[Buildroot] [PATCH] ruby: bump version to 2.4.0

[Buildroot] [PATCH] ruby: bump version to 2.4.0

[Vicente Olivert Riera]

The problem addressed by 0001 patch has been fixed upstream and is that
fix is included in this release:
  https://github.com/ruby/ruby/commit/aa107497cd379b713eba8cecdb9a882bb1e0dd89

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
---
 ...-process.c-fix-rb_spawn_process-for-nommu.patch | 34 ----------------------
 package/ruby/ruby.hash                             |  4 +--
 package/ruby/ruby.mk                               |  6 ++--
 3 files changed, 5 insertions(+), 39 deletions(-)
 delete mode 100644 package/ruby/0001-process.c-fix-rb_spawn_process-for-nommu.patch

diff --git a/package/ruby/0001-process.c-fix-rb_spawn_process-for-nommu.patch b/package/ruby/0001-process.c-fix-rb_spawn_process-for-nommu.patch
deleted file mode 100644
index 701c105..0000000
--- a/package/ruby/0001-process.c-fix-rb_spawn_process-for-nommu.patch
+++ /dev/null
@@ -1,34 +0,0 @@
-From 24e6d5bcf791a5c3f46191e544731420ff8b1312 Mon Sep 17 00:00:00 2001
-From: Gustavo Zacarias <gustavo@zacarias.com.ar>
-Date: Thu, 19 May 2016 11:10:02 -0300
-Subject: [PATCH] process.c: fix rb_spawn_process() for nommu
-
-rb_spawn_process() in process.c tries different solutions for when fork
-and/or spawnv are/aren't available.
-The last resort when both aren't is to use the system() call which
-stores the value in the status variable, which isn't declared.
-
-Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
----
-Patch status: reported https://bugs.ruby-lang.org/issues/12398
-
- process.c | 3 +++
- 1 file changed, 3 insertions(+)
-
-diff --git a/process.c b/process.c
-index e196639..40967f7 100644
---- a/process.c
-+++ b/process.c
-@@ -3897,6 +3897,9 @@ rb_spawn_process(struct rb_execarg *eargp, char *errmsg, size_t errmsg_buflen)
-     VALUE prog;
-     struct rb_execarg sarg;
- #endif
-+#if !defined HAVE_WORKING_FORK || !USE_SPAWNV
-+    int status;
-+#endif
- 
- #if defined HAVE_WORKING_FORK && !USE_SPAWNV
-     pid = rb_fork_async_signal_safe(NULL, rb_exec_atfork, eargp, eargp->redirect_fds, errmsg, errmsg_buflen);
--- 
-2.7.3
-
diff --git a/package/ruby/ruby.hash b/package/ruby/ruby.hash
index 8182e2c..19722a3 100644
--- a/package/ruby/ruby.hash
+++ b/package/ruby/ruby.hash
@@ -1,2 +1,2 @@
-# From https://www.ruby-lang.org/en/news/2016/11/21/ruby-2-3-3-released/
-sha256 1a4fa8c2885734ba37b97ffdb4a19b8fba0e8982606db02d936e65bac07419dc  ruby-2.3.3.tar.xz
+# From https://www.ruby-lang.org/en/news/2016/12/25/ruby-2-4-0-released/
+sha256 3a87fef45cba48b9322236be60c455c13fd4220184ce7287600361319bb63690  ruby-2.4.0.tar.xz
diff --git a/package/ruby/ruby.mk b/package/ruby/ruby.mk
index b8e03ac..4c6459b 100644
--- a/package/ruby/ruby.mk
+++ b/package/ruby/ruby.mk
@@ -4,9 +4,9 @@
 #
 ################################################################################
 
-RUBY_VERSION_MAJOR = 2.3
-RUBY_VERSION = $(RUBY_VERSION_MAJOR).3
-RUBY_VERSION_EXT = 2.3.0
+RUBY_VERSION_MAJOR = 2.4
+RUBY_VERSION = $(RUBY_VERSION_MAJOR).0
+RUBY_VERSION_EXT = 2.4.0
 RUBY_SITE = http://cache.ruby-lang.org/pub/ruby/$(RUBY_VERSION_MAJOR)
 RUBY_SOURCE = ruby-$(RUBY_VERSION).tar.xz
 RUBY_DEPENDENCIES = host-pkgconf host-ruby
-- 
2.10.2

[Buildroot] [PATCH] ruby: bump version to 2.4.0

[Peter Korsgaard]

>>>>> "Vicente" == Vicente Olivert Riera <Vincent.Riera@imgtec.com> writes:

 > The problem addressed by 0001 patch has been fixed upstream and is that
 > fix is included in this release:
 >   https://github.com/ruby/ruby/commit/aa107497cd379b713eba8cecdb9a882bb1e0dd89

 > Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>

Committed, thanks.

-- 
Bye, Peter Korsgaard

[Buildroot] [PATCH] ruby: bump version to 2.4.0

[Peter Korsgaard]

>>>>> "Peter" == Peter Korsgaard <peter@korsgaard.com> writes:

>>>>> "Vicente" == Vicente Olivert Riera <Vincent.Riera@imgtec.com> writes:
 >> The problem addressed by 0001 patch has been fixed upstream and is that
 >> fix is included in this release:
 >> https://github.com/ruby/ruby/commit/aa107497cd379b713eba8cecdb9a882bb1e0dd89

 >> Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>

 > Committed, thanks.

This seems to cause a number of build issues. Care to take a look?

http://autobuild.buildroot.net/?reason=ruby-2.4.0

-- 
Bye, Peter Korsgaard

[Buildroot] [PATCH] ruby: bump version to 2.4.0

[Bernd Kuhls]

Am Sun, 22 Jan 2017 22:07:21 +0100 schrieb Peter Korsgaard:

> This seems to cause a number of build issues. Care to take a look?
> 
> http://autobuild.buildroot.net/?reason=ruby-2.4.0

Hi,

apparantly configure has problems detecting math functions of uClibc:

http://autobuild.buildroot.net/
results/2aa/2aa668247b6925ff1290182a67092156b5bcc1b2//build-end.log

checking for finite... no
[...]
checking for isinf... no
checking for isnan... no

http://autobuild.buildroot.net/
results/2aa/2aa668247b6925ff1290182a67092156b5bcc1b2//ruby-2.4.0/
config.log

configure:20069: checking for finite
configure:20069: /home/test/autobuild/run/instance-1/output/host/usr/bin/
m68k-linux-gcc -o conftest -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -
D_FILE_OFFSET_BITS=64  -Os  -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -
D_FILE_OFFSET_BITS=64  conftest.c -lgmp -ldl -lcrypt -lm  >&5
conftest.c:238:6: warning: conflicting types for built-in function 
'finite'
 char finite ();
      ^
/tmp/ccip45ij.o: In function `main':
conftest.c:(.text.startup+0x6): undefined reference to `finite'
collect2: error: ld returned 1 exit status

although uClibc supports these functions.

Hacking ruby.mk like this

RUBY_CONF_ENV += \
       ac_cv_func_finite=yes \
       ac_cv_func_isinf=yes \
       ac_cv_func_isnan=yes
endif

fixes the build:

checking for finite... (cached) yes
[...]
checking for isinf... (cached) yes
checking for isnan... (cached) yes

Shall I send a patch with this hack?

Regards, Bernd

[Buildroot] [PATCH] ruby: bump version to 2.4.0

[Waldemar Brodkorb]

Hi,
Bernd Kuhls wrote,

> Am Sun, 22 Jan 2017 22:07:21 +0100 schrieb Peter Korsgaard:
> 
> > This seems to cause a number of build issues. Care to take a look?
> > 
> > http://autobuild.buildroot.net/?reason=ruby-2.4.0
> 
> Hi,
> 
> apparantly configure has problems detecting math functions of uClibc:

Because configure checks for functions, but isfinite, isinf and isnan
are implemented as macros. For Dragonfly they explicitely set
isinf/isnan as available:
https://github.com/ruby/ruby/blob/trunk/configure.in#L1278

Shouldn't this be reported upstream?
I think this is an autotool detection bug.
 
> m68k-linux-gcc -o conftest -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -
> D_FILE_OFFSET_BITS=64  -Os  -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -
> D_FILE_OFFSET_BITS=64  conftest.c -lgmp -ldl -lcrypt -lm  >&5
> conftest.c:238:6: warning: conflicting types for built-in function 
> 'finite'
>  char finite ();
>       ^
> /tmp/ccip45ij.o: In function `main':
> conftest.c:(.text.startup+0x6): undefined reference to `finite'
> collect2: error: ld returned 1 exit status
> 
> although uClibc supports these functions.

finite seems obsoleted and isfinite should be used. 

It appears musl implements finite as function for backward compatibility.
Therefore the check does not fail for musl toolchains.
https://git.musl-libc.org/cgit/musl/tree/src/math/finite.c

For glibc the check fails and the replacement functions for
finite, isinf and isnan are used.

For uclibc-ng the check fails, but the replacement functions cannot
be used, because finite is already defined as macro for backward
compatibility. (added a while ago for inetd..)

> Hacking ruby.mk like this
> 
> RUBY_CONF_ENV += \
>        ac_cv_func_finite=yes \
>        ac_cv_func_isinf=yes \
>        ac_cv_func_isnan=yes
> endif

I think this is the best thing todo inside buildroot.

I think ruby's configure script should detect isfinite (function or macro) and
only fallback trying to detect finite otherwise.
 
best regards
 Waldemar

[Buildroot] [PATCH] ruby: bump version to 2.4.0

[Peter Korsgaard]

>>>>> "Waldemar" == Waldemar Brodkorb <wbx@openadk.org> writes:

Hi,

 >> apparantly configure has problems detecting math functions of uClibc:

 > Because configure checks for functions, but isfinite, isinf and isnan
 > are implemented as macros. For Dragonfly they explicitely set
 > isinf/isnan as available:
 > https://github.com/ruby/ruby/blob/trunk/configure.in#L1278

 > Shouldn't this be reported upstream?
 > I think this is an autotool detection bug.

 >> although uClibc supports these functions.

 > finite seems obsoleted and isfinite should be used. 

 > It appears musl implements finite as function for backward compatibility.
 > Therefore the check does not fail for musl toolchains.
 > https://git.musl-libc.org/cgit/musl/tree/src/math/finite.c

 > For glibc the check fails and the replacement functions for
 > finite, isinf and isnan are used.

 > For uclibc-ng the check fails, but the replacement functions cannot
 > be used, because finite is already defined as macro for backward
 > compatibility. (added a while ago for inetd..)

Indeed, finite is a function on musl. On my glibc version they are also
(weakly) defined (and configure correctly finds them):

nm -D /lib/x86_64-linux-gnu/libc.so.6|grep -wE 'finite|isinf|isnan'
0000000000032310 W finite
00000000000322a0 W isinf
00000000000322e0 W isnan

I'm not sure what the spec (if any exists) for these legacy BSD
functions say about if the unprefixed symbols really need to be
available, but it is certainly a difference between uClibc-ng and
glibc/musl.

 >> Hacking ruby.mk like this
 >> 
 >> RUBY_CONF_ENV += \
 >> ac_cv_func_finite=yes \
 >> ac_cv_func_isinf=yes \
 >> ac_cv_func_isnan=yes
 >> endif

 > I think this is the best thing todo inside buildroot.

Yes, that might be the way to go. Notice that musl implements isinf /
isnan as macros, so the detection fails. The reason why it doesn't cause
linker errors is that include/ruby/missing.h checks if isnan is a macro
before providing a local declaration:

#ifndef isnan
# ifndef HAVE_ISNAN
RUBY_EXTERN int isnan(double);
# endif
#endif

vs:

#ifndef HAVE_FINITE
RUBY_EXTERN int finite(double);
#endif

So another solution is to add an #ifndef finite around it. That fix
could probably be upstreamed as they already do it for a number of other
functions.

-- 
Bye, Peter Korsgaard

[Buildroot] [PATCH] ruby: bump version to 2.4.0

[Thomas Petazzoni]

Hello,

On Mon, 30 Jan 2017 10:18:39 +0100, Peter Korsgaard wrote:

> Yes, that might be the way to go. Notice that musl implements isinf /
> isnan as macros, so the detection fails. The reason why it doesn't cause
> linker errors is that include/ruby/missing.h checks if isnan is a macro
> before providing a local declaration:
> 
> #ifndef isnan
> # ifndef HAVE_ISNAN
> RUBY_EXTERN int isnan(double);
> # endif
> #endif
> 
> vs:
> 
> #ifndef HAVE_FINITE
> RUBY_EXTERN int finite(double);
> #endif
> 
> So another solution is to add an #ifndef finite around it. That fix
> could probably be upstreamed as they already do it for a number of other
> functions.

Vicente, since you provided the patch bumping ruby to 2.4.0, could you
provide a patch that fixes the build issues, as suggested by Peter?

Thanks!

Thomas
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com