[Bug 1107543] New: Perl core-dumps if a hash is tied to SDBM_File before spawning a thread
by Red Hat Bugzilla
https://bugzilla.redhat.com/show_bug.cgi?id=1107543
Bug ID: 1107543
Summary: Perl core-dumps if a hash is tied to SDBM_File before
spawning a thread
Product: Fedora
Version: 19
Component: perl
Severity: medium
Assignee: jplesnik(a)redhat.com
Reporter: ppisar(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: cweyl(a)alumni.drew.edu, iarnell(a)gmail.com,
jplesnik(a)redhat.com, kasal(a)ucw.cz,
perl-devel(a)lists.fedoraproject.org, ppisar(a)redhat.com,
psabata(a)redhat.com, rc040203(a)freenet.de,
tbowling(a)redhat.com, tcallawa(a)redhat.com
+++ This bug was initially created as a clone of Bug #1107542 +++
+++ This bug was initially created as a clone of Bug #1104827 +++
Description of problem:
Perl script using SDBM_File module is core dumping. Seems to match this
upstream bug: https://rt.perl.org/Public/Bug/Display.html?id=61912#txn-515026
[...]
Steps to Reproduce:
Create reproducer test script sdbm_test.pl containing the following lines, as
described in the upstream bug report:
#!/usr/bin/perl
use strict;
use Fcntl;
use SDBM_File;
use threads;
use threads::shared;
my %dbtest;
tie(%dbtest, 'SDBM_File', "test.db", O_RDWR|O_CREAT, 0666);
for (1 .. 2)
{
my $thr = threads->new(\&testThread, $_);
$thr->detach();
}
sleep 4;
sub testThread
{
my $n = shift;
print "thread #" . $n . " started\n";
}
Make script executable and run which produces the following output:
[root@util6vm ~]# chmod u+x sdbm_test.pl
[root@util6vm ~]# ./sdbm_test.pl
Expected results:
No errors.
Actual results:
thread #1 started
thread #2 started
*** glibc detected *** /usr/bin/perl: double free or corruption (out):
0x0000000000e2c2c0 ***
======= Backtrace: =========
/lib64/libc.so.6[0x3d2ca76166]
/lib64/libc.so.6[0x3d2ca78c93]
/usr/lib64/perl5/auto/SDBM_File/SDBM_File.so(XS_SDBM_File_DESTROY+0xc0)[0x7f9d58fb06f0]
[...]
--- Additional comment from Petr Pisar on 2014-06-05 05:56:58 GMT ---
This happens with any perl, even the development version.
----
All Fedoras are affected.
--
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=sO59ARuC6M&a=cc_unsubscribe
8 years, 5 months
[Bug 783346] New: Perl-MIME-Lite is already in RHEL6
by Red Hat Bugzilla
Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug.
Summary: Perl-MIME-Lite is already in RHEL6
https://bugzilla.redhat.com/show_bug.cgi?id=783346
Summary: Perl-MIME-Lite is already in RHEL6
Product: Fedora EPEL
Version: el6
Platform: Unspecified
OS/Version: Unspecified
Status: NEW
Severity: unspecified
Priority: unspecified
Component: perl-MIME-Lite
AssignedTo: steve.traylen(a)cern.ch
ReportedBy: bochecha(a)fedoraproject.org
QAContact: extras-qa(a)fedoraproject.org
CC: fedora-perl-devel-list(a)redhat.com,
mmcgrath(a)redhat.com, steve.traylen(a)cern.ch
Classification: Fedora
Story Points: ---
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
Description of problem:
This package is already in RHEL6, as such, shouldn't it be dropped from EPEL6?
# yum whatprovides 'perl(MIME::Lite)'
Loaded plugins: changelog, merge-conf, rhnplugin
perl-MIME-Lite-3.027-2.el6.noarch : MIME::Lite - low-calorie MIME generator
Repo : epel
Matched from:
Other : perl(MIME::Lite)
perl-MIME-Lite-3.027-2.el6.noarch : MIME::Lite - low-calorie MIME generator
Repo : rhel-x86_64-server-optional-6
Matched from:
Other : perl(MIME::Lite)
--
Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
8 years, 5 months
[Bug 1062576] New: memory leak when including a file with "use utf8"
by Red Hat Bugzilla
https://bugzilla.redhat.com/show_bug.cgi?id=1062576
Bug ID: 1062576
Summary: memory leak when including a file with "use utf8"
Product: Fedora
Version: 19
Component: perl
Severity: medium
Assignee: jplesnik(a)redhat.com
Reporter: lav(a)yars.free.net
QA Contact: extras-qa(a)fedoraproject.org
CC: cweyl(a)alumni.drew.edu, iarnell(a)gmail.com,
jplesnik(a)redhat.com, kasal(a)ucw.cz,
perl-devel(a)lists.fedoraproject.org, ppisar(a)redhat.com,
psabata(a)redhat.com, rc040203(a)freenet.de,
tcallawa(a)redhat.com
Description of problem:
Every time perl includes a file with "use utf8" and text constants it leaks
memory.
Version-Release number of selected component (if applicable):
perl-5.16.3-266.fc19.x86_64
How reproducible:
always
Steps to Reproduce:
1. run the test program below
2. watch as memory usage grows
Here is the test program:
-------------------------
my $mem = 0;
sub report_memory () {
my $next_mem = qx[ps -p $$ -o size=]+0;
printf "%+8d = %8d K\n", $next_mem-$mem, $next_mem; $mem = $next_mem;
}
for (0..1e5) {
do 'x.inc';
report_memory unless $_ % 1e4;
}
-------------------------
x.inc contains:
-------------------------
use utf8;
$x='x';
-------------------------
--
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=l7ghbj1UQE&a=cc_unsubscribe
8 years, 6 months
[Bug 1129850] New: Perl debugger commands that depend on Padwalker no longer work
by Red Hat Bugzilla
https://bugzilla.redhat.com/show_bug.cgi?id=1129850
Bug ID: 1129850
Summary: Perl debugger commands that depend on Padwalker no
longer work
Product: Fedora
Version: 20
Component: perl
Severity: high
Assignee: jplesnik(a)redhat.com
Reporter: alan(a)clueserver.org
QA Contact: extras-qa(a)fedoraproject.org
CC: cweyl(a)alumni.drew.edu, iarnell(a)gmail.com,
jplesnik(a)redhat.com, kasal(a)ucw.cz,
perl-devel(a)lists.fedoraproject.org, ppisar(a)redhat.com,
psabata(a)redhat.com, rc040203(a)freenet.de,
tcallawa(a)redhat.com
Description of problem:
When using the Perl command line debugger the "y" and "X" commands no longer
work. (They fail silently.)
Version-Release number of selected component (if applicable):
perl-5.18.2-289.fc20
perl-PadWalker-1.98-1.fc20.x86_64
How reproducible:
Every time.
Steps to Reproduce:
1. Debug a test program with "perl -d testprog.pl
2. Use "n" and enter a few lines until a variable you want to examine is
loaded.
3. Use "y" or "X" on that variable.
Actual results:
Nothing returned.
Expected results:
The contents of the variable should be displayed.
Additional info:
The "p" command will print a variable, it is very narrow in scope. It will not
display the contents of an entire hash variable like "X" or "y" will. ("y"
shows the contents of the variable for all types in the current package. "X"
does the same for global variables. Neither work at this point.
--
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=VpjaTqvTg7&a=cc_unsubscribe
8 years, 6 months
[Bug 839612] New: FTBFS perl-Unix-Statgrab-0.04-14.fc18
by Red Hat Bugzilla
https://bugzilla.redhat.com/show_bug.cgi?id=839612
Bug ID: 839612
QA Contact: extras-qa(a)fedoraproject.org
Severity: unspecified
URL: https://koji.fedoraproject.org/koji/taskinfo?taskID=42
36283
Version: rawhide
Priority: unspecified
CC: oliver(a)linux-kernel.at,
perl-devel(a)lists.fedoraproject.org, steve(a)silug.org
Assignee: steve(a)silug.org
Summary: FTBFS perl-Unix-Statgrab-0.04-14.fc18
Regression: ---
Story Points: ---
Classification: Fedora
OS: Unspecified
Reporter: ppisar(a)redhat.com
Type: Bug
Documentation: ---
Hardware: Unspecified
Mount Type: ---
Status: NEW
Component: perl-Unix-Statgrab
Product: Fedora
perl-Unix-Statgrab-0.04-14.fc18 does not build in F18:
Executing(%check): /bin/sh -e /var/tmp/rpm-tmp.FI62R4
+ umask 022
+ cd /builddir/build/BUILD
+ cd Unix-Statgrab-0.04
+ unset DISPLAY
+ make test
PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0,
'blib/lib', 'blib/arch')" t/*.t
t/0_pod.t ........... ok
t/0_pod_coverage.t .. ok
t/Unix-Statgrab.t ...
Failed 1/22 subtests
Test Summary Report
-------------------
t/Unix-Statgrab.t (Wstat: 11 Tests: 21 Failed: 0)
Non-zero wait status: 11
Parse errors: Bad plan. You planned 22 tests but ran 21.
Files=3, Tests=23, 0 wallclock secs ( 0.04 usr 0.00 sys + 0.15 cusr 0.03
csys = 0.22 CPU)
Result: FAIL
Failed 1/3 test programs. 0/23 subtests failed.
make: *** [test_dynamic] Error 255
This happens with perl 5.16.0. I guess the test plan is wrong or the perl
segfaults.
--
You are receiving this mail because:
You are on the CC list for the bug.
8 years, 7 months
[Bug 750805] New: Fails to build on ARM, needs to use default setjmp not ucontext
by Red Hat Bugzilla
Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug.
Summary: Fails to build on ARM, needs to use default setjmp not ucontext
https://bugzilla.redhat.com/show_bug.cgi?id=750805
Summary: Fails to build on ARM, needs to use default setjmp not
ucontext
Product: Fedora
Version: 15
Platform: Unspecified
OS/Version: Unspecified
Status: NEW
Severity: unspecified
Priority: unspecified
Component: perl-Coro
AssignedTo: ppisar(a)redhat.com
ReportedBy: henrik(a)henriknordstrom.net
QAContact: extras-qa(a)fedoraproject.org
CC: fedora-perl-devel-list(a)redhat.com,
mmaslano(a)redhat.com, kwizart(a)gmail.com,
bochecha(a)fedoraproject.org, ppisar(a)redhat.com
Classification: Fedora
Story Points: ---
Type: ---
Description of problem:
ARM do not implement the needed ucontext functions, and need to use the default
setjmp method. This is normally the default, except that fedora patches it to
hardwire ucontext as default method..
Version-Release number of selected component (if applicable):
perl-Coro-5.372-3.fc15
How reproducible:
always
Steps to Reproduce:
1. Try to rebuild perl-Coro on arm
2.
3.
Actual results:
failed build, crashing in testsuite
Expected results:
successful build
Additional info:
Trivial spec file patch attached.
Please apply patch and sumbit a F15 koji build from which arm can pull the
srpm. update request is not stricty needed if the only change relative to
current F15 build is this patch.
--
Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
8 years, 9 months