Re: Re: Flat Short of MEMORY_BLOCK

Brian Heilig:

Thanks for those tests!

> The times seem to indicate that Franck and Berend's tests were 
> invalid because `test' was optimized out.

Not with gcc: I set the repeat rate to take a few seconds to
execute on my machine, and it varied proportionally when I changed 
the number of repeats. Assembly checks confirm it looks like it's 
doing what is expected, so gcc is less clever than MS C here.

> memcpy (&buffer[i], &a_byte, how);
[...]
> array 1.490002
> memcpy 42.758552

Similar results with gcc.

> Inspection of the disassembly indicates that memcpy was not inlined. 
> My original fear was that eiffel code would also suffer the same 
> fate, that is, it would not be inlined. As you can see if memcpy is 
> not optimized out the results can be dramatic.

OK, this is the sort of case where it sounds genuinely 
useful :-).

> I suppose the best  answer would be to compare the benchmarks 
> of two MANAGED_POINTERs that are implemented using array 
> access and memcpy.

Sounds like a good idea. If array is faster, you may then 
look at using an auxiliary array (SPECIAL/NATIVE_ARRAY) 
inside the memory block implementation, maybe in addition 
to the MANAGED_POINTER, which may get the same result 
(compiling to a C array access) without backend 
dependencies.



To Post a message, send it to:   [email protected]
To Unsubscribe, send a blank message to: [email protected] 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
     http://groups.yahoo.com/group/gobo-eiffel/

<*> To unsubscribe from this group, send an email to:
     [email protected]

<*> Your use of Yahoo! Groups is subject to:
     http://docs.yahoo.com/info/terms/