lkml.org 
[lkml]   [2021]   [Nov]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRE: [PATCH v2] drm/dp: Fix aux->transfer NULL pointer dereference on drm_dp_dpcd_access
Date
[AMD Official Use Only]

Hi Jani:


> -----Original Message-----
> From: Jani Nikula <jani.nikula@linux.intel.com>
> Sent: Wednesday, November 3, 2021 7:31 PM
> To: Yuan, Perry <Perry.Yuan@amd.com>; Maarten Lankhorst
> <maarten.lankhorst@linux.intel.com>; Maxime Ripard
> <mripard@kernel.org>; Thomas Zimmermann <tzimmermann@suse.de>;
> David Airlie <airlied@linux.ie>; Daniel Vetter <daniel@ffwll.ch>
> Cc: dri-devel@lists.freedesktop.org; linux-kernel@vger.kernel.org; Huang,
> Shimmer <Xinmei.Huang@amd.com>; Huang, Ray <Ray.Huang@amd.com>
> Subject: RE: [PATCH v2] drm/dp: Fix aux->transfer NULL pointer dereference
> on drm_dp_dpcd_access
>
> [CAUTION: External Email]
>
> On Wed, 03 Nov 2021, "Yuan, Perry" <Perry.Yuan@amd.com> wrote:
> > [AMD Official Use Only]
> >
> > Hi Jani:
> >
> >> -----Original Message-----
> >> From: Jani Nikula <jani.nikula@linux.intel.com>
> >> Sent: Tuesday, November 2, 2021 4:40 PM
> >> To: Yuan, Perry <Perry.Yuan@amd.com>; Maarten Lankhorst
> >> <maarten.lankhorst@linux.intel.com>; Maxime Ripard
> >> <mripard@kernel.org>; Thomas Zimmermann <tzimmermann@suse.de>;
> David
> >> Airlie <airlied@linux.ie>; Daniel Vetter <daniel@ffwll.ch>
> >> Cc: dri-devel@lists.freedesktop.org; linux-kernel@vger.kernel.org;
> >> Huang, Shimmer <Xinmei.Huang@amd.com>; Huang, Ray
> <Ray.Huang@amd.com>
> >> Subject: RE: [PATCH v2] drm/dp: Fix aux->transfer NULL pointer
> >> dereference on drm_dp_dpcd_access
> >>
> >> [CAUTION: External Email]
> >>
> >> On Tue, 02 Nov 2021, "Yuan, Perry" <Perry.Yuan@amd.com> wrote:
> >> > [AMD Official Use Only]
> >> >
> >> > Hi Jani:
> >> > Thanks for your comments.
> >> >
> >> >> -----Original Message-----
> >> >> From: Jani Nikula <jani.nikula@linux.intel.com>
> >> >> Sent: Monday, November 1, 2021 9:07 PM
> >> >> To: Yuan, Perry <Perry.Yuan@amd.com>; Maarten Lankhorst
> >> >> <maarten.lankhorst@linux.intel.com>; Maxime Ripard
> >> >> <mripard@kernel.org>; Thomas Zimmermann
> <tzimmermann@suse.de>;
> >> David
> >> >> Airlie <airlied@linux.ie>; Daniel Vetter <daniel@ffwll.ch>
> >> >> Cc: Yuan, Perry <Perry.Yuan@amd.com>;
> >> >> dri-devel@lists.freedesktop.org; linux- kernel@vger.kernel.org;
> >> >> Huang, Shimmer <Xinmei.Huang@amd.com>; Huang, Ray
> >> <Ray.Huang@amd.com>
> >> >> Subject: Re: [PATCH v2] drm/dp: Fix aux->transfer NULL pointer
> >> >> dereference on drm_dp_dpcd_access
> >> >>
> >> >> [CAUTION: External Email]
> >> >>
> >> >> On Mon, 01 Nov 2021, Perry Yuan <Perry.Yuan@amd.com> wrote:
> >> >> > Fix below crash by adding a check in the drm_dp_dpcd_access
> >> >> > which ensures that aux->transfer was actually initialized earlier.
> >> >>
> >> >> Gut feeling says this is papering over a real usage issue
> >> >> somewhere else. Why is the aux being used for transfers before
> >> >> ->transfer has been set? Why should the dp helper be defensive
> >> >> against all kinds of
> >> misprogramming?
> >> >>
> >> >>
> >> >> BR,
> >> >> Jani.
> >> >>
> >> >
> >> > The issue was found by Intel IGT test suite, graphic by pass test case.
> >> >
> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fg
> >> > itl
> >> > ab.freedesktop.org%2Fdrm%2Figt-gpu-
> >> tools&amp;data=04%7C01%7CPerry.Yuan
> >> > %40amd.com%7C83d011acfe65437c0fa808d99ddc65b0%7C3dd8961fe4
> 884e6
> >> 08e11a8
> >> >
> >>
> 2d994e183d%7C0%7C0%7C637714392203200313%7CUnknown%7CTWFpbG
> Zsb
> >> 3d8eyJWIj
> >> >
> >>
> oiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1
> 00
> >> 0&am
> >> >
> >>
> p;sdata=snPpRYLGeJtTpNGle1YHZAvevcABbgLkgOsffiNzQPw%3D&amp;reser
> ved
> >> =0
> >> > normally use case will not see the issue.
> >> > To avoid this issue happy again when we run the test case , it will
> >> > be nice to
> >> add a check before the transfer is called.
> >> > And we can see that it really needs to have a check here to make
> >> > ITG &kernel
> >> happy.
> >>
> >> You're missing my point. What is the root cause? Why do you have the
> >> aux device or connector registered before ->transfer function is
> >> initialized. I don't think you should do that.
> >>
> >> BR,
> >> Jani.
> >>
> >
> > One potential IGT fix patch to resolve the test case failure is:
> >
> > tests/amdgpu/amd_bypass.c
> > data->pipe_crc = igt_pipe_crc_new(data->drm_fd, data->pipe_id,
> > - AMDGPU_PIPE_CRC_SOURCE_DPRX);
> > + INTEL_PIPE_CRC_SOURCE_AUTO);
> > The kernel panic error gone after change "dprx" to "auto" in the IGT test.
> >
> > In my view ,the IGT amdgpu bypass test will do some common setup work
> including crc piple, source.
> > When the IGT sets up a new CRC pipe capture source for amdgpu bypass
> test, the SOURCE was set as "dprx" instead of "auto"
> > It makes "amdgpu_dm_crtc_set_crc_source()" failed to set correct AUX
> and it's transfer function invalid .
> > The system I tested use HDMI port connected to monitor .
> >
> > amdgpu_dm_crtc_set_crc_source-> (aux = (aconn->port) ? &aconn-
> >port->aux : &aconn->dm_dp_aux.aux;)
> > drm_dp_start_crc ->
> > drm_dp_dpcd_readb-> aux->transfer is NULL, issue here.
> > The fix will use the "auto" keyword, which will let the driver select a
> default source of frame CRCs for this CRTC.
> >
> > Correct me if have some wrong points.
>
> Apparently I'm completely failing to communicate my POV to you.
>
> If you have a kernel oops, the fix needs to be in the kernel, not IGT.
>
> The question is, why is it possible for IGT (or any userspace) to trigger AUX
> communication when the ->transfer function is not set? In my opinion, the
> kernel driver should not have exposed the interface at all if the ->transfer
> function is not set. The interface is useless without the ->transfer function.
> IMO, that's the bug.
>

Yes , you are correct , the transfer shouldn't be called before it is ready !

Let me explain more details in my view .
Maybe the root cause is not why the aux->transfer is not called before it is registered in this case.
I suppose the issue was triggered by wrong CRC pipe source .

Actually the aux->transfer has been registered when amdgpu DM registered at kernel boot.
IGT test was run when system login to Gnome desktop.

amdgpu_dm_initialize_dp_connector->
aconnector->dm_dp_aux.aux.transfer = dm_dp_aux_transfer;

The test case failed when the IGT set an "DPRX" CRC pipe source while the HDMI connected to monitor only.
At this time, the aux->transfer is NULL, and dp helper did not check the transfer pointer NULL or not.
It calls the transfers to DPCD read, then you see the kernel panic log.

amdgpu_dm_crtc_funcs-> amdgpu_dm_crtc_set_crc_source-> drm_dp_start_crc

* And if the DP cable connected only, the issue will not happen. Test will pass.
* If I change the CRC source to "auto", kernel will not see the panic as well.
Maybe the failed test case need to run on the DP instead of HDMI, I am not sure at now.


Hopping this info can help.

Perry.


>
> BR,
> Jani.
>
> >
> > Thank you!
> > Perry.
> >
> >>
> >> >
> >> > Perry.
> >> >
> >> >>
> >> >> >
> >> >> > BUG: kernel NULL pointer dereference, address: 0000000000000000
> >> >> > PGD
> >> >> > 0 P4D 0
> >> >> > Oops: 0010 [#1] SMP NOPTI
> >> >> > RIP: 0010:0x0
> >> >> > Code: Unable to access opcode bytes at RIP 0xffffffffffffffd6.
> >> >> > RSP: 0018:ffffa8d64225bab8 EFLAGS: 00010246
> >> >> > RAX: 0000000000000000 RBX: 0000000000000020 RCX:
> >> >> > ffffa8d64225bb5e
> >> >> > RDX: ffff93151d921880 RSI: ffffa8d64225bac8 RDI:
> >> >> > ffff931511a1a9d8
> >> >> > RBP: ffffa8d64225bb10 R08: 0000000000000001 R09:
> >> >> > ffffa8d64225ba60
> >> >> > R10: 0000000000000002 R11: 000000000000000d R12:
> >> >> > 0000000000000001
> >> >> > R13: 0000000000000000 R14: ffffa8d64225bb5e R15:
> >> >> > ffff931511a1a9d8
> >> >> > FS: 00007ff8ea7fa9c0(0000) GS:ffff9317fe6c0000(0000)
> >> >> > knlGS:0000000000000000
> >> >> > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> >> >> > CR2: ffffffffffffffd6 CR3: 000000010d5a4000 CR4:
> >> >> > 0000000000750ee0
> >> >> > PKRU: 55555554
> >> >> > Call Trace:
> >> >> > drm_dp_dpcd_access+0x72/0x110 [drm_kms_helper]
> >> >> > drm_dp_dpcd_read+0xb7/0xf0 [drm_kms_helper]
> >> >> > drm_dp_start_crc+0x38/0xb0 [drm_kms_helper]
> >> >> > amdgpu_dm_crtc_set_crc_source+0x1ae/0x3e0 [amdgpu]
> >> >> > crtc_crc_open+0x174/0x220 [drm]
> >> >> > full_proxy_open+0x168/0x1f0
> >> >> > ? open_proxy_open+0x100/0x100
> >> >> > do_dentry_open+0x156/0x370
> >> >> > vfs_open+0x2d/0x30
> >> >> >
> >> >> > v2: fix some typo
> >> >> >
> >> >> > Signed-off-by: Perry Yuan <Perry.Yuan@amd.com>
> >> >> > ---
> >> >> > drivers/gpu/drm/drm_dp_helper.c | 4 ++++
> >> >> > 1 file changed, 4 insertions(+)
> >> >> >
> >> >> > diff --git a/drivers/gpu/drm/drm_dp_helper.c
> >> >> > b/drivers/gpu/drm/drm_dp_helper.c index
> >> >> > 6d0f2c447f3b..76b28396001a
> >> >> > 100644
> >> >> > --- a/drivers/gpu/drm/drm_dp_helper.c
> >> >> > +++ b/drivers/gpu/drm/drm_dp_helper.c
> >> >> > @@ -260,6 +260,10 @@ static int drm_dp_dpcd_access(struct
> >> >> > drm_dp_aux
> >> >> *aux, u8 request,
> >> >> > msg.buffer = buffer;
> >> >> > msg.size = size;
> >> >> >
> >> >> > + /* No transfer function is set, so not an available DP connector */
> >> >> > + if (!aux->transfer)
> >> >> > + return -EINVAL;
> >> >> > +
> >> >> > mutex_lock(&aux->hw_mutex);
> >> >> >
> >> >> > /*
> >> >>
> >> >> --
> >> >> Jani Nikula, Intel Open Source Graphics Center
> >>
> >> --
> >> Jani Nikula, Intel Open Source Graphics Center
>
> --
> Jani Nikula, Intel Open Source Graphics Center

\
 
 \ /
  Last update: 2021-11-05 08:35    [W:0.096 / U:0.092 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site