Luca Salucci <lucasalu...@libero.it> wrote on 01/28/2014 07:33:30 AM:
>
> Hello! I am experimenting with X10 and I am mainly focused on the
> GPU support. My project has multiple files and some of them contain
> methods annotated with the @CUDA directive, so they should be
> compiled with the -NO_CHECKS -STATIC_CHECKS option. On the other
> hand I DO NOT want to compile my main .x10 file (which uses those
> with the @CUDA) with these options.
>
> What is the proper way to achieve this?
>
> I tried to compile the @CUDA files at first then the main without -
> NO_CHECKS -STATIC_CHECKS, but during the second compilation it
> recompiles everything.
>
Hi,
In many cases it will actually work just fine to compile the CUDA
files normally (without -NO_CHECKS and -STATIC_CHECKS). This is the
simplest thing to do, so I would try that first.
We actually could (should) make the compiler a little bit smarter
about how it is generating CUDA kernels and then we would automatically
disable generation of checks inside the kernels. I may try to get this
fixed for the next release of X10.
--dave------------------------------------------------------------------------------
WatchGuard Dimension instantly turns raw network data into actionable
security intelligence. It gives you real-time visual feedback on key
security issues and trends. Skip the complicated setup - simply import
a virtual appliance and go from zero to informed in seconds.
http://pubads.g.doubleclick.net/gampad/clk?id=123612991&iu=/4140/ostg.clktrk
_______________________________________________
X10-users mailing list
X10-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/x10-users
