Re: [PATCH v2 1/4] ipack: tpci200: fix many double free issues in tpci200_pci_probe

[Dongliang Mu]

On Tue, Aug 10, 2021 at 2:12 PM Greg Kroah-Hartman
<gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
>
> On Tue, Aug 10, 2021 at 07:41:55AM +0800, Dongliang Mu wrote:
> > On Tue, Aug 10, 2021 at 7:08 AM Dongliang Mu <mudongliangabcd@xxxxxxxxx> wrote:
> > >
> > > On Mon, Aug 9, 2021 at 11:32 PM Greg Kroah-Hartman
> > > <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> > > >
> > > > On Mon, Aug 09, 2021 at 10:30:26PM +0800, Dongliang Mu wrote:
> > > > > The function tpci200_register called by tpci200_install and
> > > > > tpci200_unregister called by tpci200_uninstall are in pair. However,
> > > > > tpci200_unregister has some cleanup operations not in the
> > > > > tpci200_register. So the error handling code of tpci200_pci_probe has
> > > > > many different double free issues.
> > > > >
> > > > > Fix this problem by moving those cleanup operations out of
> > > > > tpci200_unregister, into tpci200_pci_remove and reverting
> > > > > the previous commit 9272e5d0028d ("ipack/carriers/tpci200:
> > > > > Fix a double free in tpci200_pci_probe").
> > > > >
> > > > > Reported-by: Dongliang Mu <mudongliangabcd@xxxxxxxxx>
> > > > > Fixes: 9272e5d0028d ("ipack/carriers/tpci200: Fix a double free in tpci200_pci_probe")
> > > > > Signed-off-by: Dongliang Mu <mudongliangabcd@xxxxxxxxx>
> > > > > ---
> > > > > v1->v2: revise PATCH 2/3, 3/3, not depending on PATCH 1/3; move the
> > > > > location change of tpci_unregister into one separate patch;
> > > >
> > > > Also needs to go to the stable trees, right?
> > >
> > > Yes, this needs to go to the stable trees.
> >
> > Hi gregkh,
> >
> > Let me clarify more. In my series, PATCH 3/4 4/4 depends on PATCH 1/4
> > and PATCH 2/4. And also PATCH 2/4 depends on PATCH 1/4 as they are
> > closely related.
> >
> > But from your reply, the last 2 patches should not depend on the first
> > 2 patches. I don't quite understand as I don't send some patch series
> > before. For a patch series, the latter ones should depend on the
> > former ones, right? If I have any misunderstanding, please let me
> > know.
>
> Yes, they can depend on previous patches, but if some patches are to be
> merged today for 5.14-final, and others later for 5.15-rc1, then ideally
> they will be separate series of changes as those go to two different
> branches in my tree at the moment.
>
> > BTW, PATCH 3/4 has some compilation issues. I will fix it in the next version.
>
> As you haven't even tested these, I'm really hesitant to take them at
> all.
>
> Please just send the first two patches, fixed up, as a series after you
> have tested them, and then after they are merged into Linus's tree, you
> can send the cleanup patches, as they are just "nice" to have.
>

That's good. I will send the first two patches. After they are merged,
then I will send the rest patches.

> thanks,
>
> greg k-h