| 1 |
James wrote: |
| 2 |
> Hello, |
| 3 |
> |
| 4 |
> |
| 5 |
> Current CFLAGS="-O2 -march=i686 -pipe" |
| 6 |
> |
| 7 |
> |
| 8 |
> I want to add "-fomit-frame-pointer" to my |
| 9 |
> CFLAGS on an existing system that has been running |
| 10 |
> for months. |
| 11 |
> |
| 12 |
> |
| 13 |
> Is this safe or do I have to rebuild everything with |
| 14 |
> somelike "emerge -e" ? |
| 15 |
> |
| 16 |
> |
| 17 |
> Should it be avoided completely? |
| 18 |
> |
| 19 |
> James |
| 20 |
> |
| 21 |
> |
| 22 |
> |
| 23 |
> |
| 24 |
> |
| 25 |
Hi, |
| 26 |
|
| 27 |
I am not too good in English but I alway understood that: |
| 28 |
|
| 29 |
|-fomit-frame-pointer| |
| 30 |
Don't keep the frame pointer in a register for functions that don't |
| 31 |
need one. This avoids the instructions to save, set up and restore |
| 32 |
frame pointers; it also makes an extra register available in many |
| 33 |
functions. *It also makes debugging impossible on some machines.* |
| 34 |
|
| 35 |
On some machines, such as the VAX, this flag has no effect, because |
| 36 |
the standard calling sequence automatically handles the frame |
| 37 |
pointer and nothing is saved by pretending it doesn't exist. The |
| 38 |
machine-description macro |FRAME_POINTER_REQUIRED| controls whether |
| 39 |
a target machine supports this flag. See Register Usage |
| 40 |
<http://gcc.gnu.org/onlinedocs/gcc-4.1.2/gccint/Registers.html#Registers>. |
| 41 |
|
| 42 |
|
| 43 |
Enabled at levels -O, -O2, -O3, -Os. |
| 44 |
|
| 45 |
|
| 46 |
means, that fomit-frame-pointer is already active when -o ... is choose. |
| 47 |
|
| 48 |
see |
| 49 |
http://gcc.gnu.org/onlinedocs/gcc-4.1.2/gcc/Optimize-Options.html#Optimize-Options |
| 50 |
|
| 51 |
-- |
| 52 |
gentoo-user@l.g.o mailing list |
Archives