New subject: [Bug 1288611] epel7 ghc-glib ftbfs on aarch64

4 Dec 2015


      https://bugzilla.redhat.com/show_bug.cgi?id=1288611
Bug ID: 1288611
           Summary: epel7 ghc-glib ftbfs on aarch64
           Product: Fedora EPEL
           Version: epel7
         Component: gtk2hs-buildtools
          Assignee: petersen@redhat.com
          Reporter: yselkowi@redhat.com
        QA Contact: extras-qa@fedoraproject.org
                CC: haskell-devel@lists.fedoraproject.org,
                    petersen@redhat.com
        Depends On: 1118232
            Blocks: 1285484 (epel7aarch64)
+++ This bug was initially created as a clone of Bug #1118232 +++
ghc-glib-0.12.5.0-1.fc21 is ftbfs on aarch64
http://arm.koji.fedoraproject.org/koji/taskinfo?taskID=2485540
++ ghc --info
++ grep '"Booter version","7.6.3"'
+ '[' '!' ' ,("Booter version","7.6.3")' ']'
+ '[' -x Setup ']'
+ ghc --make -no-user-package-db Setup
[1 of 2] Compiling SetupWrapper     ( SetupWrapper.hs, SetupWrapper.o )
[2 of 2] Compiling Main             ( Setup.hs, Setup.o )
Linking Setup ...
+ ./Setup configure --prefix=/usr --libdir=/usr/lib64
--docdir=/usr/share/doc/ghc-glib '--libsubdir=$compiler/$pkgid'
'--datasubdir=$pkgid' --ghc --ghc-option=-O2 -p
--htmldir=/usr/share/doc/ghc/html/libraries/glib --global
[1 of 2] Compiling Gtk2HsSetup      ( Gtk2HsSetup.hs,
dist/setup-wrapper/Gtk2HsSetup.o )
[2 of 2] Compiling Main             ( SetupMain.hs, dist/setup-wrapper/Main.o )
Linking dist/setup-wrapper/setup ...
Configuring glib-0.12.5.0...
+ '[' -x Setup ']'
+ ./Setup build
Building glib-0.12.5.0...
Preprocessing library glib-0.12.5.0...
gtk2hsC2hs: Error in C header file.
/usr/include/asm/sigcontext.h:53: (column 2) [FATAL]
...
...
...
Syntax error!
The symbol `__uint128_t' does not fit here.
error: Bad exit status from /var/tmp/rpm-tmp.YQQi78 (%build)
    Bad exit status from /var/tmp/rpm-tmp.YQQi78 (%build)
RPM build errors:
--- Additional comment from Jens Petersen on 2014-09-30 23:53:43 EDT ---
I started looking at this: here is where the error occurs:
$ LANG=en_US.utf8 ./Setup build -v
creating dist/build
creating dist/build/autogen
Building glib-0.12.5.0...
Preprocessing library glib-0.12.5.0...
creating dist/build/System/Glib
/usr/bin/gtk2hsC2hs --include=dist/build
--include=/usr/lib64/ghc-7.6.3/base-4.6.0.1
--include=/usr/lib64/ghc-7.6.3/containers-0.5.0.0
--include=/usr/lib64/ghc-7.6.3/utf8-string-0.3.7 --cpp=/usr/bin/gcc
--cppopts=-E --cppopts=-Idist/build/autogen --cppopts=-Idist/build
--cppopts=-ISystem/Glib --cppopts=-I/usr/include/glib-2.0
--cppopts=-I/usr/lib64/glib-2.0/include --cppopts=-DUSE_GCLOSURE_SIGNALS_IMPL
--cppopts=-D__GLASGOW_HASKELL__=706 --output-dir=dist/build/System/Glib/
--output=GError.hs --precomp=dist/build/precompchs.bin glib-object.h
./System/Glib/GError.chs
gtk2hsC2hs: Error in C header file.
/usr/include/asm/sigcontext.h:53: (column 2) [FATAL]
...
...
...
Syntax error!
The symbol `__uint128_t' does not fit here.
I may try to see if things are better under ghc-7.8 before investigating
further...
--- Additional comment from Peter Robinson on 2014-10-01 12:56:34 EDT ---
...
I may try to see if things are better under ghc-7.8 before investigating
further...
Is 7.8 going to land in F-21?
--- Additional comment from Jens Petersen on 2014-10-01 22:34:07 EDT ---
(In reply to Peter Robinson from comment #2)
...
...
I may try to see if things are better under ghc-7.8 before investigating
further...
Is 7.8 going to land in F-21?
It was originally planned but we're behind schedule
so I don't think so at this point.
--- Additional comment from Peter Robinson on 2014-10-02 05:09:02 EDT ---
(In reply to Jens Petersen from comment #3)
...
(In reply to Peter Robinson from comment #2)
...
...
I may try to see if things are better under ghc-7.8 before investigating
further...
Is 7.8 going to land in F-21?
It was originally planned but we're behind schedule
so I don't think so at this point.
Can we investigate the issue further then? We need this fixed in F-21 too
--- Additional comment from Jens Petersen on 2014-10-02 22:51:14 EDT ---
(In reply to Peter Robinson from comment #4)
...
Can we investigate the issue further then? We need this fixed in F-21 too
Yes, I plan to, but first I was to know if it is fixed with latest
ghc and ghc-glib etc.  The problem seems related to preprocessing
of the headers and looks tricky to me.
Unfortunately getting ghc-7.8.3 to build with arm64 also seems non-trivial
so more fun in the pipeline... ;o)
--- Additional comment from Jens Petersen on 2014-10-08 04:56:33 EDT ---
Okay, I took the long scenic route, and have confirmed that Haskell glib builds
okay
on aarch64 with newer gtk2hs-buildtools.
Referenced Bugs:
https://bugzilla.redhat.com/show_bug.cgi?id=1118232
[Bug 1118232] ghc-glib ftbfs on aarch64
https://bugzilla.redhat.com/show_bug.cgi?id=1285484
[Bug 1285484] Tracker for enabling aarch64 in EPEL-7
-- 
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug https://bugzilla.redhat.com/token.cgi?t=tbAfiiHhny&a=cc_unsubscribe

[Bug 1288611] New: epel7 ghc-glib ftbfs on aarch64