Re: [PATCH] wireless: ipw2x00: Refine the error handling of ipw2100_pci_init_one()

[Zheyu Ma]

On Sat, Apr 23, 2022 at 3:25 AM Stanislav Yakovlev
<stas.yakovlev@xxxxxxxxx> wrote:
>
> Hi Zheyu,
>
> On 18/04/2022, Zheyu Ma <zheyuma97@xxxxxxxxx> wrote:
> > On Thu, Apr 14, 2022 at 2:40 AM Stanislav Yakovlev
> > <stas.yakovlev@xxxxxxxxx> wrote:
> >>
> >> On Sat, 9 Apr 2022 at 02:25, Zheyu Ma <zheyuma97@xxxxxxxxx> wrote:
> >> >
> >> > The driver should release resources in reverse order, i.e., the
> >> > resources requested first should be released last, and the driver
> >> > should adjust the order of error handling code by this rule.
> >> >
> >> > Signed-off-by: Zheyu Ma <zheyuma97@xxxxxxxxx>
> >> > ---
> >> >  drivers/net/wireless/intel/ipw2x00/ipw2100.c | 34 +++++++++-----------
> >> >  1 file changed, 16 insertions(+), 18 deletions(-)
> >> >
> >> [Skipped]
> >>
> >> > @@ -6306,9 +6303,13 @@ static int ipw2100_pci_init_one(struct pci_dev
> >> > *pci_dev,
> >> >  out:
> >> >         return err;
> >> >
> >> > -      fail_unlock:
> >> > +fail_unlock:
> >> >         mutex_unlock(&priv->action_mutex);
> >> > -      fail:
> >> > +fail:
> >> > +       pci_release_regions(pci_dev);
> >> > +fail_disable:
> >> > +       pci_disable_device(pci_dev);
> >> We can't move these functions before the following block.
> >>
> >> > +fail_dev:
> >> >         if (dev) {
> >> >                 if (registered >= 2)
> >> >                         unregister_netdev(dev);
> >> This block continues with a function call to ipw2100_hw_stop_adapter
> >> which assumes that device is still accessible via pci bus.
> >
> > Thanks for your reminder, but the existing error handling does need to
> > be revised, I got the following warning when the probing fails at
> > pci_resource_flags():
> >
> > [   20.712160] WARNING: CPU: 1 PID: 462 at lib/iomap.c:44
> > pci_iounmap+0x40/0x50
> > [   20.716583] RIP: 0010:pci_iounmap+0x40/0x50
> > [   20.726342]  <TASK>
> > [   20.726550]  ipw2100_pci_init_one+0x101/0x1ee0 [ipw2100]
> >
> > Since I am not familiar with the ipw2100, could someone give me some
> > advice to fix this.
>
> Could you please rebuild the kernel with IPW2100_DEBUG config option
> enabled, rerun the test and post your results here? Also, please post
> the output of "lspci -v" here.

Sorry for the late response.
I have rebuilt the kernel with IPW2100_DEBUG, and got the following result:

[   29.469624] libipw: 802.11 data/management/control stack, git-1.1.13
[   29.470034] libipw: Copyright (C) 2004-2005 Intel Corporation
<jketreno@xxxxxxxxxxxxxxx>
[   29.477455] ipw2100: Intel(R) PRO/Wireless 2100 Network Driver, git-1.2.2
[   29.477833] ipw2100: Copyright(c) 2003-2006 Intel Corporation
[   29.478197] ipw2100Error calling ioremap.
[   29.478398] ------------[ cut here ]------------
[   29.478630] Bad IO access at port 0x0 ()
[   29.478834] WARNING: CPU: 0 PID: 304 at lib/iomap.c:44 pci_iounmap+0x40/0x50
[   29.481116] RIP: 0010:pci_iounmap+0x40/0x50
[   29.485282] Call Trace:
[   29.485407]  <TASK>
[   29.485514]  ipw2100_pci_init_one+0x192/0x20c0 [ipw2100]
[   29.487496]  local_pci_probe+0x13f/0x200

Actually, I made a special virtual device to test this driver, not
real hardware, so the "lspci" results are irrelevant here.
I injected a software fault in the pci_iomap() function to force the
driver to fail here, and then the driver goto "fail" tag, and under
the "fail" tag the pci_iounmap() is called, resulting in a crash.
In fact such error handling is incorrect and we should not release
resources that have not been requested yet.

Thanks,
Zheyu Ma