Re: [sc-dev] Building From Source on Raspberry Pi

[Dan Stowell <danstowell+sc3@xxxxxxxxx>]

2013/11/5 Patrick Beard <pcbeard@xxxxxxx>:
>
> On Nov 5, 2013, at 1:33 AM, Dan Stowell <danstowell+sc3@xxxxxxxxx> wrote:
>
> 2013/11/5 Patrick Beard <pcbeard@xxxxxxx>:
>
> When I delete the .o files, and rebuild them on the pi, the programs don’t
> crash.
>
> I might try the swap-file route. Thanks for the pointers. My secondary goal
> is to get a more recent build of the llvm toolchain built.
>
> I just noticed these:
>
> http://packages.debian.org/search?keywords=sclang1
> http://packages.debian.org/search?keywords=scsynth1
>
> These might just be the ticket.
>
>
> Those are old. These are newer:
> http://packages.debian.org/search?keywords=supercollider-language
> http://packages.debian.org/search?keywords=supercollider-server
>
>
> Thanks!
>
> Do they included dependencies for working versions of jack? I have yet to
> make a combination of jack + alsa_out make even the slightest chirp.

The dependencies are standard "jackd" which can be provided by jack1
or jack2. We do nothing special beyond that.

> I’ve been independently trying to test jack with ecasound. I can play a .wav
> file with ecasound like this:
>
> ecasound -i test.wav -o alsa
>
> But with
>
> ecasound -i test.wav -o jack
>
> and jackd is running like this:
>
> jackd -p 8 -d dummy
>
> and alsa_out is running like this:
>
> alsa_out -q1
>
> No audio ever comes out.

I had that exact same problem on arm a while back. I'm afraid I just
stopped at that point. But that was before rpi even existed, so the
people playing with rpi probably know the answer.

> By the way, to build on the pi, I had to make the following change:
>
> diff --git a/CMakeLists.txt b/CMakeLists.txt
> index dd4fc47..57463e2 100644
> --- a/CMakeLists.txt
> +++ b/CMakeLists.txt
> @@ -108,8 +108,8 @@ option(FINAL_BUILD "Build as single source file." OFF)
>  option(FFT_GREEN "Use internal 'Green' FFT lib rather than FFTW. (Not
> recommended.)" OFF)
>
> -option(SSE "Compile with support for SSE instructions." ON)
> -option(SSE2 "Compile with support for SSE2 instructions." ON)
> +option(SSE "Compile with support for SSE instructions." OFF)
> +option(SSE2 "Compile with support for SSE2 instructions." OFF)
>
> Looks like somebody hard-coded Intel flags, which of course make gcc barf on
> arm. Perhaps those could be made conditional on the architecture?

Patches welcome - conditional would be nice :)

Dan
-- 
http://www.mcld.co.uk

_______________________________________________
sc-dev mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: https://listarc.bham.ac.uk/marchives/sc-dev/
search: https://listarc.bham.ac.uk/lists/sc-dev/search/