<br><br><div class="gmail_quote">On Tue, Jan 26, 2010 at 6:44 AM, Jason Dagit <span dir="ltr"><<a href="mailto:dagit@codersbase.com">dagit@codersbase.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<br><br><div class="gmail_quote"><div class="im">On Mon, Jan 25, 2010 at 9:50 AM, Eric Kow <span dir="ltr"><<a href="mailto:kowey@darcs.net" target="_blank">kowey@darcs.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I'm arbitrarily declaring this stage 1 of performance testing<br>
(we're not entirely there yet, though)<br>
<br>
The official repo has moved to<br>
<a href="http://code.haskell.org/darcs/darcs-benchmark/" target="_blank">http://code.haskell.org/darcs/darcs-benchmark/</a><br>
<br>
Could you guys use this for further benchmarking? I've<br>
- ignored errors removing the stats file<br>
- added benchmarks for revert/unrevert (interestingly,<br>
this one takes time; perhaps conflicts in the unrevert<br>
patch?)<br>
<br>
Thanks!<br></blockquote><div><br></div></div><div>I left it running overnight and it's not finished yet, but here is the output:</div><div><br></div></div></blockquote><div>Yay it finished:</div><div><br></div><div>=== darcs ===</div>
<div><br></div><div> || darcs-2.2.0 | darcs-2.2.1 | darcs-2.3.1 | darcs-2.3.98.1 | darcs-2.3.98.2 </div><div>===================++====================+====================+====================+=======================+======================</div>
<div> get (full) || 229.2s 10.0M | 217.3s 10.0M | 210.8s 10.0M | 213.6s 10.0M | 214.0s 11.0M</div><div> get (lazy, x10) || 95.5s 2.0M | 98.3s 2.0M | 96.8s 2.0M | 95.9s 2.0M | 95.9s 2.0M</div>
<div> pull 100 || 3.8s 21.0M | 3.8s 29.0M | 4.0s 16.0M | 5.2s 23.0M | 9.1s 17.0M</div><div> annotate || 11.4s 180.0M | 11.5s 180.0M | 11.5s 180.0M | 11.7s 181.0M | 11.6s 181.0M</div>
<div> wh x50 || 2.3s 0.0M | 2.3s 0.0M | 0.8s 0.0M | 1.7s 0.0M | 1.7s 0.0M</div><div> wh mod x50 || 4.3s 3.0M | 4.4s 3.0M | 2.9s 2.0M | 3.8s 2.0M | 3.8s 2.0M</div>
<div> wh -l x20 || 2.3s 0.0M | 2.3s 0.0M | 2.3s 0.0M | 1.3s 0.0M | 1.3s 0.0M</div><div> record mod x10 || 7.1s 28.0M | 6.8s 28.0M | 6.8s 28.0M | 6.8s 25.0M | 6.4s 25.0M</div>
<div> revert mod x50 || 12.4s 6.0M | 10.2s 6.0M | 10.3s 7.0M | 8.8s 4.0M | 8.0s 4.0M</div><div>(un)revert mod x10 || 9.7s 25.0M | 7.9s 25.0M | 7.7s 25.0M | 7.1s 24.0M | 7.0s 24.0M</div>
<div> check || 20.1s 182.0M | 11.8s 181.0M | 11.0s 50.0M | 39.4s 258.0M | 28.2s 233.0M</div><div> repair || 15.9s 182.0M | 11.5s 182.0M | 11.1s 53.0M | 39.4s 258.0M | 27.7s 234.0M</div>
<div> pull 1000 || 23.0s 44.0M | 23.7s 44.0M | 23.2s 45.0M | 25.6s 45.0M | 20.9s 45.0M</div><div><br></div><div>=== ghc-hashed ===</div><div><br></div><div> || darcs-2.2.0 | darcs-2.2.1 | darcs-2.3.1 | darcs-2.3.98.1 | darcs-2.3.98.2 </div>
<div>===================++=====================+=====================+====================+=======================+======================</div><div> get (full) || 917.7s 51.0M | 788.3s 60.0M | 820.6s 70.0M | 767.4s 70.0M | 795.5s 70.0M</div>
<div> get (lazy, x10) || 364.3s 8.0M | 360.4s 8.0M | 361.5s 8.0M | 373.0s 6.0M | 351.5s 6.0M</div><div> pull 100 || 11.5s 44.0M | 11.2s 49.0M | 10.9s 48.0M | 10.9s 37.0M | 10.2s 46.0M</div>
<div> annotate || - - | - - | - - | - - | - -</div><div> wh x50 || 5.4s 0.0M | 5.4s 0.0M | 1.4s 0.0M | 3.4s 0.0M | 3.5s 0.0M</div>
<div> wh mod x50 || 9.0s 5.0M | 8.9s 5.0M | 4.7s 3.0M | 6.5s 3.0M | 6.5s 3.0M</div><div> wh -l x20 || 6.3s 0.0M | 6.3s 0.0M | 6.3s 0.0M | 3.0s 0.0M | 3.0s 0.0M</div>
<div> record mod x10 || 21.9s 62.0M | 21.7s 55.0M | 20.6s 63.0M | 20.3s 62.0M | 20.6s 63.0M</div><div> revert mod x50 || 25.5s 15.0M | 25.4s 15.0M | 24.9s 12.0M | 16.3s 9.0M | 16.0s 9.0M</div>
<div>(un)revert mod x10 || 25.8s 50.0M | 24.1s 50.0M | 23.0s 43.0M | 19.7s 48.0M | 21.8s 48.0M</div><div> check || 178.5s 1201.0M | 180.9s 1372.0M | 153.4s 145.0M | - - | - -</div>
<div> repair || 174.0s 1204.0M | 178.2s 1207.0M | 125.8s 146.0M | 1180.7s 354.0M | 1082.7s 355.0M</div><div> pull 1000 || 67.9s 202.0M | 66.4s 202.0M | 66.2s 201.0M | 68.6s 202.0M | 69.7s 202.0M</div>
<div> </div></div>