On 2018-09-12 at 15:24, Erik Skultety wrote:
>On Tue, Sep 11, 2018 at 11:49:11PM +0800, Shi Lei wrote:
>> On 2018-09-11 at 20:50, Erik Skultety wrote:
>> >On Mon, Sep 10, 2018 at 11:47:49AM +0800, Shi Lei wrote:
>> >> This series introduce VIR_AUTOCLOSE macro which can force close fd of the file
>> >> automatically when the fd goes out of scope. It's used to eliminate VIR_FORCE_CLOSE
>> >> in cleanup sections. The new macro consults VIR_AUTOFREE and VIR_AUTOPTR.
>> >>
>> >> Shi Lei (6):
>> >> util: file: introduce VIR_AUTOCLOSE macro to close fd of the file
>> >> automatically
>> >> util: file: use VIR_AUTOCLOSE instead of VIR_FORCE_CLOSE in cleanup
>> >> sections
>> >> util: netdevbridge: use VIR_AUTOCLOSE instead of VIR_FORCE_CLOSE in
>> >> cleanup sections
>> >> util: netdev: use VIR_AUTOCLOSE instead of VIR_FORCE_CLOSE in cleanup
>> >> sections
>> >> phyp: driver: use VIR_AUTOCLOSE instead of VIR_FORCE_CLOSE in cleanup
>> >> sections
>> >> uml: conf: use VIR_AUTOCLOSE instead of VIR_FORCE_CLOSE in cleanup
>> >> sections
>> >>
>> >> src/phyp/phyp_driver.c | 29 ++---
>> >> src/uml/uml_conf.c | 13 +-
>> >> src/util/virfile.c | 21 +--
>> >> src/util/virfile.h | 20 ++-
>> >> src/util/virnetdev.c | 253 +++++++++++++------------------------
>> >> src/util/virnetdevbridge.c | 120 ++++++------------
>> >
>> >This touches only a handful of util modules and random drivers, there should be
>> >some consistency here, so if we're introducing something like this, then we
>> >should do it in a similar manner as we did for VIR_AUTO{FREE,PTR}, i.e. util/
>> >first (btw. we still haven't even finished that one yet).
>> >
>> >Erik
>>
>> Okay. But how to do in the v2 series.
>>
>> Whether v2 should only deal with files in the util/ ?
>
>Yeah, I think that's the best approach we can take, since we still haven't
>finished converting util to the other macros, we should start with the those
>modules within util/ that are already completely switched to AUTO{FREE/PTR} and
>similarly do it in batches, as reviewing 30+ patches on one go is always
>demanding...
>
>Thanks,
>Erik
Okay. I see.
Thanks.
Shi Lei
--
libvir-list mailing list
libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list
