On Thu, Jul 10, 2014 at 2:36 PM, Ben Pfaff <b...@nicira.com> wrote:
> On Thu, Jul 10, 2014 at 02:01:27PM -0700, Gurucharan Shetty wrote:
>> On Wed, Jul 9, 2014 at 3:45 PM, Ben Pfaff <b...@nicira.com> wrote:
>> > We noticed that the unit tests sometimes fail on XenServer inside glibc's
>> > memory checker, in the free_check() function.  It turns out that the
>> > glibc memory checker in glibc 2.11 and earlier had an internal race that
>> > caused false positives in multithreaded programs.
>> >
>> > This commit avoids the problem by disabling the glibc memory checker in
>> > glibc 2.11 and earlier.
>> >
>> > VMware-BZ: #1267127
>> > Signed-off-by: Ben Pfaff <b...@nicira.com>
>> > ---
>> >  tests/atlocal.in |   22 ++++++++++++++++++----
>> >  1 file changed, 18 insertions(+), 4 deletions(-)
>> >
>> > diff --git a/tests/atlocal.in b/tests/atlocal.in
>> > index 2f816fc..825d289 100644
>> > --- a/tests/atlocal.in
>> > +++ b/tests/atlocal.in
>> > @@ -52,10 +52,24 @@ fi
>> >  # Enable malloc debugging features.
>> >  case `uname` in
>> >  Linux)
>> > -    MALLOC_CHECK_=2
>> > -    MALLOC_PERTURB_=165
>> > -    export MALLOC_CHECK_
>> > -    export MALLOC_PERTURB_
>> > +    MALLOC_PERTURB_=165; export MALLOC_PERTURB_
>> > +
>> > +    # Before glibc 2.11, the feature enabled by MALLOC_CHECK_ was not
>> > +    # thread-safe.  See 
>> > https://bugzilla.redhat.com/show_bug.cgi?id=585674 and
>> > +    # in particular the patch attached there, which was applied to glibc 
>> > CVS as
>> > +    # "Restore locking in free_check." between 1.11 and 1.11.1.
>> > +    vswitchd=$abs_top_builddir/vswitchd/ovs-vswitchd
>> > +    glibc=`ldd $vswitchd | sed -n 's/^ libc\.[^ ]* => \([^ ]*\) .*/\1/p'`
>> > +    glibc_version=`$glibc | sed -n '1s/.*version 
>> > \([0-9]\{1,\}\.[0-9]\{1,\}\).*/\1/p'`
>> > +    case $glibc_version in
>> > +        2.[0-9] | 2.1[01]) mcheck=disabled ;;
>> Did you mean 2.0[0-9] instead of 2.[0-9]?
>
> I don't think so.
>
> I'm trying to match versions 2.0 through 2.11 and set mcheck=disabled
> in that case.  Did I get it wrong?
No. I made a bad assumption that the version numbers are of the form
2.0, 2.01 .. 2.11
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev

Reply via email to