Re: [COMMITTERS] pgsql: Remove unsafe calling of WSAStartup and WSA Cleanup from DllMain.
От | Dave Page |
---|---|
Тема | Re: [COMMITTERS] pgsql: Remove unsafe calling of WSAStartup and WSA Cleanup from DllMain. |
Дата | |
Msg-id | 45F173F4.9010609@postgresql.org обсуждение исходный текст |
Ответ на | Re: [COMMITTERS] pgsql: Remove unsafe calling of WSAStartup and WSA Cleanup from DllMain. (Magnus Hagander <magnus@hagander.net>) |
Ответы |
Re: Re: [COMMITTERS] pgsql: Remove unsafe calling of WSAStartup and WSA Cleanup from DllMain.
|
Список | pgsql-hackers |
Magnus Hagander wrote: > In which case it can simply be because I was building against a libpq > built with MSVC = it had the broken startup code, and you used a mingw > one, which didn't have it. Maybe - but it does imply it's potentially easy to break code with this change. > Per the docs, an application like slon *should* make it's own call to > WSAStartup() because it uses socket functions diretly (port/pipe.c)... > > > Another question related to backpatching - should I backpatch this to > 8.1 and 8.0 as well? I know we said we more or less don't maintain the > win32 port back there because it was too new, but this is all code in > the client libpq, which has been around no win32 much longer. The reason > I'm asking is that the original reporter of this problem is on 8.1... > > I'm leaning towards yes, but would like to hear further comments... I'm far from convinced it should be backpatched at all. /D
В списке pgsql-hackers по дате отправления: