Re: Review: DTrace probes (merged version) ver_03
От | Alvaro Herrera |
---|---|
Тема | Re: Review: DTrace probes (merged version) ver_03 |
Дата | |
Msg-id | 20080801131947.GC4321@alvh.no-ip.org обсуждение исходный текст |
Ответ на | Re: Review: DTrace probes (merged version) ver_03 (Robert Lor <Robert.Lor@Sun.COM>) |
Ответы |
Re: Review: DTrace probes (merged version) ver_03
Re: Review: DTrace probes (merged version) ver_03 |
Список | pgsql-hackers |
Robert Lor wrote: > I made some changes to the sed script so it works with the sed on > Solaris & OS X. I tested this patch on both Solaris and OS X with DTrace > enabled and disabled and also verified that the sed script works with > GNU sed. I hope this is the final change for this patch. Thanks for > catching all the issues, and my bad for not testing with DTrace disabled. Applied. Something seems wrong with this argument: > + /* The following probe declarations cause compilation errors > + * on Mac OS X but not on Solaris. Need further investigation. > + * probe buffer__read__start(BlockNumber, Oid, Oid, Oid, bool); > + * probe buffer__read__done(BlockNumber, Oid, Oid, Oid, bool, bool); > + */ > + probe buffer__read__start(unsigned int, unsigned int, unsigned int, unsigned int, bool); > + probe buffer__read__done(unsigned int, unsigned int, unsigned int, unsigned int, bool, bool); > + > + probe buffer__flush__start(Oid, Oid, Oid); > + probe buffer__flush__done(Oid, Oid, Oid); How come "Oid" works for FLUSH_START but not READ_START and READ_DONE? Also, I wonder if there's any proof that this works at all on Mac OS X, given that the rule to create probes.o from probes.d is conditionally pulled in only for Solaris? -- Alvaro Herrera http://www.CommandPrompt.com/ The PostgreSQL Company - Command Prompt, Inc.
В списке pgsql-hackers по дате отправления: