#raku-dev on 7 February 2026 - Raku Programming Language Log

[Tux]

Rakudo v2026.01-11-gb7a4e1a9e (v6.d) on MoarVM 2026.01-1-g6a80a5d43

csv-ip5xs	0.252	-	0.255
csv-ip5xs-20	1.075	-	1.087
csv-parser	1.099	-	1.143
csv-test-xs-20	0.117	-	0.121
test	1.797	-	1.872
test-t	0.448	-	0.454
test-t --race	0.279	-	0.287
test-t-20	5.965	-	5.993
test-t-20 --race	1.403	-	1.426

09:22

csv-test-xs 0.015 - 0.019

tux.nl/Talks/CSV6/speed4-20.html / tux.nl/Talks/CSV6/speed4.html tux.nl/Talks/CSV6/speed.log

10:09 librasteve_ joined

lizmat

another weird benchmark result:

12:12

m: sub a() { }; a() for ^50000000; say now - ENTER now

camelia

0.271946201

lizmat

m: sub a() { 42 }; a() for ^50000000; say now - ENTER now

camelia

0.235296138

lizmat

m: sub a() { }; a() for ^50000000; say now - ENTER now

12:13

camelia

0.277552336

lizmat

m: sub a() { 42 }; a() for ^50000000; say now - ENTER now

camelia

0.238863633

lizmat

some more benchmarks:

12:53

m: class A { has $.a; method BUILD(:$!a = 42) { } }; A.new for ^10000000; say now - ENTER now

camelia

2.08846937

lizmat

m: class A { has $.a; method BUILD(:$a = 42) { $!a = $a } }; A.new for ^10000000; say now - ENTER now

camelia

0.894904514

lizmat

apparently directly assigning to an attribute in a signature is relatively expensive