On Wed, Jan 27, 2010 at 11:17:32AM +0100, Jim Meyering wrote:
Daniel P. Berrange wrote:
On Wed, Jan 27, 2010 at 10:38:55AM +0100, Jim Meyering wrote:
Actually, the preceding patch fixed only the one leak that had been introduced in the last month or so. Looking at the many other functions that do the same sort of thing (call qemuMonitorJSONMakeCommand, and later virJSONValueFree), I saw that they all had exactly the same leak. So this amended patch fixes all of them:
From 28f820354dcae9950cad042ea78a893fd9475830 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering@redhat.com> Date: Wed, 27 Jan 2010 09:58:12 +0100 Subject: [PATCH] qemu_monitor_json.c: avoid many unconditional leaks
The real bug is the virJSONValueFree() itself which is missing the final VIR_FREE(value) call. By doing the free in the caller, we still leak data for compound array/hash values.
Putting the VIR_FREE in virJSONValueFree was my first reflex, too, but since coverity detected no leak for the adjacent "virJSONValueFree(reply);" use, I assumed that doing so would cause a problem:
virJSONValueFree(cmd); VIR_FREE(cmd); virJSONValueFree(reply);
I think coverity must be confused then :-) The virJSONValueFree() method is definitely intended to free any memory allocated during one of the virJSONValueNewXXXX() methods. Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|