[LAD] Floating point Denormals: C++ and Waf

David Robillard d at drobilla.net
Thu Aug 2 16:02:10 UTC 2012

On Thu, 2012-08-02 at 11:28 -0400, Paul Davis wrote:
> On Thu, Aug 2, 2012 at 10:46 AM, David Robillard <d at drobilla.net>
> wrote:
>         Is there a reason Jack can't do this for everything?  I am not
>         really
>         keen on putting a bunch of mysterious assembler crap in a host
>         meant to
>         be a relatively clean example, and it's even worse to make
>         plugins have
>         to do this...
> JACK doesn't get a chance to handle the output of a plugin until the
> host hands it over to JACK and the process cycle ends. Just have a
> reverb plugin outputting denormals and then have the host apply a gain
> value and boom .. JACK can't play a role in that signal flow. 

I mean set the flush-to-zero flag (and/or any other processor state
stuff required).

I know Ardour 'injects' non-zero data to fix denormals, but last I
tinkered with this, setting that flag made the problems go away.  This
was many processor generations ago, though.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part
URL: <http://lists.linuxaudio.org/pipermail/linux-audio-dev/attachments/20120802/8833b282/attachment.pgp>

More information about the Linux-audio-dev mailing list