Compilation fixes #70
Conversation
| @@ -1,21 +1,25 @@ | |||
| NUM_NODES = 10 | |||
| WORLD_SIZE = 1000 | |||
|
|
|||
| COMMON_CFLAGS = -g -std=gnu99 -O2 -mcpu=native -fomit-frame-pointer -Wall -Wextra | |||
| COMMON_CFLAGS = -std=gnu99 -O2 -march=native -fomit-frame-pointer -Wall -Wextra | |||
There was a problem hiding this comment.
-march=native is not equivalent to -mcpu=native. It needs to be -march=native -mtune=native.
There was a problem hiding this comment.
-march implies -mtune, see https://gcc.gnu.org/onlinedocs/gcc-4.3.2/gcc/i386-and-x86_002d64-Options.html
There was a problem hiding this comment.
It's not documented to work like that in the GCC ARM docs. Also, the options accepted by -march are generic architectures, while the options accepted by -mtune are specific cores.
There was a problem hiding this comment.
But previously the makefile didn't even specify -mtune. -march at least implies -mcpu.
|
@lgeek, I apologise for not making myself more clear. My intention with this pr was not to invest to much time in your LPathBench project. I just wanted to share the changes I needed to make it work well for me. Feel free to reject this pr, apply it or do whatever you want with it. Sorry, but I don't have either the time or the interest in working more on it. |
|
@bjourne Sorry if this is wasn't clear, but this is not my project. I just wrote the |
|
I think |
A few little compiler warnings cleanups and makefile improvements. I also think you should consider changing
-O2to-O3as that makes the c_fast benchmark much faster when compiled with clang. Clang in general beats gcc on my machine.