Activity
From 08/06/2021 to 08/12/2021
08/12/2021
-
10:04 PM Bug #17052: Ruby with LTO enabled has issues with SIGSEGV handler
- I noticed that `addr2line.c` misinterpreted the size of `DW_FORM_ref_addr` as the size of an address. In fact it should be interpreted as the size of an offset (32 bits in 32-bit DWARF and 64 bits in 64-bit DWARF) [1]. A patch is attache...
-
03:59 PM Bug #17052 (Open): Ruby with LTO enabled has issues with SIGSEGV handler
- Changing state back to open, because this is unresolved and I am repeatedly hitting this issue in various reincarnations. So this is the debug session on PPC64LE:
~~~
$ gdb --args ./miniruby -e'Process.kill("SIGSEGV",$$)'
GNU gdb (G... -
06:35 PM Bug #18000: have_library doesn't work when ruby is compiled with --disable-shared --disable-install-static-library
- While it is possible to compile with only the headers, that only works if you know what the headers should contain. If you aren't sure what the headers contain, and you have to test, you really need to test both compiling and linking to...
-
06:16 PM Bug #18073: test/ruby/test_jit.rb: "error: invalid use of '__builtin_va_arg_pack ()'" on Ruby 2.7.4 on gcc 4.8.5
- jaruga (Jun Aruga) wrote in #note-4:
> I executed `git bisect` between `v2_7_3` and `v2_7_4`. And I found the following commit causes these test failures.
> ...
As a temporary workaround to build Ruby 2.7.4 on RHEL7, I am applying a pa... -
- @xtkoba thanks for the info. I confirmed the value of `CPP` on RHEL 7 autoconf version 2.69.
The following `configure.ac` works fine.
```
$ cat configure.ac
AC_INIT
ac_cv_prog_cc_c99=-std=gnu99
AC_PROG_CC_C99
AC_PREPROC_IFELS... -
- The problem described in #note-24 does not look like very much related to the other part. Maybe worth reopening #17052?
-
- Just testing the issue from [comment 24](https://bugs.ruby-lang.org/issues/18062#note-24) and there is not difference with or without `-fno-strict-aliasing`
-
12:37 PM Bug #18062: Ruby with enabled LTO segfaults during build
- I was able to reduce the `Hash.ruby2_keywords_hash?(1)` repro and report it to GCC.
It looks like we are hitting an optimizer bug with LTO: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101868
For workarounds, putting `__attribute__ (... -
- vo.x (Vit Ondruch) wrote in #note-26:
> BTW, In your experience, do you think trying something like [1] could help catch such issues earlier?
I'm sorry but I'm not familiar with techniques for detecting strict aliasing violation. It ... -
- xtkoba (Tee KOBAYASHI) wrote in #note-25:
> I would humbly advise that the `-fno-strict-aliasing` flag be appended to `CFLAGS` by default. See also #17540.
I have not find any guidelines about/against `-fno-strict-aliasing` in Fedora... -
01:31 PM Bug #17806: Bad interaction between method cache, prepend, and refinements
- Thanks @alanwu!
I forgot to watch this issue so I was never notified of your response. Sorry for the delay.
> I would try running these two lines before requiring the test file that contains the | call:
> ...
Thank you so much -- ... -
10:55 AM Revision 5534698b (git): [ruby/reline] Version 0.2.7
- https://github.com/ruby/reline/commit/040d29b4c0
-
- https://github.com/ruby/irb/commit/98610236ec
-
- https://github.com/ruby/irb/commit/5a0c8dc9e6
-
- This reverts commit c66b27efbb27c6b6b410457282ac841ead1dd132.
-
-
04:46 AM Revision 7ac44091 (git): Separate jobs conditions
-
02:29 AM Revision 1cdc7510 (git): RBIMPL_ATTR_DEPRECATED: enable for GCC 10.3+
- They fixed the bug.
08/11/2021
- 11:47 PM Revision e1fe209d (git): * 2021-08-12 [ci skip]
-
-
- This issue is due to a race condition because the MatchData object is passed indirectly through the backref. I've submitted a pull request to fix it: https://github.com/ruby/ruby/pull/4734
I think it's likely that similar Regexp meth... -
- jaruga (Jun Aruga) wrote in #note-7:
> Seeing Debian and Ubuntu ruby packages, the "ruby" deb package depends on "ruby-rubygems" deb package. I am curious about the situation of the "ruby" package on other Linux/UNIX platforms such as A... -
- Just for the context, this issue was triggered when looking at [1]. And as long as there is `ruby --disable-gems` as well as `./configure --disable-rubygems`, I believe that it is correct to fix this.
[1]: https://github.com/rubygems/... -
- @vo.x I would just share how the motivation of some maintainers look like to fix this kind of issue, as I chat with some maintainers on Slack today.
I understand that gems moving from bundled gem to default gem as StdLib tend to depen... -
- duerst (Martin Dürst) wrote in #note-4:
> I'm not sure how strongly this is related (and I'll create a new issue if necessary), but when I use a very simple script such as `ruby -e 'puts 3+4'`, I get the following output:
> ...
Under w... -
- > I'm not sure how strongly this is related (and I'll create a new issue if necessary), but when I use a very simple script such as ruby -e 'puts 3+4', I get the following output:
> ...
I tried to reproduce what you did on the master co... -
- I would humbly advise that the `-fno-strict-aliasing` flag be appended to `CFLAGS` by default. See also #17540.
-
- So I have tried to get more details about the #17052:
~~~
$ gdb --args ./miniruby -e'Process.kill("SIGSEGV",$$)'
GNU gdb (GDB) Fedora 10.2-6.fc35
Copyright (C) 2021 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 ... -
- nobu (Nobuyoshi Nakada) wrote in #note-21:
> Probably a memory barrier would work.
I confirmed that the `make` and `make check` work with the patch applied on the latest master commit:79cc566ab4cdf75f125ecf413a27d353a9756c08 on my lo... -
- > But it's dead in the process of make test with wrong result "green".
Sorry, my mistake. The `make test` just finished without error on the Ubuntu container. So, the "green" is correct. The steps after `make install` were skipped due... -
- Probably a memory barrier would work.
```diff
diff --git a/include/ruby/internal/value_type.h b/include/ruby/internal/value_type.h
index 52b284cb4a2..f6231d2dcac 100644
--- a/include/ruby/internal/value_type.h
+++ b/include/ruby/i... -
- > Just curious, would it make sense to update the Ruby CI to test LTO? Maybe even by default? Would there be any downside?
Yes, I think it makes sense to me, and the `.github/workflows/compilers.yml` is a right place to add. Actually I ... -
- Just curious, would it make sense to update the Ruby CI to test LTO? Maybe even by default? Would there be any downside?
-
- I could reproduce the segfault with the master commit:c59da370a5d, just by `Hash.ruby2_keywords_hash?(1)`.
```
* thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=1, address=0x3)
frame #0: 0x00000001000c... -
- As https://github.com/ruby/ruby/pull/4716 itself seems to make things better at least, I merged it for now.
-
- Applied in changeset commit:git|a0a8f2abf533702b2cd96e79f700ce5b9cd94f50.
----------
Get rid of type-punning pointer casts [Bug #18062] -
05:58 PM Bug #18053: Crashes and infinite loops when generating partial backtraces in Ruby 3.0+
- Thanks Jeremy! Just confirmed that the fix in ruby_3_0 solves all the crashes and hangs we were seeing.
-
- It seems that in `configure.ac`, `AC_PROG_CC_C99` must come before `AC_PREPROC_IFELSE` to set `CPP` correctly. That is,
```
AC_INIT
ac_cv_prog_cc_c99=-std=gnu99
AC_PROG_CC_C99
AC_PREPROC_IFELSE([AC_LANG_SOURCE([[]])],[],[])
echo ... -
- > It should be the default, unless you set CFLAGS at the command line or as the environment variable.
As we chat, I confirmed the environment variable `CFLAGS` is not set before running `autoconf`, `./configure` again. The `CFLAGS: ${... -
- jaruga (Jun Aruga) wrote in #note-10:
> @nobu OK. Could you provide a patch to change only from `S["CFLAGS"]="-g -O2 -fPIC"` to `S["CFLAGS"]="${cflags} -fPIC"` in `config.status`? Then I am happy to test the patch or you can test it on... -
- > `${cflags}` should contain ${optflags} ${debugflags} ${warnflags} by the default.
> ...
@nobu OK. Could you provide a patch to change only from `S["CFLAGS"]="-g -O2 -fPIC"` to `S["CFLAGS"]="${cflags} -fPIC"` in `config.status`? Then ... -
- jaruga (Jun Aruga) wrote in #note-8:
> However the essential difference is the following part.
> ...
`${cflags}` should contain `${optflags} ${debugflags} ${warnflags}` by the default.
That means the latter `CFLAGS` lacks these flags.... -
- This appears to have been resolved in the master branch. I ran a similar out-of-place build and ran the grep, and it looks like the transcoders were only built once.
```
$ grep '^generating .*trans.*\.c ' z.make.log
generating enc/... - 01:26 PM Revision 79cc566a (git): Make during_compacting flag in GC one bit
- Commit c32218de1ba094223420a4ea017707f48d0009c5 turned during_compacting
flag to 2 bits to support the case when there is no write barrier. But
commit 32b7dcfb56a417c1d1c354102351fc1825d653bf changed compaction to
always enable the write... -
-
07:59 AM Bug #16116: ruby2.6.3(x86/x64) use win32ole could not new Object, return 0x80070057
- I have the same issue with Ruby 3.0.2 and .NET COM assemblies. The problem seems to be related to how the assembly is registered: If InprocServer32/Default contains the full path to mscoree.dll (C:\Windows\system32\mscoree.dll) there is...
-
-
- To make the meaning of assertion easier to understand,
-
-
-
-
- Because it's sometimes nil due to race condition.
-
-
-
04:41 AM Bug #12540 (Closed): test failures when SHARABLE_MIDDLE_SUBSTRING=1
- Applied in changeset commit:git|84bf4d2ce55e16a4fb51b407a8c9c79e583596b3.
----------
Term fill in String#{,l,r}strip! even when SHARABLE_MIDDLE_SUBSTRING
Each of these methods calls str_modify_keep_cr before
term filling, which should ... -
- Each of these methods calls str_modify_keep_cr before
term filling, which should ensure the backing string
uses private memory, and therefore term filling should
not affect other strings.
Skipping the term filling was added in
a707ab4bc... -
- 01:54 AM Revision 7560bf6f (git): * 2021-08-11 [ci skip]
08/10/2021
-
11:49 PM Bug #18066: Load did_you_mean/error_highlight even with --disable-gems
- I'm not sure how strongly this is related (and I'll create a new issue if necessary), but when I use a very simple script such as `ruby -e 'puts 3+4'`, I get the following output:
```
`RubyGems' were not loaded.
`error_highlight' wa... -
- I think this change makes sense. It requires quite a few test changes, though. I submitted a pull request to implement it: https://github.com/ruby/ruby/pull/4729
-
- I think this was caused by commit:a707ab4bc8a29241440f56696098efa2f7f3ff45, which removed the term filling if SHARABLE_MIDDLE_SUBSTRING was enabled. I submitted a pull request that basically reverts that commit: https://github.com/ruby/r...
-
07:11 PM Feature #18035: Introduce general model/semantic for immutability.
- Many things discussed in the description here.
I think it's important to differentiate shallow frozen (`Kernel#frozen?`) and deep frozen (= immutable), and not try to change their meaning.
So for example overriding freeze to deep fre... -
- I identified which compiler flags make this error, segmentation fault. I tested it on the latest master commit:28d03ee776af4d630556fcaad7305340cbcf9cde . I think this info helps for developers to reproduce this error on their machine.
M... -
- nobu (Nobuyoshi Nakada) wrote in #note-13:
> That patch seems fine
Thx for confirmation
> ...
If would definitely make my life easier, but nobody have requested the backport of #17553 yet and I have not met the issue myself. For t... -
- That patch seems fine, but maybe commit:f4ce78d5c139 and commit:9c769575bfa2 also need to be backported.
-
-
-
- > This commit is for configure only.
> ...
I did ssh login to RubyCI CentOS7 server, and checked the files between commits `fd95a18059` (previous commit for `29bbad9399`) and `29bbad9399` (first failure),
```
config.status
./.ext/i... -
- jaruga (Jun Aruga) wrote in #note-4:
> I executed `git bisect` between `v2_7_3` and `v2_7_4`. And I found the following commit causes these test failures.
> ...
This commit is for configure only.
How the results of `configure` (config... -
- I executed `git bisect` between `v2_7_3` and `v2_7_4`. And I found the following commit causes these test failures.
Fix 2.7 build
https://github.com/ruby/ruby/commit/29bbad939939c6dceb804aac667ba372fdee4ef5
---
I checked like t... -
- I confirmed this failures are reproduced at RubyCI - CentOS 7.6(1810) x86_64 - 2.7 case.
http://rubyci.s3.amazonaws.com/centos7/ruby-2.7/log/20210810T041843Z.fail.html.gz
-
- It has not been used since 1b82c877dfa72e8505ded149fd0e3ba956529d3f.
-
- Thanks for the fix! I confirmed I was able to build on the latest master.
-
- Applied in changeset commit:git|e3dcb6204d66c62aedaf94cac336d93d0275ce2c.
----------
Run only directories or *_spec.rb files only by mspec [Bug #18072]
Not to match *.c files under spec/ruby/optional/capi/ext, in the
case of in-place b... -
- jaruga (Jun Aruga) wrote:
> ```
> ...
The target at defs/gmake.mk:359 is `spec/%`, seems miss-firing `*.c` files.
I've missed the in-place build case.
Eregon (Benoit Daloze) wrote in #note-2:
> @nobu What's the reason to build C A... -
06:47 AM Revision 42b6dc84 (git): add some lines to delete unused TracePoint
- `TracePoint.stat` returns the "to be deleted" TP numbers, and
http://rubyci.s3.amazonaws.com/graviton2/ruby-master/log/20210810T030003Z.fail.html.gz
shows there is a "to be deleted" TP.
This patch uses only :line event and add some line... -
- A method is bound to the internal states of the receiver.
I don't think that removing the receiver from a method makes sense. -
- Not to match *.c files under spec/ruby/optional/capi/ext, in the
case of in-place build. -
02:33 AM Bug #18074 (Closed): ARGF.read(length) exhibits short rea - Applied in changeset commit:git|8df1ace64a7695c855bf0a774e3fd70edfab0bf3.
----------
Fix ARGF.read(length) short read [Bug #18074] - 02:32 AM Revision 8df1ace6 (git): Fix ARGF.read(length) short read [Bug #18074]
-
- It seems LLVM-14 begun.
See also
https://github.com/llvm/llvm-project/commit/08c766a7318ab37bf1d77e0c683cd3b00e700877
08/09/2021
-
11:28 PM Bug #18074: ARGF.read(length) exhibits short rea
- Proposed fix: https://github.com/ruby/ruby/pull/4727
-
- It was commit [4f5976c](https://github.com/ruby/ruby/commit/4f5976cbb826df462c62a0a8e72ebe91c15a7ba4) that introduced this bug:
```
commit 4f5976cbb826df462c62a0a8e72ebe91c15a7ba4
Author: matz <matz@b2dd03c8-39d4-4d8f-98ff-823fe69b0... -
- [`argf_read`](https://github.com/ruby/ruby/blob/v3_0_2/io.c#L12307-L12351) pseudocodized:
``` c
/* 1 */ argf_read(VALUE length)
/* 2 */ {
/* 3 */ long len = NUM2LONG(length);
/* 4 */ VALUE str = StringValue();
/* 5 */ retry:
/* ... -
- If ruby is invoked with three file name arguments (which refer to extant non-empty regular files), and `ARFG.read` is called with a length argument that exceeds the combined size of the first two files[1], then only the content of the fi...
-
- @nobu What's the reason to build C API specs extensions eagerly?
Can't we let ruby/spec build them at runtime as normal? -
- I see the build error happens on the following commit. While it works on the previous commit. Maybe the commit causes the build error. Do you have any idea to fix it?
Build rubyspec CAPI extensions
https://github.com/ruby/ruby/commit... -
- I just hit the build error by `make` on my machine Fedora 34 on the latest master 9ed196e4876dd15991feee27e7dce1b67783060f .
```
$ cat /etc/fedora-release
Fedora release 34 (Thirty Four)
$ rpm -q gcc
gcc-11.2.1-1.fc34.x86_64
$ g... - 08:32 PM Revision 6771316d (git): .travis.yml: Check failures for Travis arm64 again.
- It seems the concurrent jobs are stable recently on arm64 pipeline.
Keep allowing to fail on arm32 due to a failure. -
- * [Bug #18036] Pthread fibers become invalid on fork - different from normal fibers. (jeremyevans0)
* `fork` terminates other threads.
* When using pthread coroutine, all fibers are threads, therefore `fork` terminates all other fi... -
- I don't think we should make a range of literal ranges a parse error. While a range of ranges is not useful by default, a developer could define `Range#<=>` and `Range#succ` such that a range of ranges could be useful:
```ruby
class... -
- I don't think the current behavior is a bug. `Method#to_proc` is currently equivalent to:
```ruby
class Method
def to_proc
method = self
->(*args, **kwargs, &block) do
method.call(*args, **kwargs, &block)
end... -
- I hit the following JIT test failures causing the following error on v2_7_4 tag (Ruby 2.7.4) with gcc 4.8.5 on RHEL 7. The tests work on Ruby 2.7.3 and also works on Ruby 3.0.2.
```
/mnt/git/ruby/ruby/.ext/include/x86_64-linux/rb_mji... -
- I think this is expected. You get the same error if you compose the proc manually, and the proc composition operators should reflect the same behavior as manual composition:
```ruby
func1 = -> (x) { x + y }
class A
def y
... -
- 05:38 PM Revision 71c20c56 (git): * 2021-08-10 [ci skip]
-
-
-
- I compared the results between the latest release of kpeg gem (1.1.0)
and HEAD of kpeg gem with `time` command, and the execution time is
reduced from 38.35s sec to 35.06 sec.
ref. https://github.com/evanphx/kpeg/pull/43
https://github... -
- https://github.com/ruby/rdoc/commit/4e3e635afa
-
-
- https://github.com/ruby/rdoc/commit/a177377b6f
-
- https://github.com/ruby/rdoc/commit/e4b68d414c
-
-
-
05:19 PM Bug #18071: Bug in ruby interpreter or extension library
- I believe that this is the same issue as #17862, #17922, #17988, and https://github.com/typhoeus/ethon/issues/193.
-
-
11:26 AM Bug #18071: Bug in ruby interpreter or extension library
- The backtrace shows the crash happens in the `ethon` gem (calling FFI functions), so it's very likely to be a third party bug, and should be reported there: https://github.com/typhoeus/ethon
-
11:19 AM Bug #18071 (Third Party's Issue): Bug in ruby interpreter or extension library
- ecofoote@Matthews-Mac-mini ios % pod install
Adding a custom script phase for Pod RNFBApp: [RNFB] Core Configuration
Auto-linking React Native modules for target `EcoFoote`: RNCAsyncStorage, RNCMaskedView, RNDateTimePicker, RNFBApp... -
- > Thanks for the patch! Let me check it.
> ...
I tried the 2nd patch on an older master commit 1cb5a669d39fe741388be7ac45fc88c336b097cf , as I couldn't build on the latest master commit due to #18072 . But I still see similar or same re... -
- > That patch contains too many unrelated changes.
You mean mine? Or the original one?
If mine, most of the diff is because I renamed a bunch of `gc_prof_*` in just `gc_*` as they are no longer `GC::Profiler` specific. Also the link... -
- That patch contains too many unrelated changes.
And my concern is if `:time` is clear enough. Can you tell its unit by that name? -
- I would just share my update for this ticket. I wanted to check this ticket after fixing another issue #18002 that is needed to print the result of the Ubuntu s390x on [rubyci.org](https://rubyci.org/). However I may check it in parallel...
-
- And it seems that the DWARF is in better shape:
~~~
$ nm ./miniruby | grep -w rb_f_kill
00000000001ddcf0 T rb_f_kill
000000000002bd88 t rb_f_kill.cold
$ nm ./miniruby | grep -w rb_f_kill | addr2line -e ./miniruby
/builddir/buil... -
- This is still unresolved :(
-
-
-
- Unfortunately, testing on other arches, the #17052 is still an issue :(
-
- Testing with Ruby 3.0.2, I had to modify the patch slightly and I was able to build the Ruby as well as pass test suite. However, I'd like if you can review the modifications I did for Ruby 3.0.2 in the vm_method.c hunks (I have replaced...
-
-
- 05:03 AM Revision 6f6a84f2 (git): Extended logging for debugging readline failures.
-
-
-
-
- * bitwise operation between different enumeration types
('ruby_value_type' and 'ruby_fl_type') is deprecated
[-Wdeprecated-enum-enum-conversion]
* volatile-qualified parameter type 'volatile int' is deprecated
[-Wdeprecated-volatile] -
-
08/08/2021
- 11:41 PM Revision dc2eb6f7 (git): * 2021-08-09 [ci skip]
- 11:40 PM Revision 48c43f77 (git): Rework the readline test to be more robust.
- - Capture that the child is started by initial log line.
- More robust handling of child status reaping.
- Direct exit without sucess mesage if `#readline` receives input. -
06:33 PM Feature #18070 (Open): `attr` should be removed
- I'm digging through some older Ruby versions and discovered that `attr` was deprecated in Ruby 1.9. We still are seeing support for it in Ruby 3.0
Unless there are internals that require its existence, is it time to remove this deprec... -
02:30 PM Feature #18069 (Open): `instance_exec` is just ignored when the block is originally a method
I know you can't `instance_exec` a proc which is generated by `Method#to_proc` because it has its original instance's context. But, in such a case, raising `ArgumentError` would be the ideal behavior.
```ruby
f = -> (x) { a + x }
...-
02:21 PM Feature #18068 (Closed): Silence LoadError only if it is for rubygems itself
- **Background**
The "Optionally load did_you_mean and RubyGems" PR (https://github.com/ruby/ruby/pull/2764) attempted to prevent raising `LoadError` when requiring `rubygems` because it might not be available optionally when Ruby is ... -
- ```ruby
func1 = -> (x) { x + y }
class A
def y
10
end
end
A.new.instance_exec(1, &func1) # => 11
func2 = -> (x) { x * 2 }
f = func1 >> func2
A.new.instance_exec(1, &f) # => undefined local variable or method `y' f... - 10:12 AM Revision 3a8cadcf (git): Reduce chance to receive EBADF when closing an IO from another thread.
-
-
-
-
04:01 AM Revision 3922394c (git): Bump patchlevel.
-
- ruby_3_0 c07545bbf82068f4fd92a5ccfa2b09bb96b39adb.
-
- Hello Jeremy,
Thank you for your great work on the partial backtrace optimization and deep understanding for the stable branch maintenance. I'd like to adopt your 3rd option. I will merge your PR after confirming the CI passed with it. -
- This fixes multiple bugs found in the partial backtrace
optimization added in 3b24b79.
These bugs occurs when passing a start argument to caller where
the start argument lands on a iseq frame without a pc.
Before this commit, the follow... -
- ruby_3_0 d6b8b37a2608a0fbb8bcfd10356bb25f9ebbe1fe merged revision(s) e5dd40b1f3a11f48d566413ab347ce0cfdd94960.
-
- Stop force-recycling evacuated array [Bug #18065]
---
hash.c | 1 -
1 file changed, 1 deletion(-) -
- 12:25 AM Revision c6398d3d (git): * 2021-08-08 [ci skip]
08/07/2021
-
- However, I am not sure what is the impact in case Ruby can be build with `--disable-gems` configuration option (if that is possible).
-
- I guess that did_you_mean was not possible to load without RubyGems, when it used to be bundled gem. Since it is default gem, therefore part of StdLib and always available on load path, it should not be related to --disable-gems anymore....
-
- Eregon (Benoit Daloze) wrote in #note-8:
> Their description should still match the tested behavior though,
> ...
OK, I pushed another commit for that. -
06:18 PM Bug #15790: Strange interaction between autoload and $LOADED_FEATURES
- BTW, Zeitwerk has its own "Ruby compatibility suite" for language details the implementation relies on. Like, you can override autoloads, remove_const does not execute autoloads, etc.: https://github.com/fxn/zeitwerk/blob/main/test/lib/z...
-
- Indeed, to me, this belongs to the contract of these APIs and should be documented.
The documentation of `Module#autoload` should say what happens if the file is not found, if the file is found but raises, if the file is found and is ... -
- Thanks, it's fine to modify these specs.
Their description should still match the tested behavior though,
for "does not remove the constant from Module#constants if the loaded file does not define it, but leaves it as 'undefined'"
s... -
- jeremyevans0 (Jeremy Evans) wrote in #note-6:
> I updated the pull request to remove the constant instead of just deleting the autoload. If it passes CI, I'm fine with that approach.
Removing the constant breaks 4 specs:
```
ruby... -
- I updated the pull request to remove the constant instead of just deleting the autoload. If it passes CI, I'm fine with that approach.
-
- Indeed, the concept of undefined constants for me is dubious. An autoload claims that the constant should be in the given file, but you don't know until the file is evaluated. So, it has been always surprising to me that the constants AP...
-
- I deleted my previous comment, misunderstood the situation.
Yeah, in my view if you fire an autoload, the autoload should be removed regardless of the outcome. Additionally, if the constant has not been defined, it should be removed (... -
- Maybe we should remove the constant as well if we remove the autoload?
I think that would be natural and simpler semantically.
Currently there is some really strange semantics around that, which lead to a lot of complexity (e.g., the c... -
-
-
- byroot (Jean Boussier) wrote in #note-6:
> Ideally we'd return a float, but as noted by @jasonrclark, it's not easy to return anything else than `size_t` without breaking the `size_t rb_gc_stat(VALUE)` C API.
If that precision is needed... -
-
05:18 AM Revision ac7986f4 (git): Remove unneeded rb_fiber_transfer_kw declaration
-
- * Hide read function warning in string_spec_RSTRING_PTR_read function
* The type of `read` may be `ssize_t`
Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org> -
- Applied in changeset commit:git|e5dd40b1f3a11f48d566413ab347ce0cfdd94960.
----------
Stop force-recycling evacuated array [Bug #18065] -
- It looks likely related to commit:84d9a9afc0b49d095541acb9832f8b12fb506e19.
-
12:03 AM Bug #18065 (Closed): 3.0.2 - possible memory leak in Hash#transform_keys!
- I found what may be a memory leak in Hash#transform_keys! in version 3.0.2
This sample code will reproduce the issue. It quickly consumes gigabytes of memory when running on 3.0.2, but has very normal memory usage on 3.0.1. I test... -
-
- OK, how about this?
https://github.com/ruby/ruby/pull/4716
08/06/2021
-
- This appears to still be an issue in the master branch. I submitted an upstream pull request to fix it: https://github.com/ruby/rinda/pull/1
-
10:12 PM Bug #18061: Execshield test: libruby.so.N.N.N: FAIL: property-note test because no .note.gnu.property section found
- ```
On another side, we have an Intel CET [2], [3]:
"Control-flow Enforcement Technology (CET) provides the following capabilities to defend
against ROP/JOP style control-flow subversion attacks:
- Shadow Stack - return address prote... -
- Thanks for the patch! Let me check it.
-
- That seems the same as a section generated by GCC.
-
- Applied in changeset commit:git|d16b68cb2204eeb5af8bd39149202b630374c67f.
----------
Use Rational for Float#round with ndigits > 14
ndigits higher than 14 can result in values that are slightly too
large due to floating point limitatio... -
- Applied in changeset commit:git|d16b68cb2204eeb5af8bd39149202b630374c67f.
----------
Use Rational for Float#round with ndigits > 14
ndigits higher than 14 can result in values that are slightly too
large due to floating point limitatio... -
- ndigits higher than 14 can result in values that are slightly too
large due to floating point limitations. Converting to rational
for the calculation and then back to float fixes these issues.
Fixes [Bug #14635]
Fixes [Bug #17183]
Co-... -
- This is still an issue in the master branch. One way to fix this is to delete the autoload if the constant isn't present after the file is loaded. I submitted a pull request for that: https://github.com/ruby/ruby/pull/4715
-
- > FWIW TruffleRuby and JRuby both have a :time value in GC.stat and it represents the total GC time, in milliseconds
Ah that's good to know. Thank you!
In this case I think the patch could look like this: https://github.com/Shopif... -
- FWIW TruffleRuby and JRuby both have a `:time` value in `GC.stat` and it represents the total GC time, in milliseconds (`GarbageCollectorMXBean#getCollectionTime` provides this data on the JVM)
(No GC::Profiler needed to query it)
If thi... -
- I'd like to restart the discussion on this as I recently hit the same issue.
> With incremental GC (incremental marking and lazy sweep) consume very short time for each step.
Indeed. However when running a web service, even if ind... -
-
- I committed my rewrite of backtrace generation (commit:1a05dc03f953830564c272665c47a61e53550f3e), which fixes these issues in master/3.1. I took the examples from this issue and added them as tests in that commit.
For Ruby 3.0, there ... -
- Applied in changeset commit:git|1a05dc03f953830564c272665c47a61e53550f3e.
----------
Make backtrace generation work outward from current frame
This fixes multiple bugs found in the partial backtrace
optimization added in 3b24b7914c1693... - 05:15 PM Revision 0d3520b0 (git): * 2021-08-07 [ci skip]
-
- This fixes multiple bugs found in the partial backtrace
optimization added in 3b24b7914c16930bfadc89d6aff6326a51c54295.
These bugs occurs when passing a start argument to caller where
the start argument lands on a iseq frame without a pc... -
-
07:32 AM Bug #18064: Backport fix for bug 7844 (include/prepend satisfiable module dependencies are not satisfied)
- Upgrading Ruby from 2.7 to 3.0 is too much work just for this bug fix
Especially when many existing gems don't work properly in 3.x
This backport is required to address
[an issue about usage of a gem](https://github.com/panorama-ed... -
- Bug:
https://bugs.ruby-lang.org/issues/7844
Already merged PR with fix (merged into 3.0):
https://github.com/ruby/ruby/pull/3796
I have already created a PR for backport:
https://github.com/ruby/ruby/pull/4711 -
- nobu (Nobuyoshi Nakada) wrote in #note-8:
Thx, this helps. Howeverer, there is more:
~~~
(gdb) r -I./lib -I. -I.ext/common ./tool/runruby.rb --extout=.ext -- --disable-gems "./libexec/rdoc" --root "." --encoding=UTF-8 --all --ri -... -
- vo.x (Vit Ondruch) wrote in #note-5:
> ~~~
> ...
I've missed here.
-
- I should probably include a few more lines from the log which include the build options:
~~~
gcc -O2 -flto=auto -ffat-lto-objects -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-... -
- nobu (Nobuyoshi Nakada) wrote in #note-4:
> Maybe a strict aliasing issue?
The patch changes the error:
~~~
./miniruby -I./lib -I. -I.ext/common ./tool/generic_erb.rb -c -o encdb.h ./template/encdb.h.tmpl ./enc enc
<internal:ra... -
- Eregon (Benoit Daloze) wrote in #note-5:
> `make clean` is sometimes needed when building CRuby (correct C files dependencies is still a mess to this day, not CRuby-specific).
> ...
Thanks, that helped. -
- duerst (Martin Dürst) wrote in #note-4:
> ```
> ...
This file should be updated, but actually wasn't.
Could you try again after removing that file? -
- `make clean` is sometimes needed when building CRuby (correct C files dependencies is still a mess to this day, not CRuby-specific).
Would that help here? -
- nobu (Nobuyoshi Nakada) wrote in #note-3:
> You can remove the old directory, and `ext/-test-/exts.mk` file.
Many thanks for your help. I removed both the directory and the file. Now I'm getting the following:
```
transdb.h uncha... -
-
- * [Feature #10917] Add GC.stat[:total_time] (byroot)
* Time spent in GC is a very important metric for application performance monitoring.
* Right now only `GC::Profiler` can be used, but it is ill suited when there are multiple call... -
- Thanks @jeremyevans0 for informing the backport request before I create the backport ticket
A backport ticket is created:
https://bugs.ruby-lang.org/issues/18064
I have provided the reason for requesting this backport to 2.7 in th... -
-
-
- http://rubyci.s3.amazonaws.com/solaris11-gcc/ruby-master/log/20210806T000008Z.fail.html.gz
```
1) Error:
Reline::Terminfo::Test#test_tigetstr:
Reline::Terminfo::TerminfoError: The terminfo database could not be found.
/export/home/...