[MLton-devel] New release of MLton: call-graphs

[Joe Hurd]

On Wed, 2 Apr 2003, Stephen Weeks wrote:

> So, the plan is to split everything in mlton and keep the precise
> call graph.  While the program runs, keep counts for both the split
> and unsplit versions of each function.  Then, in mlprof, run -graph
> on the split call graph to compute the set of nodes and solid and
> dotted edges.  Then, merge together nodes for each function that
> isn't -profile-split, using the count for the unsplit version of the
> function.

Yes, but with one small difference. I was imagining that the -graph
predicate would be evaluated for each function (rather than each split
of a function) by looking at the aggregate counts (and for the
reachability properties, declaring that a function satisfies the
property if at least one split of the function does).

But given that you saw things differently, how about this compromise
to compute the set of graph nodes:  using aggregate counts by default
for all functions, except those specified by the -profile-split
regular expression.

> The only problem I see with this is that with -graph computed on the
> split call graph, some versions of a function may be ignored while
> others are kept.

Then the only instances of this phenomenon will be functions that the
user has declared to be -profile-split (which seems quite intuitive).

But perhaps you think this isn't a good overloading of the
-profile-split argument? After all, it won't be possible to profile
each split of a function f separately, but display f as a combined
node in the resulting graph. [Confusion: would this represent a
combination of (i) the splits of f that satisfied the graph predicate,
or (ii) all splits of f?]

> I don't think it will happen that often, since most of the time
> either all of the splits of a function will be ignored or none of
> them will.

Good.

Joe




-------------------------------------------------------
This SF.net email is sponsored by: ValueWeb: 
Dedicated Hosting for just $79/mo with 500 GB of bandwidth! 
No other company gives more support or power for your dedicated server
http://click.atdmt.com/AFF/go/sdnxxaff00300020aff/direct/01/
_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel