Re: [PATCH] async: Suppress GCC13 false positive in aio_bh_poll()

[Juan Quintela <quintela@redhat.com>]

Paolo Bonzini <pbonzini@redhat.com> wrote:
> Il mar 25 apr 2023, 15:31 Daniel P. Berrangé <berrange@redhat.com> ha
> scritto:
>
>> > > -    BHListSlice slice;
>> > > +    /*
>> > > +     * gcc13 complains about putting a local variable
>> > > +     * in a global list, so put it on the heap.
>> > > +     */
>> > > +    g_autofree BHListSlice *slice = g_new(BHListSlice, 1);
>> > >      BHListSlice *s;
>> > >      int ret = 0;
>> > >
>> >
>> > This must be a memory leak since you're adding a g_new but not
>> > adding any g_free
>>
>> Sorry, I'm failing to read properly today. It uses g_autofree
>> so there is no leak.
>>
>
> On the other hand, if the pointer to the heap-allocated BHListSlice
> escaped, this would be a dangling pointer as well—just not the kind that
> the new GCC warning can report.

I don't agree here.
If with my patch it becomes a dangling pointer because we free it.
With Cedric patch it is a local variable that gets exited out of the
function that created it.

Choose your poison.  One thing is bad and the other is worse.

> So this patch is also doing nothing but shut up the compiler; but it's
> doing so in an underhanded manner and with a runtime cost, and as such it's
> worse than Cedric's patch.

Ok.  I don't care (enogouh) about this to continue a discussion.. Can we
get Cedric patch upstream?

Thanks, Juan.