[Bf-committers] How much ram is needed for compiling cuda kernels ?

[Sergey Sharybin]

The thing -- i gave up building sm_13 on 32bit environment for linux and
was using 64bit platform to build sm_13 32bit kernel. It just requires more
memory than 32bit platform could physically address. So afraid currently no
way you could compile sm_13 on a 32bit platform without tweaking the kernel.


On Thu, Apr 18, 2013 at 6:14 PM, Brecht Van Lommel <
brechtvanlommel at pandora.be> wrote:

> Hi,
>
> On Thu, Apr 18, 2013 at 1:17 PM, Marko Radojcic <sambucuself at gmail.com>
> wrote:
> > I regularly compile "my own" version of blender from svn and use prebuild
> > versions with CUDA and OpenCL support.
> >
> > Now I tried to compile the CUDA build myself.
> >
> > Although I have set the variable to use only "sm_21" kernel type, the
> build
> > step said it was building for compute capability 1.3, which is fine in
> > general, but it produces "memory allocation error" - at that moment the
> > process monitor said that "be" compiler is using 2.5 GB of RAM.
>
> We dropped official support for sm 1.3, partly because it is so
> difficult to compile for. Which build system are you using? Perhaps
> this is not properly configured.
>
> > How much RAM is required to build the kernels? - that is my main
> question.
> >
> > Is 2.5GB limit within the OS or the build environment ?
> >
> > I'm using gcc 4.6.2 and CUDA 5.0
>
> I don't know the number, it changes each time we add more features.
> For sm 2.x or newer RAM usage is a lot lower than sm 1.3 so I suggest
> just trying to find out how to configure things to avoid the issue, as
> building sm 1.3 on 32 bit is something we don't support anymore.
>
> Brecht.
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at blender.org
> http://lists.blender.org/mailman/listinfo/bf-committers
>



-- 
With best regards, Sergey Sharybin