This is the mail archive of the gcc@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]

Re: How to avoid de-optimization


On Sun, Aug 26, 2001 at 05:07:29PM +0200, Jan Hubicka wrote:
> 
> Also many shift+add sequences simplifies back to leas.
> 
> Perhaps we can re-try on spec2000 benchmark, as it is larger scale, but I guess
> that in such benchmark the differences will not be measurable.
> 
> Do you have some testcases?
>
They are not public available.


Another problem I saw is that jumps are "exploding".

Normally on Intel jumps (JMP, Jcc) are 2 byte long.
If the distance become bigger than -128...+127 the jump instruction becomes
5 (JMP) or 6 (Jcc) byte long. This enlarges code so that other Jumps also
become bigger and so on ...

Especially if you have code with lots of if/while/for statements and less
arithmetic this can increase the code by additional 50 per cent.

-- 
Frank Klemm


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