[m-dev.] efficiency situation with file offset tracking

Robert_William HUTTON rwh at ecr.mu.oz.au
Thu Feb 10 11:26:09 AEDT 2000


Well, after much stuffing around and many recompilations due to
home/mercury0 being full, I have managed to profile cat.m with and without
file offset tracking in io.m.  The top parts of the profiles follow.

With file offset tracking:

   %  cumulative    self              self    total
 time    seconds  seconds    calls  ms/call  ms/call name
 85.3       1.75     1.75   240802     0.00     0.00 <predicate
io:'read_line_as_string'/4 mode 0> [5] 
 12.1       2.00     0.25   240801     0.00     0.00 <predicate
io:'write_string'/3 mode 0> [6] 
  2.4       2.05     0.05        1    50.00  2050.00 <predicate
cat:'cat'/2 mode 0> [3] 
  0.0       2.05     0.00   240802     0.00     0.00 <predicate
io:'read_line_as_string'/3 mode 0> [4] 

Without file offset tracking:

   %  cumulative    self              self    total
 time    seconds  seconds    calls  ms/call  ms/call name
 84.2       1.60     1.60   240802     0.00     0.00 <predicate
io:'read_line_as_string'/4 mode 0> [5] 
 13.1       1.85     0.25   240801     0.00     0.00 <predicate
io:'write_string'/3 mode 0> [6] 
  2.6       1.90     0.05        1    50.00  1900.00 <predicate
cat:'cat'/2 mode 0> [3] 
  0.0       1.90     0.00   240802     0.00     0.00 <predicate
io:'read_line_as_string'/3 mode 0> [4] 

Is this too much of an efficiency loss?  If it's acceptable, I'll go ahead
and commit the change.

-- 
Robert Hutton <rwh at ecr.mu.oz.au>     | War doesn't demonstrate who's
WWW: <http://www.ecr.mu.oz.au/~rwh>  | right... just who's left.

--------------------------------------------------------------------------
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
--------------------------------------------------------------------------



More information about the developers mailing list