On 04/06/2018 12:53 PM, John Ferlan wrote:
Rather than using VIR_ALLOC, use the New API since we already
use the virDomainChrSourceDefFree function when done.
Signed-off-by: John Ferlan <jferlan(a)redhat.com>
---
src/conf/domain_conf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
index aacd06a87a..caf3f47c63 100644
--- a/src/conf/domain_conf.c
+++ b/src/conf/domain_conf.c
@@ -11138,7 +11138,7 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt,
goto error;
}
- if (VIR_ALLOC(def->data.vhostuser) < 0)
+ if (!(def->data.vhostuser = virDomainChrSourceDefNew(xmlopt)))
goto error;
def->data.vhostuser->type = VIR_DOMAIN_CHR_TYPE_UNIX;
Reviewed-by: Laine Stump <laine(a)laine.org>
There have been cases in the past where a *New() or *Free() function was
added but not consistently used, leading to leaks or (worse, but at
least easier to detect) crashes due to null pointer dereferences. It
would really be nice if there was some way to automate the auditing of
all the VIR_ALLOCs, but I don't think it's possible with the simple make
syntax-check target (that's supposed to be a challenge to someone :-),
and there's 1223 uses of VIR_ALLOC() (6421 of VIR_FREE()), so even a one
time manual audit is really out of the question (and the results would
be immediately obsolete as soon as a new VIR_ALLOC or VIR_FREE was added).