[LTP] [PATCH 4/4] syscalls/fanotify09: Add test cases for merge of ignore mask

Jan Kara jack@suse.cz
Tue Jun 21 10:31:55 CEST 2022


On Tue 21-06-22 06:02:24, Amir Goldstein wrote:
> On Mon, Jun 20, 2022 at 11:35 PM Jan Kara <jack@suse.cz> wrote:
> >
> > On Mon 20-06-22 19:59:48, Amir Goldstein wrote:
> > > On Mon, Jun 20, 2022 at 6:20 PM Jan Kara <jack@suse.cz> wrote:
> > > >
> > > > On Mon 20-06-22 16:27:37, Amir Goldstein wrote:
> > > > > 1. Verify that an ignore mask that does not survive modify event,
> > > > >    does survive a modify event on child, if parent is not watching
> > > > >    events on children.
> > > > >
> > > > > 2. Verify that an ignore mask on parent does not ignore close events
> > > > >    sent to mount mark, if parent is not watching events on children.
> > > > >
> > > > > The behavior of these corner cases of ignore mask on parent dir have
> > > > > always been undefined, so do not run the test for kernel < v5.19.
> > > > >
> > > > > Signed-off-by: Amir Goldstein <amir73il@gmail.com>
> > > >
> > > > Hum, I was looking into the testcase. What does generate a modify event
> > > > there and checks that ignore mask does not survive it?
> > >
> > > I think this does:
> > >
> > >         /*
> > >          * generate MODIFY event and no FAN_CLOSE_NOWRITE event.
> > >          */
> > >         SAFE_FILE_PRINTF(fname, "1");
> >
> > Yeah, I was looking at that but then I'm somewhat confused because this
> > gets called in .setup() hook while the notification groups get created only
> > in .test() hook that gets called later. What am I missing?
> 
> There is one in setup().
> The one with the comment above is from test_fanotify().

Bah, I must have been blind... Thanks for directing me.

								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR


More information about the ltp mailing list