Crash in acpi_ns_validate_handle triggered by soundwire on Linux 5.10

[Rafael J. Wysocki]

On Thu, Jan 28, 2021 at 1:13 PM Rafael J. Wysocki <rafael at kernel.org> wrote:
>
> On Wed, Jan 27, 2021 at 8:19 PM Marcin Ślusarz <marcin.slusarz at gmail.com> wrote:
> >
> > śr., 27 sty 2021 o 18:28 Pierre-Louis Bossart
> > <pierre-louis.bossart at linux.intel.com> napisał(a):
> > > > Weird, I can't reproduce this problem with my self-compiled kernel :/
> > > > I don't even see soundwire modules loaded in. Manually loading them of course
> > > > doesn't do much.
> > > >
> > > > Previously I could boot into the "faulty" kernel by using "recovery mode", but
> > > > I can't do that anymore - it crashes too.
> > > >
> > > > Maybe there's some kind of race and this bug depends on some specific
> > > > ordering of events?
> > >
> > > missing Kconfig?
> > > You need CONFIG_SOUNDWIRE and CONFIG_SND_SOC_SOF_INTEL_SOUNDWIRE
> > > selected to enter this sdw_intel_acpi_scan() routine.
> >
> > It was a PEBKAC, but a slightly different one. I won't bore you with
> > (embarrassing) details ;).
> >
> > I reproduced the problem, tested both your and Rafael's patches
> > and the kernel still crashes, with the same stack trace.
> > (Yes, I'm sure I booted the right kernel :)
> >
> > Why "recovery mode" stopped working (or worked previously) is still a mystery.
>
> So for clarity, you've tried this:
>
> static int snd_intel_dsp_check_soundwire(struct pci_dev *pci)
> {
>     struct sdw_intel_acpi_info info;
>     acpi_handle handle;
>     int ret;
>
>     handle = ACPI_HANDLE(&pci->dev);
>     if (!handle)
>         return -ENODEV;
>
> and it has not made a difference?
>
> And the relevant part of the trace is:
>
> RIP: 0010:acpi_ns_validate_handle+0x1a/0x23
> Code: 00 48 83 c4 10 5b 5d 41 5c 41 5d 41 5e 41 5f c3 0f 1f 44 00 00
> 48 8d 57 ff 48 89 f8 48 83 fa fd 76 08 48 8b 05 0c b8 67 01 c3 <80> 7f
> 08 0f 74 02 31 c0 c3 0f 1f 44 00 00 48 8b 3d f6 b7 67 01 e8
> RSP: 0000:ffffc388807c7b20 EFLAGS: 00010213
> RAX: 0000000000000048 RBX: ffffc388807c7b70 RCX: 0000000000000000
> RDX: 0000000000000047 RSI: 0000000000000246 RDI: 0000000000000048
> RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000
> R10: ffffffffc0f5f4d1 R11: ffffffff8f0cb268 R12: 0000000000001001
> R13: ffffffff8e33b160 R14: 0000000000000048 R15: 0000000000000000
> FS:  00007f24548288c0(0000) GS:ffff9f781fb80000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 0000000000000050 CR3: 0000000106158004 CR4: 0000000000770ee0
> PKRU: 55555554
> Call Trace:
>  acpi_get_data_full+0x4d/0x92
>  acpi_bus_get_device+0x1f/0x40
>  sdw_intel_acpi_scan+0x59/0x230 [soundwire_intel]
>  ? strstr+0x22/0x60
>  ? dmi_matches+0x76/0xe0
>  snd_intel_dsp_driver_probe.cold+0xaf/0x163 [snd_intel_dspcfg]
>  azx_probe+0x7a/0x970 [snd_hda_intel]
>  local_pci_probe+0x42/0x80
>  ? _cond_resched+0x16/0x40
>  pci_device_probe+0xfd/0x1b0
>
> so it looks like we got to sdw_intel_acpi_scan() with a non-NULL, but
> otherwise invalid parent_handle which then was passed to
> acpi_bus_get_device().  Subsequently it got to acpi_get_data_full()
> and acpi_ns_validate_handle() that crashed, because it tried to
> dereference it via ACPI_GET_DESCRIPTOR_TYPE().

But interestingly enough, sdw_intel_acpi_cb() calls
acpi_evaluate_integer() on the same handle that is passed to
acpi_bus_get_device() later and it also calls
acpi_ns_validate_handle() on that handle which doesn't crash.

Moreover, it asks _ADR to be evaluated with respect to that handle and
because it gets to the acpi_bus_get_device() call at all, this appears
to have been successful.

The only explanation for that I can think about (and which does not
involve supernatural intervention so to speak) is a stack corruption
occurring between these two calls in sdw_intel_acpi_cb().  IOW,
something scribbles on the handle in the meantime, but ATM I have no
idea what that can be.

Marcin, please boot with ACPICA deubg (level = ACPI_LV_INFO and
component = ACPI_NAMESPACE | ACPI_BUS_COMPONENT) enabled (see
Documentation/firmware-guide/acpi/debug.rst for instructions) and
collect the log.