[SSSD] [PATCH] CI: Suppress memory errors from poptGetNextOpt

Lukas Slebodnik lslebodn at redhat.com
Thu Dec 4 23:40:00 UTC 2014


On (04/12/14 22:53), Jakub Hrozek wrote:
>Hi,
>
>Using this patch and the one that Lukas has pending on the list, the only
>error in CI is the RHEL-6 build failure:
>    http://sssd-ci.duckdns.org/logs/job/4/70/summary.html
>which is tracked by:
>    https://fedorahosted.org/sssd/ticket/2518
>
>I'm not thrilled by the use of ellipsis, but it was the only notation I
>could come up with that would work on both Debian and all RHELs.

>From 96913027b7e2babab659ae649d12a18f48a6ed14 Mon Sep 17 00:00:00 2001
>From: Jakub Hrozek <jhrozek at redhat.com>
>Date: Thu, 4 Dec 2014 21:12:45 +0100
>Subject: [PATCH] CI: Suppress memory errors from poptGetNextOpt
>
>---
> contrib/ci/sssd.supp | 31 +++++++++++++++++++++++++++++++
> 1 file changed, 31 insertions(+)
>
>diff --git a/contrib/ci/sssd.supp b/contrib/ci/sssd.supp
>index 5bd8c8d699d8807c260966f7e08dee312a0938c6..7ecc09d0def07de8595ce166db8ad84eb8d4049b 100644
>--- a/contrib/ci/sssd.supp
>+++ b/contrib/ci/sssd.supp
>@@ -122,3 +122,34 @@
>    fun:exit
>    fun:(below main)
> }
>+
>+# long-standing memory leak in popt
>+{
>+   popt-memleak-from-poptGetNextOpt-malloc
>+   Memcheck:Leak
>+   match-leak-kinds: definite
>+   fun:malloc
>+   ...
>+   fun:poptGetNextOpt
>+   fun:main
>+}
>+
>+# long-standing memory leak in popt
>+{
>+   popt-memleak-from-poptGetNextOpt-realloc
>+   Memcheck:Leak
>+   match-leak-kinds: definite
>+   fun:realloc
>+   ...
>+   fun:poptGetNextOpt
>+   fun:main
>+}
I do not acare about memory leaks. So I'm fine with it.

But the next explanation is pretty vague at least for error
"Invalid read of size 4" -> Memcheck:Addr4
I cannot see such warning with libpotp 1.16
Please add comment which distro is problematic (the oldest one)
and that it is fixed in new version of libpopt

>+
>+# popt is not good with access either..
>+{
>+    popt-suppress-invalid-read
>+    Memcheck:Addr4
>+    ...
>+    fun:poptGetNextOpt
>+    fun:main
>+}

LS



More information about the sssd-devel mailing list