(j3.2006) More on "contiguous"

Bill Long longb
Tue May 15 17:59:07 EDT 2007


It is certainly useful to know that more that the first dimension is 
contiguous.  This allows the compiler to collapse loops (actual or 
implied).  This is one of the primary reasons for contiguous.   If you 
know that the leftmost dimension of the array D is contiguous, but other 
dimensions might not qualify, then you could associate a contiguous 
pointer with D(:,i) and reference the pointer.

I'd be skeptical of the claim by gprof  that the statement D(:,i) = 0 
consumes 8.8% of a program's time unless the value of n is really large 
and there is not much else happening in the program.  The performance 
for that statement is probably limited by the hardware memory bandwidth, 
pretty much independent of how much optimizing a reasonable compiler is 
doing.

Cheers,
Bill


Van Snyder wrote:

>Does "contiguous" make any sense and confer any benefit for other than
>the first dimension?
>
>I have a dummy argument array D(:,:).  I know the actual argument is
>contiguous in the first dimension because it's a section A(:n,:) of an
>automatic variable.
>
>The line-by-line profile says I'm spending 8.8% of a 25 hour run on one
>statement "D(:,i) = 0.0" -- unless I'm not interpreting the profiler's
>output correctly.  I assume that by reporting line 312, gprof is telling
>me the basic block beginning (or ending?) at that line is what's being
>reported.  Since the previous line is a DO statement and the next one is
>an IF statement, I assume the reported time is all spent at this line.
>
>The array D wouldn't be contiguous as a whole according to our current
>definition, but I suspect it would be helpful if the processor knew the
>first dimension were contiguous.  Is it ever helpful to know that more
>dimensions are contiguous?  If not, can/should we change the definition
>of "contiguous" to apply only to the first dimension?
>
>  
>

-- 
Bill Long                                   longb at cray.com
Fortran Technical Support    &              voice: 651-605-9024
Bioinformatics Software Development         fax:   651-605-9142
Cray Inc., 1340 Mendota Heights Rd., Mendota Heights, MN, 55120

            





More information about the J3 mailing list