| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Bruce Momjian <bruce(at)momjian(dot)us> |
| Cc: | Gurjeet Singh <gurjeet(at)singh(dot)im>, Robert Haas <robertmhaas(at)gmail(dot)com>, Peter Eisentraut <peter_e(at)gmx(dot)net>, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>, PGSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Shave a few instructions from child-process startup sequence |
| Date: | 2014-03-08 16:26:42 |
| Message-ID: | 17586.1394296002@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Bruce Momjian <bruce(at)momjian(dot)us> writes:
> On Sun, Dec 1, 2013 at 12:07:21PM -0500, Gurjeet Singh wrote:
>> Yes, this is a performance patch, but as the subject says, it saves a few
>> instructions. I don't know how to write a test case that can measure savings of
>> skipping a few instructions in a startup sequence that potentially takes
>> thousands, or even millions, of instructions.
> Are we saying we don't want this patch?
I don't --- I think it makes the code less robust for no gain worthy
of the name.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2014-03-08 16:28:21 | Re: Is SPI safe to use in multi-threaded PL/Java? |
| Previous Message | Atri Sharma | 2014-03-08 15:32:01 | Re: Selection of join algorithm. |