Re: [COMMITTERS] pgsql: Allow copydir() to be interrupted.

On Fri, Jul 2, 2010 at 12:38 PM, Robert Haas <robertmhaas@gmail.com> wrote:
> On Fri, Jul 2, 2010 at 11:07 AM, Robert Haas <robertmhaas@gmail.com> wrote:
>> On Fri, Jul 2, 2010 at 10:21 AM, Robert Haas <robertmhaas@gmail.com> wrote:
>>> On Fri, Jul 2, 2010 at 10:18 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>>> Robert Haas <robertmhaas@gmail.com> writes:
>>>>>>> This appears to have broken MinGW and Cygwin builds on the buildfarm.
>>>>>
>>>>> Well, that's not awesome. IM-ing with Magnus now.  I'm wondering if
>>>>> this is a link-ordering problem of some kind.
>>>>
>>>> Possibly an #ifndef FRONTEND will fix it.
>>>
>>> What's failing to link is postgres.exe
>>
>> I suspect that moving copydir.c into the backend will fix this, but I
>> don't have an appropriate build environment to test.  Can someone
>> please test the attached patch?
>
> Andrew confirms that this works on mingw, so I'm going to go ahead and
> check it in and see what happens.

hamerkop [Windows Server 2008 R2 (64bit) Visual C++ 2005 AMD64] is not
happy with this.

http://pgbuildfarm.org/cgi-bin/show_log.pl?nm=hamerkop&dt=2010-07-02%2018:45:44

I don't really understand most of the log messages, but the problem
seems to be here (some garbage removed for clarity):

LINK : fatal error LNK1104: '.\Release\postgres\src_port_copydir.obj'
LIB : fatal error LNK1181: '.\Release\libpgport\copydir.obj'

What doesn't make sense about that is that that file should not have
existed in the snapshot hamerkop ran against.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise Postgres Company