| age |
author |
description |
| Sun, 09 Sep 2012 20:50:02 -0700 |
Sean Halle |
chgd _brch__perf_tuning_paper back to _brch__default.. in default branchdefault tip |
| Wed, 18 Apr 2012 15:51:01 +0200 |
Nina Engelhardt |
Added tag lots of cores version for changeset 0d7177551bffperf_tuning_paper |
| Wed, 18 Apr 2012 15:50:14 +0200 |
Nina Engelhardt |
optimization for large number of cores: dedicate core 0 to create/receiveperf_tuning_paper lots of cores version |
| Sun, 15 Jul 2012 01:27:39 -0700 |
Sean Halle |
Merge with perf_tuning_paper branch, which grabs fixes and best performing version |
| Tue, 17 Apr 2012 20:13:26 +0200 |
Nina Engelhardt |
slicing factor 0.5 actually gives better results in this verperf_tuning_paper |
| Tue, 17 Apr 2012 19:58:33 +0200 |
Nina Engelhardt |
slicing factor 0.4, which gives good results for dedicated ver, gives bad result hereperf_tuning_paper |
| Tue, 17 Apr 2012 19:52:50 +0200 |
Nina Engelhardt |
try to adjust slicing, pick too small task size (creation becomes bottleneck)perf_tuning_paper |
| Tue, 17 Apr 2012 19:46:38 +0200 |
Nina Engelhardt |
move receivePr to core 0perf_tuning_paper |
| Tue, 17 Apr 2012 19:45:47 +0200 |
Nina Engelhardt |
Added tag portable version for changeset b7d8cf6680a9perf_tuning_paper |
| Tue, 17 Apr 2012 19:39:18 +0200 |
Nina Engelhardt |
portable ver: start division at core 1perf_tuning_paper portable version |
| Tue, 17 Apr 2012 19:20:46 +0200 |
Nina Engelhardt |
Performance debug series: fix divider bugperf_tuning_paper |
| Tue, 17 Apr 2012 19:18:57 +0200 |
Nina Engelhardt |
Performance debug series: original code (with divider bug)perf_tuning_paper |
| Tue, 17 Apr 2012 19:32:08 +0200 |
Nina Engelhardt |
open branch for perf tuning paperperf_tuning_paper |
| Mon, 16 Apr 2012 18:27:12 +0200 |
Nina Engelhardt |
better load balance |
| Mon, 19 Mar 2012 10:29:43 -0700 |
Some Random Person |
changed to DEBUG__printf to be consistent and easier to remember |
| Wed, 14 Mar 2012 23:22:00 -0700 |
Some Random Person |
New states for sub-repos, almost works, just bug in shutdown to fix still |
| Tue, 13 Mar 2012 18:31:04 -0700 |
Some Random Person |
More progress towards getting common_ancestor working |
| Tue, 13 Mar 2012 10:07:00 -0700 |
Some Random Person |
Compiles, but does not yet run properly |
| Sat, 10 Mar 2012 20:35:37 -0800 |
Some Random Person |
more changes to vms_impl and ssr_impl |
| Fri, 09 Mar 2012 22:23:31 -0800 |
Some Random Person |
added __brch__default |
| Fri, 09 Mar 2012 22:21:26 -0800 |
Some Random Person |
update to common ancestor (VirtProcr to SlaveVP) and updated include paths |
| Fri, 09 Mar 2012 19:03:28 +0100 |
Nina Engelhardt |
update header files |
| Tue, 07 Feb 2012 14:07:38 -0800 |
Me |
Changed dir structure to new project structure |
| Wed, 07 Sep 2011 13:06:25 +0200 |
Nina Engelhardt |
fix use after free |
| Mon, 29 Aug 2011 13:50:04 +0200 |
Nina Engelhardt |
Add variables for histogram |
| Thu, 12 May 2011 14:27:17 +0200 |
Merten Sach |
fixed: uninitialized variable |
| Sun, 14 Nov 2010 11:07:52 -0800 |
Me |
Updated singleton -- works, and added measurement of Request Handler time |
| Mon, 08 Nov 2010 04:04:19 -0800 |
Me |
Moved zeroing out results array from results Pr to Divider |
| Thu, 04 Nov 2010 17:44:15 -0700 |
Me |
Fixed last bugs in matrix multiply code -- gives correct answers consistently |
| Tue, 02 Nov 2010 17:00:50 -0700 |
Me |
Fixed concurrency bug -- added singleton to SSR -- works! 3.4x speedup |
| Thu, 14 Oct 2010 17:10:17 -0700 |
Me |
added .hgignore |
| Thu, 14 Oct 2010 17:09:22 -0700 |
Me |
Either working correctly, or close.. |
| Tue, 05 Oct 2010 10:00:11 -0700 |
Me |
init commit of copy of sliced matrix mult |
