[FFmpeg-devel] [RFC] Runtime-inited versus Hardcoded tables.

[Zdenek Kabelac]

2008/1/31, Reimar D?ffinger <Reimar.Doeffinger at stud.uni-karlsruhe.de>:
> Hello,
> On Thu, Jan 31, 2008 at 10:03:18AM +0100, Zdenek Kabelac wrote:
> > 2008/1/31, Reimar D?ffinger <Reimar.Doeffinger at stud.uni-karlsruhe.de>:
> > > On Thu, Jan 31, 2008 at 02:12:23AM +0100, Michael Niedermayer wrote:
> > > > $free -m
> > > >              total       used       free     shared    buffers     cached
> > > > Mem:           181        177          3          0          0         49
> > > > -/+ buffers/cache:        127         53
> > > > Swap:          980        357        622
> > > >
> > > > And which OSs dont support COW?
> > >
> > > What does COW even matter here? This is about .rodata, with the right
> > > linker hacks that can even be shared on MMU-less systems.
> > > And even the experimental OS I wrote during my time as a student,
> > > despite not getting further than running a very basic shell without a
> > > proper filesystem supported COW...
> >
> > And now I'm getting curious, how do you want to make COW with MMU-less
> > system - was your system running in some emulator/valgrind-like mode
> > ?
>
> No, I am not aware of a way to implement COW without MMU, I was trying
> to say that for one of the approaches (hard coded constant tables) no
> COW and no MMU is needed.

Yes - that's obvious - but I think distribution of some large binary
tables (either in sources or executables) is by faaaaaar more painful
then having them initialized on the runtime and only when needed.

Zdenek