[Bug other/21319] Please add the feature of switching target at runtime

2005-05-01 Thread pinskia at gcc dot gnu dot org

--- Additional Comments From pinskia at gcc dot gnu dot org  2005-05-01 
16:02 ---
This was a very long term goal of the hooking GCC but then it was proved that 
it slowed down the 
compiler too much.

-- 
   What|Removed |Added

 Status|UNCONFIRMED |NEW
 Ever Confirmed||1
   Last reconfirmed|-00-00 00:00:00 |2005-05-01 16:02:19
   date||


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21319


[Bug other/21319] Please add the feature of switching target at runtime

2005-05-01 Thread radek at podgorny dot cz

--- Additional Comments From radek at podgorny dot cz  2005-05-01 16:13 
---
Slowed down? I'm not talking about compliling ALL the stuff in, just what I
need. Like configure --with-arch-i686 --with-arch-arm...


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21319


[Bug other/21319] Please add the feature of switching target at runtime

2005-05-01 Thread pinskia at gcc dot gnu dot org

--- Additional Comments From pinskia at gcc dot gnu dot org  2005-05-01 
16:23 ---
(In reply to comment #2)
 Slowed down? I'm not talking about compliling ALL the stuff in, just what I
 need. Like configure --with-arch-i686 --with-arch-arm...

Yes that would slow down the compiler as right now we have macros for some 
target dependent 
controls and then they have to be changed to function calls which slow down the 
compiler.

I think it slowed down the compiler about 1-5% IIRC.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21319


[Bug other/21319] Please add the feature of switching target at runtime

2005-05-01 Thread radek at podgorny dot cz

--- Additional Comments From radek at podgorny dot cz  2005-05-01 16:35 
---
Oh, I see. Anyway, is 1-5% a major slowdown? I think it's still a good penalty
if it saves all the headaches...


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21319


[Bug other/21319] Please add the feature of switching target at runtime

2005-05-01 Thread radek at podgorny dot cz

--- Additional Comments From radek at podgorny dot cz  2005-05-01 17:05 
---
...anyway, I'm not too educated at this, but wouldn't inlined function solve 
this?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21319