Re: [dataflow] PATCH removal of more of flow.c.

[Paolo Bonzini <paolo dot bonzini at lu dot unisi dot ch>]

> It is hard to see how this could be correct unless the uses are called
> immediately after combine or scheduling since no other passes even try
> to keep this up to date.

It is not -- and indeed it is incorrect.  I seem to recall, from the 
last time I look at sh_reorg, that the LOG_LINKS are only taken as 
shortcuts, because there are parts that work around LOG_LINKS that "must 
be there but are not".  However, your work will probably spur me (or 
better, an SH backend maintainer) to remove that code and only rely on 
the "work arounds".

Paolo