Issue 749041: code review 749041: arm: disable some problematic tests

Issue 749041: code review 749041: arm: disable some problematic tests (Closed)

Can't Edit
Can't Publish+Mail
Start Review

Created:
15 years, 3 months ago by dpx

Modified:
15 years, 3 months ago

Reviewers:

CC:
rsc, kaib, golang-dev

Visibility:
Public.

Description

arm: disable some problematic tests test/64bit.go: segfaults on sheevaplug(armv5) The following have long runtimes when GC is on. Sample run time in seconds for android emulator and sheevaplug test/stack.go: 4934s 1780s test/ken/chan.go: 860s 296s test/gc1.go: 218s 69s

Patch Set 1 #

Patch Set 2 : code review 749041: arm: disable some problematic tests #

Created: 15 years, 3 months ago

Download [raw] [tar.bz2]

		Unified diffs	Side-by-side diffs	Delta from patch set	Stats (+0 lines, -7 lines)			Patch
	M	test/arm-pass.txt	View		4 chunks	+0 lines, -4 lines	0 comments	Download
	M	test/golden-arm.out	View		1 chunk	+0 lines, -3 lines	0 comments	Download

Messages

Total messages: 6

Expand All Messages | Collapse All Messages

dpx

Hello rsc, kaib (cc: golang-dev@googlegroups.com), I'd like you to review this change.

15 years, 3 months ago (2010-03-25 05:54:32 UTC) #1

rsc

*** Submitted as http://code.google.com/p/go/source/detail?r=6c356775e908 *** arm: disable some problematic tests test/64bit.go: segfaults on sheevaplug(armv5) The ...

15 years, 3 months ago (2010-03-25 07:23:10 UTC) #3

kaib

While searching for a recent CL from Russ I noticed all these review requests had ...

15 years, 3 months ago (2010-04-02 13:55:48 UTC) #4

dpx

after clpatch 884041 This is what i am getting. but a lot of tests are ...

15 years, 3 months ago (2010-04-02 14:50:32 UTC) #5

after clpatch 884041

This is what i am getting.  but a lot of tests are failing for me in makemap
right now. I'll let you know when i have more info.

=========== 64bit.go
runtime.makemap+0x38 :0
    runtime.makemap(0x6026c, 0x70288, 0x0, 0x37340, 0x6026c, ...)
once.Init·+0x84 :0
    once.Init·()
os.Init·+0x6c $GOROOT/src/pkg/reflect/deepequal.go:0
    os.Init·()
main.init+0x68 $GOROOT/test/./64bit.go:94
    main.init()
mainstart+0x8 $GOROOT/src/pkg/runtime/linux/thread.c:585
    mainstart()
initdone+0x54 $GOROOT/src/pkg/runtime/print.c:808
    initdone()
SIGSEGV: segmentation violation
Faulting address: 0x64
PC=0x33bcc

tmp.go:2: package statement must be first
BUG: 64bit


On Fri, Apr 2, 2010 at 7:55 AM, Kai Backman <kaib@golang.org> wrote:

> While searching for a recent CL from Russ I noticed all these review
> requests had been filtered into a folder that I read less often. My
> apology for the non-existent responses.
>
> I'm surprised that 64bit is causing a segfault, it definitely should
> not. Are you able to produce more details about the offending part
> Dean? Given how central it's to everything I'd rather not want to
> regressing.
>
>  Kai
>
> On Thu, Mar 25, 2010 at 9:23 AM,  <rsc@golang.org> wrote:
> > *** Submitted as
> > http://code.google.com/p/go/source/detail?r=6c356775e908 ***
> >
> > arm: disable some problematic tests
> >
> > test/64bit.go:  segfaults on sheevaplug(armv5)
> >
> > The following have long runtimes when GC is on.
> > Sample run time in seconds for android emulator
> > and sheevaplug
> > test/stack.go:          4934s   1780s
> > test/ken/chan.go:        860s    296s
> > test/gc1.go:             218s     69s
> >
> > R=rsc, kaib
> > CC=golang-dev
> > http://codereview.appspot.com/749041
> >
> > Committer: Russ Cox <rsc@golang.org>
> >
> >
> > http://codereview.appspot.com/749041/show
> >
>

rsc

15 years, 3 months ago (2010-04-02 16:39:06 UTC) #6

yes, lots of things are dying with the current 884041.
but other things work, so one step at a time.  ;-)

On Fri, Apr 2, 2010 at 07:50, Dean Prichard <dean.prichard@gmail.com> wrote:
> after clpatch 884041
>
> This is what i am getting.  but a lot of tests are failing for me in makemap
> right now. I'll let you know when i have more info.
>
> =========== 64bit.go
> runtime.makemap+0x38 :0
>     runtime.makemap(0x6026c, 0x70288, 0x0, 0x37340, 0x6026c, ...)
> once.Init·+0x84 :0
>     once.Init·()
> os.Init·+0x6c $GOROOT/src/pkg/reflect/deepequal.go:0
>     os.Init·()
> main.init+0x68 $GOROOT/test/./64bit.go:94
>     main.init()
> mainstart+0x8 $GOROOT/src/pkg/runtime/linux/thread.c:585
>     mainstart()
> initdone+0x54 $GOROOT/src/pkg/runtime/print.c:808
>     initdone()
> SIGSEGV: segmentation violation
> Faulting address: 0x64
> PC=0x33bcc
>
> tmp.go:2: package statement must be first
> BUG: 64bit
>
>
> On Fri, Apr 2, 2010 at 7:55 AM, Kai Backman <kaib@golang.org> wrote:
>>
>> While searching for a recent CL from Russ I noticed all these review
>> requests had been filtered into a folder that I read less often. My
>> apology for the non-existent responses.
>>
>> I'm surprised that 64bit is causing a segfault, it definitely should
>> not. Are you able to produce more details about the offending part
>> Dean? Given how central it's to everything I'd rather not want to
>> regressing.
>>
>>  Kai
>>
>> On Thu, Mar 25, 2010 at 9:23 AM,  <rsc@golang.org> wrote:
>> > *** Submitted as
>> > http://code.google.com/p/go/source/detail?r=6c356775e908 ***
>> >
>> > arm: disable some problematic tests
>> >
>> > test/64bit.go:  segfaults on sheevaplug(armv5)
>> >
>> > The following have long runtimes when GC is on.
>> > Sample run time in seconds for android emulator
>> > and sheevaplug
>> > test/stack.go:          4934s   1780s
>> > test/ken/chan.go:        860s    296s
>> > test/gc1.go:             218s     69s
>> >
>> > R=rsc, kaib
>> > CC=golang-dev
>> > http://codereview.appspot.com/749041
>> >
>> > Committer: Russ Cox <rsc@golang.org>
>> >
>> >
>> > http://codereview.appspot.com/749041/show
>> >
>
>

Expand All Messages | Collapse All Messages