The package rpms/ruby.git has added or updated architecture specific content in its
spec file (ExclusiveArch/ExcludeArch or %ifarch/%ifnarch) in commit(s):
https://src.fedoraproject.org/cgit/rpms/ruby.git/commit/?id=4979be53acdcf....
Change:
-%ifarch s390x
Thanks.
Full change:
============
commit 4979be53acdcfd0d6021c4f209403c2e88fae58e
Author: Vít Ondruch <vondruch(a)redhat.com>
Date: Mon Jul 27 17:16:51 2020 +0200
Avoid possible timeout errors in TestBugReporter#test_bug_reporter_add.
diff --git a/ruby-2.7.1-Timeout-the-test_bug_reporter_add-witout-raising-err.patch
b/ruby-2.7.1-Timeout-the-test_bug_reporter_add-witout-raising-err.patch
new file mode 100644
index 0000000..64e2114
--- /dev/null
+++ b/ruby-2.7.1-Timeout-the-test_bug_reporter_add-witout-raising-err.patch
@@ -0,0 +1,34 @@
+From 9b42fce32bff25e0569581f76f532b9d57865aef Mon Sep 17 00:00:00 2001
+From: =?UTF-8?q?V=C3=ADt=20Ondruch?= <vondruch(a)redhat.com>
+Date: Mon, 27 Jul 2020 14:56:05 +0200
+Subject: [PATCH] Timeout the test_bug_reporter_add witout raising error.
+
+While timeouting the threads might be still good idea, it does not seems
+the timeout impacts the TestBugReporter#test_bug_reporter_add result,
+because the output of the child process has been already collected
+earlier.
+
+It seems that when the system is under heavy load, the thread might not
+be sheduled to finish its processing. Even finishing the child process
+might take tens of seconds and therefore the test case finish might take
+a while.
+---
+ test/-ext-/bug_reporter/test_bug_reporter.rb | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/test/-ext-/bug_reporter/test_bug_reporter.rb
b/test/-ext-/bug_reporter/test_bug_reporter.rb
+index 628fcd0340..2c677cc8a7 100644
+--- a/test/-ext-/bug_reporter/test_bug_reporter.rb
++++ b/test/-ext-/bug_reporter/test_bug_reporter.rb
+@@ -21,7 +21,7 @@ def test_bug_reporter_add
+ args = ["--disable-gems", "-r-test-/bug_reporter",
+ "-C", tmpdir]
+ stdin = "register_sample_bug_reporter(12345); Process.kill :SEGV, $$"
+- assert_in_out_err(args, stdin, [], expected_stderr, encoding:
"ASCII-8BIT")
++ assert_in_out_err(args, stdin, [], expected_stderr, encoding:
"ASCII-8BIT", timeout_error: nil)
+ ensure
+ FileUtils.rm_rf(tmpdir) if tmpdir
+ end
+--
+2.27.0
+
diff --git a/ruby.spec b/ruby.spec
index bda2afd..02a6027 100644
--- a/ruby.spec
+++ b/ruby.spec
@@ -173,6 +173,9 @@ Patch16: rubygems-3.1.3-Improve-require.patch
Patch17: rubygems-3.1.3-Revert-Exclude-empty-suffix-from-I-require-loop.patch
#
https://github.com/rubygems/rubygems/pull/3639
Patch18: rubygems-3.1.3-Fix-correctness-and-performance-regression-in-require.patch
+# Avoid possible timeout errors in TestBugReporter#test_bug_reporter_add.
+#
https://bugs.ruby-lang.org/issues/16492
+Patch19: ruby-2.7.1-Timeout-the-test_bug_reporter_add-witout-raising-err.patch
# Add support for .include directive used by OpenSSL config files.
#
https://github.com/ruby/openssl/pull/216
@@ -594,6 +597,7 @@ rm -rf ext/fiddle/libffi*
%patch16 -p1
%patch17 -p1
%patch18 -p1
+%patch19 -p1
%patch22 -p1
# Provide an example of usage of the tapset:
@@ -877,13 +881,6 @@ MSPECOPTS="$MSPECOPTS -P 'File.utime allows Time instances
in the far future to
MSPECOPTS="$MSPECOPTS -P 'File.lchmod returns false from
\#respond_to?'"
MSPECOPTS="$MSPECOPTS -P 'File.lchmod raises a NotImplementedError when
called'"
-# Increase timeout for TestBugReporter#test_bug_reporter_add test, which fails
-# quite often.
-#
https://bugs.ruby-lang.org/issues/16492
-%ifarch s390x
-sed -i '/assert_in_out_err/ s/)/, timeout: 30)/'
test/-ext-/bug_reporter/test_bug_reporter.rb
-%endif
-
make check TESTS="-v $DISABLE_TESTS" MSPECOPT="-fs $MSPECOPTS"
%files
@@ -1299,6 +1296,7 @@ make check TESTS="-v $DISABLE_TESTS" MSPECOPT="-fs
$MSPECOPTS"
%changelog
* Mon Jul 27 2020 Vít Ondruch <vondruch(a)redhat.com> - 2.7.1-133
- Disable LTO, which appear to cause issues with SIGSEV handler.
+- Avoid possible timeout errors in TestBugReporter#test_bug_reporter_add.
* Wed Jun 24 2020 Jun Aruga <jaruga(a)redhat.com> - 2.7.1-132
- Add ruby-default-gems dependency on irb.
commit d461276d489a6f8f6b130c93762d98ac5e269b07
Author: Vít Ondruch <vondruch(a)redhat.com>
Date: Mon Jul 27 17:15:32 2020 +0200
Disable LTO, which appear to cause issues with SIGSEV handler.
https://src.fedoraproject.org/rpms/redhat-rpm-config/c/5baaf4a99cc77572d3...
diff --git a/ruby.spec b/ruby.spec
index d5977cb..bda2afd 100644
--- a/ruby.spec
+++ b/ruby.spec
@@ -22,7 +22,7 @@
%endif
-%global release 132
+%global release 133
%{!?release_string:%define release_string
%{?development_release:0.}%{release}%{?development_release:.%{development_release}}%{?dist}}
# The RubyGems library has to stay out of Ruby directory tree, since the
@@ -81,6 +81,10 @@
%bcond_without hardening_test
%endif
+# LTO appears to cause some issue to SEGV handler.
+#
https://bugs.ruby-lang.org/issues/17052
+%define _lto_cflags %{nil}
+
Summary: An interpreter of object-oriented scripting language
Name: ruby
Version: %{ruby_version}
@@ -1293,6 +1297,9 @@ make check TESTS="-v $DISABLE_TESTS" MSPECOPT="-fs
$MSPECOPTS"
%changelog
+* Mon Jul 27 2020 Vít Ondruch <vondruch(a)redhat.com> - 2.7.1-133
+- Disable LTO, which appear to cause issues with SIGSEV handler.
+
* Wed Jun 24 2020 Jun Aruga <jaruga(a)redhat.com> - 2.7.1-132
- Add ruby-default-gems dependency on irb.
Resolves: rhbz#1850541