Bug 71618 - bencher script should support remote benchmarking
Summary: bencher script should support remote benchmarking
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Tools / Tests (show other bugs)
Version: 528+ (Nightly build)
Hardware: All All
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2011-11-05 18:06 PDT by Filip Pizlo
Modified: 2011-11-10 15:47 PST (History)
0 users

See Also:

Attachments
the patch (19.67 KB, patch)
2011-11-05 18:20 PDT, Filip Pizlo 		ggaren: review+
Details | Formatted Diff | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Filip Pizlo 2011-11-05 18:06:00 PDT 
Currently, when you do a long benchmarking run with bencher, you have to either wait or go use another machine until after bencher finishes, to ensure that performance is unperturbed.  Also, if you want to do benchmarking on multiple types of hardware to reduce systematics, you have to move your builds around manually and manually invoke bencher multiple times on multiple machines.  Bencher should be able to do a remote benchmarking run (complete with moving your builds for you) automatically.
Comment 1 Filip Pizlo 2011-11-05 18:20:37 PDT 
Created attachment 113774 [details]
the patch
Comment 2 Geoffrey Garen 2011-11-10 15:37:26 PST 
Comment on attachment 113774 [details]
the patch

rs=me
Comment 3 Filip Pizlo 2011-11-10 15:47:53 PST 
Landed in http://trac.webkit.org/changeset/99909