Re: Signal emission far to slow

[Alex Larsson <alexl redhat com>]

On Sun, 27 May 2001, Alex Larsson wrote:

> Following up to myself here. After an irc discussion with jamesh I added
> to the test an extra signal much like the "string" one, but this one had
> the argument using G_SIGNAL_TYPE_STATIC_SCOPE, and boy does this make a
> difference:

Embarassing. I forgot to connect to the sstring signal, that's why it was
so fast, if i connect to it i get slightly slower speeds, but the points i
made before are still valid:

Testing signal emissions with no installed handlers
===================================================
100000 nop emits by name in 1590 milliseconds
100000 integer emits by name in 1992 milliseconds
100000 string emits by name in 1972 milliseconds
100000 nop emits by id in 1023 milliseconds
100000 integer emits by id in 1268 milliseconds
100000 string emits by id in 1763 milliseconds
100000 sstring emits by id in 1277 milliseconds

Testing signal emissions with one null handler
==============================================
100000 nop emits by name in 1870 milliseconds
100000 integer emits by name in 2208 milliseconds
100000 string emits by name in 2320 milliseconds
100000 nop emits by id in 1268 milliseconds
100000 integer emits by id in 1641 milliseconds
100000 string emits by id in 2329 milliseconds
100000 sstring emits by id in 1692 milliseconds

Testing signal emissions with ten null handlers
===============================================
100000 nop emits by name in 3125 milliseconds
100000 integer emits by name in 4091 milliseconds
100000 string emits by name in 4244 milliseconds
100000 nop emits by id in 2525 milliseconds
100000 integer emits by id in 3485 milliseconds
100000 string emits by id in 4207 milliseconds
100000 sstring emits by id in 1686 milliseconds

/ Alex