Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [cdt-dev] M7

Looks like the internal build break is intermittent and pretty random which makes debugging painful, but a bit less severe. The bigger issue on MinGW is the lack of scanner discovery anyway.

I'll publish the latest build and plug away a bit more.

But this does point out how poor our test coverage is and that we need to get stricter on test failures, and possibly our code review process to make sure quality doesn't suffer like this in the future.

Doug

From: Doug Schaefer <dschaefer@xxxxxxx>
Date: Wednesday, 9 May, 2012 1:44 PM
To: "CDT General developers list." <cdt-dev@xxxxxxxxxxx>
Subject: Re: [cdt-dev] M7

That test is related to options work that Mentor did. It shouldn't affect the build.

I'm still trying to find the guilty commit that broke the internal builder. It's somewhere between M5 and M6.

Doug

From: Sergey Prigogin <eclipse.sprigogin@xxxxxxxxx>
Reply-To: "CDT General developers list." <cdt-dev@xxxxxxxxxxx>
Date: Wednesday, 9 May, 2012 1:42 PM
To: "CDT General developers list." <cdt-dev@xxxxxxxxxxx>
Subject: Re: [cdt-dev] M7



On Wed, May 9, 2012 at 10:34 AM, Vivian Kong <vivkong@xxxxxxxxxx> wrote:

Here is the new build: https://hudson.eclipse.org/hudson/job/cdt-nightly/1080/

The 2 managed build JUnit failures are gone. There are some new failures in the PDOM tests though.


We're testing the build now. If this is good, can you please help us promote it to the aggregator please Doug?


The errors in the pdom tests are random. The consistently failing test is  org.eclipse.cdt.managedbuilder.core.tests.ManagedBuildCoreTests.testTreeOptions.

-sergey 


Regards,

Vivian Kong
IBM Eclipse CDT
IBM Canada Toronto Lab


Inactive hide details for Doug Schaefer ---05/09/2012 11:30:14 AM---The cut off has usually been to get the build into the aggrDoug Schaefer ---05/09/2012 11:30:14 AM---The cut off has usually been to get the build into the aggregator site by 5 p.m. EDT. From: Vivian K


From: Doug Schaefer <dschaefer@xxxxxxx>
To: "CDT General developers list." <cdt-dev@xxxxxxxxxxx>,
Date: 05/09/2012 11:30 AM

Subject: Re: [cdt-dev] M7
Sent by: cdt-dev-bounces@xxxxxxxxxxx




The cut off has usually been to get the build into the aggregator site by 5 p.m. EDT.

From: Vivian Kong <vivkong@xxxxxxxxxx>
Reply-To:
"CDT General developers list." <cdt-dev@xxxxxxxxxxx>
Date:
Wednesday, 9 May, 2012 11:15 AM
To:
"CDT General developers list." <cdt-dev@xxxxxxxxxxx>
Subject:
Re: [cdt-dev] M7

Sorry about all this.

We are working on fixing this today. When is the last chance for respin?

Regards,

Vivian Kong
IBM Eclipse CDT
IBM Canada Toronto Lab


Inactive hide details for Doug Schaefer ---05/09/2012 09:57:49 AM---And, yes, turning off Build Automatically fixes it. That alDoug Schaefer ---05/09/2012 09:57:49 AM---And, yes, turning off Build Automatically fixes it. That almost points to a platform resource proble

From:
Doug Schaefer <dschaefer@xxxxxxx>
To:
"CDT General developers list." <cdt-dev@xxxxxxxxxxx>,
Date:
05/09/2012 09:57 AM
Subject:
Re: [cdt-dev] M7
Sent by:
cdt-dev-bounces@xxxxxxxxxxx





And, yes, turning off Build Automatically fixes it. That almost points to a platform resource problem. Needs to see if the delta is messed.


Doug


From:
Doug Schaefer <dschaefer@xxxxxxx>
Date:
Wednesday, 9 May, 2012 9:38 AM
To:
"CDT General developers list." <cdt-dev@xxxxxxxxxxx>
Subject:
Re: [cdt-dev] M7

I'll take a look at the Build Automatically and try and get MinGW building at least. The missing scanner discovery for MinGW is also pretty bad since Codan marks everything as an error.


+1 for reverting the RefreshScopeManager. Rule number one of CDT-club is "Don't break anything", unless Vivian can fix it today.


We have one more chance at a respin. Let's see if we can get thing sane enough.


Doug


From:
Sergey Prigogin <eclipse.sprigogin@xxxxxxxxx>
Reply-To:
"CDT General developers list." <cdt-dev@xxxxxxxxxxx>
Date:
Wednesday, 9 May, 2012 2:00 AM
To:
"CDT General developers list." <cdt-dev@xxxxxxxxxxx>
Subject:
Re: [cdt-dev] M7



On Tue, May 8, 2012 at 10:02 PM, Marc-Andre Laperle <
malaperle@xxxxxxxxxxxxx> wrote:
    On 12-05-08 11:56 PM, Doug Schaefer wrote:
      Hey gang, is it just me or does M7 seem like it has a lot of issues. Almost all my smoke tests see some sort of issue, mostly in build/CModel. The worst being the internal builder, which is used by the MinGW toolchain on Windows, doesn't seem to be picking up resource changes and thus never builds.
    It looks like this happens when "Build automatically" is used, I don't see the problem when it's turned off. If you turn that off does it work OK for you?
      I'm also seeing the refresh not being done on the initial project build at project creation time and thus not finding any binaries breaking the initial debug.
    This might be related to the recent changes to RefreshScopeManager (per configuration refresh, bug 374442, 375859 ). It seems obvious to me that this one wasn't reviewed or tested enough before getting committed, I have found many problems with the limited testing I did. I just think we should test more in general.

Should we revert these commits until the problems are addressed?


    Marc-Andre

-sergey
_______________________________________________
cdt-dev mailing list

cdt-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/cdt-dev
_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/cdt-dev

[attachment "graycol.gif" deleted by Vivian Kong/Toronto/IBM]


_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/cdt-dev


Attachment: graycol.gif
Description: graycol.gif


Back to the top