Bug #98745 for Devel-PeekPoke: Test suite started to fail with perl 5.20 and newer

Wed Sep 10 01:14:03 2014 SREZIC [...] cpan.org - Ticket created

Subject:

Test suite started to fail with perl 5.20 and newer

See http://matrix.cpantesters.org/?dist=Devel-PeekPoke%200.04

Wed Sep 10 01:22:51 2014 ribasushi [...] leporine.io - Correspondence added

Subject:	Re: [rt.cpan.org #98745] Test suite started to fail with perl 5.20 and newer
Date:	Wed, 10 Sep 2014 07:22:36 +0200
To:	bug-Devel-PeekPoke [...] rt.cpan.org
From:	Peter Rabbitson <ribasushi [...] cpan.org>

On 09/10/2014 07:14 AM, Slaven_Rezic via RT wrote: Show quoted text

> Wed Sep 10 01:14:03 2014: Request 98745 was acted upon. > Transaction: Ticket created by SREZIC > Queue: Devel-PeekPoke > Subject: Test suite started to fail with perl 5.20 and newer > Broken in: 0.04 > Severity: (no value) > Owner: Nobody > Requestors: SREZIC@cpan.org > Status: new > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=98745 > > > > See http://matrix.cpantesters.org/?dist=Devel-PeekPoke%200.04

Sigh. Yes, reproduced locally. I will not be able to look at this for at least 1.5 weeks. This is still a low-prio module, so I am content with it languishing (like it did during 0.03). It is due for a thorough re-gut anyway. If someone else beats me to diagnosing the 5.20 intermittent fail - it's be grand ;)

Wed Sep 10 01:22:51 2014 The RT System itself - Status changed from 'new' to 'open'

Wed Sep 10 07:28:44 2014 RURBAN [...] cpan.org - Correspondence added

CC:	pp <perl5-porters [...] perl.org>
Subject:	Re: [rt.cpan.org #98745] Test suite started to fail with perl 5.20 and newer
Date:	Wed, 10 Sep 2014 06:28:30 -0500
To:	bug-Devel-PeekPoke [...] rt.cpan.org
From:	Reini Urban <rurban [...] cpan.org>

I suspect a serious upstream regression in destructive substr handling with certain string lengths 10-14. (which is incidently exactly the area I tried to fix earlier) This is a normal pure-perl substr, which is failing in certain string lengths. See https://github.com/rurban/Devel-PeekPoke/commit/8099054c964dd0a62805f27be06bbe49bbe7f452 Please hold 5.20.1 until this is analyzed. On Wed, Sep 10, 2014 at 12:22 AM, Peter Rabbitson via RT <bug-Devel-PeekPoke@rt.cpan.org> wrote: Show quoted text

> Queue: Devel-PeekPoke > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=98745 > > > On 09/10/2014 07:14 AM, Slaven_Rezic via RT wrote:

>> Wed Sep 10 01:14:03 2014: Request 98745 was acted upon. >> Transaction: Ticket created by SREZIC >> Queue: Devel-PeekPoke >> Subject: Test suite started to fail with perl 5.20 and newer >> Broken in: 0.04 >> Severity: (no value) >> Owner: Nobody >> Requestors: SREZIC@cpan.org >> Status: new >> Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=98745 > >> >> >> See http://matrix.cpantesters.org/?dist=Devel-PeekPoke%200.04

> Sigh. Yes, reproduced locally. I will not be able to look at this for at > least 1.5 weeks. This is still a low-prio module, so I am content with > it languishing (like it did during 0.03). It is due for a thorough > re-gut anyway. > > If someone else beats me to diagnosing the 5.20 intermittent fail - it's > be grand ;) >

Download 03torture.result
application/octet-stream 32.3k

Message body not shown because it is not plain text.

Wed Sep 10 10:10:46 2014 RURBAN [...] cpan.org - Correspondence added

CC:	pp <perl5-porters [...] perl.org>
Subject:	Re: [rt.cpan.org #98745] Test suite started to fail with perl 5.20 and newer
Date:	Wed, 10 Sep 2014 09:10:30 -0500
To:	bug-Devel-PeekPoke [...] rt.cpan.org
From:	Reini Urban <rurban [...] cpan.org>

Ok, I can clear the warning. No upstream problem. Devel::PeekPoke writes to buffers manually via substr, but got the logic with new cases with certain COW strings wrong. (len >= 10) 5.20.1 is good to go. Just Devel::PeekPoke::PP needs to adjust to COW, it cannot even detect COW strings in its current API. I'm working on it. (easy in XS but hard with PP) On Wed, Sep 10, 2014 at 6:28 AM, Reini Urban <rurban@cpan.org> wrote: Show quoted text

> I suspect a serious upstream regression in destructive substr handling > with certain string lengths 10-14. (which is incidently exactly the > area I tried to fix earlier) > > This is a normal pure-perl substr, which is failing in certain string lengths. > > See https://github.com/rurban/Devel-PeekPoke/commit/8099054c964dd0a62805f27be06bbe49bbe7f452 > > Please hold 5.20.1 until this is analyzed. > > On Wed, Sep 10, 2014 at 12:22 AM, Peter Rabbitson via RT > <bug-Devel-PeekPoke@rt.cpan.org> wrote:

>> Queue: Devel-PeekPoke >> Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=98745 > >> >> On 09/10/2014 07:14 AM, Slaven_Rezic via RT wrote:

>>> Wed Sep 10 01:14:03 2014: Request 98745 was acted upon. >>> Transaction: Ticket created by SREZIC >>> Queue: Devel-PeekPoke >>> Subject: Test suite started to fail with perl 5.20 and newer >>> Broken in: 0.04 >>> Severity: (no value) >>> Owner: Nobody >>> Requestors: SREZIC@cpan.org >>> Status: new >>> Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=98745 > >>> >>> >>> See http://matrix.cpantesters.org/?dist=Devel-PeekPoke%200.04

>> Sigh. Yes, reproduced locally. I will not be able to look at this for at >> least 1.5 weeks. This is still a low-prio module, so I am content with >> it languishing (like it did during 0.03). It is due for a thorough >> re-gut anyway. >> >> If someone else beats me to diagnosing the 5.20 intermittent fail - it's >> be grand ;)

Wed Nov 04 22:08:17 2015 ribasushi [...] leporine.io - Reference to ticket #108478 added

Wed Nov 04 22:09:58 2015 ribasushi [...] leporine.io - Correspondence added

Please see https://rt.cpan.org/Ticket/Display.html?id=108478 for an update. Closing this ticket without merging.

Wed Nov 04 22:10:04 2015 ribasushi [...] leporine.io - Status changed from 'open' to 'rejected'