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]
Other format: [Raw text]

Re: gcc compile-time performance


In message <m3bsbd1m65.fsf@merlin.nerim.net>, Gabriel Dos Reis writes:
 > dewar@gnat.com (Robert Dewar) writes:
 > 
 > | > Whether your Ada front-end got faster overthe year doesn't imply that
 > | > g++ also did.  Actually the C++ front-end got slower.  That is a fact;
 > | > and people keep making report about that.  You can't declare they're
 > | > dreaming just because your Ada front-end got faster.
 > | 
 > | 
 > | No, you missed what I said, which is that assuming I am always working on
 > | latest hardware, the peformance of all parts of GCC improves  overall.
 > 
 > I didn't miss it.  I even point out that the assumption of working on
 > the lastest hardware is an unrealistic.  What you're missing is that
 > we have hard data from reality.  
Even if you are working on the latest and greatest hardware, compile time is
an issue.  Red Hat has customers that have large, expensive, multi-processor,
multi-machine build farms for their software and it still takes on the order
of days to rebuild their systems. 

jeff


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