- ruby-core - ml.ruby-lang.org

[ruby-core:115410] [Ruby master Bug#20007] prism test begain to fail on s390x
by mtasaka (Mamoru Tasaka) 18 Nov '23

18 Nov '23

Issue #20007 has been reported by mtasaka (Mamoru Tasaka). ---------------------------------------- Bug #20007: prism test begain to fail on s390x https://bugs.ruby-lang.org/issues/20007 * Author: mtasaka (Mamoru Tasaka) * Status: Open * Priority: Normal * ruby -v: ruby 3.3.0dev (2023-11-18 master ef72970a04) [s390x-linux] * Backport: 3.0: UNKNOWN, 3.1: UNKNOWN, 3.2: UNKNOWN ---------------------------------------- As of commit:3bc41f4f0b testsuite was successful on both x86_64 and s390x, but as of commit:ef72970a04 testsuite fails on s390x (note that s390x is big endian): ``` .... .... 87) Skipped: TestIRB::DebugCommandTest#test_multi_irb_commands_are_not_available_after_activating_the_debugger [/builddir/build/BUILD/ruby-3.3.0-ef72970a04/test/irb/helper.rb:94]: This test works only under ruby/irb 88) Skipped: TestIRB::DebugCommandTest#test_prompt_irb_name_is_kept [/builddir/build/BUILD/ruby-3.3.0-ef72970a04/test/irb/helper.rb:94]: This test works only under ruby/irb 89) Skipped: TestIRB::DebugCommandTest#test_delete [/builddir/build/BUILD/ruby-3.3.0-ef72970a04/test/irb/helper.rb:94]: This test works only under ruby/irb 90) Skipped: TestIRB::DebugCommandTest#test_long_stepping [/builddir/build/BUILD/ruby-3.3.0-ef72970a04/test/irb/helper.rb:94]: This test works only under ruby/irb 91) Error: Prism::ParseTest#test_filepath_seattlerb/call_dot_parens.txt: ArgumentError: offset outside of string <internal:pack>:29:in `unpack1' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:216:in `load_constant' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:233:in `load_required_constant' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:284:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `block in load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `initialize' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `new' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:475:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:150:in `load_nodes' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:154:in `load_result' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:41:in `load' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism.rb:65:in `load' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/test/prism/parse_test.rb:152:in `block (2 levels) in <class:ParseTest>' 92) Error: Prism::ParseTest#test_filepath_unparser/corpus/literal/yield.txt: ArgumentError: offset outside of string <internal:pack>:29:in `unpack1' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:216:in `load_constant' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:233:in `load_required_constant' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:284:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:256:in `block in load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:256:in `initialize' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:256:in `new' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:256:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:183:in `load_optional_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:535:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `block in load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `initialize' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `new' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:475:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:150:in `load_nodes' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:154:in `load_result' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:41:in `load' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism.rb:65:in `load' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/test/prism/parse_test.rb:152:in `block (2 levels) in <class:ParseTest>' 93) Error: Prism::ParseTest#test_filepath_seattlerb/f_kw__required.txt: ArgumentError: offset outside of string <internal:pack>:29:in `unpack1' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:216:in `load_constant' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:233:in `load_required_constant' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:333:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `block in load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `initialize' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `new' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:511:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:475:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:150:in `load_nodes' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:154:in `load_result' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:41:in `load' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism.rb:65:in `load' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/test/prism/parse_test.rb:152:in `block (2 levels) in <class:ParseTest>' 94) Error: Prism::ParseTest#test_filepath_whitequark/masgn.txt: ArgumentError: offset outside of string <internal:pack>:29:in `unpack1' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:216:in `load_constant' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:233:in `load_required_constant' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:475:in `block in load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:475:in `initialize' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:475:in `new' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:475:in `load_node' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:150:in `load_nodes' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:154:in `load_result' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism/serialize.rb:41:in `load' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/lib/prism.rb:65:in `load' /builddir/build/BUILD/ruby-3.3.0-ef72970a04/test/prism/parse_test.rb:152:in `block (2 levels) in <class:ParseTest>' ... ... Finished tests in 608.001668s, 43.2581 tests/s, 10825.4473 assertions/s. 26301 tests, 6581890 assertions, 0 failures, 774 errors, 90 skips ruby -v: ruby 3.3.0dev (2023-11-18 master ef72970a04) [s390x-linux] make: Leaving directory '/builddir/build/BUILD/ruby-3.3.0-ef72970a04/redhat-linux-build' RPM build errors: make: *** [uncommon.mk:945: yes-test-all] Error 6 ``` Maybe due to commit:7c99e43c3f050244b06dbd18de4f605ea70d234c ? -- https://bugs.ruby-lang.org/

3 4

[ruby-core:115404] Ruby 3.2.2 - rbconfig.rb's MAKEFILE_CONFIG
by Jay Mav 17 Nov '23

17 Nov '23

Hello Ruby Dev Team, With Ruby 3.2.2, gem install fails because in rbconfig.rb, MAKEFILE_CONFIG's $(target_os) is not expanded (whereas CONFIG's is expanded correctly). The inline doc in mkconfig.rb states that the caller can expand MAKEFILE_CONFIG's values by calling RbConfig::expand(). However, ruby-shadow gem's Makefile generator -- extconf.rb -- does not call expand and directly uses mkmf.rb's CONFIG instance (which is RbConfig::MAKEFILE_CONFIG), because of which $(target_os) does not get expanded. This causes ruby-shadow gem install to fail with Ruby 3.2.2. Would the Ruby Dev Team accept the following proposal to fix mkconfig.rb, so that the generated rbconfig.rb instantiates MAKEFILE_CONFIG AFTER CONFIG's values have been expanded? # Existing code... CONFIG.each_value do |val| RbConfig::expand(val) end # Move instantiation of MAKEFILE_CONFIG after expanding CONFIG's values. MAKEFILE_CONFIG = {} CONFIG.each{|k,v| MAKEFILE_CONFIG[k] = v.dup} If we fix it in mkconfig.rb, we can avoid callers having to explicitly call expand for MAKEFILE_CONFIG's values. Thanks.

2 3

[ruby-core:115403] [Ruby master Bug#7968] Poor UDPSocket#send performance in ruby 2.0.0 on windows
by cs96and (Alan Davies) 16 Nov '23

16 Nov '23

Issue #7968 has been updated by cs96and (Alan Davies). k0kubun (Takashi Kokubun) wrote in #note-14: > Just in case, could you measure the time spent on the loop instead of a whole Ruby command? The prelude script sometimes takes a lot of time, the performance varies across different versions, and `time` command would benchmark that as well. You can just call `Time.new` before and after the loop, subtract them, and report the result for each version. I also increased the loop by a factor of 10 (to 1,000,000) so it runs for longer. Times are in seconds... **1.9.3 send:** ``` ruby-1.9.3\bin\ruby.exe socketsendtest.rb 10.495193 ``` **1.9.3 print:** ``` ruby-1.9.3\bin\ruby.exe socketsendtest.rb 10.384035 ``` **3.2.2 send** ``` ruby-3.2.2\bin\ruby.exe socketsendtest.rb 17.3447397 ``` **3.2.2 print** ``` ruby-3.2.2\bin\ruby.exe socketsendtest.rb 9.8053804 ``` Out of interest I ran the code through the profile module as well... **1.9.3 send** ``` ruby-1.9.3\bin\ruby.exe -rprofile socketsendtest.rb 19.520433 % cumulative self self total time seconds seconds calls ms/call ms/call name 56.95 6.03 6.03 1000000 0.01 0.01 UDPSocket#send 43.05 10.59 4.56 1 4561.00 10594.00 Integer#times 0.00 10.59 0.00 1 0.00 0.00 UDPSocket#bind 0.00 10.59 0.00 10 0.00 0.00 Class#inherited 0.00 10.59 0.00 40 0.00 0.00 BasicObject#singleton_method_added 0.00 10.59 0.00 104 0.00 0.00 Module#method_added 0.00 10.59 0.00 2 0.00 0.00 Kernel.gem_original_require 0.00 10.59 0.00 2 0.00 0.00 Kernel.require 0.00 10.59 0.00 4 0.00 0.00 Module#private 0.00 10.59 0.00 2 0.00 0.00 Module#attr_reader 0.00 10.59 0.00 1 0.00 0.00 UDPSocket#initialize 0.00 10.59 0.00 1 0.00 0.00 IO#new 0.00 10.59 0.00 2 0.00 0.00 Hash#empty? 0.00 10.59 0.00 1 0.00 0.00 UDPSocket#connect 0.00 10.59 0.00 2 0.00 0.00 Time#initialize 0.00 10.59 0.00 2 0.00 0.00 Class#new 0.00 10.59 0.00 4 0.00 0.00 IO#set_encoding 0.00 10.59 0.00 2 0.00 0.00 Gem.unresolved_deps 0.00 10.59 0.00 1 0.00 0.00 Float#/ 0.00 10.59 0.00 1 0.00 0.00 Float#quo 0.00 10.59 0.00 1 0.00 0.00 Time#- 0.00 10.59 0.00 1 0.00 0.00 Float#to_s 0.00 10.59 0.00 2 0.00 0.00 IO#write 0.00 10.59 0.00 1 0.00 0.00 IO#puts 0.00 10.59 0.00 1 0.00 0.00 Kernel.puts 0.00 10.59 0.00 1 0.00 10594.00 #toplevel ``` **1.9.3 print** ``` ruby-1.9.3\bin\ruby.exe -rprofile socketsendtest.rb 25.818462 % cumulative self self total time seconds seconds calls ms/call ms/call name 38.19 6.18 6.18 1000000 0.01 0.01 IO#print 33.27 11.56 5.38 1000002 0.01 0.01 IO#write 28.45 16.16 4.60 1 4600.00 16156.00 Integer#times 0.09 16.17 0.01 2 7.50 15.00 Kernel.require 0.00 16.17 0.00 1 0.00 0.00 UDPSocket#bind 0.00 16.17 0.00 104 0.00 0.00 Module#method_added 0.00 16.17 0.00 2 0.00 7.50 Kernel.gem_original_require 0.00 16.17 0.00 4 0.00 0.00 IO#set_encoding 0.00 16.17 0.00 4 0.00 0.00 Module#private 0.00 16.17 0.00 2 0.00 0.00 Module#attr_reader 0.00 16.17 0.00 1 0.00 0.00 UDPSocket#initialize 0.00 16.17 0.00 1 0.00 0.00 IO#new 0.00 16.17 0.00 40 0.00 0.00 BasicObject#singleton_method_added 0.00 16.17 0.00 1 0.00 0.00 UDPSocket#connect 0.00 16.17 0.00 2 0.00 0.00 Time#initialize 0.00 16.17 0.00 2 0.00 0.00 Class#new 0.00 16.17 0.00 10 0.00 0.00 Class#inherited 0.00 16.17 0.00 2 0.00 0.00 Hash#empty? 0.00 16.17 0.00 2 0.00 0.00 Gem.unresolved_deps 0.00 16.17 0.00 1 0.00 0.00 Float#/ 0.00 16.17 0.00 1 0.00 0.00 Float#quo 0.00 16.17 0.00 1 0.00 0.00 Time#- 0.00 16.17 0.00 1 0.00 0.00 Float#to_s 0.00 16.17 0.00 1 0.00 0.00 IO#puts 0.00 16.17 0.00 1 0.00 0.00 Kernel.puts 0.00 16.17 0.00 1 0.00 16171.00 #toplevel ``` **3.2.2 send** ``` ruby-3.2.2\bin\ruby.exe -rprofile socketsendtest.rb 32.9209883 % cumulative self self total time seconds seconds calls ms/call ms/call name 39.03 7.72 7.72 1000000 0.01 0.02 nil# 38.17 15.27 7.55 1000000 0.01 0.01 UDPSocket#send 22.72 19.77 4.50 1 4495.00 19766.00 Integer#times 0.08 19.78 0.02 4 4.00 16.00 Kernel#require 0.00 19.78 0.00 2 0.00 0.00 Kernel#respond_to? 0.00 19.78 0.00 2 0.00 0.00 Hash#[] 0.00 19.78 0.00 2 0.00 0.00 Gem.find_unresolved_default_spec 0.00 19.78 0.00 2 0.00 0.00 Gem::Specification.unresolved_deps 0.00 19.78 0.00 2 0.00 0.00 Hash#empty? 0.00 19.78 0.00 2 0.00 0.00 Monitor#exit 0.00 19.78 0.00 378 0.00 0.00 Module#const_added 0.00 19.78 0.00 13 0.00 0.00 Class#inherited 0.00 19.78 0.00 50 0.00 0.00 BasicObject#singleton_method_added 0.00 19.78 0.00 141 0.00 0.00 Module#method_added 0.00 19.78 0.00 2 0.00 0.00 Monitor#enter 0.00 19.78 0.00 1 0.00 0.00 Module#method_defined? 0.00 19.78 0.00 1 0.00 0.00 Module#protected 0.00 19.78 0.00 1 0.00 0.00 String#=~ 0.00 19.78 0.00 3 0.00 0.00 Module#private 0.00 19.78 0.00 2 0.00 0.00 Module#attr_reader 0.00 19.78 0.00 1 0.00 0.00 UDPSocket#initialize 0.00 19.78 0.00 1 0.00 0.00 IO.new 0.00 19.78 0.00 1 0.00 0.00 UDPSocket#bind 0.00 19.78 0.00 1 0.00 0.00 UDPSocket#connect 0.00 19.78 0.00 2 0.00 0.00 Time#initialize 0.00 19.78 0.00 2 0.00 0.00 Class#new 0.00 19.78 0.00 2 0.00 0.00 Gem.discover_gems_on_require 0.00 19.78 0.00 4 0.00 0.00 IO#set_encoding 0.00 19.78 0.00 1 0.00 0.00 TracePoint#enable 0.00 19.78 0.00 1 0.00 0.00 Float#/ 0.00 19.78 0.00 1 0.00 0.00 Float#fdiv 0.00 19.78 0.00 1 0.00 0.00 Time#- 0.00 19.78 0.00 1 0.00 0.00 Float#to_s 0.00 19.78 0.00 1 0.00 0.00 IO#write 0.00 19.78 0.00 1 0.00 0.00 IO#puts 0.00 19.78 0.00 1 0.00 0.00 Kernel#puts 0.00 19.78 0.00 1 0.00 0.00 TracePoint#disable 0.00 19.78 0.00 1 0.00 19782.00 #toplevel ``` **3.2.2 print** ``` ruby-3.2.2\bin\ruby.exe -rprofile socketsendtest.rb 29.3270973 % cumulative self self total time seconds seconds calls ms/call ms/call name 33.07 6.50 6.50 1000000 0.01 0.01 IO#print 27.12 11.84 5.34 1000000 0.01 0.02 nil# 21.52 16.08 4.23 1 4234.00 19672.00 Integer#times 18.28 19.67 3.60 1000001 0.00 0.00 IO#write 0.00 19.67 0.00 2 0.00 0.00 Kernel#respond_to? 0.00 19.67 0.00 2 0.00 0.00 Hash#[] 0.00 19.67 0.00 2 0.00 0.00 Gem.find_unresolved_default_spec 0.00 19.67 0.00 2 0.00 0.00 Gem::Specification.unresolved_deps 0.00 19.67 0.00 2 0.00 0.00 Hash#empty? 0.00 19.67 0.00 2 0.00 0.00 Monitor#exit 0.00 19.67 0.00 378 0.00 0.00 Module#const_added 0.00 19.67 0.00 13 0.00 0.00 Class#inherited 0.00 19.67 0.00 50 0.00 0.00 BasicObject#singleton_method_added 0.00 19.67 0.00 141 0.00 0.00 Module#method_added 0.00 19.67 0.00 2 0.00 0.00 Gem.discover_gems_on_require 0.00 19.67 0.00 1 0.00 0.00 Module#method_defined? 0.00 19.67 0.00 1 0.00 0.00 Module#protected 0.00 19.67 0.00 1 0.00 0.00 String#=~ 0.00 19.67 0.00 3 0.00 0.00 Module#private 0.00 19.67 0.00 2 0.00 0.00 Module#attr_reader 0.00 19.67 0.00 1 0.00 0.00 UDPSocket#initialize 0.00 19.67 0.00 1 0.00 0.00 IO.new 0.00 19.67 0.00 1 0.00 0.00 UDPSocket#bind 0.00 19.67 0.00 1 0.00 0.00 UDPSocket#connect 0.00 19.67 0.00 2 0.00 0.00 Time#initialize 0.00 19.67 0.00 2 0.00 0.00 Class#new 0.00 19.67 0.00 2 0.00 0.00 Monitor#enter 0.00 19.67 0.00 4 0.00 0.00 IO#set_encoding 0.00 19.67 0.00 4 0.00 0.00 Kernel#require 0.00 19.67 0.00 1 0.00 0.00 TracePoint#enable 0.00 19.67 0.00 1 0.00 0.00 Float#/ 0.00 19.67 0.00 1 0.00 0.00 Float#fdiv 0.00 19.67 0.00 1 0.00 0.00 Time#- 0.00 19.67 0.00 1 0.00 0.00 Float#to_s 0.00 19.67 0.00 1 0.00 0.00 IO#puts 0.00 19.67 0.00 1 0.00 0.00 Kernel#puts 0.00 19.67 0.00 1 0.00 0.00 TracePoint#disable 0.00 19.67 0.00 1 0.00 19672.00 #toplevel ``` 3.2.2 seems to be spending a fair amount of time in a function called `nil#`. Though the 3.2.2 print() version does as well, but doesn't suffer the slowdown. ---------------------------------------- Bug #7968: Poor UDPSocket#send performance in ruby 2.0.0 on windows https://bugs.ruby-lang.org/issues/7968#change-105341 * Author: cs96and (Alan Davies) * Status: Assigned * Priority: Normal * Assignee: windows * ruby -v: ruby 2.0.0p0 (2013-02-24) [x64-mingw32] ---------------------------------------- I have noticed that the performance of UDPSocket#send on ruby 2.0.0 on windows is much poorer than that of 1.9.3 or 1.8.7. Running the attahced script on 2.0.0 gives the following... d:\scripts>bash -c "ruby --version" ruby 2.0.0p0 (2013-02-24) [x64-mingw32] d:\scripts>bash -c "time ruby socketsendtest.rb" real 0m2.572s user 0m0.000s sys 0m0.016s However, running the same test with 1.9.3 is much faster... d:\scripts>pik 193 d:\scripts>bash -c "ruby --version" ruby 1.9.3p374 (2013-01-15) [i386-mingw32] d:\scripts>bash -c "time ruby socketsendtest.rb" real 0m0.993s user 0m0.015s sys 0m0.016s Additionally, if I change the send call to a print (commented out in the script), then the performance is fine on 2.0.0.... d:\scripts>pik 200 d:\scripts>bash -c "ruby --version" ruby 2.0.0p0 (2013-02-24) [x64-mingw32] d:\scripts>bash -c "time ruby socketsendtest.rb" real 0m0.907s user 0m0.000s sys 0m0.015s What is send() doing that print() doesn't do that is causing the massive performance drop? Thanks Alan. ---Files-------------------------------- socketsendtest.rb (266 Bytes) -- https://bugs.ruby-lang.org/

1 0

[ruby-core:115401] [Ruby master Bug#7968] Poor UDPSocket#send performance in ruby 2.0.0 on windows
by k0kubun (Takashi Kokubun) 16 Nov '23

16 Nov '23

Issue #7968 has been updated by k0kubun (Takashi Kokubun). Just in case, could you measure the time spent on the loop instead of a whole Ruby command? The prelude script sometimes takes a lot of time, the performance varies across different versions, and `time` command would benchmark that as well. You can just call `Time.new` before and after the loop, subtract them, and report the result for each version. ---------------------------------------- Bug #7968: Poor UDPSocket#send performance in ruby 2.0.0 on windows https://bugs.ruby-lang.org/issues/7968#change-105338 * Author: cs96and (Alan Davies) * Status: Assigned * Priority: Normal * Assignee: windows * ruby -v: ruby 2.0.0p0 (2013-02-24) [x64-mingw32] ---------------------------------------- I have noticed that the performance of UDPSocket#send on ruby 2.0.0 on windows is much poorer than that of 1.9.3 or 1.8.7. Running the attahced script on 2.0.0 gives the following... d:\scripts>bash -c "ruby --version" ruby 2.0.0p0 (2013-02-24) [x64-mingw32] d:\scripts>bash -c "time ruby socketsendtest.rb" real 0m2.572s user 0m0.000s sys 0m0.016s However, running the same test with 1.9.3 is much faster... d:\scripts>pik 193 d:\scripts>bash -c "ruby --version" ruby 1.9.3p374 (2013-01-15) [i386-mingw32] d:\scripts>bash -c "time ruby socketsendtest.rb" real 0m0.993s user 0m0.015s sys 0m0.016s Additionally, if I change the send call to a print (commented out in the script), then the performance is fine on 2.0.0.... d:\scripts>pik 200 d:\scripts>bash -c "ruby --version" ruby 2.0.0p0 (2013-02-24) [x64-mingw32] d:\scripts>bash -c "time ruby socketsendtest.rb" real 0m0.907s user 0m0.000s sys 0m0.015s What is send() doing that print() doesn't do that is causing the massive performance drop? Thanks Alan. ---Files-------------------------------- socketsendtest.rb (266 Bytes) -- https://bugs.ruby-lang.org/

1 0

[ruby-core:115399] [Ruby master Bug#7968] Poor UDPSocket#send performance in ruby 2.0.0 on windows
by cs96and (Alan Davies) 15 Nov '23

15 Nov '23

Issue #7968 has been updated by cs96and (Alan Davies). JesseJohnson (Jesse Johnson) wrote in #note-12: > Can anyone replicate this on Ruby 3 or later? Still poor performance in 3.1.3... **1.9.3 send()** ``` $ ./ruby-1.9.3-p551-i386-mingw32/bin/ruby.exe --version ruby 1.9.3p551 (2014-11-13) [i386-mingw32] $ time ./ruby-1.9.3-p551-i386-mingw32/bin/ruby.exe d:/socketsendtest.rb real 0m1.086s user 0m0.000s sys 0m0.000s ``` **1.9.3 print()** ``` $ time ./ruby-1.9.3-p551-i386-mingw32/bin/ruby.exe d:/socketsendtest.rb real 0m1.108s user 0m0.000s sys 0m0.000s ``` **3.1.3 send()** ``` $ /c/tools/ruby31/bin/ruby --version ruby 3.1.3p185 (2022-11-24 revision 1a6b16756e) [x64-mingw-ucrt] $ time /c/tools/ruby31/bin/ruby d:/socketsendtest.rb real 0m1.927s user 0m0.000s sys 0m0.015s ``` **3.1.3 print()** ``` $ time /c/tools/ruby31/bin/ruby d:/socketsendtest.rb real 0m1.273s user 0m0.000s sys 0m0.015s ``` I also tried 3.2.2, however it runs incredibly quickly, and looking in Wireshark, I cannot see it actually generating any traffic. ---------------------------------------- Bug #7968: Poor UDPSocket#send performance in ruby 2.0.0 on windows https://bugs.ruby-lang.org/issues/7968#change-105336 * Author: cs96and (Alan Davies) * Status: Assigned * Priority: Normal * Assignee: windows * ruby -v: ruby 2.0.0p0 (2013-02-24) [x64-mingw32] ---------------------------------------- I have noticed that the performance of UDPSocket#send on ruby 2.0.0 on windows is much poorer than that of 1.9.3 or 1.8.7. Running the attahced script on 2.0.0 gives the following... d:\scripts>bash -c "ruby --version" ruby 2.0.0p0 (2013-02-24) [x64-mingw32] d:\scripts>bash -c "time ruby socketsendtest.rb" real 0m2.572s user 0m0.000s sys 0m0.016s However, running the same test with 1.9.3 is much faster... d:\scripts>pik 193 d:\scripts>bash -c "ruby --version" ruby 1.9.3p374 (2013-01-15) [i386-mingw32] d:\scripts>bash -c "time ruby socketsendtest.rb" real 0m0.993s user 0m0.015s sys 0m0.016s Additionally, if I change the send call to a print (commented out in the script), then the performance is fine on 2.0.0.... d:\scripts>pik 200 d:\scripts>bash -c "ruby --version" ruby 2.0.0p0 (2013-02-24) [x64-mingw32] d:\scripts>bash -c "time ruby socketsendtest.rb" real 0m0.907s user 0m0.000s sys 0m0.015s What is send() doing that print() doesn't do that is causing the massive performance drop? Thanks Alan. ---Files-------------------------------- socketsendtest.rb (266 Bytes) -- https://bugs.ruby-lang.org/

1 0

[ruby-core:115398] [Ruby master Bug#8444] Regexp vars $~ and friends are not thread local
by Dan0042 (Daniel DeLorme) 15 Nov '23

15 Nov '23

Issue #8444 has been updated by Dan0042 (Daniel DeLorme). nobu (Nobuyoshi Nakada) wrote in #note-9: > "Thread-local" in that sentence was wrong, and removed at commit:932dd9f10e684fa99b059054fbc934607d85b45a, but "method-local" is removed too. Indeed, the documentation used to say "Regexp Global Variables" or "Special Global Variables" and this has been reduced to just "Global Variables" which I think is quite misleading. Sure they have the `$` sigil, but they're not global at all. If anything, "Special Variables" would be more accurate. ---------------------------------------- Bug #8444: Regexp vars $~ and friends are not thread local https://bugs.ruby-lang.org/issues/8444#change-105335 * Author: jamespharaoh (James Pharaoh) * Status: Open * Priority: Normal * Assignee: ko1 (Koichi Sasada) * ruby -v: trunk ---------------------------------------- In the docs for the Regexp special variables, `$~` and friends, it says "These global variables are thread-local and method-local variables". However the following gives an unexpected result: ```ruby def get_proc proc do |str| str =~ /(.+)/ sleep 0.1 puts "got #{$1} from #{str}\n" end end proc = get_proc t1 = Thread.new { proc.call "abc" } t2 = Thread.new { proc.call "def" } t1.join t2.join ``` This outputs the following: ``` got abc from abc got abc from def ``` The expected result is of course: ``` got abc from abc got def from def ``` Clearly the variables are being scoped to the `get_proc` method and are being shared by both threads. This runs contrary to the documentation and also to expectations. This behaviour should either be changed, or the documentation updated to reflect the actual behaviour. Interestingly, the following does work as expected: ```ruby proc = proc do |str| str =~ /(.+)/ sleep 0.1 puts "got #{$1} from #{str}\n" end t1 = Thread.new { proc.call "abc" } t2 = Thread.new { proc.call "def" } t1.join t2.join ``` -- https://bugs.ruby-lang.org/

1 0

[ruby-core:115389] [Ruby master Bug#20006] [Windows] failed to allocate memory (NoMemoryError) - test/ruby/test_ast.rb
by MSP-Greg (Greg L) 14 Nov '23

14 Nov '23

Issue #20006 has been reported by MSP-Greg (Greg L). ---------------------------------------- Bug #20006: [Windows] failed to allocate memory (NoMemoryError) - test/ruby/test_ast.rb https://bugs.ruby-lang.org/issues/20006 * Author: MSP-Greg (Greg L) * Status: Open * Priority: Normal * Backport: 3.0: UNKNOWN, 3.1: UNKNOWN, 3.2: UNKNOWN ---------------------------------------- On GitHub Actions, all ruby-loco windows builds are intermittently failing with logs similar to the below: ``` D:\a\ruby-loco\ruby-loco\ruby-mingw\bin\ruby.exe: failed to allocate memory (NoMemoryError) D:\a\ruby-loco\ruby-loco\ruby-mingw\bin\ruby.exe: failed to allocate memory (NoMemoryError) D:\a\ruby-loco\ruby-loco\ruby-mingw\bin\ruby.exe: failed to allocate memory (NoMemoryError) D:\a\ruby-loco\ruby-loco\ruby-mingw\bin\ruby.exe: failed to allocate memory (NoMemoryError) /ruby/tool/lib/test/unit/testcase.rb:173:in `rescue in run': failed to allocate memory (NoMemoryError) < removed /ruby/tool/lib/test/unit/ lines > running file: /ruby/test/ruby/test_ast.rb ``` Of course, I cannot reproduce locally. These errors have been happening for several days. I have not tried to force the test to run single, as it's running parallel currently. Thanks, Greg -- https://bugs.ruby-lang.org/

2 1

[ruby-core:115387] [Ruby master Bug#8444] Regexp vars $~ and friends are not thread local
by nobu (Nobuyoshi Nakada) 14 Nov '23

14 Nov '23

Issue #8444 has been updated by nobu (Nobuyoshi Nakada). Description updated "Thread-local" in that sentence was wrong, and removed at commit:932dd9f10e684fa99b059054fbc934607d85b45a, but "method-local" is removed too. ---------------------------------------- Bug #8444: Regexp vars $~ and friends are not thread local https://bugs.ruby-lang.org/issues/8444#change-105324 * Author: jamespharaoh (James Pharaoh) * Status: Open * Priority: Normal * Assignee: ko1 (Koichi Sasada) * ruby -v: trunk ---------------------------------------- In the docs for the Regexp special variables, `$~` and friends, it says "These global variables are thread-local and method-local variables". However the following gives an unexpected result: ```ruby def get_proc proc do |str| str =~ /(.+)/ sleep 0.1 puts "got #{$1} from #{str}\n" end end proc = get_proc t1 = Thread.new { proc.call "abc" } t2 = Thread.new { proc.call "def" } t1.join t2.join ``` This outputs the following: ``` got abc from abc got abc from def ``` The expected result is of course: ``` got abc from abc got def from def ``` Clearly the variables are being scoped to the `get_proc` method and are being shared by both threads. This runs contrary to the documentation and also to expectations. This behaviour should either be changed, or the documentation updated to reflect the actual behaviour. Interestingly, the following does work as expected: ```ruby proc = proc do |str| str =~ /(.+)/ sleep 0.1 puts "got #{$1} from #{str}\n" end t1 = Thread.new { proc.call "abc" } t2 = Thread.new { proc.call "def" } t1.join t2.join ``` -- https://bugs.ruby-lang.org/

1 0

[ruby-core:115385] [Ruby master Bug#17037] rounding of Rational#to_f
by JesseJohnson (Jesse Johnson) 13 Nov '23

13 Nov '23

Issue #17037 has been updated by JesseJohnson (Jesse Johnson). Test case: ``` def test_rational_with_large_integer_components_to_f r = 450359962737049649/450359962737049600r assert_equal(1.0, r.to_f) end ``` The issue is that large integers, abs > 2 ** 53 - 1, are truncated when converting to double precision floating point and this affects which floating point value is closest after division. This can be fixed by using BigDecimal but I'm not sure that is an appropriate dependency for Rational. ---------------------------------------- Bug #17037: rounding of Rational#to_f https://bugs.ruby-lang.org/issues/17037#change-105321 * Author: akr (Akira Tanaka) * Status: Open * Priority: Normal * Backport: 2.5: UNKNOWN, 2.6: UNKNOWN, 2.7: UNKNOWN ---------------------------------------- I found a doubtful rounding behavior of Rational#to_f. ``` % ./ruby -ve ' a = 1r e = Float::EPSILON.to_r puts "e=#{e}" n = 100 (0..n).each {|i| r = a + e * i / n f = r.to_f p [i, f, r] } ' ruby 2.8.0dev (2020-07-20T06:39:31Z master 935d0b3d05) [x86_64-linux] e=1/4503599627370496 [0, 1.0, (1/1)] [1, 1.0, (450359962737049601/450359962737049600)] [2, 1.0, (225179981368524801/225179981368524800)] [3, 1.0, (450359962737049603/450359962737049600)] [4, 1.0, (112589990684262401/112589990684262400)] [5, 1.0, (90071992547409921/90071992547409920)] [6, 1.0, (225179981368524803/225179981368524800)] [7, 1.0, (450359962737049607/450359962737049600)] [8, 1.0, (56294995342131201/56294995342131200)] [9, 1.0, (450359962737049609/450359962737049600)] [10, 1.0, (45035996273704961/45035996273704960)] [11, 1.0, (450359962737049611/450359962737049600)] [12, 1.0, (112589990684262403/112589990684262400)] [13, 1.0, (450359962737049613/450359962737049600)] [14, 1.0, (225179981368524807/225179981368524800)] [15, 1.0, (90071992547409923/90071992547409920)] [16, 1.0, (28147497671065601/28147497671065600)] [17, 1.0, (450359962737049617/450359962737049600)] [18, 1.0, (225179981368524809/225179981368524800)] [19, 1.0, (450359962737049619/450359962737049600)] [20, 1.0, (22517998136852481/22517998136852480)] [21, 1.0, (450359962737049621/450359962737049600)] [22, 1.0, (225179981368524811/225179981368524800)] [23, 1.0, (450359962737049623/450359962737049600)] [24, 1.0, (56294995342131203/56294995342131200)] [25, 1.0, (18014398509481985/18014398509481984)] [26, 1.0, (225179981368524813/225179981368524800)] [27, 1.0, (450359962737049627/450359962737049600)] [28, 1.0, (112589990684262407/112589990684262400)] [29, 1.0, (450359962737049629/450359962737049600)] [30, 1.0, (45035996273704963/45035996273704960)] [31, 1.0, (450359962737049631/450359962737049600)] [32, 1.0, (14073748835532801/14073748835532800)] [33, 1.0000000000000002, (450359962737049633/450359962737049600)] [34, 1.0000000000000002, (225179981368524817/225179981368524800)] [35, 1.0, (90071992547409927/90071992547409920)] [36, 1.0000000000000002, (112589990684262409/112589990684262400)] [37, 1.0000000000000002, (450359962737049637/450359962737049600)] [38, 1.0000000000000002, (225179981368524819/225179981368524800)] [39, 1.0000000000000002, (450359962737049639/450359962737049600)] [40, 1.0, (11258999068426241/11258999068426240)] [41, 1.0000000000000002, (450359962737049641/450359962737049600)] [42, 1.0000000000000002, (225179981368524821/225179981368524800)] [43, 1.0000000000000002, (450359962737049643/450359962737049600)] [44, 1.0000000000000002, (112589990684262411/112589990684262400)] [45, 1.0000000000000002, (90071992547409929/90071992547409920)] [46, 1.0000000000000002, (225179981368524823/225179981368524800)] [47, 1.0000000000000002, (450359962737049647/450359962737049600)] [48, 1.0000000000000002, (28147497671065603/28147497671065600)] [49, 1.0000000000000002, (450359962737049649/450359962737049600)] [50, 1.0, (9007199254740993/9007199254740992)] [51, 1.0000000000000002, (450359962737049651/450359962737049600)] [52, 1.0000000000000002, (112589990684262413/112589990684262400)] [53, 1.0000000000000002, (450359962737049653/450359962737049600)] [54, 1.0000000000000002, (225179981368524827/225179981368524800)] [55, 1.0000000000000002, (90071992547409931/90071992547409920)] [56, 1.0000000000000002, (56294995342131207/56294995342131200)] [57, 1.0000000000000002, (450359962737049657/450359962737049600)] [58, 1.0000000000000002, (225179981368524829/225179981368524800)] [59, 1.0000000000000002, (450359962737049659/450359962737049600)] [60, 1.0000000000000002, (22517998136852483/22517998136852480)] [61, 1.0000000000000002, (450359962737049661/450359962737049600)] [62, 1.0000000000000002, (225179981368524831/225179981368524800)] [63, 1.0000000000000002, (450359962737049663/450359962737049600)] [64, 1.0000000000000002, (7036874417766401/7036874417766400)] [65, 1.0000000000000002, (90071992547409933/90071992547409920)] [66, 1.0000000000000002, (225179981368524833/225179981368524800)] [67, 1.0000000000000002, (450359962737049667/450359962737049600)] [68, 1.0000000000000002, (112589990684262417/112589990684262400)] [69, 1.0000000000000002, (450359962737049669/450359962737049600)] [70, 1.0000000000000002, (45035996273704967/45035996273704960)] [71, 1.0000000000000002, (450359962737049671/450359962737049600)] [72, 1.0000000000000002, (56294995342131209/56294995342131200)] [73, 1.0000000000000002, (450359962737049673/450359962737049600)] [74, 1.0000000000000002, (225179981368524837/225179981368524800)] [75, 1.0000000000000002, (18014398509481987/18014398509481984)] [76, 1.0000000000000002, (112589990684262419/112589990684262400)] [77, 1.0000000000000002, (450359962737049677/450359962737049600)] [78, 1.0000000000000002, (225179981368524839/225179981368524800)] [79, 1.0000000000000002, (450359962737049679/450359962737049600)] [80, 1.0000000000000002, (5629499534213121/5629499534213120)] [81, 1.0000000000000002, (450359962737049681/450359962737049600)] [82, 1.0000000000000002, (225179981368524841/225179981368524800)] [83, 1.0000000000000002, (450359962737049683/450359962737049600)] [84, 1.0000000000000002, (112589990684262421/112589990684262400)] [85, 1.0000000000000002, (90071992547409937/90071992547409920)] [86, 1.0000000000000002, (225179981368524843/225179981368524800)] [87, 1.0000000000000002, (450359962737049687/450359962737049600)] [88, 1.0000000000000002, (56294995342131211/56294995342131200)] [89, 1.0000000000000002, (450359962737049689/450359962737049600)] [90, 1.0000000000000002, (45035996273704969/45035996273704960)] [91, 1.0000000000000002, (450359962737049691/450359962737049600)] [92, 1.0000000000000002, (112589990684262423/112589990684262400)] [93, 1.0000000000000002, (450359962737049693/450359962737049600)] [94, 1.0000000000000002, (225179981368524847/225179981368524800)] [95, 1.0000000000000002, (90071992547409939/90071992547409920)] [96, 1.0000000000000002, (14073748835532803/14073748835532800)] [97, 1.0000000000000002, (450359962737049697/450359962737049600)] [98, 1.0000000000000002, (225179981368524849/225179981368524800)] [99, 1.0000000000000002, (450359962737049699/450359962737049600)] [100, 1.0000000000000002, (4503599627370497/4503599627370496)] ``` This sample program tries to convert rationals between 1.0 and 1.0 + Float::EPSILON to float. Since there is no representable float values between them (except 1.0 and 1.0 + Float::EPSILON), I expect that r.to_f returns 1.0 for r < X and 1.0 + Float::EPSILON for r > X for some X. But my expectation is not valid: ``` [34, 1.0000000000000002, (225179981368524817/225179981368524800)] [35, 1.0, (90071992547409927/90071992547409920)] ``` I.e. `1 + Float::EPSILON.to_r * 34 / 100 < 1 + Float::EPSILON.to_r * 35 / 100` but `(1 + Float::EPSILON.to_r * 34 / 100).to_f > (1 + Float::EPSILON.to_r * 35 / 100).to_f` I guess Rational#to_f should round the rational value to a nearest representable float value. -- https://bugs.ruby-lang.org/

1 0

[ruby-core:115384] [Ruby master Bug#18966] Strange behavior when numbered parameters and method definition are both used in a block
by fcheung (Frederick Cheung) 13 Nov '23

13 Nov '23

Issue #18966 has been updated by fcheung (Frederick Cheung). Ruby 3.2.2 exhibits the same behaviour as above but on master ( as of b1f345b1 ) all of these now evaluate to 'a' ---------------------------------------- Bug #18966: Strange behavior when numbered parameters and method definition are both used in a block https://bugs.ruby-lang.org/issues/18966#change-105320 * Author: tompng (tomoya ishida) * Status: Open * Priority: Normal * ruby -v: ruby 3.2.0dev (2022-08-19T04:20:20Z :detached: 17d0e5bee7) [x86_64-linux] * Backport: 2.7: UNKNOWN, 3.0: UNKNOWN, 3.1: UNKNOWN ---------------------------------------- ~~~ruby 'a'.tap { p _1; def f()=42 } #=> "a" 'a'.tap { p _1; def f(a)=42 } #=> nil 'a'.tap { def f()=42; p _1 } #=> "a" 'a'.tap { def f(a)=42; p _1 } # Syntax Error -:1: ordinary parameter is defined ~~~ -- https://bugs.ruby-lang.org/

1 0