| 1 |
Thank you. I'll stick to -march=native for now |
| 2 |
|
| 3 |
Walter Dnes <waltdnes@××××××××.org> wrote: |
| 4 |
|
| 5 |
>On Fri, Dec 19, 2014 at 02:38:48AM +0400, German wrote |
| 6 |
>> Couldn't find those in documentation. Thank you |
| 7 |
> |
| 8 |
> If you're building on the target machine, use the "native" CFLAG. It |
| 9 |
>has been around for a while. It detects the CPU, and builds for it |
| 10 |
>"automagically". You don't have to do any more grunt work, figuring out |
| 11 |
>the flags for your CPU. Computers are supposed to do the hard work. I |
| 12 |
>use... |
| 13 |
> |
| 14 |
>FLAGS="-O2 -march=native -mfpmath=sse -fomit-frame-pointer -pipe -fno-unwind-tables -fno-asynchronous-unwind-tables" |
| 15 |
>CXXFLAGS="${CFLAGS}" |
| 16 |
> |
| 17 |
> Mind you, if you're cross-compiling on another system, and then moving |
| 18 |
>the binaries over, you will have to figure out the correct flags. See |
| 19 |
>below for a method. |
| 20 |
> |
| 21 |
> Another problem is that there are also cpu-specific USE flags. You |
| 22 |
>can get a good start on figuring them out, as well as CFLAGS, by running |
| 23 |
> |
| 24 |
>grep flags /proc/cpuinfo |
| 25 |
> |
| 26 |
>on the target machine. There will be one line of output for each core. |
| 27 |
>You'll have multiple identical lines of output. |
| 28 |
> |
| 29 |
>-- |
| 30 |
>Walter Dnes <waltdnes@××××××××.org> |
| 31 |
>I don't run "desktop environments"; I run useful applications |
| 32 |
> |
Archives