Horton

Re: Performance improvements for src/port/snprintf.c

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Andres Freund <andres@anarazel.de>
Cc: Alexander Kuzmenkov <a.kuzmenkov@postgrespro.ru>, pgsql-hackers@lists.postgresql.org
Date: 2018-10-03T18:36:54Z
Lists: pgsql-hackers 
I wrote:
>> Hm. I guess that'd be a bit better, but I'm not sure it's worth it. How
>> about we simply add a static assert that long long isn't bigger than
>> int64?

> WFM, I'll make it happen.

Actually, while writing a comment to go with that assertion, I decided
this was dumb.  If we're expecting the compiler to have "long long",
and if we're convinced that no platforms define "long long" as wider
than 64 bits, we may as well go with the s/int64/long long/g solution.
That should result in no code change on any platform today.  And it
will still work correctly, if maybe a bit inefficiently, on some
hypothetical future platform where long long is wider.  We (or our
successors) can worry about optimizing that when the time comes.

			regards, tom lane

Commits

  1. Improve snprintf.c's handling of NaN, Infinity, and minus zero.

  2. Rationalize snprintf.c's handling of "ll" formats.

  3. Provide fast path in snprintf.c for conversion specs that are just "%s".

  4. Make assorted performance improvements in snprintf.c.

  5. Set snprintf.c's maximum number of NL arguments to be 31.

  6. Always use our own versions of *printf().