Hi Russel,
That's really interesting to note. Would you be willing to submit a
performance test that demonstrates this (after submitting a
contributor's agreement, of course)? Guidelines for how to do so can be
found in $CHPL_HOME/doc/developer/bestPractices/TestSystem.txt under the
heading "Creating Performance tests".
Thanks,
Lydia
On 10/16/2015 01:58 AM, Russel Winder wrote:
Vassily,
Thank you much for reporting this bug. We will fix it in Git.
Meanwhile feel free to apply this patch and rebuild the compiler -
your program should compile correctly:
[...]
Please let us know if this works for you.
I just pulled from the Git mainline and it includes the change you
suggest (with a bit of explanatory text as well :-) I have recompiled
and reinstalled Chapel, and recompiled my (very tiny) Chapel codes, and
everything is now compiling.
I wonder if it might be worthy of note that this construct seems to
create parallel loops that appear to be 10 times faster than any other
Chapel, C++, Fortran,... construct.
------------------------------------------------------------------------------
_______________________________________________
Chapel-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/chapel-bugs
------------------------------------------------------------------------------
_______________________________________________
Chapel-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/chapel-bugs
