's are lowered into integer arithmetic. Many things can only be optimized after lowering in the code generator, and LICM of simp...

.../lists.llvm.org/pipermail/llvm-dev/2018-February/121295.html) that the 6.0.0 final tag won't be made by the 21st of Feb (as in the original release schedule), but the hope is to have rc3 tagged by Friday with [release blockers](llvm.org/PR35804) fixed. ## LLVM commits * The width of the GEP (getelemenptr) index used in address calculation can now be specified using the data layout string. [r325102](http://reviews.llvm.org/rL325102). * LLVM will now try harder to report a fatal error when out of memory (when memory allocation fails). [r325426](http://reviews.llvm.org/rL325426). * The X86 disassemb...