Re: removing some dead "keep compiler quiet" code - Mailing list pgsql-hackers

From Tom Lane
Subject Re: removing some dead "keep compiler quiet" code
Date
Msg-id 25398.1353384584@sss.pgh.pa.us
Whole thread Raw
In response to removing some dead "keep compiler quiet" code  (Peter Eisentraut <peter_e@gmx.net>)
List pgsql-hackers
Peter Eisentraut <peter_e@gmx.net> writes:
> Since ereport() can now communicate to the compiler whether it returns
> or not, a fair amount of "keep compiler quiet" code is dead.  Since the
> method that ereport() uses is not dependent on any compiler-specific
> attributes, I think this code can just be removed.  I propose the
> attached patch.

Meh.  This will only work if the compiler understands that abort()
doesn't return --- compilers that don't know that will start bleating
about uninitialized variables again.  And the ones that do know it
will be able to throw away the dead code anyway.

I doubt this is a good idea.
        regards, tom lane



pgsql-hackers by date:

Previous
From: Peter Eisentraut
Date:
Subject: removing some dead "keep compiler quiet" code
Next
From: Jeevan Chalke
Date:
Subject: Re: too much pgbench init output