Rietveld Code Review Tool
Help | Bug tracker | Discussion group | Source code | Sign in
(7621)

Issue 14526048: code review 14526048: net: fix TestDialFailPDLeak to work when GOMAXPROCS is large (Closed)

Can't Edit
Can't Publish+Mail
Start Review
Created:
10 years, 6 months ago by iant
Modified:
10 years, 6 months ago
Reviewers:
mikio, albert.strasheim
CC:
golang-dev, mikio
Visibility:
Public.

Description

net: fix TestDialFailPDLeak to work when GOMAXPROCS is large Fixes issue 6553.

Patch Set 1 #

Patch Set 2 : diff -r bfdf7b2f16a4 https://code.google.com/p/go #

Unified diffs Side-by-side diffs Delta from patch set Stats (+3 lines, -2 lines) Patch
M src/pkg/net/dial_test.go View 2 chunks +3 lines, -2 lines 0 comments Download

Messages

Total messages: 6
iant
Hello golang-dev@googlegroups.com, I'd like you to review this change to https://code.google.com/p/go
10 years, 6 months ago (2013-10-09 13:15:33 UTC) #1
mikio
lgtm
10 years, 6 months ago (2013-10-09 13:17:08 UTC) #2
iant
*** Submitted as https://code.google.com/p/go/source/detail?r=3f38e968e072 *** net: fix TestDialFailPDLeak to work when GOMAXPROCS is large Fixes ...
10 years, 6 months ago (2013-10-09 20:52:32 UTC) #3
albert.strasheim
Hello This test still seems flaky somehow + go version go version devel +3f38e968e072 Wed ...
10 years, 6 months ago (2013-10-10 14:36:25 UTC) #4
iant
On 2013/10/10 14:36:25, albert.strasheim wrote: > Hello > > This test still seems flaky somehow ...
10 years, 6 months ago (2013-10-10 16:01:42 UTC) #5
albert.strasheim
10 years, 6 months ago (2013-10-14 12:31:15 UTC) #6
Message was sent while issue was closed.
On 2013/10/10 16:01:42, iant wrote:
> On 2013/10/10 14:36:25, albert.strasheim wrote:
> > Hello
> > 
> > This test still seems flaky somehow
> > 
> > + go version
> > go version devel +3f38e968e072 Wed Oct 09 13:52:29 2013 -0700 linux/amd64
> > 
> > + go test -v -cpu 1,2,4 -timeout=60m std
> > 
> > --- FAIL: TestDialFailPDLeak (0.10 seconds)
> > dial_test.go:476: detected possible memory leak in runtime
> 
> The lack of a number (not TestDialFailPDLeak-4) implies that the test failed
> with GOMAXPROCS == 1.  I have no explanation for that.  I'm unable to repeat
it.
>  How repeatable is it for you?

$ GOMAXPROCS=1 go test net -run=TestDialFailPDLeak
ok  	net	0.042s
$ GOMAXPROCS=1 go test net -run=TestDialFailPDLeak
ok  	net	0.042s
$ GOMAXPROCS=1 go test net -run=TestDialFailPDLeak
ok  	net	0.042s
$ GOMAXPROCS=2 go test net -run=TestDialFailPDLeak
--- FAIL: TestDialFailPDLeak-2 (0.02 seconds)
	dial_test.go:476: detected possible memory leak in runtime
FAIL
FAIL	net	0.029s
$ GOMAXPROCS=2 go test net -run=TestDialFailPDLeak
ok  	net	0.031s
$ GOMAXPROCS=2 go test net -run=TestDialFailPDLeak
ok  	net	0.038s
$ GOMAXPROCS=2 go test net -run=TestDialFailPDLeak
--- FAIL: TestDialFailPDLeak-2 (0.02 seconds)
	dial_test.go:476: detected possible memory leak in runtime
FAIL
FAIL	net	0.029s
$ GOMAXPROCS=2 go test net -run=TestDialFailPDLeak
ok  	net	0.031s
$ GOMAXPROCS=2 go test net -run=TestDialFailPDLeak
ok  	net	0.031s
$ GOMAXPROCS=2 go test net -run=TestDialFailPDLeak
--- FAIL: TestDialFailPDLeak-2 (0.02 seconds)
	dial_test.go:476: detected possible memory leak in runtime
FAIL
FAIL	net	0.027s

go version devel +f1545db4a9c4 Mon Oct 14 00:18:46 2013 -0400 linux/amd64

Cheers

Albert
Sign in to reply to this message.

Powered by Google App Engine
RSS Feeds Recent Issues | This issue
This is Rietveld f62528b