[FFmpeg-devel] [PATCH] Don't needlessly reinitialize ff_cos_## tables.

[wm4]

On Fri, 23 Oct 2015 06:40:29 -0400
"Ronald S. Bultje" <rsbultje at gmail.com> wrote:

> Hi,
> 
> On Fri, Oct 23, 2015 at 12:48 AM, Dale Curtis <dalecurtis at chromium.org>
> wrote:
> 
> > On Wed, Oct 21, 2015 at 6:52 PM, Dale Curtis <dalecurtis at chromium.org>
> > wrote:
> >  
> > > On Tue, Oct 20, 2015 at 11:50 PM, Michael Niedermayer <  
> > > michael at niedermayer.cc> wrote:
> > >>
> > >> the last element to be written should be checked, so that if
> > >> initialization is done by 2 threads at the same time, neither can
> > >> return from this function without initialization having finished
> > >>
> > >> also the race detectors are broken if they complain about cases where
> > >> a variable that has value a is set to value a, that cannot be part of
> > >> a race, not even if a is written byte per byte instead of atomically
> > >> unless theres something iam missing
> > >> Is this something that can be fixed or disabled on the side of the
> > >> race detectors?
> > >> It might reduce false positives in FFmpeg and potentially other
> > >> tools.
> > >>  
> > >
> > > We can suppress it, which I think is more reasonable then the overhead
> > > it'd take to make this "race" go away. I notice the sin tables are
> > > initialized within the fft context so there's no "race." Is there a  
> > reason  
> > > the cosine tables aren't done this way?
> > >  
> >
> > Actually it looks like sin may suffer from the same issues -- it doesn't
> > make a copy like I was thinking it did. One of the TSAN folk detailed why
> > suppression isn't a good idea here
> > https://codereview.chromium.org/1412123007/#msg7 -- which roughly says we
> > can't rely on the compiler to do the right thing here.
> >
> > What are your thoughts about using ff_thread_once() to build the full sin
> > and cos tables all at once?  
> 
> 
> That sounds right. The reason that wasn't initially done is probably
> because cos/sin tables existed before ff_thread_once did.

Yep, it's a recent addition. We also want to stop the practice of
initializing global tables without any form of synchronization. The
existing codebase doesn't reflect how things should be done.

There are 2 possibilities:
1. Using ff_thread_once()
2. Not using global tables, but moving them to per-instance contexts

Both should be fine, but I think 2. is preferred.