SIGFPE integer divide by zero in S_make_trie #14945

p5pRT · 2015-09-27T16:35:24Z

Migrated from rt.perl.org#126206 (status was 'resolved')

Searchable as RT126206$

p5pRT · 2015-09-27T16:35:24Z

From @dcollinsn

Greetings Porters,

I have compiled bleadperl with the afl-gcc compiler using:

./Configure -Dusedevel -Dprefix='/usr/local/perl-afl' -Dcc='ccache afl-gcc' -Duselongdouble -Duse64bitint -Doptimize=-g -Uversiononly -Uman1dir -Uman3dir -des
AFL_HARDEN=1 make && make test

And then fuzzed the resulting binary using:

AFL_NO_VAR_CHECK=1 afl-fuzz -i in -o out bin/perl @@

After reducing testcases using `afl-tmin` and performing additional minimization by hand, I have located the following testcase that triggers a floating point exception in the perl interpreter. The testcase is the attached 11-character file:

dcollins@nagios:~/perl$ od -c /usr/local/perl-afl/out/allcrash/f3i000002
0000000 t . / \ N { } 337 | / i

This is very similar to [perl #125990], however this testcase never triggered that bug, and was presumably missed when the fuzzer found it alongside the many testcases that did. A debugging perl produces no other interesting output.

The version of 5.20.2 that shipped with my Debian does not have this bug. The result of a bisect run is shown below, and the problem does arise from the same commit, where \N{} forces /d to /u.

**GDB**

dcollins@nagios:~/perl$ gdb --args ./perl /usr/local/perl-afl/out/allcrash/f3i000002
GNU gdb (GDB) 7.10
Copyright (C) 2015 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "i686-pc-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from ./perl...done.
(gdb) run
Starting program: /home/dcollins/perl/perl /usr/local/perl-afl/out/allcrash/f3i000002
bt[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/libthread_db.so.1".

Program received signal SIGFPE, Arithmetic exception.
0x08119636 in S_make_trie (pRExC_state=0xbfffed64, startbranch=0x8477d8c,
first=0x8477d8c, last=0x8477da4, tail=0x8477da4, word_count=2, flags=35,
depth=1) at regcomp.c:2775
2775 accept_state = TRIE_NODENUM( state );
(gdb) bt
#0 0x08119636 in S_make_trie (pRExC_state=0xbfffed64, startbranch=0x8477d8c,
first=0x8477d8c, last=0x8477da4, tail=0x8477da4, word_count=2, flags=35,
depth=1) at regcomp.c:2775
#1 0x08123811 in S_study_chunk (pRExC_state=0xbfffed64, scanp=0xbfffeb54,
minlenp=0xbfffeb5c, deltap=0xbfffeb68, last=0x8477da8, data=0xbfffece4,
stopparen=-1, recursed_depth=0, and_withp=0x8478f60, flags=12288, depth=0)
at regcomp.c:4368
#2 0x08132920 in Perl_re_op_compile (patternp=0x0, pat_count=1,
expr=0x8477cdc, eng=0x83917e0 <PL_core_reg_engine>, old_re=0x0,
is_bare_re=0x0, orig_rx_flags=4, pm_flags=4) at regcomp.c:7296
#3 0x08071183 in Perl_pmruntime (o=0x8477ca4, expr=0x8477cdc, repl=0x0,
isreg=true, floor=0) at op.c:5579
#4 0x08107385 in Perl_yyparse (gramtype=258) at perly.y:1038
#5 0x0809b04e in S_parse_body (env=0x0, xsinit=0x80612f8 <xs_init>)
at perl.c:2304
#6 0x0809983a in perl_parse (my_perl=0x845d008, xsinit=0x80612f8 <xs_init>,
argc=2, argv=0xbffff4a4, env=0x0) at perl.c:1634
#7 0x08061252 in main (argc=2, argv=0xbffff4a4, env=0xbffff4b0)
at perlmain.c:114
(gdb) info locals
uc = 0x8477d94 "\002%\004"
e = 0x8477d94 "\002%\004"
state = 1
accept_state = 0
wordlen = 0
noper = 0x8477d90
charid = 0
trie = 0x8478fe0
widecharmap = 0x0
revcharmap = 0x8470614
cur = 0x8477d8c
len = 0
uvc = 0
curword = 0
next_alloc = 1
jumper = 0x0
nextbranch = 0x0
convert = 0x8477d8c
prev_states = 0x8477d40
folder = 0x842e540 <PL_fold_latin1> ""
data_slot = 0
trie_words = 0x0
re_trie_maxbuff = 0x8470600
re_debug_flags = 0
__PRETTY_FUNCTION__ = "S_make_trie"

**VALGRIND**

dcollins@nagios:~/perl$ valgrind ./perl /usr/local/perl-afl/out/allcrash/f3i000002
==16731== Memcheck, a memory error detector
==16731== Copyright (C) 2002-2013, and GNU GPL'd, by Julian Seward et al.
==16731== Using Valgrind-3.10.1 and LibVEX; rerun with -h for copyright info
==16731== Command: ./perl /usr/local/perl-afl/out/allcrash/f3i000002
==16731==
==16731==
==16731== Process terminating with default action of signal 8 (SIGFPE)
==16731== Integer divide by zero at address 0x6356C3A2
==16731== at 0x811963B: S_make_trie (regcomp.c:2775)
==16731== by 0x8123810: S_study_chunk (regcomp.c:4368)
==16731== by 0x813291F: Perl_re_op_compile (regcomp.c:7296)
==16731== by 0x8071182: Perl_pmruntime (op.c:5579)
==16731== by 0x8107384: Perl_yyparse (perly.y:1038)
==16731== by 0x809B04D: S_parse_body (perl.c:2304)
==16731== by 0x8099839: perl_parse (perl.c:1634)
==16731== by 0x8061251: main (perlmain.c:114)
==16731==
==16731== HEAP SUMMARY:
==16731== in use at exit: 108,665 bytes in 658 blocks
==16731== total heap usage: 752 allocs, 94 frees, 113,684 bytes allocated
==16731==
==16731== LEAK SUMMARY:
==16731== definitely lost: 156 bytes in 1 blocks
==16731== indirectly lost: 2,593 bytes in 37 blocks
==16731== possibly lost: 0 bytes in 0 blocks
==16731== still reachable: 105,916 bytes in 620 blocks
==16731== suppressed: 0 bytes in 0 blocks
==16731== Rerun with --leak-check=full to see details of leaked memory
==16731==
==16731== For counts of detected and suppressed errors, rerun with: -v
==16731== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
Floating point exception

**BISECT**

75697d6 is the first bad commit
commit 75697d6
Author: Karl Williamson <khw@cpan.org>
Date: Sun Dec 21 22:02:30 2014 -0700

Empty \N{} in regex pattern should force /d to /u

\N{} is for Unicode names, even if the name is actually omitted.
(Accepting an empty name is, I believe, an accident, and now is
supported only for backwards compatibility.)

:100644 100644 e6ddb803aae9f37f00d2f0f38fd109f05bf0bcb6 c2521a97ccee8d15370822c3b975be2628fcddb2 M regcomp.c
:040000 040000 218758eff11e9ad6989600b1a4a93ca23c3da3cf a6a3a35946820a1f6e9b6f99f414edc78099b63b M t
bisect run success

**PERL -V**

dcollins@nagios:~/perl$ ./perl -V
Summary of my perl5 (revision 5 version 23 subversion 4) configuration:
Commit id: 7a36e61
Platform:
osname=linux, osvers=2.6.32-5-686, archname=i686-linux-64int-ld
uname='linux nagios 2.6.32-5-686 #1 smp tue may 13 16:33:32 utc 2014 i686 gnulinux '
config_args='-Dusedevel -Dprefix=/usr/local/perl-afl -Dcc=ccache afl-gcc -Duselongdouble -Duse64bitint -Doptimize=-g -Uversiononly -Uman1dir -Uman3dir -des'
hint=recommended, useposix=true, d_sigaction=define
useithreads=undef, usemultiplicity=undef
use64bitint=define, use64bitall=undef, uselongdouble=define
usemymalloc=n, bincompat5005=undef
Compiler:
cc='ccache afl-gcc', ccflags ='-fwrapv -fno-strict-aliasing -pipe -fstack-protector-strong -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64',
optimize='-g',
cppflags='-fwrapv -fno-strict-aliasing -pipe -fstack-protector-strong -I/usr/local/include'
ccversion='', gccversion='5.2.0', gccosandvers=''
intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=12345678, doublekind=3
d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=12, longdblkind=3
ivtype='long long', ivsize=8, nvtype='long double', nvsize=12, Off_t='off_t', lseeksize=8
alignbytes=4, prototype=define
Linker and Libraries:
ld='ccache afl-gcc', ldflags =' -fstack-protector-strong -L/usr/local/lib'
libpth=/usr/local/lib /usr/local/lib/gcc/i686-pc-linux-gnu/5.2.0/include-fixed /usr/lib /lib/../lib /usr/lib/../lib /lib /usr/lib/i486-linux-gnu /usr/lib64
libs=-lpthread -lnsl -ldl -lm -lcrypt -lutil -lc
perllibs=-lpthread -lnsl -ldl -lm -lcrypt -lutil -lc
libc=libc-2.22.so, so=so, useshrplib=false, libperl=libperl.a
gnulibc_version='2.22'
Dynamic Linking:
dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-Wl,-E'
cccdlflags='-fPIC', lddlflags='-shared -g -L/usr/local/lib -fstack-protector-strong'

Characteristics of this binary (from libperl):
Compile-time options: DEBUGGING HAS_TIMES PERLIO_LAYERS PERL_COPY_ON_WRITE
PERL_DONT_CREATE_GVSV
PERL_HASH_FUNC_ONE_AT_A_TIME_HARD PERL_MALLOC_WRAP
PERL_PRESERVE_IVUV PERL_USE_DEVEL USE_64_BIT_INT
USE_LARGE_FILES USE_LOCALE USE_LOCALE_COLLATE
USE_LOCALE_CTYPE USE_LOCALE_NUMERIC USE_LOCALE_TIME
USE_LONG_DOUBLE USE_PERLIO USE_PERL_ATOF
Built under linux
Compiled at Sep 25 2015 21:59:04
@INC:
/usr/local/perl-afl/lib/site_perl/5.23.4/i686-linux-64int-ld
/usr/local/perl-afl/lib/site_perl/5.23.4
/usr/local/perl-afl/lib/5.23.4/i686-linux-64int-ld
/usr/local/perl-afl/lib/5.23.4
/usr/local/perl-afl/lib/site_perl/5.23.3
/usr/local/perl-afl/lib/site_perl/5.23.2
/usr/local/perl-afl/lib/site_perl
.

p5pRT · 2015-09-27T16:35:24Z

From @dcollinsn

f3i000002

p5pRT · 2015-10-05T20:14:04Z

From @dcollinsn

A simpler testcase:

/\N{}Ss|/i

** GDB **

Starting program: /usr/local/perl-afl/bin/perl -e /\\N\{\}Ss\|/i
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".

Program received signal SIGFPE, Arithmetic exception.
0x00000000007497ce in S_make_trie (pRExC_state=0x7fffffffddc0, startbranch=0x0, first=0x0, last=0x1e, tail=0x3, word_count=18933704, flags=35, depth=1) at regcomp.c:2847
2847 const U32 laststate = TRIE_NODENUM( next_alloc );
(gdb) bt
#0 0x00000000007497ce in S_make_trie (pRExC_state=0x7fffffffddc0, startbranch=0x0, first=0x0, last=0x1e, tail=0x3, word_count=18933704, flags=35, depth=1) at regcomp.c:2847
#1 0x0000000000769d6c in S_study_chunk (pRExC_state=0x7fffffffdce0, scanp=0x0, minlenp=0x0, deltap=0x1e, last=0x3, data=0x120e7b0, stopparen=-1, recursed_depth=0, flags=10240, depth=0, and_withp=<optimized out>) at regcomp.c:4368
#2 0x00000000007998f8 in Perl_re_op_compile (patternp=0x120eae0, pat_count=0, expr=0x120e7b0, eng=0x1e, old_re=0x7, is_bare_re=0x7fffffffdd80, orig_rx_flags=4, pm_flags=4) at regcomp.c:7296
#3 0x00000000004e8872 in Perl_pmruntime (o=0x120e978, expr=0x120e938, repl=0x0, isreg=30, floor=7938400) at op.c:5572
#4 0x000000000066f2ed in Perl_yyparse (gramtype=18934136) at perly.y:1038
#5 0x000000000053a8d5 in S_parse_body (env=env@entry=0x0, xsinit=xsinit@entry=0x42c840 <xs_init>) at perl.c:2304
#6 0x0000000000542653 in perl_parse (my_perl=<optimized out>, xsinit=xsinit@entry=0x42c840 <xs_init>, argc=<optimized out>, argv=<optimized out>, env=env@entry=0x0) at perl.c:1634
#7 0x000000000042c468 in main (argc=3, argv=0x7fffffffe648, env=0x7fffffffe668) at perlmain.c:114
(gdb) info locals
state = 18933704
trie = 0x120fdd0
len = 0
uvc = 0
next_alloc = 1
folder = 0xfbab40 <PL_fold_latin1> ""
re_debug_flags = 0
__PRETTY_FUNCTION__ = "S_make_trie"

** VALGRIND **

==15719== Memcheck, a memory error detector
==15719== Copyright (C) 2002-2015, and GNU GPL'd, by Julian Seward et al.
==15719== Using Valgrind-3.11.0 and LibVEX; rerun with -h for copyright info
==15719== Command: ../bin/perl -e /\\N{}Ss|/i
==15719==
==15719==
==15719== Process terminating with default action of signal 8 (SIGFPE)
==15719== Integer divide by zero at address 0x80418779B
==15719== at 0x7497BC: S_make_trie (regcomp.c:2847)
==15719== by 0x769D6B: S_study_chunk.constprop.51 (regcomp.c:4368)
==15719== by 0x7998F7: Perl_re_op_compile (regcomp.c:7296)
==15719== by 0x4E8871: Perl_pmruntime (op.c:5572)
==15719== by 0x66F2EC: Perl_yyparse (perly.y:1038)
==15719== by 0x53A8D4: S_parse_body (perl.c:2304)
==15719== by 0x542652: perl_parse (perl.c:1634)
==15719== by 0x42C467: main (perlmain.c:114)
==15719==
==15719== HEAP SUMMARY:
==15719== in use at exit: 122,004 bytes in 573 blocks
==15719== total heap usage: 662 allocs, 89 frees, 130,332 bytes allocated
==15719==
==15719== LEAK SUMMARY:
==15719== definitely lost: 176 bytes in 1 blocks
==15719== indirectly lost: 1,993 bytes in 20 blocks
==15719== possibly lost: 16 bytes in 1 blocks
==15719== still reachable: 119,819 bytes in 551 blocks
==15719== suppressed: 0 bytes in 0 blocks
==15719== Rerun with --leak-check=full to see details of leaked memory
==15719==
==15719== For counts of detected and suppressed errors, rerun with: -v
==15719== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
Floating point exception

** PERL -V **

Summary of my perl5 (revision 5 version 23 subversion 4) configuration:
Commit id: 94757bf
Platform:
osname=linux, osvers=3.16.0-4-amd64, archname=x86_64-linux-ld
uname='linux nightshade64 3.16.0-4-amd64 #1 smp debian 3.16.7-ckt11-1+deb8u4 (2015-09-19) x86_64 gnulinux '
config_args='-Dusedevel -Dprefix=/usr/local/perl-afl -Dcc=ccache afl-gcc -Duselongdouble -Duse64bitall -Doptimize=-g -Uversiononly -Uman1dir -Uman3dir -DDEBUGGING -DPERL_POISON -des'
hint=recommended, useposix=true, d_sigaction=define
useithreads=undef, usemultiplicity=undef
use64bitint=define, use64bitall=define, uselongdouble=define
usemymalloc=n, bincompat5005=undef
Compiler:
cc='ccache afl-gcc', ccflags ='-fwrapv -DDEBUGGING -fno-strict-aliasing -pipe -fstack-protector-strong -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64',
optimize='-g',
cppflags='-fwrapv -DDEBUGGING -fno-strict-aliasing -pipe -fstack-protector-strong -I/usr/local/include'
ccversion='', gccversion='4.9.2', gccosandvers=''
intsize=4, longsize=8, ptrsize=8, doublesize=8, byteorder=12345678, doublekind=3
d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16, longdblkind=3
ivtype='long', ivsize=8, nvtype='long double', nvsize=16, Off_t='off_t', lseeksize=8
alignbytes=16, prototype=define
Linker and Libraries:
ld='ccache afl-gcc', ldflags =' -fstack-protector-strong -L/usr/local/lib'
libpth=/usr/local/lib /usr/lib/gcc/x86_64-linux-gnu/4.9/include-fixed /usr/include/x86_64-linux-gnu /usr/lib /lib/x86_64-linux-gnu /lib/../lib /usr/lib/x86_64-linux-gnu /usr/lib/../lib /lib
libs=-lpthread -lnsl -ldl -lm -lcrypt -lutil -lc
perllibs=-lpthread -lnsl -ldl -lm -lcrypt -lutil -lc
libc=libc-2.19.so, so=so, useshrplib=false, libperl=libperl.a
gnulibc_version='2.19'
Dynamic Linking:
dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-Wl,-E'
cccdlflags='-fPIC', lddlflags='-shared -g -L/usr/local/lib -fstack-protector-strong'

Characteristics of this binary (from libperl):
Compile-time options: DEBUGGING HAS_TIMES PERLIO_LAYERS PERL_COPY_ON_WRITE
PERL_DONT_CREATE_GVSV
PERL_HASH_FUNC_ONE_AT_A_TIME_HARD PERL_MALLOC_WRAP
PERL_PRESERVE_IVUV PERL_USE_DEVEL USE_64_BIT_ALL
USE_64_BIT_INT USE_LARGE_FILES USE_LOCALE
USE_LOCALE_COLLATE USE_LOCALE_CTYPE
USE_LOCALE_NUMERIC USE_LOCALE_TIME USE_LONG_DOUBLE
USE_PERLIO USE_PERL_ATOF
Built under linux
Compiled at Sep 30 2015 16:39:59
@INC:
/usr/local/perl-afl/lib/site_perl/5.23.4/x86_64-linux-ld
/usr/local/perl-afl/lib/site_perl/5.23.4
/usr/local/perl-afl/lib/5.23.4/x86_64-linux-ld
/usr/local/perl-afl/lib/5.23.4
.

p5pRT · 2016-03-22T17:44:14Z

@khwilliamson - Status changed from 'new' to 'open'

p5pRT · 2016-04-02T10:01:45Z

From @demerphq

\N{} was creating an EMPTY node, which was tickling a secondary issue with trie construction from alternations with EXACTFU_SS nodes in them.

commit 944e05e
Author: Yves Orton <demerphq@gmail.com>
Date: Wed Mar 9 20:00:53 2016 +0100

Fix Perl #126206: handle NOTHING regops and EXACTFU_SS regops in make_trie() properly

... and avoid dereffing non-EXACT nodes unnecessarily at the same
time. This fixes https://rt-archive.perl.org/perl5/Ticket/Display.html?id=126206

p5pRT · 2016-04-02T10:01:48Z

@demerphq - Status changed from 'open' to 'resolved'

p5pRT closed this Apr 2, 2016

p5pRT added the Severity Low label Oct 19, 2019

Perl / perl5

SIGFPE integer divide by zero in S_make_trie #14945

SIGFPE integer divide by zero in S_make_trie #14945

p5pRT commented Sep 27, 2015

p5pRT commented Sep 27, 2015

p5pRT commented Sep 27, 2015

p5pRT commented Oct 5, 2015

p5pRT commented Mar 22, 2016

p5pRT commented Apr 2, 2016

p5pRT commented Apr 2, 2016

Perl / perl5

SIGFPE integer divide by zero in S_make_trie #14945

SIGFPE integer divide by zero in S_make_trie #14945

Comments

p5pRT commented Sep 27, 2015

p5pRT commented Sep 27, 2015

From @dcollinsn

p5pRT commented Sep 27, 2015

From @dcollinsn

p5pRT commented Oct 5, 2015

From @dcollinsn

p5pRT commented Mar 22, 2016

p5pRT commented Apr 2, 2016

From @demerphq

p5pRT commented Apr 2, 2016