[MPlayer-dev-eng] threaded cache

[Reimar Döffinger]

On Mon, Dec 05, 2011 at 08:55:58AM +0100, Dan Oscarsson wrote:
> Though, do we have to make things work correctly when system clock is
> changed during play?

Not necessarily, but replacing code that works ok and is portable
by code that is less portable, adds complexity and is in some very minor
ways buggy with no big benefits is hardly a very convincing argument,
or?
If I couldn't have come up with a way to make the cache process quit
that would have been a convincing argument to accept such minor issues,
but what are the arguments now to accept a pthread cache that is not
"perfect"?