Re: SP-GiST micro-optimizations
| От | Tom Lane |
|---|---|
| Тема | Re: SP-GiST micro-optimizations |
| Дата | |
| Msg-id | 6524.1346179347@sss.pgh.pa.us обсуждение |
| Ответ на | Re: SP-GiST micro-optimizations (Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>) |
| Ответы |
Re: SP-GiST micro-optimizations
|
| Список | pgsql-hackers |
Heikki Linnakangas <heikki.linnakangas@enterprisedb.com> writes:
> On 28.08.2012 20:30, Tom Lane wrote:
>> Fascinating. I'd been of the opinion that modern compilers would inline
>> memset() for themselves and MemSet was probably not better than what the
>> compiler could do these days. What platform are you testing on?
> x64, gcc 4.7.1, running Debian.
> The assembly generated for the MemSet is:
> [ pretty darn tight ]
> while the corresponding memset code is:
> [ not so good ]
Seems like that's down to the CPU not doing "rep stosq" particularly
quickly, which might well be chip-specific.
Anyway, IIRC there are similar memsets for all the SPGiST opclass
invocation calls, so I guess you should switch them all not just these
two.
regards, tom lane
В списке pgsql-hackers по дате отправления: