Re: BUG #18083: not compile PostgreSQL module in Qt with GCC 11.2 compiler

Поиск
Список
Период
Сортировка
От Daniel Gustafsson
Тема Re: BUG #18083: not compile PostgreSQL module in Qt with GCC 11.2 compiler
Дата
Msg-id 160C3F41-16C0-4DCF-AC30-222F42905058@yesql.se
обсуждение исходный текст
Ответ на Re: BUG #18083: not compile PostgreSQL module in Qt with GCC 11.2 compiler  (Thomas Munro <thomas.munro@gmail.com>)
Список pgsql-bugs
> On 5 Sep 2023, at 10:57, Thomas Munro <thomas.munro@gmail.com> wrote:
>
> On Tue, Sep 5, 2023 at 2:25 PM Thomas Munro <thomas.munro@gmail.com> wrote:
>> What is C:/Prg/PostgreSQL/15/include/pthread.h, though?
>>
>> Searching the web finds other complaints like this, and in one case it
>> involved the Chocolatey postgresql package, and peering at
>> https://github.com/majkinetor/au-packages/blob/master/postgresql/tools/chocolateyInstall.ps1
>> reveals that Choco fetches the EnterpriseDB build.  I know that a long
>> time ago PostgreSQL used to use the pthread-win32 package...  is it
>> possible that the EntrepriseDB installer is still installing that
>> project's pthread.h?  That theory might be a long shot, but
>> *something* seems fishy here.
>
> Navadvipa confirmed privately that it was the EDB installer (via
> postgresql.org), and it's clearly the pthread.h from
> https://sourceware.org/pthreads-win32/.  CCing Sandeep who might know
> who to ask why that's being installed and check if it's a mistake.
> Nothing in PostgreSQL itself should need <pthread.h> on Windows AFAIK
> (I think that changed some time around 2007, commit 96b17190?).

The 15.4 ZIP archive from EDB does contain pthread.h in ./include, but the
16rc1 ZIP archive does not.  The 16rc1 does not contain all the additional
packages though so it might come from there?

--
Daniel Gustafsson




В списке pgsql-bugs по дате отправления:

Предыдущее
От: Branko Radovanovic
Дата:
Сообщение: Re: BUG #18081: Spurious "function with OID ###### does not exist" error
Следующее
От: PG Bug reporting form
Дата:
Сообщение: BUG #18087: pg_visibility: false positives in pg_check_visible