summaryrefslogtreecommitdiff
path: root/benchmark/bm_io_select2.rb
diff options
context:
space:
mode:
authork0kubun <k0kubun@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>2018-07-10 13:05:29 +0000
committerk0kubun <k0kubun@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>2018-07-10 13:05:29 +0000
commit8068d34cfe4f74784cba29e981e2d8437ff02a9f (patch)
tree85b1f81c676f3fecdf3c425a94f341f63797eec4 /benchmark/bm_io_select2.rb
parentbd12037e50a216bda5896676b678bafee5eacb49 (diff)
Revert "benchmark/*.yml: convert from benchmark/bm_*.rb"
This reverts r63900. Having single-execution benchmark as a normal Ruby script is preferred by ko1. I'm not a big fan of having inconsistent benchmark formats, but I can understand some benefits of it. common.mk: remove obsolsted benchmark-each PHONY declaration, support running Ruby scripts added by this commit. README.md: follow ARGS change git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63926 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Diffstat (limited to 'benchmark/bm_io_select2.rb')
-rw-r--r--benchmark/bm_io_select2.rb22
1 files changed, 22 insertions, 0 deletions
diff --git a/benchmark/bm_io_select2.rb b/benchmark/bm_io_select2.rb
new file mode 100644
index 0000000..10e37d7
--- /dev/null
+++ b/benchmark/bm_io_select2.rb
@@ -0,0 +1,22 @@
+# IO.select performance. worst case of single fd.
+
+ios = []
+nr = 1000000
+if defined?(Process::RLIMIT_NOFILE)
+ max = Process.getrlimit(Process::RLIMIT_NOFILE)[0]
+else
+ max = 64
+end
+puts "max fd: #{max} (results not apparent with <= 1024 max fd)"
+
+((max / 2) - 10).times do
+ ios.concat IO.pipe
+end
+
+last = [ ios[-1] ]
+puts "last IO: #{last[0].inspect}"
+
+nr.times do
+ IO.select nil, last
+end
+