Bus error with (unshared) dir handles and threads

[p5pRT]

Migrated from rt.perl.org#77934 (status was 'resolved')

Searchable as RT77934$

[p5pRT]

From @cpansprout

I was wondering whether readdir is thread-safe, so I wrote this script​:

#!perl -l

chdir '/tmp';
mkdir "toberead$$";
chdir "toberead$$";
open thing, ">$_" and close thing or die "$_​: $!" for a..z;

use threads;
$thr = async {
opendir dir, ".";
$start = telldir dir;
readdir dir or seekdir dir, 0 while 1;
} if $ARGV[0];

opendir dir, ".";
for(1..26) {
print scalar readdir dir;
select undef, undef, undef, .25;
}

$thr->kill(KILL) if $ARGV[0];
__END__

Pass an argument to it to start the other thread.

What surprised me was that the output is all correct (. .. a b c ... v w x), but then there is a bus error.

My patch for 75154 (cloning dir handles) doesn’t fix it, so this is a separate bug.

---

Flags​:
  category=core
  severity=low

---

Site configuration information for perl 5.13.4​:

Configured by sprout at Mon Sep 13 22​:18​:01 PDT 2010.

Summary of my perl5 (revision 5 version 13 subversion 4 patch v5.13.4-224-g25222ff) configuration​:
  Snapshot of​: 25222ff
  Platform​:
  osname=darwin, osvers=10.4.0, archname=darwin-2level
  uname='darwin pint.local 10.4.0 darwin kernel version 10.4.0​: fri apr 23 18​:28​:53 pdt 2010; root​:xnu-1504.7.4~1release_i386 i386 '
  config_args='-de -Dusedevel -DDEBUGGING'
  hint=recommended, useposix=true, d_sigaction=define
  useithreads=undef, usemultiplicity=undef
  useperlio=define, d_sfio=undef, uselargefiles=define, usesocks=undef
  use64bitint=undef, use64bitall=undef, uselongdouble=undef
  usemymalloc=n, bincompat5005=undef
  Compiler​:
  cc='cc', ccflags ='-fno-common -DPERL_DARWIN -no-cpp-precomp -DDEBUGGING -fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include',
  optimize='-O3 -g',
  cppflags='-no-cpp-precomp -fno-common -DPERL_DARWIN -no-cpp-precomp -DDEBUGGING -fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include'
  ccversion='', gccversion='4.2.1 (Apple Inc. build 5664)', gccosandvers=''
  intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=1234
  d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16
  ivtype='long', ivsize=4, nvtype='double', nvsize=8, Off_t='off_t', lseeksize=8
  alignbytes=8, prototype=define
  Linker and Libraries​:
  ld='env MACOSX_DEPLOYMENT_TARGET=10.3 cc', ldflags =' -fstack-protector -L/usr/local/lib'
  libpth=/usr/local/lib /usr/lib
  libs=-ldbm -ldl -lm -lutil -lc
  perllibs=-ldl -lm -lutil -lc
  libc=/usr/lib/libc.dylib, so=dylib, useshrplib=false, libperl=libperl.a
  gnulibc_version=''
  Dynamic Linking​:
  dlsrc=dl_dlopen.xs, dlext=bundle, d_dlsymun=undef, ccdlflags=' '
  cccdlflags=' ', lddlflags=' -bundle -undefined dynamic_lookup -L/usr/local/lib -fstack-protector'

Locally applied patches​:
 

---

@​INC for perl 5.13.4​:
  /usr/local/lib/perl5/site_perl/5.13.4/darwin-2level
  /usr/local/lib/perl5/site_perl/5.13.4
  /usr/local/lib/perl5/5.13.4/darwin-2level
  /usr/local/lib/perl5/5.13.4
  /usr/local/lib/perl5/site_perl
  .

---

Environment for perl 5.13.4​:
  DYLD_LIBRARY_PATH (unset)
  HOME=/Users/sprout
  LANG=en_US.UTF-8
  LANGUAGE (unset)
  LD_LIBRARY_PATH (unset)
  LOGDIR (unset)
  PATH=/usr/bin​:/bin​:/usr/sbin​:/sbin​:/usr/local/bin​:/usr/X11/bin​:/usr/local/bin
  PERL_BADLANG (unset)
  SHELL=/bin/bash

[p5pRT]

From @jkeenan

On Mon, 20 Sep 2010 01​:23​:36 GMT, sprout wrote​:

I was wondering whether readdir is thread-safe, so I wrote this
script​:

#!perl -l

chdir '/tmp';
mkdir "toberead$$";
chdir "toberead$$";
open thing, ">$_" and close thing or die "$_​: $!" for a..z;

use threads;
$thr = async {
opendir dir, ".";
$start = telldir dir;
readdir dir or seekdir dir, 0 while 1;
} if $ARGV[0];

opendir dir, ".";
for(1..26) {
print scalar readdir dir;
select undef, undef, undef, .25;
}

$thr->kill(KILL) if $ARGV[0];
__END__

Pass an argument to it to start the other thread.

What surprised me was that the output is all correct (. .. a b c ... v
w x), but then there is a bus error.

My patch for 75154 (cloning dir handles) doesn’t fix it, so this is a
separate bug.

What I got running the program today on a threaded perl near HEAD on blead​:

#####
$ ./bin/perl -Ilib ~/learn/perl/p5p/77934-threads.pl 7
m
c
u
k
g
e
x
v
j
a
o
w
f
q
i
y
b
r
l
d
..
z
h
t
n
p
Signal KILL received in thread 1, but no signal handler set. at /home/jkeenan/learn/perl/p5p/77934-threads.pl line 21.
Perl exited with active threads​:
  1 running and unjoined
  0 finished and unjoined
  0 running and detached
#####

--
James E Keenan (jkeenan@​cpan.org)

[p5pRT]

From @jkeenan

Summary of my perl5 (revision 5 version 25 subversion 9) configuration​:
  Commit id​: 9065925
  Platform​:
  osname=linux
  osvers=4.4.0-57-generic
  archname=x86_64-linux-thread-multi
  uname='linux zareason 4.4.0-57-generic #78-ubuntu smp fri dec 9 23​:50​:32 utc 2016 x86_64 x86_64 x86_64 gnulinux '
  config_args='-des -Dusedevel -Dusethreads -Uversiononly -Dprefix=/home/jkeenan/testing/blead -Dman1dir=none -Dman3dir=none'
  hint=recommended
  useposix=true
  d_sigaction=define
  useithreads=define
  usemultiplicity=define
  use64bitint=define
  use64bitall=define
  uselongdouble=undef
  usemymalloc=n
  bincompat5005=undef
  Compiler​:
  cc='cc'
  ccflags ='-D_REENTRANT -D_GNU_SOURCE -fwrapv -fno-strict-aliasing -pipe -fstack-protector-strong -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64'
  optimize='-O2'
  cppflags='-D_REENTRANT -D_GNU_SOURCE -fwrapv -fno-strict-aliasing -pipe -fstack-protector-strong -I/usr/local/include'
  ccversion=''
  gccversion='5.4.0 20160609'
  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='double'
  nvsize=8
  Off_t='off_t'
  lseeksize=8
  alignbytes=8
  prototype=define
  Linker and Libraries​:
  ld='cc'
  ldflags =' -fstack-protector-strong -L/usr/local/lib'
  libpth=/usr/local/lib /usr/lib/gcc/x86_64-linux-gnu/5/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 /lib64 /usr/lib64
  libs=-lpthread -lnsl -ldb -ldl -lm -lcrypt -lutil -lc
  perllibs=-lpthread -lnsl -ldl -lm -lcrypt -lutil -lc
  libc=libc-2.23.so
  so=so
  useshrplib=false
  libperl=libperl.a
  gnulibc_version='2.23'
  Dynamic Linking​:
  dlsrc=dl_dlopen.xs
  dlext=so
  d_dlsymun=undef
  ccdlflags='-Wl,-E'
  cccdlflags='-fPIC'
  lddlflags='-shared -O2 -L/usr/local/lib -fstack-protector-strong'

Characteristics of this binary (from libperl)​:
  Compile-time options​:
  HAS_TIMES
  MULTIPLICITY
  PERLIO_LAYERS
  PERL_COPY_ON_WRITE
  PERL_DONT_CREATE_GVSV
  PERL_IMPLICIT_CONTEXT
  PERL_MALLOC_WRAP
  PERL_OP_PARENT
  PERL_PRESERVE_IVUV
  PERL_USE_DEVEL
  USE_64_BIT_ALL
  USE_64_BIT_INT
  USE_ITHREADS
  USE_LARGE_FILES
  USE_LOCALE
  USE_LOCALE_COLLATE
  USE_LOCALE_CTYPE
  USE_LOCALE_NUMERIC
  USE_LOCALE_TIME
  USE_PERLIO
  USE_PERL_ATOF
  USE_REENTRANT_API
  Built under linux
  Compiled at Jan 6 2017 21​:04​:04
  %ENV​:
  PERLBREW_BASHRC_VERSION="0.67"
  PERLBREW_HOME="/home/jkeenan/.perlbrew"
  PERLBREW_MANPATH="/home/jkeenan/perl5/perlbrew/perls/perl-5.24.0/man"
  PERLBREW_PATH="/home/jkeenan/perl5/perlbrew/bin​:/home/jkeenan/perl5/perlbrew/perls/perl-5.24.0/bin"
  PERLBREW_PERL="perl-5.24.0"
  PERLBREW_ROOT="/home/jkeenan/perl5/perlbrew"
  PERLBREW_VERSION="0.67"
  PERL_WORKDIR="/home/jkeenan/gitwork/perl"
  @​INC​:
  lib
  /home/jkeenan/testing/blead/lib/perl5/site_perl/5.25.9/x86_64-linux-thread-multi
  /home/jkeenan/testing/blead/lib/perl5/site_perl/5.25.9
  /home/jkeenan/testing/blead/lib/perl5/5.25.9/x86_64-linux-thread-multi
  /home/jkeenan/testing/blead/lib/perl5/5.25.9
  .

[p5pRT]

The RT System itself - Status changed from 'new' to 'open'

[p5pRT]

From @Leont

On Sat, Jan 7, 2017 at 4​:32 PM, James E Keenan via RT <
perlbug-followup@​perl.org> wrote​:

On Mon, 20 Sep 2010 01​:23​:36 GMT, sprout wrote​:

I was wondering whether readdir is thread-safe, so I wrote this
script​:

#!perl -l

chdir '/tmp';
mkdir "toberead$$";
chdir "toberead$$";
open thing, ">$_" and close thing or die "$_​: $!" for a..z;

use threads;
$thr = async {
opendir dir, ".";
$start = telldir dir;
readdir dir or seekdir dir, 0 while 1;
} if $ARGV[0];

opendir dir, ".";
for(1..26) {
print scalar readdir dir;
select undef, undef, undef, .25;
}

$thr->kill(KILL) if $ARGV[0];
__END__

Pass an argument to it to start the other thread.

What surprised me was that the output is all correct (. .. a b c ... v
w x), but then there is a bus error.

My patch for 75154 (cloning dir handles) doesn’t fix it, so this is a
separate bug.

What I got running the program today on a threaded perl near HEAD on blead​:

#####
$ ./bin/perl -Ilib ~/learn/perl/p5p/77934-threads.pl 7
m
c
u
k
g
e
x
v
j
a
o
w
f
q
i
y
b
r
l
d
..
z
h
t
n
p
Signal KILL received in thread 1, but no signal handler set. at
/home/jkeenan/learn/perl/p5p/77934-threads.pl line 21.
Perl exited with active threads​:
1 running and unjoined
0 finished and unjoined
0 running and detached
#####

Also tried it, not seeing anything suggesting that the bug is still present.

Leon

[p5pRT]

From @jkeenan

On Sun, 08 Jan 2017 12​:53​:28 GMT, LeonT wrote​:

On Sat, Jan 7, 2017 at 4​:32 PM, James E Keenan via RT <
perlbug-followup@​perl.org> wrote​:

On Mon, 20 Sep 2010 01​:23​:36 GMT, sprout wrote​:

I was wondering whether readdir is thread-safe, so I wrote this
script​:

#!perl -l

chdir '/tmp';
mkdir "toberead$$";
chdir "toberead$$";
open thing, ">$_" and close thing or die "$_​: $!" for a..z;

use threads;
$thr = async {
opendir dir, ".";
$start = telldir dir;
readdir dir or seekdir dir, 0 while 1;
} if $ARGV[0];

opendir dir, ".";
for(1..26) {
print scalar readdir dir;
select undef, undef, undef, .25;
}

$thr->kill(KILL) if $ARGV[0];
__END__

Pass an argument to it to start the other thread.

What surprised me was that the output is all correct (. .. a b c
... v
w x), but then there is a bus error.

My patch for 75154 (cloning dir handles) doesn’t fix it, so this is
a
separate bug.

What I got running the program today on a threaded perl near HEAD on
blead​:

#####
$ ./bin/perl -Ilib ~/learn/perl/p5p/77934-threads.pl 7
m
c
u
k
g
e
x
v
j
a
o
w
f
q
i
y
b
r
l
d
..
z
h
t
n
p
Signal KILL received in thread 1, but no signal handler set. at
/home/jkeenan/learn/perl/p5p/77934-threads.pl line 21.
Perl exited with active threads​:
1 running and unjoined
0 finished and unjoined
0 running and detached
#####

Also tried it, not seeing anything suggesting that the bug is still
present.

Leon

I built threaded perls at v5.10.1 and v5.14.4 and got the same results. I'm taking this ticket for the purpose of closing it within 7 days unless Father C wants to follow up.

Thank you very much.

--
James E Keenan (jkeenan@​cpan.org)

[p5pRT]

From @cpansprout

On Jan 8, 2017, at 9​:48 AM, "James E Keenan via RT" <perlbug-followup@​perl.org> wrote​:

On Sun, 08 Jan 2017 12​:53​:28 GMT, LeonT wrote​:

On Sat, Jan 7, 2017 at 4​:32 PM, James E Keenan via RT <
perlbug-followup@​perl.org> wrote​:

On Mon, 20 Sep 2010 01​:23​:36 GMT, sprout wrote​:

I was wondering whether readdir is thread-safe, so I wrote this
script​:

#!perl -l

chdir '/tmp';
mkdir "toberead$$";
chdir "toberead$$";
open thing, ">$_" and close thing or die "$_​: $!" for a..z;

use threads;
$thr = async {
opendir dir, ".";
$start = telldir dir;
readdir dir or seekdir dir, 0 while 1;
} if $ARGV[0];

opendir dir, ".";
for(1..26) {
print scalar readdir dir;
select undef, undef, undef, .25;
}

$thr->kill(KILL) if $ARGV[0];
__END__

Pass an argument to it to start the other thread.

What surprised me was that the output is all correct (. .. a b c
... v
w x), but then there is a bus error.

My patch for 75154 (cloning dir handles) doesn’t fix it, so this is
a
separate bug.

What I got running the program today on a threaded perl near HEAD on
blead​:

#####
$ ./bin/perl -Ilib ~/learn/perl/p5p/77934-threads.pl 7
m
c
u
k
g
e
x
v
j
a
o
w
f
q
i
y
b
r
l
d
..
z
h
t
n
p
Signal KILL received in thread 1, but no signal handler set. at
/home/jkeenan/learn/perl/p5p/77934-threads.pl line 21.
Perl exited with active threads​:
1 running and unjoined
0 finished and unjoined
0 running and detached
#####

Also tried it, not seeing anything suggesting that the bug is still
present.

Leon

I built threaded perls at v5.10.1 and v5.14.4 and got the same results. I'm taking this ticket for the purpose of closing it within 7 days unless Father C wants to follow up.

Does anyone have time to do a bisect, to see when it was fixed? If it was fixed inadvertently, it might be worth adding a test.

[p5pRT]

From @jkeenan

On Sun, 08 Jan 2017 18​:54​:03 GMT, sprout wrote​:

On Jan 8, 2017, at 9​:48 AM, "James E Keenan via RT" <perlbug-
followup@​perl.org> wrote​:

On Sun, 08 Jan 2017 12​:53​:28 GMT, LeonT wrote​:

On Sat, Jan 7, 2017 at 4​:32 PM, James E Keenan via RT <
perlbug-followup@​perl.org> wrote​:

On Mon, 20 Sep 2010 01​:23​:36 GMT, sprout wrote​:

I was wondering whether readdir is thread-safe, so I wrote this
script​:

#!perl -l

chdir '/tmp';
mkdir "toberead$$";
chdir "toberead$$";
open thing, ">$_" and close thing or die "$_​: $!" for a..z;

use threads;
$thr = async {
opendir dir, ".";
$start = telldir dir;
readdir dir or seekdir dir, 0 while 1;
} if $ARGV[0];

opendir dir, ".";
for(1..26) {
print scalar readdir dir;
select undef, undef, undef, .25;
}

$thr->kill(KILL) if $ARGV[0];
__END__

Pass an argument to it to start the other thread.

What surprised me was that the output is all correct (. .. a b c
... v
w x), but then there is a bus error.

My patch for 75154 (cloning dir handles) doesn’t fix it, so this
is
a
separate bug.

What I got running the program today on a threaded perl near HEAD
on
blead​:

#####
$ ./bin/perl -Ilib ~/learn/perl/p5p/77934-threads.pl 7
m
c
u
k
g
e
x
v
j
a
o
w
f
q
i
y
b
r
l
d
..
z
h
t
n
p
Signal KILL received in thread 1, but no signal handler set. at
/home/jkeenan/learn/perl/p5p/77934-threads.pl line 21.
Perl exited with active threads​:
1 running and unjoined
0 finished and unjoined
0 running and detached
#####

Also tried it, not seeing anything suggesting that the bug is still
present.

Leon

I built threaded perls at v5.10.1 and v5.14.4 and got the same
results. I'm taking this ticket for the purpose of closing it within
7 days unless Father C wants to follow up.

Does anyone have time to do a bisect, to see when it was fixed? If it
was fixed inadvertently, it might be worth adding a test.

It's not clear what the problem was in the first place. You stated that there was a "bus error" but didn't present any output. Moreover, the 'perl -V' in your original post was from a non-threaded perl. But this problem deals with threads. So it's not clear what the bisection test would be -- at least not to me.

In addition, the program now dies due to a signal rather than, presumably, crashing with a bus error. I'm not sure how we'd invoke the bisect runner to distinguish between those two cases, as both would likely be covered by this​:

#####
  # When did this test program start crashing (any signal or coredump)?
  .../Porting/bisect.pl --crash -- ./perl -Ilib ../test_prog.pl
#####

Suggestions?

Thank you very much.
--
James E Keenan (jkeenan@​cpan.org)

[p5pRT]

From @Leont

On Sun, Jan 8, 2017 at 8​:37 PM, James E Keenan via RT <
perlbug-followup@​perl.org> wrote​:

In addition, the program now dies due to a signal rather than, presumably,
crashing with a bus error.

That is correct behavior that can be resolved by adding a handler for that
signal, it's needed anyway to quit the while loop.

Leon

[p5pRT]

From @dur-randir

On Sun, 08 Jan 2017 11​:37​:23 -0800, jkeenan wrote​:

It's not clear what the problem was in the first place.

In addition, the program now dies due to a signal rather than,
presumably, crashing with a bus error. I'm not sure how we'd invoke
the bisect runner to distinguish between those two cases, as both
would likely be covered by this​:

I'm able to reproduce segfault with the original script on 5.12.5, but not on 5.14.4. Unfortunately, I can't bisect 5.14 branch on this machine, so just those rough estimates.

[p5pRT]

From @hvds

On Sun, 08 Jan 2017 11​:37​:23 -0800, jkeenan wrote​:

In addition, the program now dies due to a signal rather than,
presumably, crashing with a bus error. I'm not sure how we'd invoke
the bisect runner to distinguish between those two cases, as both
would likely be covered by this​:

#####
# When did this test program start crashing (any signal or coredump)?
.../Porting/bisect.pl --crash -- ./perl -Ilib ../test_prog.pl
#####

Suggestions?

I'd suggest locally hacking bisect-runner.pl - look at run_report_and_exit() where it uses $options{crash} to modify behaviour, then hack it once to show what $ret is in the signal case, then a second time to set $ret to zero if you see that value (to treat that as a passing case).

Hugo

[p5pRT]

From @jkeenan

On Sun, 08 Jan 2017 20​:06​:42 GMT, LeonT wrote​:

On Sun, Jan 8, 2017 at 8​:37 PM, James E Keenan via RT <
perlbug-followup@​perl.org> wrote​:

In addition, the program now dies due to a signal rather than, presumably,
crashing with a bus error.

That is correct behavior that can be resolved by adding a handler for that
signal, it's needed anyway to quit the while loop.

Leon

With valuable assistance and code review from Jerry Hedden, I have pushed an additional test file under dist/threads that will capture any future regressions in this area.

I adapted Father C's original code sample to run both cases -- without and with command-line argument -- in one program. Running that on threaded perls built at perl-5.12.5 and perl-5.14.4 re-confirmed the range; running with a command-line argument at 5.12.5 caused the program to hang indefinitely, but running with an argument at 5.14.4 threw the previously reported exception. With Jerry's help I added a die handler, then I converted this into an additional test file. Jerry reviewed this and I also got several PASSes on a smoke-me branch.

Pushed to blead in commit afa4768. Marking ticket Resolved.

Thank you very much.

--
James E Keenan (jkeenan@​cpan.org)

[p5pRT]

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

[p5pRT]

From @cpansprout

On Jan 13, 2017, at 10​:22 AM, "James E Keenan via RT" <perlbug-followup@​perl.org> wrote​:

On Sun, 08 Jan 2017 20​:06​:42 GMT, LeonT wrote​:

On Sun, Jan 8, 2017 at 8​:37 PM, James E Keenan via RT <
perlbug-followup@​perl.org> wrote​:

In addition, the program now dies due to a signal rather than, presumably,
crashing with a bus error.

That is correct behavior that can be resolved by adding a handler for that
signal, it's needed anyway to quit the while loop.

Leon

With valuable assistance and code review from Jerry Hedden, I have pushed an additional test file under dist/threads that will capture any future regressions in this area.

I adapted Father C's original code sample to run both cases -- without and with command-line argument -- in one program. Running that on threaded perls built at perl-5.12.5 and perl-5.14.4 re-confirmed the range; running with a command-line argument at 5.12.5 caused the program to hang indefinitely, but running with an argument at 5.14.4 threw the previously reported exception. With Jerry's help I added a die handler, then I converted this into an additional test file. Jerry reviewed this and I also got several PASSes on a smoke-me branch.

Pushed to blead in commit afa4768. Marking ticket Resolved.

Thank you very much.

Thank you for your work in resolving this.