Re: [libvirt] [PATCH v1 25/40] util: usb: use VIR_AUTOPTR for aggregate types

On Wed, 25 Jul 2018 at 15:03, Erik Skultety <eskultet(a)redhat.com&gt; wrote: > > On Sat, Jul 21, 2018 at 05:36:57PM +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&gt; > > --- > > src/util/virusb.c | 3 +-- > > 1 file changed, 1 insertion(+), 2 deletions(-) > > > > diff --git a/src/util/virusb.c b/src/util/virusb.c > > index c14683f..cfeac51 100644 > > --- a/src/util/virusb.c > > +++ b/src/util/virusb.c > > @@ -508,8 +508,7 @@ void > > virUSBDeviceListDel(virUSBDeviceListPtr list, > > virUSBDevicePtr dev) > > { > > - virUSBDevicePtr ret = virUSBDeviceListSteal(list, dev); > > - virUSBDeviceFree(ret); > > + VIR_AUTOPTR(virUSBDevice) ret = virUSBDeviceListSteal(list, dev); > > } > > > > Technically, there's also a virUSBDevicePtr instance in virUSBDeviceSearch that > could be converted to VIR_AUTOPTR, but virUSBDeviceListAdd would have to take a > double pointer to @dev instead of a single pointer. A bit more background > info - the current issue is that virUSBDeviceListAdd calls our > VIR_APPEND_ELEMENT helper which does clear the original pointer which we could > utilize here, but not while passing a single pointer. > Not a deal breaker, though, it's just a nice to have, since you're already > working in this area, because I don't suppose we'd make such a change any time > soon after your assignment is over. > > (regardless) > Reviewed-by: Erik Skultety <eskultet(a)redhat.com&gt; There are many such functions: virMediatedDeviceListAdd virSCSIDeviceListAdd virPCIDeviceListAdd Making those changes would take a while and it is not directly related to our cleanup. So, I'll do the necessary after the two cleanup macros are used in all files. Is that ok?