Re: Inefficiency in parallel pg_restore with many tables
| От | Nathan Bossart |
|---|---|
| Тема | Re: Inefficiency in parallel pg_restore with many tables |
| Дата | |
| Msg-id | 20230718160713.GA1139177@nathanxps13 обсуждение исходный текст |
| Ответ на | Re: Inefficiency in parallel pg_restore with many tables (Alvaro Herrera <alvherre@alvh.no-ip.org>) |
| Ответы |
Re: Inefficiency in parallel pg_restore with many tables
|
| Список | pgsql-hackers |
On Tue, Jul 18, 2023 at 06:05:11PM +0200, Alvaro Herrera wrote: > On 2023-Jul-17, Nathan Bossart wrote: > >> @@ -35,7 +42,11 @@ binaryheap_allocate(int capacity, binaryheap_comparator compare, void *arg) >> binaryheap *heap; >> >> sz = offsetof(binaryheap, bh_nodes) + sizeof(Datum) * capacity; >> +#ifdef FRONTEND >> + heap = (binaryheap *) pg_malloc(sz); >> +#else >> heap = (binaryheap *) palloc(sz); >> +#endif > > Hmm, as I recall fe_memutils.c provides you with palloc() in the > frontend environment, so you don't actually need this one. Ah, yes it does. Thanks for the pointer. -- Nathan Bossart Amazon Web Services: https://aws.amazon.com
В списке pgsql-hackers по дате отправления: