Re: [PATCH] dtrace probes for memory manager
| От | fche@redhat.com (Frank Ch. Eigler) |
|---|---|
| Тема | Re: [PATCH] dtrace probes for memory manager |
| Дата | |
| Msg-id | y0m8wdaabl3.fsf@fche.csb обсуждение исходный текст |
| Ответ на | Re: [PATCH] dtrace probes for memory manager (Zdenek Kotala <Zdenek.Kotala@Sun.COM>) |
| Ответы |
Re: [PATCH] dtrace probes for memory manager
|
| Список | pgsql-hackers |
Zdenek Kotala <Zdenek.Kotala@Sun.COM> writes: > [...] > + header = (StandardChunkHeader *) > + ((char *) ret - STANDARDCHUNKHEADERSIZE); > + > +// TRACE_POSTGRESQL_MCXT_ALLOC(context->name, context, size, header->size, true); > + > [...] If the dormant overhead of these probes is measured or suspected to be excessive, consider using the dtrace-generated per-probe foo_ENABLED() conditional, or a postgres configuration global thusly: if (__builtin_expect(TRACE_POSTGRESQL_MCXT_ALLOC_ENABLED(), 0)) TRACE_POSTGRESQL_MCXT_ALLOC(...); so that the whole instrumentation parameter setup/call can be placed out of the hot line with gcc -freorder-blocks. - FChE
В списке pgsql-hackers по дате отправления: