longjmp()/setjmp() difference between gcc 3.3 (Apple) and gcc 3.4.x

[Andrew Pinski]

On Jan 22, 2005, at 12:41 PM, Remy X.O. Martin wrote:

> 1) Any idea why the setjmp() call is so expensive on a G5, and is 
> there a better, less expensive way I could implement the catch 
> mechanism?


This one I can answer.   Well on a G5, the registers are twice as big, 
64bit vs 32bit so
you are storing twice as much data.

The others I cannot answer right away because you have not given a good 
testcase.

I would file a bug: http://gcc.gnu.org/bugs.html and provide all the 
needed
information.

-- Pinski