This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug rtl-optimization/23523] peephole2 causes code size on i686



------- Comment #11 from pinskia at gcc dot gnu dot org  2005-11-03 02:40 -------
(In reply to comment #10)
> I am not sure what kind of answer you expect here. 
> Speed and code size are not disjoint. Think about I-cache and I-TLB misses.
But again who is using an pentiumpro machine any more.  People who really care
about preformance in code really should be tunning for their machine.  Xterm is
not a good example of a program which really needs more speed.  Also you are
going to get a lot of Icache missed anyways as the machine is going to be
switching between processes more offten than actually doing anything with this
instruction.  Now if you can prove that using -march=pentium3 (or 4 depending
on your target) will be worse then just maybe.  But otherwise this is invalid. 
Yes most distros defaultly compile for pentiumpro (i686) but that is not GCC's
fault as this is actually faster for the pentumpro.


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |RESOLVED
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23523


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]