| 1 |
Andrew Savchenko <bircoph@g.o> wrote: |
| 2 |
> |
| 3 |
> That's why kernel makes sure that no floating point instructions |
| 4 |
> sneaks in using CFLAGS, you may see a lot of -mno-${intrucion_set} |
| 5 |
> flags when running make -V. |
| 6 |
|
| 7 |
So it should be sufficient that the kernel does not use "float" |
| 8 |
or "double", shouldn't it? |
| 9 |
I can hardly imagine that otherwise the compiler converts integer |
| 10 |
or pointer arithmetic into floating point arithmetics, or is |
| 11 |
this really the case for certain flags? If yes, why should these |
| 12 |
flags *ever* be useful? |
| 13 |
I mean: The context switching happens for non-kernel code as well, |
| 14 |
doesn't it? |
Archives