On Wed, Jul 26, 2017 at 08:14:14AM -0600, Jerry James wrote:
> Failing that, does anybody with access
> to an s390x instance have time to help me? The latest bigloo release
> built successfully on all arches but s390x:
>
> https://koji.fedoraproject.org/koji/taskinfo?taskID=20740401
>
> The failure comes after the main binary has been built and linked
> successfully, but then is executed to compile some Scheme files. The
> binary is reporting that it was invoked on an empty input file.
Well it's certainly reproducible:
/home/nfs/rjones/d/fedora/bigloo/master/bigloo4.3a/bin/bigloo.sh -srfi
enable-unistring -srfi enable-resolv -srfi enable-gmp -srfi enable-pcre -unsafe
-q -mkheap -mklib -s -no-hello -L
/home/nfs/rjones/d/fedora/bigloo/master/bigloo4.3a/lib/bigloo/4.3a
Llib/make_lib.scm -heap bigloo.heap
*** ERROR:top-level:Parse error:
Illegal source file -- #eof-object
I ran the same command under gdb to get a stack trace:
Thread 1 (Thread 0x3fffdff4720 (LWP 38254)):
#0 0x000003fffd843ea8 in exit () from /lib64/libc.so.6
#1 0x000003fffdc3893e in bigloo_exit (val=val@entry=0x9) at Clib/cmain.c:118
#2 0x000003fffddb8c80 in BGl_exitz00zz__errorz00 (BgL_nz00_13=<optimized out>)
at objs/obj_s/Llib/error.c:1530
#3 0x000003fffddb9066 in BGl_raisez00zz__errorz00 (BgL_valz00_16=0x16e5480)
at objs/obj_s/Llib/error.c:1823
#4 0x000003fffddb9516 in BGl_withzd2exceptionzd2handlerz00zz__errorz00 (
BgL_handlerz00_14=<optimized out>,
BgL_thunkz00_15=BgL_thunkz00_15@entry=0x178f640)
at objs/obj_s/Llib/error.c:1595
#5 0x00000000011a3bd6 in BGl_z62zc3anonymousza31407ze3ze1zztools_errorz00 (
BgL_envz00_1676=<optimized out>) at Tools/error.c:1342
#6 0x00000000011a3e56 in BGl_withzd2dumpzd2stackz00zztools_errorz00 (
BgL_thunkz00_20=0x179d9c3) at Tools/error.c:548
#7 0x00000000011a465e in BGl_userzd2errorzd2zztools_errorz00 (
BgL_procz00_31=BgL_procz00_31@entry=0x12f0340
<BgL_bgl_string1902za700za7za7e1953za7+8>,
BgL_mesz00_32=BgL_mesz00_32@entry=0x12f0308
<BgL_bgl_string1903za700za7za7e1954za7+8>,
BgL_objz00_33=BgL_objz00_33@entry=0x802,
BgL_continuez00_34=BgL_continuez00_34@entry=0x2) at Tools/error.c:1041
#8 0x000000000100c84c in BGl_compilerz00zzengine_compilerz00 ()
at Engine/compiler.c:802
#9 0x0000000001233244 in BGl_mainz00zzinit_mainz00 (
BgL_argvz00_3=BgL_argvz00_3@entry=0x1517d43) at Init/main.c:274
#10 0x000000000123331e in bigloo_main (BgL_argvz00_46=0x1517d43)
at Init/main.c:144
#11 0x000003fffdc38bb4 in _bigloo_main (argc=<optimized out>,
argv=<optimized out>, env=<optimized out>,
bigloo_main=bigloo_main@entry=0x12332c8 <bigloo_main>,
libinit=libinit@entry=0x1232f28 <bigloo_libinit>, uheapsize=0)
at Clib/cmain.c:229
#12 0x000000000100befc in main (argc=<optimized out>, argv=<optimized out>,
env=<optimized out>) at Init/main.c:107
Unfortunately once I started to look at the code it seems to be some
kind of generated C which is almost impossible to follow, eg:
(gdb) frame 8
#8 0x000000000100c84c in BGl_compilerz00zzengine_compilerz00 ()
at Engine/compiler.c:802
802
BGl_userzd2errorzd2zztools_errorz00
(gdb) list
797
MAKE_YOUNG_PAIR(BINT(((long) 1)), BNIL);
798
BGl_exitz00zz__errorz00(BgL_list1426z00_1638);
799
}
800
else
801
{/*
Engine/compiler.scm 123 */
802
BGl_userzd2errorzd2zztools_errorz00
803
(BGl_string1902z00zzengine_compilerz00,
804
BGl_string1903z00zzengine_compilerz00, BgL_srcz00_1635,
805
BNIL);
806
}
> Are there any test s390x instances available for a peon like me to use
> to diagnose a build failure?
You can ask for access to a machine on #fedora-s390x (Freenode).
Rich.
--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-builder quickly builds VMs from scratch
http://libguestfs.org/virt-builder.1.html
_______________________________________________
devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
