[gmpi] Re: Revised Topic 7.2: In-place processing
- From: "Angus F. Hewlett" <amulet@xxxxxxxxxxxxx>
- To: GMPI Mailing list <gmpi@xxxxxxxxxxxxx>
- Date: Fri, 6 Jun 2003 08:48:53 -0400 (EDT)
On Fri, 6 Jun 2003, Tim Hockin wrote:
> Topic 7.2: In-place processing
> ------------------------------------------
>
> Q: Do we do in-place processing?
>
> a) Yes. A plugin-global flag indicates a plugins ability to do this.
> This tells the host that ANY input buffer may be re-used on ANY output
> (perhaps withing some higher-level grouping of stuff (think MIDI
> channel?)). If a plug can not accept this rule, it must not use
> in-place processing.
>
> b) Yes. A plugin statically mnaps inputs to outputs. The host may
> re-use an input buffer for an output if they map together.
>
> c) Yes. A plugin is given inputs and produces it's own output buffers
> based on internal knowledge or other variables. The host is not
> involved, except to furnish new buffers as requested. The plugin must
> re-use or release all input buffers. All inputs and outputs are
> unique.
>
> d) Yes. Option C, with the following change: buffers can be marked
> read-only, in which case a plugin must copy the buffer. Input buffers
> can be non-unique, so long as no more than one instance of the buffer
> is writable (zero-filled buffer can be re-used, for example). Ouput
> buffers are always unique.
> e) Yes. Just like option D, but inputs and outputs may be non-unique.
> Shared buffers must be read-only.
e), the first instance of the buffer (input and output) is writable.
This is a nice, high performance solution, we just have to make the
documentation and examples very clear to avoid bugs creeping in.
Regards,
Angus.
----------------------------------------------------------------------
Generalized Music Plugin Interface (GMPI) public discussion list
Participation in this list is contingent upon your abiding by the
following rules: Please stay on topic. You are responsible for your own
words. Please respect your fellow subscribers. Please do not
redistribute anyone else's words without their permission.
Archive: //www.freelists.org/archives/gmpi
Email gmpi-request@xxxxxxxxxxxxx w/ subject "unsubscribe" to unsubscribe
Other related posts: