[m-rev.] for review: document the decl debugger

[Zoltan Somogyi]

I wrote:
> We should consider generating all events even with --trace deep by default.

On 25-Aug-2002, Fergus Henderson <fjh at cs.mu.OZ.AU> wrote:
> I think that is a good idea.  How much extra code space / compilation time
> will it cost?

I measured this over the weekend. (Except for the required updates to the
expected output of the debugger tests, the change is literally a one-liner.)
The executable size increased by a bit more than 5%, and the time taken to
compile stage two increased by a bit less than 8%.

I think the increased simplicity is worth this much. A relevant quote:

	You cannot make a pussycat out of a lion
	by describing its claws in a reference manual.

If users deem the cost excessive, they can get back the existing behavior
using an existing mechanism, --suppress-trace (which is at the moment
undocumented).

I will post a diff, including documentation of --suppress-trace, later this
week.

Zoltan.
--------------------------------------------------------------------------
mercury-reviews mailing list
post:  mercury-reviews at cs.mu.oz.au
administrative address: owner-mercury-reviews at cs.mu.oz.au
unsubscribe: Address: mercury-reviews-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-reviews-request at cs.mu.oz.au Message: subscribe
--------------------------------------------------------------------------