[m-dev.] trail usage optimization for lowlevel backend

[Julien Fischer]

Here are some results of running speedtests with and without the trail usage
optimization in the various LLDS grades.

none.gc vs. none.gc.tr (no opt)
	- 2.9% increase in time
	- 14.8% increase in space

none.gc vs. none.gc.tr (with opt)
	- 1.2% increase in time
	- 11.8% increase in space

reg.gc vs. reg.gc.tr (no opt)
	- 4.5% increase in time
	- 16.4% increase in space

reg.gc vs. reg.gc.tr (with opt)
	 - 1.5% increase in time
	 - 13.1% increase in space

asm_fast.gc vs. asm_fast.gc.tr (no opt)
	- 10.4% increase in time
	- 17.8% increase in space

asm_fast.gc vs. asm_fast.gc.tr (with opt)
	- 3.6% increase in time
	- 12.9% increase in space
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at cs.mu.oz.au
Administrative Queries: owner-mercury-developers at cs.mu.oz.au
Subscriptions:          mercury-developers-request at cs.mu.oz.au
--------------------------------------------------------------------------