Re: [Tinycc-devel] i just can't get over how farging fast tcc is...

[Stephan Beal]

On Wed, Feb 2, 2011 at 4:20 AM, Stephan Beal <address@hidden> wrote:

On Wed, Feb 2, 2011 at 4:09 AM, Gary Birkett <address@hidden> wrote:

Why not test it? :)

Okay, since i was provoked ;)...

Here's a test which connects to a mysql db, creates a small table, puts a few records in it, and performs some sanity checks on the DB API (the code i'm developing - a db access abstraction layer)...

~> time tcc -run -Iinclude -I. -rdynamic -L/usr/lib/mysql -lmysqlclient -lsqlite3 -L$HOME/lib tcc-test.c -m
Testing against DSN [mysql5:dbname=cpdo;host=localhost;port=3306;autocommit=true;fieldbuffersize=128;enablenamedparams=true;ignored=hi]...
... big snip ...

real    0m0.080s
user    0m0.020s
sys    0m0.010s


With gcc:

real    0m0.070s
user    0m0.000s
sys    0m0.000s

only 1/10th of a second slower.

Interestingly... the same test against an sqlite3 database (exact same SQL operations, different back-end):

~> time tcc -run -Iinclude -I. -rdynamic -L/usr/lib/mysql -lmysqlclient -lsqlite3 -L$HOME/lib tcc-test.c -s
Testing against DSN [sqlite3:my.sq3]...
<snip>

real    0m0.240s
user    0m0.030s
sys    0m0.000s

gcc:

real    0m0.199s
user    0m0.000s
sys    0m0.000s

gcc is still only about 1/5th faster. (What i cannot explain, though, is HTF MySQL is faster (and so much faster!) than sqlite? The MySQL server and the sqlite3 db are on the same hard drive, so drive/system speed differences aren't a factor.)

-- 
----- stephan beal
http://wanderinghorse.net/home/stephan/