[m-dev.] for review: cleanup of liveness info around calls

Tyson Dowd trd at cs.mu.OZ.AU
Wed Nov 11 18:33:38 AEDT 1998


On 11-Nov-1998, Zoltan Somogyi <zs at cs.mu.OZ.AU> wrote:
> 
> This is for Tyson.
> 
> I do not intend to commit this change before the release
> (unless Tyson insists).
> compiler/call_gen.m:
> compiler/code_info.m:
> 	General cleanup of the code dealing with (a) the livevals annotation
> 	before calls, which says what lvals are live before a call; and (b)
> 	what lvals are live containing what values after a call. This involves
> 	a rewrite of the appropriate submodule in code_info, including much
> 	more meaningful predicate names, and related changes in call_gen.
> 	It also involves adding a new, non-exported predicate in another
> 	submodule of code_info, and removing some code duplication; some
> 	of the work of the code_info submodule was duplicated in call_gen.
> 
> 	There is no change in functionality.
> 

I don't see any problems with this change, but I agree with your
intention to wait until after the release.

> Zoltan.
> 
> I include a unified diff for call_gen and a context diff for code_info,
> since these are significantly easier to read.

Perhaps it's just a matter of experience, but I can read unified diffs
relatively easily, while context diffs require my full attention just
to see what has changed.

Especially when the --- number,number --- lines are easily obscured
by % ----------- lines like they are in mercury source.
With the original separated from the final by hundreds of lines,
I also find it difficult to compare the changes. 

After trying to review the context diff, I am more convinced than
ever that context diffs offer no advantages over unified diffs,
and that unified diffs are easier to review.  (although in side-by-side
context would be great).

So I just went to your workspace and looked at the unified diff.
It was like switching to "see john run" after War and Peace ;-)

-- 

Because I dislike being quoted I lie almost constantly when talking 
about my work.
		-- Terry Gilliam

Tyson Dowd   <tyson at tyse.net>   http://tyse.net/tyson 




More information about the developers mailing list