[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gcc-patches
Subject:    Re: [PATCH] libfortran/29121 -- Add real(10) and real(16) cpu_time routines
From:       Steve Kargl <sgk () troutmask ! apl ! washington ! edu>
Date:       2006-09-29 21:48:03
Message-ID: 20060929214803.GA57761 () troutmask ! apl ! washington ! edu
[Download RAW message or body]

On Fri, Sep 29, 2006 at 11:40:53PM +0200, FX Coudert wrote:
> >I don't know a portable way to test cpu_time.
> 
> I was thinking about simply a compile & linking test, with cpu_time  
> called for both real and double precision, as well as a large real  
> kind if available. Getting actual values from cpu_time would be, as  
> you say, a pain.
> 

Ah, I see.  I think that such a test is of very limited usefulness.
In the end, gfortran is being used by more and more people, so
problems with intrinsic procedures will be caught.  Once fixed
it shouldn't regress.  Of course, if someone wants to write 
compile and link test for *every* intrinsic procedure that 
gfortran supports, I won't object. 

-- 
Steve
[prev in list] [next in list] [prev in thread] [next in thread]