Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 03 Jul 2012 23:20:03 +0400
From:      Ruslan Mahmatkhanov <cvs-src@yandex.ru>
To:        Chris Rees <crees@FreeBSD.org>
Cc:        Koop Mast <kwm@freebsd.org>, Rodrigo OSORIO <rodrigo@bebik.net>, freebsd-ports@freebsd.org
Subject:   Re: unbroken dev/icu in current
Message-ID:  <4FF345E3.3050604@yandex.ru>
In-Reply-To: <CADLo83-voqfZ0%2BE%2Bn-1LdsfJjBr65ziOmqKA-ohfwzgB8Ob6_g@mail.gmail.com>
References:  <20120703115119.GB49312@oldfaithful.bebik.local> <CADLo839Pxgr21SopyhyDV5KefSvp0HM2P58BJYG5Cmsz0uGtCw@mail.gmail.com> <4FF33B8E.1000001@yandex.ru> <CADLo83-voqfZ0%2BE%2Bn-1LdsfJjBr65ziOmqKA-ohfwzgB8Ob6_g@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Chris Rees wrote on 03.07.2012 22:53:
> On 3 July 2012 19:35, Ruslan Mahmatkhanov <cvs-src@yandex.ru> wrote:
>> Chris Rees wrote on 03.07.2012 22:22:
>>
>>> On 3 July 2012 12:51, Rodrigo OSORIO <rodrigo@bebik.net> wrote:
>>>>
>>>>
>>>> Hi,
>>>>
>>>> The port dev/icu fails to build in current due to an lock issue
>>>> caused by the use of threads api.
>>>> Disabling threads for icu in current solves the problem  and the port
>>>> can be build successfully and AFAIK this unbroke some of the ports who
>>>> depends on icu.
>>>>
>>>> I don't know if this can be an acceptable short-term workaround for this
>>>> por ?
>>>
>>>
>>> I've been reliably informed that icu builds just fine on CURRENT from
>>> 1/July.  I'll update my Tinderbox and let you know.
>>>
>>> Chris
>>
>>
>> Can confirm that it builds without an issue on r237936 (Sun Jul  1 19:07:45
>> 2012) with default option set (threads enabled).
>>
>
> Great.... now I need to find an appropriate OSVERSION for marking
> BROKEN on; I'm tempted to start with 1000000 and end with 1000014; if
> no-one yells at me I'll commit it in an hour or so.
>
> Chris

Why to not just ask Rodrigo to update? I think this BROKEN is only 
needed if icu doesn't builds on a system, that was installed from 
official FreeBSD -current snapshot. If memory serves me well there is 
many extra cautions in docs that using -current, csuped in wrong time, 
may produce various problems, so "It's a -current, babe" should sound 
reasonable, I believe.
Is there any extra info (error logs), or was you able to reproduce it on 
some of your -current system?
Rodrigo, can you please update your system and check if problem is gone?

-- 
Regards,
Ruslan

Tinderboxing kills... the drives.





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4FF345E3.3050604>