On Mon, May 17, 2010 at 01:04:24PM +0200, Jim Meyering wrote:
The following theoretical possibility of a NULL dereference has been in the code since April 1 (commit 6e41f30efcac08e50b21d9c943d6d27e90555951). It's theoretical, because if that vm = NULL statement is ever executed, the very next one, calling virDomainObjUnlock would dereference that now-NULL "vm". Hence, I think we can conclude the vm = NULL statement is effectively dead code. That conclusion is in line with the "should" in the preceding comment.
From f88969b986a1c88985671c9d6fa9cb1dc449ed74 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering@redhat.com> Date: Mon, 17 May 2010 12:10:52 +0200 Subject: [PATCH] qemu_driver: avoid NULL dereference
* src/qemu/qemu_driver.c (qemudDomainStart): After setting vm to NULL, goto cleanup, rather than dereferencing the NULL pointer. --- src/qemu/qemu_driver.c | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 8f69b5a..3559e36 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -6489,8 +6489,11 @@ static int qemudDomainStart(virDomainPtr dom) { * We should still have a reference left to vm but * one should check for 0 anyway */ - if (qemuDomainObjEndJob(vm) == 0) + if (qemuDomainObjEndJob(vm) = 0) { vm = NULL; + goto cleanup; + } + virDomainObjUnlock(vm); qemuDriverUnlock(driver); ret = qemudDomainRestore(dom->conn, managed_save);
ACK Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://deltacloud.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|