aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* support builtin for Kernel#FloatS.H2020-04-223-25/+32
| | | | | | | | | | | | | # Iteration per second (i/s) | |compare-ruby|built-ruby| |:------------|-----------:|---------:| |float | 30.395M| 38.314M| | | -| 1.26x| |float_true | 3.833M| 27.322M| | | -| 7.13x| |float_false | 4.182M| 24.938M| | | -| 5.96x|
* .github/workflows/mingw.yml: add `shell: bash` for PR checkoutYusuke Endoh2020-04-221-0/+1
| | | | as well as e62aead26909e83f2c8b940186047f0a88b9f2d6
* configure.ac: Automatically add -D__ANDROID_API=ver on AndroidYusuke Endoh2020-04-221-0/+24
|
* * 2020-04-22 [ci skip]git2020-04-221-1/+1
|
* test/ruby/test_fiber.rb: Skip the problematic test_stack_size on riscvYusuke Endoh2020-04-221-0/+1
| | | | | | | Since it is impossible to detect stack overflow of C functions robustly, the test is a bit unreasonable. https://rubyci.org/logs/rubyci.s3.amazonaws.com/debian-riscv64/ruby-master/log/20200421T120018Z.fail.html.gz
* LEGAL: fixed links to the old-style BSD license [ci skip]Nobuyoshi Nakada2020-04-211-3/+3
|
* * remove trailing spaces. [ci skip]git2020-04-211-19/+19
|
* LEGAL: moved the old-style BSD license and linked [ci skip]Nobuyoshi Nakada2020-04-211-83/+34
|
* LEGAL: moved the MIT license and linked [ci skip]Nobuyoshi Nakada2020-04-211-23/+6
|
* LEGAL: made license names themselves links [ci skip]Nobuyoshi Nakada2020-04-211-8/+2
|
* eval_error.c: default nil as "reverse" is now false [Feature #8661]Nobuyoshi Nakada2020-04-212-6/+4
|
* NEWS.md: the order of backtrace [Feature #8661] [ci skip]Nobuyoshi Nakada2020-04-211-0/+5
|
* eval_error.c: revert the "reversed" backtrace [Feature #8661]Yusuke Endoh2020-04-211-1/+1
| | | | | | | | | | | | | | | | | | Now, the order is good, old-fashioned style: ``` $ ./local/bin/ruby -e 'def foo; raise; end def bar; foo; end def baz; bar; end def qux; baz; end qux ' -e:1:in `foo': unhandled exception from -e:2:in `bar' from -e:3:in `baz' from -e:4:in `qux' from -e:5:in `<main>' ```
* rb_memerror: abort immediately卜部昌平2020-04-211-0/+3
| | | | | Ditto for adab82b9a71f60ad1c7f4f8c134a5ae9198ab32a. TRY_WITH_GC was found innocent.
* Fixed a typo [ci skip]Nobuyoshi Nakada2020-04-211-1/+1
|
* reroute redefinition of NDEBUG卜部昌平2020-04-214-0/+4
| | | | | NDEBUG can be defined via a command-line argument. Should take care of such situations.
* __GNUC__ is too lax卜部昌平2020-04-212-3/+3
| | | | Ditto for 4b853932eaa7fa4acf8a0f0c9b7c695bb4f5e76d
* TRY_WITH_GC: abort immediately卜部昌平2020-04-211-4/+36
| | | | | NoMemoryError is observed on icc but I fail to reproduce so far. Let me see the backtrace on CI.
* Fix typos [ci skip]Kazuhiro NISHIYAMA2020-04-213-3/+3
|
* Fixed incorrect man path with ruby installation pathHiroshi SHIBATA2020-04-211-0/+2
| | | | [Bug #15359][ruby-core:90164]
* test/ruby/test_refinement.rb: extend the timeoutYusuke Endoh2020-04-211-1/+1
| | | | https://rubyci.org/logs/rubyci.s3.amazonaws.com/debian-riscv64/ruby-master/log/20200420T083601Z.fail.html.gz
* * 2020-04-21 [ci skip]git2020-04-211-1/+1
|
* vm_dump.c: Do not show C backtrace on riscvYusuke Endoh2020-04-201-1/+1
| | | | | Currently, objdump -W miniruby emits some errors on riscv, so I guess that DWARF is corrupted.
* Skip JIT tests on riscv64 due to SEGV of cc1Yusuke Endoh2020-04-201-0/+2
|
* [sync_default_gems.rb] Force reset conflict files to be ignoredNobuyoshi Nakada2020-04-201-3/+6
| | | | [ci skip]
* Revert irrelevant change [ci skip]Nobuyoshi Nakada2020-04-201-7/+3
|
* Removed phony atomic operations for void* and VALUENobuyoshi Nakada2020-04-202-35/+11
|
* * 2020-04-20 [ci skip]git2020-04-201-1/+1
|
* Bail out if no atomic operation foundNobuyoshi Nakada2020-04-201-24/+1
| | | | | As atomic operations are mandatory now, not-working phony fallback definitions are not only useless but confusing and harmful.
* Removed useless VMDEBUG definition [ci skip]Nobuyoshi Nakada2020-04-191-6/+0
| | | | VMDEBUG is always defined as defaulted to 0 in vm_core.h.
* Update VMDEBUG reference doc [ci skip]Nguyễn Quang Minh2020-04-191-1/+1
| | | Since this commit (https://github.com/ruby/ruby/commit/9e1b06e17d27fb4ddf51e9244f205417e9c4dd5c), the VM Debug Level constant is moved from `vm_insnhelper.h` to `vm_core.h`. This PR is a super tiny update to reflect that change so that people won't waste time on searching in a wrong file.
* Skip Process#clock_getres specs on AndroidYusuke Endoh2020-04-191-2/+2
| | | | ... just like AIX and OpenBSD.
* Skip TestRequire#test_loading_fifo_fd_leak on AndroidYusuke Endoh2020-04-191-0/+2
| | | | | The test fails due to unknown reason. Need to debug in future, but tentatively skipped.
* The pdb header error is printed at stdoutTakashi Kokubun2020-04-181-2/+2
| | | | https://ci.appveyor.com/project/ruby/ruby/builds/32278754/job/90jmky2jq2k0wjv8
* Update workflows/mingw.yml - use setup-ruby-pkgs (#3042)MSP-Greg2020-04-181-12/+5
| | | MSP-Greg/actions-ruby is deprecated...
* * 2020-04-19 [ci skip]git2020-04-191-1/+1
|
* Env values removed by ENV.clear are not usedNobuyoshi Nakada2020-04-181-4/+3
|
* Bypass env key encoding conversion if unnecessaryNobuyoshi Nakada2020-04-181-9/+13
|
* Hoisted out reset_by_modified_envNobuyoshi Nakada2020-04-181-10/+12
|
* Environment variable values are not case-insensitiveNobuyoshi Nakada2020-04-181-2/+2
| | | | Only the names are case-sensitive.
* [ruby/reline] Add ed_search_next_historyaycabta2020-04-182-0/+126
| | | | https://github.com/ruby/reline/commit/ca750b676b
* [ruby/reline] Add ed_search_prev_historyaycabta2020-04-182-0/+116
| | | | https://github.com/ruby/reline/commit/e9ae288825
* [ruby/reline] Rename wrong test nameaycabta2020-04-181-1/+1
| | | | https://github.com/ruby/reline/commit/8480db575b
* [ruby/reline] Use vi_search_{prev,next} to incremental searchaycabta2020-04-184-14/+14
| | | | https://github.com/ruby/reline/commit/cab312f584
* [ruby/reline] Rename search_history with incremental_search_historyaycabta2020-04-181-3/+3
| | | | https://github.com/ruby/reline/commit/d563063ea0
* Compare environment variable names in those manor [Bug #16798]Nobuyoshi Nakada2020-04-182-3/+21
|
* power_assert repository is transferred to ruby/power_assertKazuki Tsujimoto2020-04-181-1/+1
|
* Make sure newarraykwsplat accesses a correct indexTakashi Kokubun2020-04-182-1/+11
| | | | | | | on stack when local_stack_p is enabled. This fixes `RB_FL_TEST_RAW:"RB_FL_ABLE(obj)"` assertion failure on power_assert's test with JIT enabled.
* RUBY_SPECIAL_SHIFT and RUBY_FL_USHIFT are mandatory for rp in lldb_cruby.pyNobuyoshi Nakada2020-04-183-6/+6
|
* Super TestJIT#setup to skip unsupported envsTakashi Kokubun2020-04-181-3/+1
| | | | Slightly simplifying 8d6aa06620b316904fd10d0cab1b85e07f2fbf67