On Fri, Aug 03, 2018 at 06:46:28PM +0530, Sukrit Bhatnagar wrote:
On Fri, 3 Aug 2018 at 18:41, Erik Skultety
<eskultet(a)redhat.com> wrote:
>
> On Fri, Aug 03, 2018 at 06:38:30PM +0530, Sukrit Bhatnagar wrote:
> > On Fri, 3 Aug 2018 at 18:32, Erik Skultety <eskultet(a)redhat.com> wrote:
> > >
> > > On Sat, Jul 28, 2018 at 11:31:28PM +0530, Sukrit Bhatnagar wrote:
> > > > By making use of GNU C's cleanup attribute handled by the
> > > > VIR_AUTOPTR macro for declaring aggregate pointer variables,
> > > > majority of the calls to *Free functions can be dropped, which
> > > > in turn leads to getting rid of most of our cleanup sections.
> > > >
> > > > Signed-off-by: Sukrit Bhatnagar <skrtbhtngr(a)gmail.com>
> > > > ---
> > > > src/util/virnetdevip.c | 55
+++++++++++++++++++++-----------------------------
> > > > 1 file changed, 23 insertions(+), 32 deletions(-)
> > > >
> > > > diff --git a/src/util/virnetdevip.c b/src/util/virnetdevip.c
> > > > index 8f1081b..ca206e2 100644
> > > > --- a/src/util/virnetdevip.c
> > > > +++ b/src/util/virnetdevip.c
> > > > @@ -634,19 +634,22 @@ virNetDevIPCheckIPv6Forwarding(void)
> > > > }
> > > >
> > > > if (!valid) {
> > > > - virBuffer buf = VIR_BUFFER_INITIALIZER;
> > > > + VIR_AUTOPTR(virBuffer) buf = NULL;
> > > > +
> > > > + if (VIR_ALLOC(buf) < 0)
> > > > + goto cleanup;
> > >
> > > Hmm, this will actually leak memory because @buf is never going to be
freed,
> > > worse, we'll assign NULL to it.
> >
> > But since @buf is declared as AUTOPTR, virBufferFreeAndReset will be called
> > when it exits the scope, right?
> > If I were using virBufferContentAndReset, then it might be the case.
>
> How does virBufferFreeAndReset free @buf? It frees buf->content, but keeps
> @buf.
Got it. I actually made a lot of such changes, have to revert them all.
Luckily I don't see any of that merged yet, so at least we don't have to hunt
it down in the release.
Erik