Strange test suite behaviour since zsh 5.2-dev-1: C02cond.ztst no more reports its finishing

Strange test suite behaviour since zsh 5.2-dev-1: C02cond.ztst no more reports its finishing

[Axel Beckert]

Hi,

I've seen this locally only once (on x86_64) once (IIRC when enabling
-fPIE for one test build), but on x86 (32 bit) the following issue has
been observed on our (Debian Zsh Packaging Team) Jenkins since I pushd
5.2-dev-1
(http://jenkins.grml.org/view/Debian/job/zsh-binaries/architecture=i386/):

>From the log of the test suite (examples taken from
http://jenkins.grml.org/view/Debian/job/zsh-binaries/architecture=i386/251/console):

[…]
20:56:44 Running test: type of variable when created in arithmetic context
20:56:44 Test successful.
20:56:44 ../../Test/C01arith.ztst: all tests successful.
20:56:44 ../../Test/C02cond.ztst: starting.
20:56:44 Running test: -a cond
20:56:44 Test successful.
[…]
20:56:46 Running test: test with two arguments and a token
20:56:46 Test successful.
20:56:46 Running test: yet more old-fashioned test fix ups: prefer comparison to parentheses
20:56:46 ../../Test/C03traps.ztst: starting.
20:56:46 Running test: Nested `trap ... EXIT'
20:56:46 Test successful.
[…]

What I miss is the "Test successful." for "yet more old-fashioned test
fix ups: prefer comparison to parentheses" as well as the
"../../Test/C02cond.ztst: all tests successful.".

Even if "yet more old-fashioned test fix ups: prefer comparison to
parentheses" did fail, I'd expect that it at least reports that fact.
But it looks as if the whole test script just silently exits.

In comparison, the x86_64 test suite runs look fine (taken from
http://jenkins.grml.org/view/Debian/job/zsh-binaries/architecture=amd64/251/consoleFull):

[…]
20:47:55 Running test: type of variable when created in arithmetic context
20:47:55 Test successful.
20:47:55 ../../Test/C01arith.ztst: all tests successful.
20:47:55 ../../Test/C02cond.ztst: starting.
20:47:56 Running test: -a cond
20:47:56 Test successful.
[…]
20:47:58 Running test: yet more old-fashioned test fix ups: prefer comparison to parentheses
20:47:58 Test successful.
20:47:58 ../../Test/C02cond.ztst: all tests successful.
20:47:58 ../../Test/C03traps.ztst: starting.
20:47:58 Running test: Nested `trap ... EXIT'
20:47:58 Test successful.
[…]

Any ideas what could cause this silent exiting of a .ztst script and
why it only appears under certain (yet not 100% clear) conditions?

		Kind regards, Axel
-- 
/~\  Plain Text Ribbon Campaign                   | Axel Beckert
\ /  Say No to HTML in E-Mail and News            | abe@deuxchevaux.org  (Mail)
 X   See http://www.nonhtmlmail.org/campaign.html | abe@noone.org (Mail+Jabber)
/ \  I love long mails: http://email.is-not-s.ms/ | http://abe.noone.org/ (Web)

Re: Strange test suite behaviour since zsh 5.2-dev-1: C02cond.ztst no more reports its finishing

[Bart Schaefer]

On Jan 31,  9:19pm, Axel Beckert wrote:
}
} Any ideas what could cause this silent exiting of a .ztst script and
} why it only appears under certain (yet not 100% clear) conditions?

Silent exit occurs when the test crashes the test-harness shell.  This
is an unfortunate side-effect of the way error output is managed by the
harness; the only way to detect it is to spot coredump files, or find
temp files that have never been removed, left behind after the "make".

What circumstances are causing the crash, you'll have to try to run down.

Re: Strange test suite behaviour since zsh 5.2-dev-1: C02cond.ztst no more reports its finishing

[Axel Beckert]

Hi Bart,

Bart Schaefer wrote:
> On Jan 31,  9:19pm, Axel Beckert wrote:
> } Any ideas what could cause this silent exiting of a .ztst script and
> } why it only appears under certain (yet not 100% clear) conditions?
> 
> Silent exit occurs when the test crashes the test-harness shell.

Yeah, sounds reasonable.

> This is an unfortunate side-effect of the way error output is
> managed by the harness; the only way to detect it is to spot
> coredump files, or find temp files that have never been removed,
> left behind after the "make".
> 
> What circumstances are causing the crash, you'll have to try to run
> down.

I'll try to reproduce it locally again and see what I can find. Thanks
for the hints.

		Regards, Axel
-- 
 ,''`.  |  Axel Beckert <abe@debian.org>, http://people.debian.org/~abe/
: :' :  |  Debian Developer, ftp.ch.debian.org Admin
`. `'   |  4096R: 2517 B724 C5F6 CA99 5329  6E61 2FF9 CD59 6126 16B5
  `-    |  1024D: F067 EA27 26B9 C3FC 1486  202E C09E 1D89 9593 0EDE

C02cond.ztst segfaults (was: Re: Strange test suite behaviour since zsh 5.2-dev-1: C02cond.ztst no more reports its finishing)

[Axel Beckert]

Hi Bart,

[Taking pkg-zsh-devel@lists.alioth.debian.org out of the loop for now.]

Bart Schaefer wrote:
> On Jan 31,  9:19pm, Axel Beckert wrote:
> } Any ideas what could cause this silent exiting of a .ztst script and
> } why it only appears under certain (yet not 100% clear) conditions?
> 
> Silent exit occurs when the test crashes the test-harness shell.  This
> is an unfortunate side-effect of the way error output is managed by the
> harness; the only way to detect it is to spot coredump files, or find

Thanks for reminding me of looking for core dumps!

I remembered it right: Building zsh with -fPIE triggered the issue (or
at least a very similar one) also on x86_64. And since I got a
core dump, here's the backtrace:

Core was generated by `../Src/zsh +Z -f ../../Test/ztst.zsh ../../Test/C02cond.ztst'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0  __strcmp_sse2_unaligned () at ../sysdeps/x86_64/multiarch/strcmp-sse2-unaligned.S:30
30      ../sysdeps/x86_64/multiarch/strcmp-sse2-unaligned.S: No such file or directory.
(gdb) bt
#0  __strcmp_sse2_unaligned () at ../sysdeps/x86_64/multiarch/strcmp-sse2-unaligned.S:30
#1  0x000055e7dc4af8f4 in is_cond_binary_op (str=0x7f05376fe238 "non-empty-string") at ../../Src/text.c:63
#2  0x000055e7dc445491 in bin_test (name=0x7f05376fe228 "[", argv=0x7f05376fe260, ops=<optimized out>, func=<optimized out>) at ../../Src/builtin.c:6556
#3  0x000055e7dc4458b8 in execbuiltin (args=args@entry=0x7f05376fe198, assigns=assigns@entry=0x0, bn=bn@entry=0x55e7dc6df160 <builtins+320>) at ../../Src/builtin.c:484
#4  0x000055e7dc45390a in execcmd (state=state@entry=0x7ffcdc3096f0, input=input@entry=0, output=output@entry=0, how=<optimized out>, how@entry=2, last1=<optimized out>) at ../../Src/exec.c:3651
#5  0x000055e7dc45420e in execpline2 (state=state@entry=0x7ffcdc3096f0, pcode=<optimized out>, how=how@entry=2, input=0, output=0, last1=last1@entry=0) at ../../Src/exec.c:1749
#6  0x000055e7dc4545ec in execpline (state=state@entry=0x7ffcdc3096f0, slcode=<optimized out>, how=how@entry=2, last1=last1@entry=0) at ../../Src/exec.c:1527
#7  0x000055e7dc455eff in execlist (state=state@entry=0x7ffcdc3096f0, dont_change_job=dont_change_job@entry=1, exiting=exiting@entry=0) at ../../Src/exec.c:1293
#8  0x000055e7dc4561f0 in execode (p=0x7f05376fda20, dont_change_job=dont_change_job@entry=1, exiting=exiting@entry=0, context=context@entry=0x55e7dc4c57e8 "eval") at ../../Src/exec.c:1076
#9  0x000055e7dc43c276 in eval (argv=0x7f05376fd250) at ../../Src/builtin.c:5509
#10 0x000055e7dc4458b8 in execbuiltin (args=args@entry=0x7f05376fd148, assigns=assigns@entry=0x0, bn=bn@entry=0x55e7dc6df5e0 <builtins+1472>) at ../../Src/builtin.c:484
#11 0x000055e7dc45390a in execcmd (state=state@entry=0x7ffcdc30b000, input=input@entry=0, output=output@entry=0, how=<optimized out>, how@entry=18, last1=<optimized out>) at ../../Src/exec.c:3651
#12 0x000055e7dc45420e in execpline2 (state=state@entry=0x7ffcdc30b000, pcode=<optimized out>, how=how@entry=18, input=0, output=0, last1=last1@entry=0) at ../../Src/exec.c:1749
#13 0x000055e7dc4545ec in execpline (state=state@entry=0x7ffcdc30b000, slcode=<optimized out>, how=how@entry=18, last1=0) at ../../Src/exec.c:1527
#14 0x000055e7dc455ddc in execlist (state=state@entry=0x7ffcdc30b000, dont_change_job=dont_change_job@entry=1, exiting=exiting@entry=0) at ../../Src/exec.c:1285
#15 0x000055e7dc4561f0 in execode (p=p@entry=0x7f0537701288, dont_change_job=dont_change_job@entry=1, exiting=exiting@entry=0, context=context@entry=0x55e7dc4bff74 "shfunc") at ../../Src/exec.c:1076
#16 0x000055e7dc456ecb in runshfunc (prog=0x7f0537701288, wrap=0x0, name=0x7f05376fd028 "(anon)") at ../../Src/exec.c:5368
#17 0x000055e7dc457630 in doshfunc (shfunc=shfunc@entry=0x55e7de4d05a0, doshargs=doshargs@entry=0x7f05377012c0, noreturnval=noreturnval@entry=0) at ../../Src/exec.c:5234
#18 0x000055e7dc4579ff in execshfunc (shf=shf@entry=0x55e7de4d05a0, args=args@entry=0x7f05377012c0) at ../../Src/exec.c:4865
#19 0x000055e7dc4581f3 in execshfunc (args=0x7f05377012c0, shf=0x55e7de4d05a0) at ../../Src/exec.c:4727
#20 execfuncdef (state=state@entry=0x7ffcdc30b950, redir_prog=redir_prog@entry=0x0) at ../../Src/exec.c:4731
#21 0x000055e7dc4584aa in execsimple (state=state@entry=0x7ffcdc30b950) at ../../Src/exec.c:1130
#22 0x000055e7dc455e62 in execlist (state=state@entry=0x7ffcdc30b950, dont_change_job=dont_change_job@entry=1, exiting=exiting@entry=0) at ../../Src/exec.c:1283
#23 0x000055e7dc4561f0 in execode (p=p@entry=0x55e7de4cc5e0, dont_change_job=dont_change_job@entry=1, exiting=exiting@entry=0, context=context@entry=0x55e7dc4bff74 "shfunc") at ../../Src/exec.c:1076
#24 0x000055e7dc456ecb in runshfunc (prog=0x55e7de4cc5e0, wrap=0x0, name=0x7f0537705030 "ZTST_execchunk") at ../../Src/exec.c:5368
#25 0x000055e7dc457630 in doshfunc (shfunc=shfunc@entry=0x55e7de4ccea0, doshargs=doshargs@entry=0x7f0537884ba8, noreturnval=noreturnval@entry=0) at ../../Src/exec.c:5234
#26 0x000055e7dc4579ff in execshfunc (shf=shf@entry=0x55e7de4ccea0, args=args@entry=0x7f0537884ba8) at ../../Src/exec.c:4865
#27 0x000055e7dc45397a in execshfunc (args=0x7f0537884ba8, shf=0x55e7de4ccea0) at ../../Src/exec.c:4831
#28 execcmd (state=state@entry=0x7ffcdc310500, input=input@entry=0, output=output@entry=0, how=<optimized out>, how@entry=2, last1=2) at ../../Src/exec.c:3532
#29 0x000055e7dc45420e in execpline2 (state=state@entry=0x7ffcdc310500, pcode=<optimized out>, how=how@entry=2, input=0, output=0, last1=last1@entry=0) at ../../Src/exec.c:1749
#30 0x000055e7dc4545ec in execpline (state=state@entry=0x7ffcdc310500, slcode=<optimized out>, how=how@entry=2, last1=0) at ../../Src/exec.c:1527
#31 0x000055e7dc455ddc in execlist (state=state@entry=0x7ffcdc310500, dont_change_job=dont_change_job@entry=1, exiting=exiting@entry=0) at ../../Src/exec.c:1285
#32 0x000055e7dc479e9b in execif (state=0x7ffcdc310500, do_exec=0) at ../../Src/loop.c:565
#33 0x000055e7dc4526bd in execcmd (state=state@entry=0x7ffcdc310500, input=input@entry=0, output=output@entry=0, how=<optimized out>, how@entry=2, last1=2) at ../../Src/exec.c:3480
#34 0x000055e7dc45420e in execpline2 (state=state@entry=0x7ffcdc310500, pcode=<optimized out>, how=how@entry=2, input=0, output=0, last1=last1@entry=0) at ../../Src/exec.c:1749
#35 0x000055e7dc4545ec in execpline (state=state@entry=0x7ffcdc310500, slcode=<optimized out>, how=how@entry=2, last1=0) at ../../Src/exec.c:1527
#36 0x000055e7dc455ddc in execlist (state=state@entry=0x7ffcdc310500, dont_change_job=dont_change_job@entry=1, exiting=exiting@entry=0) at ../../Src/exec.c:1285
#37 0x000055e7dc479aa7 in execwhile (state=0x7ffcdc310500, do_exec=<optimized out>) at ../../Src/loop.c:452
#38 0x000055e7dc4526bd in execcmd (state=state@entry=0x7ffcdc310500, input=input@entry=0, output=output@entry=0, how=<optimized out>, how@entry=2, last1=2) at ../../Src/exec.c:3480
#39 0x000055e7dc45420e in execpline2 (state=state@entry=0x7ffcdc310500, pcode=<optimized out>, how=how@entry=2, input=0, output=0, last1=last1@entry=0) at ../../Src/exec.c:1749
#40 0x000055e7dc4545ec in execpline (state=state@entry=0x7ffcdc310500, slcode=<optimized out>, how=how@entry=2, last1=0) at ../../Src/exec.c:1527
#41 0x000055e7dc455ddc in execlist (state=state@entry=0x7ffcdc310500, dont_change_job=dont_change_job@entry=1, exiting=exiting@entry=0) at ../../Src/exec.c:1285
#42 0x000055e7dc4561f0 in execode (p=p@entry=0x55e7de4cd0d0, dont_change_job=dont_change_job@entry=1, exiting=exiting@entry=0, context=context@entry=0x55e7dc4bff74 "shfunc") at ../../Src/exec.c:1076
#43 0x000055e7dc456ecb in runshfunc (prog=0x55e7de4cd0d0, wrap=0x0, name=0x7f0537884030 "ZTST_test") at ../../Src/exec.c:5368
#44 0x000055e7dc457630 in doshfunc (shfunc=shfunc@entry=0x55e7de4cd7d0, doshargs=doshargs@entry=0x7f0537889560, noreturnval=noreturnval@entry=0) at ../../Src/exec.c:5234
#45 0x000055e7dc4579ff in execshfunc (shf=shf@entry=0x55e7de4cd7d0, args=args@entry=0x7f0537889560) at ../../Src/exec.c:4865
#46 0x000055e7dc45397a in execshfunc (args=0x7f0537889560, shf=0x55e7de4cd7d0) at ../../Src/exec.c:4831
#47 execcmd (state=state@entry=0x7ffcdc3150f0, input=input@entry=0, output=output@entry=0, how=<optimized out>, how@entry=2, last1=2) at ../../Src/exec.c:3532
#48 0x000055e7dc45420e in execpline2 (state=state@entry=0x7ffcdc3150f0, pcode=<optimized out>, how=how@entry=2, input=0, output=0, last1=last1@entry=0) at ../../Src/exec.c:1749
#49 0x000055e7dc4545ec in execpline (state=state@entry=0x7ffcdc3150f0, slcode=<optimized out>, how=how@entry=2, last1=0) at ../../Src/exec.c:1527
#50 0x000055e7dc455ddc in execlist (state=state@entry=0x7ffcdc3150f0, dont_change_job=dont_change_job@entry=1, exiting=0) at ../../Src/exec.c:1285
#51 0x000055e7dc47a184 in execcase (state=0x7ffcdc3150f0, do_exec=0) at ../../Src/loop.c:658
#52 0x000055e7dc4526bd in execcmd (state=state@entry=0x7ffcdc3150f0, input=input@entry=0, output=output@entry=0, how=<optimized out>, how@entry=18, last1=2) at ../../Src/exec.c:3480
#53 0x000055e7dc45420e in execpline2 (state=state@entry=0x7ffcdc3150f0, pcode=<optimized out>, how=how@entry=18, input=0, output=0, last1=last1@entry=0) at ../../Src/exec.c:1749
#54 0x000055e7dc4545ec in execpline (state=state@entry=0x7ffcdc3150f0, slcode=<optimized out>, how=how@entry=18, last1=0) at ../../Src/exec.c:1527
#55 0x000055e7dc455ddc in execlist (state=state@entry=0x7ffcdc3150f0, dont_change_job=dont_change_job@entry=1, exiting=exiting@entry=0) at ../../Src/exec.c:1285
#56 0x000055e7dc479aa7 in execwhile (state=0x7ffcdc3150f0, do_exec=<optimized out>) at ../../Src/loop.c:452
#57 0x000055e7dc4526bd in execcmd (state=state@entry=0x7ffcdc3150f0, input=input@entry=0, output=output@entry=0, how=<optimized out>, how@entry=18, last1=2) at ../../Src/exec.c:3480
#58 0x000055e7dc45420e in execpline2 (state=state@entry=0x7ffcdc3150f0, pcode=<optimized out>, how=how@entry=18, input=0, output=0, last1=last1@entry=0) at ../../Src/exec.c:1749
#59 0x000055e7dc4545ec in execpline (state=state@entry=0x7ffcdc3150f0, slcode=<optimized out>, how=how@entry=18, last1=0) at ../../Src/exec.c:1527
#60 0x000055e7dc455ddc in execlist (state=state@entry=0x7ffcdc3150f0, dont_change_job=dont_change_job@entry=0, exiting=exiting@entry=0) at ../../Src/exec.c:1285
#61 0x000055e7dc4561f0 in execode (p=p@entry=0x7f0537709eb0, dont_change_job=dont_change_job@entry=0, exiting=exiting@entry=0, context=context@entry=0x55e7dc4c1189 "toplevel") at ../../Src/exec.c:1076
#62 0x000055e7dc469f13 in loop (toplevel=toplevel@entry=1, justonce=justonce@entry=0) at ../../Src/init.c:208
#63 0x000055e7dc46d55e in zsh_main (argc=<optimized out>, argv=<optimized out>) at ../../Src/init.c:1684
#64 0x00007f05369e8870 in __libc_start_main (main=0x55e7dc438230 <main>, argc=5, argv=0x7ffcdc315448, init=<optimized out>, fini=<optimized out>, rtld_fini=<optimized out>, stack_end=0x7ffcdc315438) at libc-start.c:291
#65 0x000055e7dc438269 in _start ()

The full line with which zsh has been linked in the end was:

gcc -Wl,--as-needed -g -fPIE -pie -Wl,-z,relro -Wl,-z,now -rdynamic \
 -o zsh main.o `cat stamp-modobjs` -lgdbm -lpcre -lcap -ldl \
 -lncursesw -lrt -lm -lc

The full list of parameters passed to make seem to be in this line (originally
without the line breaks):

make -f Makemod prefix='/usr' exec_prefix='/usr' bindir='/bin'
libdir='/usr/lib/x86_64-linux-gnu'
MODDIR='/usr/lib/x86_64-linux-gnu/zsh/5.2-dev-1'
infodir='/usr/share/info' mandir='/usr/share/man' datadir='/usr/share'
fndir='/usr/share/zsh/functions' htmldir='/usr/share/zsh/htmldoc'
runhelpdir='/usr/share/zsh/5.2-dev-1/help' CC='gcc'
CPPFLAGS='-Wdate-time -D_FORTIFY_SOURCE=2 ' DEFS='-DHAVE_CONFIG_H'
CFLAGS='-g -O2 -fPIE -fstack-protector-strong -Wformat
-Werror=format-security -Wall -g' LDFLAGS='-Wl,--as-needed -g -fPIE
-pie -Wl,-z,relro -Wl,-z,now' EXTRA_LDFLAGS='-rdynamic'
DLCFLAGS='-fPIC' DLLDFLAGS='-shared' LIBLDFLAGS='' EXELDFLAGS=''
LIBS='-lgdbm -lpcre -lcap -ldl -lncursesw -lrt -lm  -lc' DL_EXT='so'
DLLD='gcc' AWK='gawk' ANSI2KNR=': ansi2knr' YODL='yodl -k -L'
YODL2TXT='yodl2txt' YODL2HTML='yodl2html' FUNCTIONS_INSTALL=''
tzsh='zsh' prep

HTH!

		Regards, Axel
-- 
 ,''`.  |  Axel Beckert <abe@debian.org>, http://people.debian.org/~abe/
: :' :  |  Debian Developer, ftp.ch.debian.org Admin
`. `'   |  4096R: 2517 B724 C5F6 CA99 5329  6E61 2FF9 CD59 6126 16B5
  `-    |  1024D: F067 EA27 26B9 C3FC 1486  202E C09E 1D89 9593 0EDE

Re: C02cond.ztst segfaults (was: Re: Strange test suite behaviour since zsh 5.2-dev-1: C02cond.ztst no more reports its finishing)

[Bart Schaefer]

On Feb 1, 12:03am, Axel Beckert wrote:
}
} Program terminated with signal SIGSEGV, Segmentation fault.
} #1  0x000055e7dc4af8f4 in is_cond_binary_op (str=0x7f05376fe238 "non-empty-string") at ../../Src/text.c:63

Try this:

diff --git a/Src/text.c b/Src/text.c
index 04acd2a..cf6d3eb 100644
--- a/Src/text.c
+++ b/Src/text.c
@@ -47,7 +47,7 @@ int text_expand_tabs;
  */
 static const char *cond_binary_ops[] = {
     "=", "!=", "<", ">", "-nt", "-ot", "-ef", "-eq",
-    "-ne", "-lt", "-gt", "-le", "-ge", "=~"
+    "-ne", "-lt", "-gt", "-le", "-ge", "=~", NULL
 };
 
 static char *tptr, *tbuf, *tlim, *tpending;

Re: C02cond.ztst segfaults (was: Re: Strange test suite behaviour since zsh 5.2-dev-1: C02cond.ztst no more reports its finishing)

[Axel Beckert]

Hi Bart,

On Sun, Jan 31, 2016 at 03:12:51PM -0800, Bart Schaefer wrote:
> On Feb 1, 12:03am, Axel Beckert wrote:
> } Program terminated with signal SIGSEGV, Segmentation fault.
> } #1  0x000055e7dc4af8f4 in is_cond_binary_op (str=0x7f05376fe238 "non-empty-string") at ../../Src/text.c:63
> 
> Try this:
> 
> diff --git a/Src/text.c b/Src/text.c
> index 04acd2a..cf6d3eb 100644
> --- a/Src/text.c
> +++ b/Src/text.c
> @@ -47,7 +47,7 @@ int text_expand_tabs;
>   */
>  static const char *cond_binary_ops[] = {
>      "=", "!=", "<", ">", "-nt", "-ot", "-ef", "-eq",
> -    "-ne", "-lt", "-gt", "-le", "-ge", "=~"
> +    "-ne", "-lt", "-gt", "-le", "-ge", "=~", NULL
>  };
>  
>  static char *tptr, *tbuf, *tlim, *tpending;

Thanks! That fixed at least my local segfault when compiling with
-fPIE. Will now push that patch to our Jenkins. Hopefully this also
fixes the 32-bit C02cond.ztst issues, at least the symptoms are the
same.

		Kind regards, Axel
-- 
/~\  Plain Text Ribbon Campaign                   | Axel Beckert
\ /  Say No to HTML in E-Mail and News            | abe@deuxchevaux.org  (Mail)
 X   See http://www.nonhtmlmail.org/campaign.html | abe@noone.org (Mail+Jabber)
/ \  I love long mails: http://email.is-not-s.ms/ | http://abe.noone.org/ (Web)

Re: C02cond.ztst segfaults (was: Re: Strange test suite behaviour since zsh 5.2-dev-1: C02cond.ztst no more reports its finishing)

[Axel Beckert]

Hi,

Axel Beckert wrote:
> On Sun, Jan 31, 2016 at 03:12:51PM -0800, Bart Schaefer wrote:
> > On Feb 1, 12:03am, Axel Beckert wrote:
> > } Program terminated with signal SIGSEGV, Segmentation fault.
> > } #1  0x000055e7dc4af8f4 in is_cond_binary_op (str=0x7f05376fe238 "non-empty-string") at ../../Src/text.c:63
> > 
> > Try this:
> > 
> > diff --git a/Src/text.c b/Src/text.c
> > index 04acd2a..cf6d3eb 100644
> > --- a/Src/text.c
> > +++ b/Src/text.c
> > @@ -47,7 +47,7 @@ int text_expand_tabs;
> >   */
> >  static const char *cond_binary_ops[] = {
> >      "=", "!=", "<", ">", "-nt", "-ot", "-ef", "-eq",
> > -    "-ne", "-lt", "-gt", "-le", "-ge", "=~"
> > +    "-ne", "-lt", "-gt", "-le", "-ge", "=~", NULL
> >  };
> >  
> >  static char *tptr, *tbuf, *tlim, *tpending;
> 
> Thanks! That fixed at least my local segfault when compiling with
> -fPIE. Will now push that patch to our Jenkins. Hopefully this also
> fixes the 32-bit C02cond.ztst issues, at least the symptoms are the
> same.

It did. Thanks again, Bart!

		Regards, Axel
-- 
 ,''`.  |  Axel Beckert <abe@debian.org>, http://people.debian.org/~abe/
: :' :  |  Debian Developer, ftp.ch.debian.org Admin
`. `'   |  4096R: 2517 B724 C5F6 CA99 5329  6E61 2FF9 CD59 6126 16B5
  `-    |  1024D: F067 EA27 26B9 C3FC 1486  202E C09E 1D89 9593 0EDE