Re: Re: [COMMITTERS] pgsql: Initialise perl library as documented in perl API.
| От | Andrew Dunstan |
|---|---|
| Тема | Re: Re: [COMMITTERS] pgsql: Initialise perl library as documented in perl API. |
| Дата | |
| Msg-id | 4A53704D.50608@dunslane.net обсуждение исходный текст |
| Ответ на | Re: Re: [COMMITTERS] pgsql: Initialise perl library as documented in perl API. (Andrew Dunstan <andrew@dunslane.net>) |
| Список | pgsql-hackers |
Andrew Dunstan wrote: > > I think we need to float a bug upstream to the perl guys on this, but > as a holding position I suggest that we alter the #ifdef test to avoid > calling PERL_SYS_INIT3() where MYMALLOC is defined. It's ugly, but I > can't think of another simple way around it (and we've done worse > things to accommodate platform weirdness ;-) ) > > It turns out that this doesn't work on Cygwin with its latest Perl (which has MYMALLOC defined). If we call PERL_SYS_INIT3() it fails in that call, and if we don't call it it fails later. We haven't noticed it earlier because my Cygwin buildfarm member's Perl is a couple of years old. I guess we're lucky that it's not a more critical platform, but there is no guarantee that this is the only place it will fail, now or in the future. I will see about following up more energetically with the Perl people. The workaround on Cygwin is to downgrade the perl installation to 5.8.8. cheers andrew
В списке pgsql-hackers по дате отправления: