<div dir="ltr">Is this a measurement of <span style="font-size:12.8px">the impact of elevation of </span><span style="font-size:12.8px">max-gcse-memory from 50MB to 1GB on the execution speed of compiled code?</span><div><span style="font-size:12.8px"><br></span></div><div><span style="font-size:12.8px">What do the first and second columns describe, etotal execution time for the 50MB and 1MB cases?</span></div><div><span style="font-size:12.8px"><br></span></div></div><div class="gmail_extra"><br><div class="gmail_quote">2015-11-08 11:50 GMT+08:00 Bradley Lucier <span dir="ltr"><<a href="mailto:lucier@math.purdue.edu" target="_blank">lucier@math.purdue.edu</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">I just rebuilt Gambit with<br>
<br>
firefly:~/programs/gambit/gambit-v4_8_1-devel/bench> ./configure 'CC=gcc<br>
-march=native -Wdisabled-optimization -Wall -W --param<br>
max-gcse-memory=1000000000' '--enable-single-host'<br>
'--enable-multiple-versions' '--enable-shared' --enable-c-opt<br>
<br>
and got the results at the bottom, which are uniformly better than the<br>
old results.<br>
<br>
The -Wdisabled-optimization line showed me that GCSE wasn't being done,<br>
and after looking at gcse.c in the gcc sources I saw that I needed to<br>
increase the max-gcse-memory from 50 MB (found in gcc/params.def) to<br>
something bigger. I'm working a machine with 32GB, so I just decided to<br>
make it roughly 1GB.<br>
<br>
Pretty good.<br>
<br>
Brad<br>
<br>
./bench -r 3 -s r6rs-fixflo-unsafe gambit all<br>
./summarize results.Gambit-C-r6rs-fixflo-unsafe<br>
(succeed 0 0)<br>
(fail )<br>
(crash )<br>
(browse 892 894)<br>
(cpstak 264 266)<br>
(ctak 344 344)<br>
(dderiv 308 311)<br>
(deriv 272 274)<br>
(destruc 292 293)<br>
(diviter 292 293)<br>
(divrec 388 390)<br>
(puzzle 144 147)<br>
(takl 140 141)<br>
(trav1 180 183)<br>
(trav2 688 689)<br>
(triangl 308 312)<br>
(fft 76 79)<br>
(fib 212 214)<br>
(fibfp 168 169)<br>
(mbrot 152 153)<br>
(nucleic 64 67)<br>
(pnpoly 60 61)<br>
(sum 88 89)<br>
(sumfp 684 684)<br>
(tak 220 220)<br>
(tfib 444 444)<br>
(ack 160 160)<br>
(array1 64 64)<br>
(cat 360 361)<br>
(string 292 292)<br>
(sum1 376 376)<br>
(sumloop 1560 1563)<br>
(tail 292 294)<br>
(wc 168 171)<br>
(conform 240 240)<br>
(dynamic 228 230)<br>
(earley 244 243)<br>
(fibc 264 263)<br>
(graphs 292 295)<br>
(lattice 516 518)<br>
(matrix 372 371)<br>
(maze 172 169)<br>
(mazefun 276 280)<br>
(nqueens 288 289)<br>
(paraffins 752 756)<br>
(peval 280 279)<br>
(pi 472 471)<br>
(primes 492 492)<br>
(ray 100 101)<br>
(scheme 332 333)<br>
(simplex 112 114)<br>
(slatex 164 165)<br>
(perm9 300 302)<br>
(nboyer 340 339)<br>
(sboyer 232 233)<br>
(gcbench 752 752)<br>
<div class="HOEnZb"><div class="h5"><br>
_______________________________________________<br>
Gambit-list mailing list<br>
<a href="mailto:Gambit-list@iro.umontreal.ca">Gambit-list@iro.umontreal.ca</a><br>
<a href="https://webmail.iro.umontreal.ca/mailman/listinfo/gambit-list" rel="noreferrer" target="_blank">https://webmail.iro.umontreal.ca/mailman/listinfo/gambit-list</a><br>
</div></div></blockquote></div><br></div>