Re: making the new if-converter not mangle IR that is already vectorizer-friendly

2015-07-23 Thread Abe
[Abe wrote:] Ideally, I/we fix the above problem -- and the rest of the regressions in the new if converter -- [Alan wrote:] OK, what are these regressions? Each of these files` test cases currently fails to be autovectorized: gcc/testsuite/gcc.dg/vect/pr61194.c gcc/testsuite/gcc.dg/ve

RE: making the new if-converter not mangle IR that is already vectorizer-friendly

2015-07-23 Thread Thomas Preud'homme
> From: gcc-ow...@gcc.gnu.org [mailto:gcc-ow...@gcc.gnu.org] On > Behalf Of Abe > Sent: Wednesday, July 22, 2015 4:20 AM > > [Abe wrote:] > >> The preceding makes me wonder: has anybody considered adding an > optimization > >> profile for GCC, […] that optimizes for the amount of energy > consumed

RETURN_ADDRESS_POINTER_REGNUM Macro

2015-07-23 Thread Ajit Kumar Agarwal
All: >From the description of the definition of the macro >RETURN_ADDRESS_POINTER_REGNUM , it is derived that this macro is used to Define a register for the above macro that helps in getting the return address from the stack or frame pointer. I could see many of the architectures supported by