BR: gcc vs gcc-c++

[Vít Ondruch]

Hi,

Ruby currently fails to build due to removal of GCC from buildroot. The
build failure is due to one test case [1] which fails now and which
should be probably fixed to pass even without GCC.

However, digging into this, I wonder if Ruby should have "BR: gcc" or
"BR: gcc-c++". The thing is, that Ruby is checking presence of C++
compiler during its configuration phase. When I was asking upstream what
is the reason for that [2], the answer was that some extension/gems
might use C++. Generally, for distribution purpose, I think that it does
not really make sense and every extension should be configured
separately, therefore we should have just "BR: gcc", i.e. no C++
compiler available during build. However, I am not really sure if this
is going to break something or not. I have tried to build
rubygem-eventmachine at it seems to build just fine, but what else might
be broken?

Anyway, for now, I'll probably stick with "BR: gcc" and we can change to
"BR: gcc-c++" later, if we discover it causes real troubles.


Vít


[1] https://bugzilla.redhat.com/show_bug.cgi?id=1606143#c4

[2] https://bugs.ruby-lang.org/issues/14590

[Mamoru TASAKA]

Vít Ondruch wrote on 07/26/2018 06:28 PM:
>  Hi,
>  
>  Ruby currently fails to build due to removal of GCC from buildroot. The
>  build failure is due to one test case [1] which fails now and which
>  should be probably fixed to pass even without GCC.
>  
>  However, digging into this, I wonder if Ruby should have "BR: gcc" or
>  "BR: gcc-c++". The thing is, that Ruby is checking presence of C++
>  compiler during its configuration phase.  
ruby-libs contains:


/usr/lib64/ruby/rbconfig.rb:180:  CONFIG["CXXFLAGS"] = "-O2 -g -pipe -Wall
-Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions
-fstack-protector-strong -grecord-gcc-switches
-specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
-specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic -fasynchronous-unwind-tables
-fstack-clash-protection -fcf-protection"
/usr/lib64/ruby/rbconfig.rb:181:  CONFIG["CXX"] = "g++"
/usr/share/ruby/mkmf.rb:74:  CXX_EXT = %w[cc mm cxx cpp]
/usr/share/ruby/mkmf.rb:76:    CXX_EXT.concat(%w[C])
/usr/share/ruby/mkmf.rb:82:  SRC_EXT = C_EXT + CXX_EXT
/usr/share/ruby/mkmf.rb:1830:        $CXXFLAGS += " " << cflags
/usr/share/ruby/mkmf.rb:1963:CXX = #{CONFIG['CXX']}
/usr/share/ruby/mkmf.rb:1984:CXXFLAGS = $(CCDLFLAGS) #$CXXFLAGS $(ARCH_FLAG)
/usr/share/ruby/mkmf.rb:2429:    command = compile_command % COMPILE_CXX
/usr/share/ruby/mkmf.rb:2430:    asm_command = compile_command.sub(/compiling/,
'translating') % ASSEMBLE_CXX
/usr/share/ruby/mkmf.rb:2431:    CXX_EXT.each do |e|
/usr/share/ruby/mkmf.rb:2458:    if
srcs.any?(&%r"\.(?:#{CXX_EXT.join('|')})\z".method(:===))
/usr/share/ruby/mkmf.rb:2510:    $CXXFLAGS = (with_config("cxxflags",
arg_config("CXXFLAGS", config["CXXFLAGS"]))||'').dup
/usr/share/ruby/mkmf.rb:2666:  COMPILE_CXX = config_string('COMPILE_CXX') ||
'$(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<'
/usr/share/ruby/mkmf.rb:2676:  ASSEMBLE_CXX = config_string('ASSEMBLE_CXX') ||
COMPILE_CXX.sub(/(?<=\s)-c(?=\s)/, '-S')

So it is quite reasonable that ruby compilation checks
CXX compiler.

Regards,
Mamoru

[Vít Ondruch]

Dne 26.7.2018 v 13:40 Mamoru TASAKA napsal(a):
>  Vít Ondruch wrote on 07/26/2018 06:28 PM:
> > Hi,
> >
> > Ruby currently fails to build due to removal of GCC from buildroot. The
> > build failure is due to one test case [1] which fails now and which
> > should be probably fixed to pass even without GCC.
> >
> > However, digging into this, I wonder if Ruby should have "BR: gcc" or
> > "BR: gcc-c++". The thing is, that Ruby is checking presence of C++
> > compiler during its configuration phase. 
>
>  ruby-libs contains:
>
>
>  /usr/lib64/ruby/rbconfig.rb:180:  CONFIG["CXXFLAGS"] = "-O2 -g -pipe
>  -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
>  -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
>  -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
>  -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
>  -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection"
>  /usr/lib64/ruby/rbconfig.rb:181:  CONFIG["CXX"] = "g++"
>  /usr/share/ruby/mkmf.rb:74:  CXX_EXT = %w[cc mm cxx cpp]
>  /usr/share/ruby/mkmf.rb:76:    CXX_EXT.concat(%w[C])
>  /usr/share/ruby/mkmf.rb:82:  SRC_EXT = C_EXT + CXX_EXT
>  /usr/share/ruby/mkmf.rb:1830:        $CXXFLAGS += " " << cflags
>  /usr/share/ruby/mkmf.rb:1963:CXX = #{CONFIG['CXX']}
>  /usr/share/ruby/mkmf.rb:1984:CXXFLAGS = $(CCDLFLAGS) #$CXXFLAGS
>  $(ARCH_FLAG)
>  /usr/share/ruby/mkmf.rb:2429:    command = compile_command % COMPILE_CXX
>  /usr/share/ruby/mkmf.rb:2430:    asm_command =
>  compile_command.sub(/compiling/, 'translating') % ASSEMBLE_CXX
>  /usr/share/ruby/mkmf.rb:2431:    CXX_EXT.each do |e|
>  /usr/share/ruby/mkmf.rb:2458:    if
>  srcs.any?(&%r"\.(?:#{CXX_EXT.join('|')})\z".method(:===))
>  /usr/share/ruby/mkmf.rb:2510:    $CXXFLAGS = (with_config("cxxflags",
>  arg_config("CXXFLAGS", config["CXXFLAGS"]))||'').dup
>  /usr/share/ruby/mkmf.rb:2666:  COMPILE_CXX =
>  config_string('COMPILE_CXX') || '$(CXX) $(INCFLAGS) $(CPPFLAGS)
>  $(CXXFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<'
>  /usr/share/ruby/mkmf.rb:2676:  ASSEMBLE_CXX =
>  config_string('ASSEMBLE_CXX') || COMPILE_CXX.sub(/(?<=\s)-c(?=\s)/,
> '-S')
>
>  So it is quite reasonable that ruby compilation checks
>  CXX compiler. 
The ticket [1] I referenced even linked to commit which introduced some
of these lines [10]. However, this says nothing about usage of this
code. Neither the ruby-dev discussion provides more information [11].

So is there some extension library which uses results of these checks?
Frankly I'd be much happier if this was removed from upstream, since why
special case C++, but not Rust for example. I am looking for some
evidence ....


V.


[10]
https://bugs.ruby-lang.org/projects/ruby-trunk/repository/revisions/8371
[11] http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-dev/26089

[Mamoru TASAKA]

Sending to mailing list:

Vít Ondruch wrote on 07/26/2018 09:18 PM:
>  
>  
>  Dne 26.7.2018 v 13:40 Mamoru TASAKA napsal(a):
> > Vít Ondruch wrote on 07/26/2018 06:28 PM:
> >> Hi,
> >>
> >> Ruby currently fails to build due to removal of GCC from buildroot. The
> >> build failure is due to one test case [1] which fails now and which
> >> should be probably fixed to pass even without GCC.
> >>
> >> However, digging into this, I wonder if Ruby should have "BR: gcc" or
> >> "BR: gcc-c++". The thing is, that Ruby is checking presence of C++
> >> compiler during its configuration phase.
> >
> > ruby-libs contains:
> >
> >
> > /usr/lib64/ruby/rbconfig.rb:180:  CONFIG["CXXFLAGS"] = "-O2 -g -pipe
> > -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
> > -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
> > -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> > -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
> > -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection"
> > /usr/lib64/ruby/rbconfig.rb:181:  CONFIG["CXX"] = "g++"
> > /usr/share/ruby/mkmf.rb:74:  CXX_EXT = %w[cc mm cxx cpp]
> > /usr/share/ruby/mkmf.rb:76:    CXX_EXT.concat(%w[C])
> > /usr/share/ruby/mkmf.rb:82:  SRC_EXT = C_EXT + CXX_EXT
> > /usr/share/ruby/mkmf.rb:1830:        $CXXFLAGS += " " << cflags
> > /usr/share/ruby/mkmf.rb:1963:CXX = #{CONFIG['CXX']}
> > /usr/share/ruby/mkmf.rb:1984:CXXFLAGS = $(CCDLFLAGS) #$CXXFLAGS
> > $(ARCH_FLAG)
> > /usr/share/ruby/mkmf.rb:2429:    command = compile_command % COMPILE_CXX
> > /usr/share/ruby/mkmf.rb:2430:    asm_command =
> > compile_command.sub(/compiling/, 'translating') % ASSEMBLE_CXX
> > /usr/share/ruby/mkmf.rb:2431:    CXX_EXT.each do |e|
> > /usr/share/ruby/mkmf.rb:2458:    if
> > srcs.any?(&%r"\.(?:#{CXX_EXT.join('|')})\z".method(:===))
> > /usr/share/ruby/mkmf.rb:2510:    $CXXFLAGS = (with_config("cxxflags",
> > arg_config("CXXFLAGS", config["CXXFLAGS"]))||'').dup
> > /usr/share/ruby/mkmf.rb:2666:  COMPILE_CXX =
> > config_string('COMPILE_CXX') || '$(CXX) $(INCFLAGS) $(CPPFLAGS)
> > $(CXXFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<'
> > /usr/share/ruby/mkmf.rb:2676:  ASSEMBLE_CXX =
> > config_string('ASSEMBLE_CXX') || COMPILE_CXX.sub(/(?<=\s)-c(?=\s)/,
> '-S')
> >
> > So it is quite reasonable that ruby compilation checks
> > CXX compiler.
>  
>  The ticket [1] I referenced even linked to commit which introduced some
>  of these lines [10]. However, this says nothing about usage of this
>  code. Neither the ruby-dev discussion provides more information [11].
>  
>  So is there some extension library which uses results of these checks?
>  Frankly I'd be much happier if this was removed from upstream, since why
>  special case C++, but not Rust for example. I am looking for some
>  evidence ....
>   
https://koji.fedoraproject.org/koji/buildinfo?buildID=1124633
https://github.com/knu/ruby-unf_ext/blob/master/ext/unf_ext/extconf.rb#L31

Regards,
Mamoru

[Vít Ondruch]

Dne 26.7.2018 v 15:11 Mamoru TASAKA napsal(a):
>  Sending to mailing list:
>
>  Vít Ondruch wrote on 07/26/2018 09:18 PM:
> >
> >
> > Dne 26.7.2018 v 13:40 Mamoru TASAKA napsal(a):
> >> Vít Ondruch wrote on 07/26/2018 06:28 PM:
> >>> Hi,
> >>>
> >>> Ruby currently fails to build due to removal of GCC from buildroot.
> >>> The
> >>> build failure is due to one test case [1] which fails now and which
> >>> should be probably fixed to pass even without GCC.
> >>>
> >>> However, digging into this, I wonder if Ruby should have "BR: gcc"
> or
> >>> "BR: gcc-c++". The thing is, that Ruby is checking presence of C++
> >>> compiler during its configuration phase.
> >>
> >> ruby-libs contains:
> >>
> >>
> >> /usr/lib64/ruby/rbconfig.rb:180:  CONFIG["CXXFLAGS"] = "-O2 -g
> -pipe
> >> -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
> >> -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
> >> -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> >> -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
> >> -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection"
> >> /usr/lib64/ruby/rbconfig.rb:181:  CONFIG["CXX"] = "g++"
> >> /usr/share/ruby/mkmf.rb:74:  CXX_EXT = %w[cc mm cxx cpp]
> >> /usr/share/ruby/mkmf.rb:76:    CXX_EXT.concat(%w[C])
> >> /usr/share/ruby/mkmf.rb:82:  SRC_EXT = C_EXT + CXX_EXT
> >> /usr/share/ruby/mkmf.rb:1830:        $CXXFLAGS += " " << cflags
> >> /usr/share/ruby/mkmf.rb:1963:CXX = #{CONFIG['CXX']}
> >> /usr/share/ruby/mkmf.rb:1984:CXXFLAGS = $(CCDLFLAGS) #$CXXFLAGS
> >> $(ARCH_FLAG)
> >> /usr/share/ruby/mkmf.rb:2429:    command = compile_command %
> >> COMPILE_CXX
> >> /usr/share/ruby/mkmf.rb:2430:    asm_command =
> >> compile_command.sub(/compiling/, 'translating') % ASSEMBLE_CXX
> >> /usr/share/ruby/mkmf.rb:2431:    CXX_EXT.each do |e|
> >> /usr/share/ruby/mkmf.rb:2458:    if
> >> srcs.any?(&%r"\.(?:#{CXX_EXT.join('|')})\z".method(:===))
> >> /usr/share/ruby/mkmf.rb:2510:    $CXXFLAGS = (with_config("cxxflags",
> >> arg_config("CXXFLAGS", config["CXXFLAGS"]))||'').dup
> >> /usr/share/ruby/mkmf.rb:2666:  COMPILE_CXX =
> >> config_string('COMPILE_CXX') || '$(CXX) $(INCFLAGS) $(CPPFLAGS)
> >> $(CXXFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<'
> >> /usr/share/ruby/mkmf.rb:2676:  ASSEMBLE_CXX =
> >> config_string('ASSEMBLE_CXX') || COMPILE_CXX.sub(/(?<=\s)-c(?=\s)/,
> >> '-S')
> >>
> >> So it is quite reasonable that ruby compilation checks
> >> CXX compiler.
> >
> > The ticket [1] I referenced even linked to commit which introduced some
> > of these lines [10]. However, this says nothing about usage of this
> > code. Neither the ruby-dev discussion provides more information [11].
> >
> > So is there some extension library which uses results of these checks?
> > Frankly I'd be much happier if this was removed from upstream, since why
> > special case C++, but not Rust for example. I am looking for some
> > evidence ....
> >
>
>  https://koji.fedoraproject.org/koji/buildinfo?buildID=1124633
>  https://github.com/knu/ruby-unf_ext/blob/master/ext/unf_ext/extconf.rb#L31
>  
Ups, sorry, I forgot test it prior issuing the latest build (I was happy
enough that I workarounded the OpenSSL test failures) :(

However testing the build now, rubygem-unf_ext builds just fine. "gem
install unf_ext" works fine as well. Also, the RbConfig contains the CXX
variables referred in the extconf.rb:

~~~
$ rpm -q ruby-libs
ruby-libs-2.5.1-94.fc29.x86_64

$ ruby -r rbconfig -e "RbConfig::CONFIG.each {|c| p c}" | grep CXX
["configure_args", " '--build=x86_64-redhat-linux-gnu'
'--host=x86_64-redhat-linux-gnu' '--program-prefix='
'--disable-dependency-tracking' '--prefix=/usr'
'--exec-prefix=/usr'
'--bindir=/usr/bin' '--sbindir=/usr/sbin' '--sysconfdir=/etc'
'--datadir=/usr/share' '--includedir=/usr/include'
'--libdir=/usr/lib64'
'--libexecdir=/usr/libexec' '--localstatedir=/var'
'--sharedstatedir=/var/lib' '--mandir=/usr/share/man'
'--infodir=/usr/share/info' '--with-rubylibprefix=/usr/share/ruby'
'--with-archlibdir=/usr/lib64' '--with-rubyarchprefix=/usr/lib64/ruby'
'--with-sitedir=/usr/local/share/ruby/site_ruby'
'--with-sitearchdir=/usr/local/lib64/ruby/site_ruby'
'--with-vendordir=/usr/share/ruby/vendor_ruby'
'--with-vendorarchdir=/usr/lib64/ruby/vendor_ruby'
'--with-rubyhdrdir=/usr/include' '--with-rubyarchhdrdir=/usr/include'
'--with-sitearchhdrdir=$(sitehdrdir)/$(arch)'
'--with-vendorarchhdrdir=$(vendorhdrdir)/$(arch)'
'--with-rubygemsdir=/usr/share/rubygems' '--with-ruby-pc=ruby.pc'
'--with-compress-debug-sections=no' '--disable-rpath'
'--enable-shared'
'--with-ruby-version=' '--enable-multiarch'
'--with-prelude=./abrt_prelude.rb' 'build_alias=x86_64-redhat-linux-gnu'
'host_alias=x86_64-redhat-linux-gnu' 'CFLAGS=-O2 -g -pipe -Wall
-Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
-Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
-grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
-specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic
-fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection'
'LDFLAGS=-Wl,-z,relro   -Wl,-z,now
-specs=/usr/lib/rpm/redhat/redhat-hardened-ld' 'CXXFLAGS=-O2 -g -pipe
-Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
-Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
-grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
-specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic
-fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection'"]
["CXXFLAGS", "-O2 -g -pipe -Wall -Werror=format-security
-Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions
-fstack-protector-strong -grecord-gcc-switches
-specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
-specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
-fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection"]
["CXX", "g++"]
["CFLAGS", "-O2 -g -pipe -Wall -Werror=format-security
-Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions
-fstack-protector-strong -grecord-gcc-switches
-specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
-specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
-fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection
-fPIC"]
~~~

Actually, this makes me even less worried, because for RPM build, we can
use %{optflags} and we are relatively safe here, but I was more worried
about pure "gem install" scenario.


Vít

[Vít Ondruch]

Dne 27.7.2018 v 12:11 Vít Ondruch napsal(a):
>  Dne 26.7.2018 v 15:11 Mamoru TASAKA napsal(a):
> > Sending to mailing list:
> >
> > Vít Ondruch wrote on 07/26/2018 09:18 PM:
> >>
> >> Dne 26.7.2018 v 13:40 Mamoru TASAKA napsal(a):
> >>> Vít Ondruch wrote on 07/26/2018 06:28 PM:
> >>>> Hi,
> >>>>
> >>>> Ruby currently fails to build due to removal of GCC from buildroot.
> >>>> The
> >>>> build failure is due to one test case [1] which fails now and which
> >>>> should be probably fixed to pass even without GCC.
> >>>>
> >>>> However, digging into this, I wonder if Ruby should have "BR:
> gcc" or
> >>>> "BR: gcc-c++". The thing is, that Ruby is checking presence of
> C++
> >>>> compiler during its configuration phase.
> >>> ruby-libs contains:
> >>>
> >>>
> >>> /usr/lib64/ruby/rbconfig.rb:180:  CONFIG["CXXFLAGS"] = "-O2 -g
> -pipe
> >>> -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
> >>> -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
> >>> -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> >>> -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
> >>> -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection"
> >>> /usr/lib64/ruby/rbconfig.rb:181:  CONFIG["CXX"] = "g++"
> >>> /usr/share/ruby/mkmf.rb:74:  CXX_EXT = %w[cc mm cxx cpp]
> >>> /usr/share/ruby/mkmf.rb:76:    CXX_EXT.concat(%w[C])
> >>> /usr/share/ruby/mkmf.rb:82:  SRC_EXT = C_EXT + CXX_EXT
> >>> /usr/share/ruby/mkmf.rb:1830:        $CXXFLAGS += " " <<
> cflags
> >>> /usr/share/ruby/mkmf.rb:1963:CXX = #{CONFIG['CXX']}
> >>> /usr/share/ruby/mkmf.rb:1984:CXXFLAGS = $(CCDLFLAGS) #$CXXFLAGS
> >>> $(ARCH_FLAG)
> >>> /usr/share/ruby/mkmf.rb:2429:    command = compile_command %
> >>> COMPILE_CXX
> >>> /usr/share/ruby/mkmf.rb:2430:    asm_command =
> >>> compile_command.sub(/compiling/, 'translating') % ASSEMBLE_CXX
> >>> /usr/share/ruby/mkmf.rb:2431:    CXX_EXT.each do |e|
> >>> /usr/share/ruby/mkmf.rb:2458:    if
> >>>
> srcs.any?(&%r"\.(?:#{CXX_EXT.join('|')})\z".method(:===))
> >>> /usr/share/ruby/mkmf.rb:2510:    $CXXFLAGS =
> (with_config("cxxflags",
> >>> arg_config("CXXFLAGS",
> config["CXXFLAGS"]))||'').dup
> >>> /usr/share/ruby/mkmf.rb:2666:  COMPILE_CXX =
> >>> config_string('COMPILE_CXX') || '$(CXX) $(INCFLAGS) $(CPPFLAGS)
> >>> $(CXXFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<'
> >>> /usr/share/ruby/mkmf.rb:2676:  ASSEMBLE_CXX =
> >>> config_string('ASSEMBLE_CXX') ||
> COMPILE_CXX.sub(/(?<=\s)-c(?=\s)/,
> >>> '-S')
> >>>
> >>> So it is quite reasonable that ruby compilation checks
> >>> CXX compiler.
> >> The ticket [1] I referenced even linked to commit which introduced some
> >> of these lines [10]. However, this says nothing about usage of this
> >> code. Neither the ruby-dev discussion provides more information [11].
> >>
> >> So is there some extension library which uses results of these checks?
> >> Frankly I'd be much happier if this was removed from upstream, since why
> >> special case C++, but not Rust for example. I am looking for some
> >> evidence ....
> >>
> > https://koji.fedoraproject.org/koji/buildinfo?buildID=1124633
> > https://github.com/knu/ruby-unf_ext/blob/master/ext/unf_ext/extconf.rb#L31
> >
>  Ups, sorry, I forgot test it prior issuing the latest build (I was happy
>  enough that I workarounded the OpenSSL test failures) :(
>
>  However testing the build now, rubygem-unf_ext builds just fine. "gem
>  install unf_ext" works fine as well. Also, the RbConfig contains the CXX
>  variables referred in the extconf.rb:
>
>  ~~~
>  $ rpm -q ruby-libs
>  ruby-libs-2.5.1-94.fc29.x86_64
>
>  $ ruby -r rbconfig -e "RbConfig::CONFIG.each {|c| p c}" | grep CXX
>  ["configure_args", " '--build=x86_64-redhat-linux-gnu'
>  '--host=x86_64-redhat-linux-gnu' '--program-prefix='
>  '--disable-dependency-tracking' '--prefix=/usr'
> '--exec-prefix=/usr'
>  '--bindir=/usr/bin' '--sbindir=/usr/sbin' '--sysconfdir=/etc'
>  '--datadir=/usr/share' '--includedir=/usr/include'
> '--libdir=/usr/lib64'
>  '--libexecdir=/usr/libexec' '--localstatedir=/var'
>  '--sharedstatedir=/var/lib' '--mandir=/usr/share/man'
>  '--infodir=/usr/share/info' '--with-rubylibprefix=/usr/share/ruby'
>  '--with-archlibdir=/usr/lib64' '--with-rubyarchprefix=/usr/lib64/ruby'
>  '--with-sitedir=/usr/local/share/ruby/site_ruby'
>  '--with-sitearchdir=/usr/local/lib64/ruby/site_ruby'
>  '--with-vendordir=/usr/share/ruby/vendor_ruby'
>  '--with-vendorarchdir=/usr/lib64/ruby/vendor_ruby'
>  '--with-rubyhdrdir=/usr/include' '--with-rubyarchhdrdir=/usr/include'
>  '--with-sitearchhdrdir=$(sitehdrdir)/$(arch)'
>  '--with-vendorarchhdrdir=$(vendorhdrdir)/$(arch)'
>  '--with-rubygemsdir=/usr/share/rubygems' '--with-ruby-pc=ruby.pc'
>  '--with-compress-debug-sections=no' '--disable-rpath'
> '--enable-shared'
>  '--with-ruby-version=' '--enable-multiarch'
>  '--with-prelude=./abrt_prelude.rb' 'build_alias=x86_64-redhat-linux-gnu'
>  'host_alias=x86_64-redhat-linux-gnu' 'CFLAGS=-O2 -g -pipe -Wall
>  -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
>  -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
>  -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
>  -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic
>  -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection'
>  'LDFLAGS=-Wl,-z,relro   -Wl,-z,now
>  -specs=/usr/lib/rpm/redhat/redhat-hardened-ld' 'CXXFLAGS=-O2 -g -pipe
>  -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
>  -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
>  -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
>  -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic
>  -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection'"]
>  ["CXXFLAGS", "-O2 -g -pipe -Wall -Werror=format-security
>  -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions
>  -fstack-protector-strong -grecord-gcc-switches
>  -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
>  -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
>  -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection"]
>  ["CXX", "g++"]
>  ["CFLAGS", "-O2 -g -pipe -Wall -Werror=format-security
>  -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions
>  -fstack-protector-strong -grecord-gcc-switches
>  -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
>  -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
>  -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection
>  -fPIC"]
>  ~~~
>
>  Actually, this makes me even less worried, because for RPM build, we can
>  use %{optflags} and we are relatively safe here, but I was more worried
>  about pure "gem install" scenario.
>
>
>  Vít
>  
This brings me to another question, shouldn't we add "cxxflags" here?

https://src.fedoraproject.org/rpms/ruby/blob/7c494340f9b797e5a8ce2f8dd8fd...


V.

[Mamoru TASAKA]

Vít Ondruch wrote on 07/27/2018 07:23 PM:
>  
>  
>  Dne 27.7.2018 v 12:11 Vít Ondruch napsal(a):
> >
> > Dne 26.7.2018 v 15:11 Mamoru TASAKA napsal(a):
> >> Sending to mailing list:
> >>
> >> Vít Ondruch wrote on 07/26/2018 09:18 PM:
> >>>
> >>> Dne 26.7.2018 v 13:40 Mamoru TASAKA napsal(a):
> >>>> Vít Ondruch wrote on 07/26/2018 06:28 PM:
> >>>>> Hi,
> >>>>>
> >>>>> Ruby currently fails to build due to removal of GCC from buildroot.
> >>>>> The
> >>>>> build failure is due to one test case [1] which fails now and which
> >>>>> should be probably fixed to pass even without GCC.
> >>>>>
> >>>>> However, digging into this, I wonder if Ruby should have "BR:
> gcc" or
> >>>>> "BR: gcc-c++". The thing is, that Ruby is checking presence
> of C++
> >>>>> compiler during its configuration phase.
> >>>> ruby-libs contains:
> >>>>
> >>>>
> >>>> /usr/lib64/ruby/rbconfig.rb:180:  CONFIG["CXXFLAGS"] =
> "-O2 -g -pipe
> >>>> -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
> >>>> -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
> >>>> -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> >>>> -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
> >>>> -fasynchronous-unwind-tables -fstack-clash-protection
> -fcf-protection"
> >>>> /usr/lib64/ruby/rbconfig.rb:181:  CONFIG["CXX"] =
> "g++"
> >>>> /usr/share/ruby/mkmf.rb:74:  CXX_EXT = %w[cc mm cxx cpp]
> >>>> /usr/share/ruby/mkmf.rb:76:    CXX_EXT.concat(%w[C])
> >>>> /usr/share/ruby/mkmf.rb:82:  SRC_EXT = C_EXT + CXX_EXT
> >>>> /usr/share/ruby/mkmf.rb:1830:        $CXXFLAGS += " " <<
> cflags
> >>>> /usr/share/ruby/mkmf.rb:1963:CXX = #{CONFIG['CXX']}
> >>>> /usr/share/ruby/mkmf.rb:1984:CXXFLAGS = $(CCDLFLAGS) #$CXXFLAGS
> >>>> $(ARCH_FLAG)
> >>>> /usr/share/ruby/mkmf.rb:2429:    command = compile_command %
> >>>> COMPILE_CXX
> >>>> /usr/share/ruby/mkmf.rb:2430:    asm_command =
> >>>> compile_command.sub(/compiling/, 'translating') % ASSEMBLE_CXX
> >>>> /usr/share/ruby/mkmf.rb:2431:    CXX_EXT.each do |e|
> >>>> /usr/share/ruby/mkmf.rb:2458:    if
> >>>>
> srcs.any?(&%r"\.(?:#{CXX_EXT.join('|')})\z".method(:===))
> >>>> /usr/share/ruby/mkmf.rb:2510:    $CXXFLAGS =
> (with_config("cxxflags",
> >>>> arg_config("CXXFLAGS",
> config["CXXFLAGS"]))||'').dup
> >>>> /usr/share/ruby/mkmf.rb:2666:  COMPILE_CXX =
> >>>> config_string('COMPILE_CXX') || '$(CXX) $(INCFLAGS)
> $(CPPFLAGS)
> >>>> $(CXXFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<'
> >>>> /usr/share/ruby/mkmf.rb:2676:  ASSEMBLE_CXX =
> >>>> config_string('ASSEMBLE_CXX') ||
> COMPILE_CXX.sub(/(?<=\s)-c(?=\s)/,
> >>>> '-S')
> >>>>
> >>>> So it is quite reasonable that ruby compilation checks
> >>>> CXX compiler.
> >>> The ticket [1] I referenced even linked to commit which introduced some
> >>> of these lines [10]. However, this says nothing about usage of this
> >>> code. Neither the ruby-dev discussion provides more information [11].
> >>>
> >>> So is there some extension library which uses results of these checks?
> >>> Frankly I'd be much happier if this was removed from upstream, since why
> >>> special case C++, but not Rust for example. I am looking for some
> >>> evidence ....
> >>>
> >> https://koji.fedoraproject.org/koji/buildinfo?buildID=1124633
> >> https://github.com/knu/ruby-unf_ext/blob/master/ext/unf_ext/extconf.rb#L31
> >>
> > Ups, sorry, I forgot test it prior issuing the latest build (I was happy
> > enough that I workarounded the OpenSSL test failures) :(
> >
> > However testing the build now, rubygem-unf_ext builds just fine. "gem
> > install unf_ext" works fine as well. Also, the RbConfig contains the CXX
> > variables referred in the extconf.rb:
> >
> > ~~~
> > $ rpm -q ruby-libs
> > ruby-libs-2.5.1-94.fc29.x86_64
> >
> > $ ruby -r rbconfig -e "RbConfig::CONFIG.each {|c| p c}" | grep CXX
> > ["configure_args", " '--build=x86_64-redhat-linux-gnu'
> > '--host=x86_64-redhat-linux-gnu' '--program-prefix='
> > '--disable-dependency-tracking' '--prefix=/usr'
> '--exec-prefix=/usr'
> > '--bindir=/usr/bin' '--sbindir=/usr/sbin'
> '--sysconfdir=/etc'
> > '--datadir=/usr/share' '--includedir=/usr/include'
> '--libdir=/usr/lib64'
> > '--libexecdir=/usr/libexec' '--localstatedir=/var'
> > '--sharedstatedir=/var/lib' '--mandir=/usr/share/man'
> > '--infodir=/usr/share/info' '--with-rubylibprefix=/usr/share/ruby'
> > '--with-archlibdir=/usr/lib64'
> '--with-rubyarchprefix=/usr/lib64/ruby'
> > '--with-sitedir=/usr/local/share/ruby/site_ruby'
> > '--with-sitearchdir=/usr/local/lib64/ruby/site_ruby'
> > '--with-vendordir=/usr/share/ruby/vendor_ruby'
> > '--with-vendorarchdir=/usr/lib64/ruby/vendor_ruby'
> > '--with-rubyhdrdir=/usr/include'
> '--with-rubyarchhdrdir=/usr/include'
> > '--with-sitearchhdrdir=$(sitehdrdir)/$(arch)'
> > '--with-vendorarchhdrdir=$(vendorhdrdir)/$(arch)'
> > '--with-rubygemsdir=/usr/share/rubygems' '--with-ruby-pc=ruby.pc'
> > '--with-compress-debug-sections=no' '--disable-rpath'
> '--enable-shared'
> > '--with-ruby-version=' '--enable-multiarch'
> > '--with-prelude=./abrt_prelude.rb'
> 'build_alias=x86_64-redhat-linux-gnu'
> > 'host_alias=x86_64-redhat-linux-gnu' 'CFLAGS=-O2 -g -pipe -Wall
> > -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
> > -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
> > -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> > -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic
> > -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection'
> > 'LDFLAGS=-Wl,-z,relro   -Wl,-z,now
> > -specs=/usr/lib/rpm/redhat/redhat-hardened-ld' 'CXXFLAGS=-O2 -g -pipe
> > -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
> > -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
> > -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> > -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic
> > -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection'"]
> > ["CXXFLAGS", "-O2 -g -pipe -Wall -Werror=format-security
> > -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions
> > -fstack-protector-strong -grecord-gcc-switches
> > -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> > -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
> > -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection"]
> > ["CXX", "g++"]
> > ["CFLAGS", "-O2 -g -pipe -Wall -Werror=format-security
> > -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions
> > -fstack-protector-strong -grecord-gcc-switches
> > -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> > -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
> > -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection
> > -fPIC"]
> > ~~~
> >
> > Actually, this makes me even less worried, because for RPM build, we can
> > use %{optflags} and we are relatively safe here, but I was more worried
> > about pure "gem install" scenario.
> >
> >
> > Vít
> >
>  
>  This brings me to another question, shouldn't we add "cxxflags" here?
>  
>
> https://src.fedoraproject.org/rpms/ruby/blob/7c494340f9b797e5a8ce2f8dd8fd...
I guess it is better.

Regards,
Mamoru

[Vít Ondruch]

Dne 30.7.2018 v 17:40 Mamoru TASAKA napsal(a):
>  Vít Ondruch wrote on 07/27/2018 07:23 PM:
> >
> >
> > Dne 27.7.2018 v 12:11 Vít Ondruch napsal(a):
> >>
> >> Dne 26.7.2018 v 15:11 Mamoru TASAKA napsal(a):
> >>> Sending to mailing list:
> >>>
> >>> Vít Ondruch wrote on 07/26/2018 09:18 PM:
> >>>>
> >>>> Dne 26.7.2018 v 13:40 Mamoru TASAKA napsal(a):
> >>>>> Vít Ondruch wrote on 07/26/2018 06:28 PM:
> >>>>>> Hi,
> >>>>>>
> >>>>>> Ruby currently fails to build due to removal of GCC from
> buildroot.
> >>>>>> The
> >>>>>> build failure is due to one test case [1] which fails now and
> which
> >>>>>> should be probably fixed to pass even without GCC.
> >>>>>>
> >>>>>> However, digging into this, I wonder if Ruby should have
> "BR:
> >>>>>> gcc" or
> >>>>>> "BR: gcc-c++". The thing is, that Ruby is checking
> presence of C++
> >>>>>> compiler during its configuration phase.
> >>>>> ruby-libs contains:
> >>>>>
> >>>>>
> >>>>> /usr/lib64/ruby/rbconfig.rb:180:  CONFIG["CXXFLAGS"] =
> "-O2 -g -pipe
> >>>>> -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
> >>>>> -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
> >>>>> -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> >>>>> -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
> >>>>> -fasynchronous-unwind-tables -fstack-clash-protection
> >>>>> -fcf-protection"
> >>>>> /usr/lib64/ruby/rbconfig.rb:181:  CONFIG["CXX"] =
> "g++"
> >>>>> /usr/share/ruby/mkmf.rb:74:  CXX_EXT = %w[cc mm cxx cpp]
> >>>>> /usr/share/ruby/mkmf.rb:76:    CXX_EXT.concat(%w[C])
> >>>>> /usr/share/ruby/mkmf.rb:82:  SRC_EXT = C_EXT + CXX_EXT
> >>>>> /usr/share/ruby/mkmf.rb:1830:        $CXXFLAGS += " "
> << cflags
> >>>>> /usr/share/ruby/mkmf.rb:1963:CXX = #{CONFIG['CXX']}
> >>>>> /usr/share/ruby/mkmf.rb:1984:CXXFLAGS = $(CCDLFLAGS) #$CXXFLAGS
> >>>>> $(ARCH_FLAG)
> >>>>> /usr/share/ruby/mkmf.rb:2429:    command = compile_command %
> >>>>> COMPILE_CXX
> >>>>> /usr/share/ruby/mkmf.rb:2430:    asm_command =
> >>>>> compile_command.sub(/compiling/, 'translating') %
> ASSEMBLE_CXX
> >>>>> /usr/share/ruby/mkmf.rb:2431:    CXX_EXT.each do |e|
> >>>>> /usr/share/ruby/mkmf.rb:2458:    if
> >>>>>
> srcs.any?(&%r"\.(?:#{CXX_EXT.join('|')})\z".method(:===))
> >>>>> /usr/share/ruby/mkmf.rb:2510:    $CXXFLAGS =
> >>>>> (with_config("cxxflags",
> >>>>> arg_config("CXXFLAGS",
> config["CXXFLAGS"]))||'').dup
> >>>>> /usr/share/ruby/mkmf.rb:2666:  COMPILE_CXX =
> >>>>> config_string('COMPILE_CXX') || '$(CXX) $(INCFLAGS)
> $(CPPFLAGS)
> >>>>> $(CXXFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<'
> >>>>> /usr/share/ruby/mkmf.rb:2676:  ASSEMBLE_CXX =
> >>>>> config_string('ASSEMBLE_CXX') ||
> COMPILE_CXX.sub(/(?<=\s)-c(?=\s)/,
> >>>>> '-S')
> >>>>>
> >>>>> So it is quite reasonable that ruby compilation checks
> >>>>> CXX compiler.
> >>>> The ticket [1] I referenced even linked to commit which introduced
> >>>> some
> >>>> of these lines [10]. However, this says nothing about usage of this
> >>>> code. Neither the ruby-dev discussion provides more information [11].
> >>>>
> >>>> So is there some extension library which uses results of these
> >>>> checks?
> >>>> Frankly I'd be much happier if this was removed from upstream,
> >>>> since why
> >>>> special case C++, but not Rust for example. I am looking for some
> >>>> evidence ....
> >>>>
> >>> https://koji.fedoraproject.org/koji/buildinfo?buildID=1124633
> >>> https://github.com/knu/ruby-unf_ext/blob/master/ext/unf_ext/extconf.rb#L31
> >>>
> >>>
> >> Ups, sorry, I forgot test it prior issuing the latest build (I was
> >> happy
> >> enough that I workarounded the OpenSSL test failures) :(
> >>
> >> However testing the build now, rubygem-unf_ext builds just fine. "gem
> >> install unf_ext" works fine as well. Also, the RbConfig contains the
> >> CXX
> >> variables referred in the extconf.rb:
> >>
> >> ~~~
> >> $ rpm -q ruby-libs
> >> ruby-libs-2.5.1-94.fc29.x86_64
> >>
> >> $ ruby -r rbconfig -e "RbConfig::CONFIG.each {|c| p c}" | grep CXX
> >> ["configure_args", " '--build=x86_64-redhat-linux-gnu'
> >> '--host=x86_64-redhat-linux-gnu' '--program-prefix='
> >> '--disable-dependency-tracking' '--prefix=/usr'
> '--exec-prefix=/usr'
> >> '--bindir=/usr/bin' '--sbindir=/usr/sbin'
> '--sysconfdir=/etc'
> >> '--datadir=/usr/share' '--includedir=/usr/include'
> >> '--libdir=/usr/lib64'
> >> '--libexecdir=/usr/libexec' '--localstatedir=/var'
> >> '--sharedstatedir=/var/lib' '--mandir=/usr/share/man'
> >> '--infodir=/usr/share/info'
> '--with-rubylibprefix=/usr/share/ruby'
> >> '--with-archlibdir=/usr/lib64'
> '--with-rubyarchprefix=/usr/lib64/ruby'
> >> '--with-sitedir=/usr/local/share/ruby/site_ruby'
> >> '--with-sitearchdir=/usr/local/lib64/ruby/site_ruby'
> >> '--with-vendordir=/usr/share/ruby/vendor_ruby'
> >> '--with-vendorarchdir=/usr/lib64/ruby/vendor_ruby'
> >> '--with-rubyhdrdir=/usr/include'
> '--with-rubyarchhdrdir=/usr/include'
> >> '--with-sitearchhdrdir=$(sitehdrdir)/$(arch)'
> >> '--with-vendorarchhdrdir=$(vendorhdrdir)/$(arch)'
> >> '--with-rubygemsdir=/usr/share/rubygems'
> '--with-ruby-pc=ruby.pc'
> >> '--with-compress-debug-sections=no' '--disable-rpath'
> '--enable-shared'
> >> '--with-ruby-version=' '--enable-multiarch'
> >> '--with-prelude=./abrt_prelude.rb'
> >> 'build_alias=x86_64-redhat-linux-gnu'
> >> 'host_alias=x86_64-redhat-linux-gnu' 'CFLAGS=-O2 -g -pipe -Wall
> >> -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
> >> -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
> >> -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> >> -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic
> >> -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection'
> >> 'LDFLAGS=-Wl,-z,relro   -Wl,-z,now
> >> -specs=/usr/lib/rpm/redhat/redhat-hardened-ld' 'CXXFLAGS=-O2 -g -pipe
> >> -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2
> >> -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong
> >> -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> >> -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic
> >> -fasynchronous-unwind-tables -fstack-clash-protection
> >> -fcf-protection'"]
> >> ["CXXFLAGS", "-O2 -g -pipe -Wall -Werror=format-security
> >> -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions
> >> -fstack-protector-strong -grecord-gcc-switches
> >> -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> >> -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
> >> -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection"]
> >> ["CXX", "g++"]
> >> ["CFLAGS", "-O2 -g -pipe -Wall -Werror=format-security
> >> -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions
> >> -fstack-protector-strong -grecord-gcc-switches
> >> -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
> >> -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -mtune=generic
> >> -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection
> >> -fPIC"]
> >> ~~~
> >>
> >> Actually, this makes me even less worried, because for RPM build, we
> >> can
> >> use %{optflags} and we are relatively safe here, but I was more worried
> >> about pure "gem install" scenario.
> >>
> >>
> >> Vít
> >>
> >
> > This brings me to another question, shouldn't we add "cxxflags" here?
> >
> >
> https://src.fedoraproject.org/rpms/ruby/blob/7c494340f9b797e5a8ce2f8dd8fd...
> >
>
>  I guess it is better. 
This is now committed:

https://src.fedoraproject.org/rpms/ruby/c/d5e84a533c0c6d5b92ea21ee5ba09b4...


V.

>  Regards,
>  Mamoru
>
>
>
>  _______________________________________________
>  ruby-sig mailing list -- ruby-sig(a)lists.fedoraproject.org
>  To unsubscribe send an email to ruby-sig-leave(a)lists.fedoraproject.org
>  Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
>  List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
>  List Archives:
>
> https://lists.fedoraproject.org/archives/list/[e-mail]...