Re: [PATCH 7/7] firmware: firmware: fix NULL pointer dereference in __fw_load_abort()

["Luis R. Rodriguez" <mcgrof@kernel.org>]

On Wed, Jan 25, 2017 at 11:52:04AM +0100, Greg KH wrote:
> On Mon, Jan 23, 2017 at 08:11:11AM -0800, Luis R. Rodriguez wrote:
> > Since commit 5d47ec02c37ea632398cb251c884e3a488dff794
> > ("firmware: Correct handling of fw_state_wait() return value")
> > fw_load_abort(fw_priv) could be called twice and lead us to a
> > kernel crash. This happens only when the firmware fallback mechanism
> > (regular or custom) is used. The fallback mechanism exposes a sysfs
> > interface for userspace to upload a file and notify the kernel when
> > the file is loaded and ready, or to cancel an upload by echo'ing -1
> > into on the loading file:
> > 
> > echo -n "-1" > /sys/$DEVPATH/loading
> > 
> > This will call fw_load_abort(). Some distributions actually have
> > a udev rule in place to *always* immediately cancel all firmware
> > fallback mechanism requests (Debian, OpenSUSE), they have:
> > 
> >   $ cat /lib/udev/rules.d/50-firmware.rules
> >   # stub for immediately telling the kernel that userspace firmware loading
> >   # failed; necessary to avoid long timeouts with CONFIG_FW_LOADER_USER_HELPER=y
> >   SUBSYSTEM=="firmware", ACTION=="add", ATTR{loading}="-1
> > 
> > This was done since udev removed the firmware fallback mechanism a while ago
> > and a long standing misunderstood issues with the timeout (but now corrected).
> > Distributions with this udev rule would run into this crash only if the
> > fallback mechanism is used. Since most distributions disable by default
> > using the fallback mechanism (CONFIG_FW_LOADER_USER_HELPER_FALLBACK), this
> > would typicaly mean only 2 drivers which *require* the fallback mechanism
> > could typically incur a crash: drivers/firmware/dell_rbu.c and the
> > drivers/leds/leds-lp55xx-common.c driver.
> > 
> > Distributions enabling CONFIG_FW_LOADER_USER_HELPER_FALLBACK are clearly
> > more exposed as every file not found through a firmware request will
> > use the fallback mechanism.
> > 
> > The crash happens because after commit 5b029624948d ("firmware: do not
> > use fw_lock for fw_state protection") and subsequent fix commit
> > 5d47ec02c37ea6 ("firmware: Correct handling of fw_state_wait() return
> > value") a race can happen between this cancelation and the firmware
> > fw_state_wait_timeout() being woken up after a state change with which
> > fw_load_abort() as that calls swake_up(). Upon error fw_state_wait_timeout()
> > will also again call fw_load_abort() and trigger a null reference.
> > 
> > At first glance we could just fix this with a !buf check on
> > fw_load_abort() before accessing buf->fw_st, however there is
> > a logical issue in having a state machine used for the fallback
> > mechanism and preventing access from it once we abort as its inside
> > the buf (buf->fw_st).
> > 
> > The firmware_class.c code is setting the buf to NULL to annotate an
> > abort has occurred. Replace this mechanism by simply using the state check
> > instead. All the other code in place already uses similar checks
> > for aborting as well so no further changes are needed.
> > 
> > An oops can be reproduced with the new fw_fallback.sh fallback
> > mechanism cancellation test. Either cancelling the fallback mechanism
> > or the custom fallback mechanism triggers a crash.
> > 
> > mcgrof@piggy ~/linux-next/tools/testing/selftests/firmware
> > (git::20170111-fw-fixes)$ sudo ./fw_fallback.sh
> > 
> > ./fw_fallback.sh: timeout works
> > ./fw_fallback.sh: firmware comparison works
> > ./fw_fallback.sh: fallback mechanism works
> > 
> > [ this then sits here when it is trying the cancellation test ]
> > 
> > Kernel log:
> > 
> > [   36.750521] test_firmware: loading 'nope-test-firmware.bin'
> > [   36.751144] misc test_firmware: Direct firmware load for nope-test-firmware.bin failed with error -2
> > [   36.752034] misc test_firmware: Falling back to user helper
> > [   36.853324] BUG: unable to handle kernel NULL pointer dereference at 0000000000000038
> > [   36.854221] IP: _request_firmware+0xa27/0xad0
> > [   36.854671] PGD 0
> > [   36.854672]
> > [   36.855081] Oops: 0000 [#1] SMP
> > [   36.855433] Modules linked in: test_firmware(E) ... etc ...
> > [   36.857802] CPU: 1 PID: 1396 Comm: fw_fallback.sh Tainted: G        W E   4.10.0-rc3-next-20170111+ #30
> > [   36.857802] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.10.1-0-g8891697-prebuilt.qemu-project.org 04/01/2014
> > [   36.857802] task: ffff9740b27f4340 task.stack: ffffbb15c0bc8000
> > [   36.857802] RIP: 0010:_request_firmware+0xa27/0xad0
> > [   36.857802] RSP: 0018:ffffbb15c0bcbd10 EFLAGS: 00010246
> > [   36.857802] RAX: 00000000fffffffe RBX: ffff9740afe5aa80 RCX: 0000000000000000
> > [   36.857802] RDX: ffff9740b27f4340 RSI: 0000000000000283 RDI: 0000000000000000
> > [   36.857802] RBP: ffffbb15c0bcbd90 R08: ffffbb15c0bcbcd8 R09: 0000000000000000
> > [   36.857802] R10: 0000000894a0d4b1 R11: 000000000000008c R12: ffffffffc0312480
> > [   36.857802] R13: 0000000000000005 R14: ffff9740b1c32400 R15: 00000000000003e8
> > [   36.857802] FS:  00007f8604422700(0000) GS:ffff9740bfc80000(0000) knlGS:0000000000000000
> > [   36.857802] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > [   36.857802] CR2: 0000000000000038 CR3: 000000012164c000 CR4: 00000000000006e0
> > [   36.857802] Call Trace:
> > [   36.857802]  request_firmware+0x37/0x50
> > [   36.857802]  trigger_request_store+0x79/0xd0 [test_firmware]
> > [   36.857802]  dev_attr_store+0x18/0x30
> > [   36.857802]  sysfs_kf_write+0x37/0x40
> > [   36.857802]  kernfs_fop_write+0x110/0x1a0
> > [   36.857802]  __vfs_write+0x37/0x160
> > [   36.857802]  ? _cond_resched+0x1a/0x50
> > [   36.857802]  vfs_write+0xb5/0x1a0
> > [   36.857802]  SyS_write+0x55/0xc0
> > [   36.857802]  ? trace_do_page_fault+0x37/0xd0
> > [   36.857802]  entry_SYSCALL_64_fastpath+0x1e/0xad
> > [   36.857802] RIP: 0033:0x7f8603f49620
> > [   36.857802] RSP: 002b:00007fff6287b788 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
> > [   36.857802] RAX: ffffffffffffffda RBX: 000055c307b110a0 RCX: 00007f8603f49620
> > [   36.857802] RDX: 0000000000000016 RSI: 000055c3084d8a90 RDI: 0000000000000001
> > [   36.857802] RBP: 0000000000000016 R08: 000000000000c0ff R09: 000055c3084d6336
> > [   36.857802] R10: 000055c307b108b0 R11: 0000000000000246 R12: 000055c307b13c80
> > [   36.857802] R13: 000055c3084d6320 R14: 0000000000000000 R15: 00007fff6287b950
> > [   36.857802] Code: 9f 64 84 e8 9c 61 fe ff b8 f4 ff ff ff e9 6b f9 ff
> > ff 48 c7 c7 40 6b 8d 84 89 45 a8 e8 43 84 18 00 49 8b be 00 03 00 00 8b
> > 45 a8 <83> 7f 38 02 74 08 e8 6e ec ff ff 8b 45 a8 49 c7 86 00 03 00 00
> > [   36.857802] RIP: _request_firmware+0xa27/0xad0 RSP: ffffbb15c0bcbd10
> > [   36.857802] CR2: 0000000000000038
> > [   36.872685] ---[ end trace 6d94ac339c133e6f ]---
> > 
> > In above case the call hierarchy that causes the crash looks as follows:
> > 
> > lib/test_firmware.c request_firmware()
> > -> fw_load_from_user_helper()
> > -> _request_firmware_load()
> > -> call fw_state_wait_timeout()
> > 
> > Some time later firmware_loading_store() scans a control value of "-1"
> > -> switch(loading) case -1: will call
> > -> fw_load_abort(fw_priv) which calls
> > -> __fw_load_abort(fw_priv->buf)
> > -> and set fw_priv->buf = NULL;
> > 
> > Upon being woken up via swake_up(), back in _request_firmware_load()
> > fw_state_wait_timeout() returns -ENOENT
> > -> since mentioned commit
> > -> fw_load_abort(fw_priv) is called a second time
> > -> and this time it would call:
> > -> __fw_load_abort(NULL /* fw_priv->buf */)
> > -> and we get: NULL->fw_st.status
> > 
> > Fixes: 5d47ec02c37e ("firmware: Correct handling of fw_state_wait() return value")
> > Reported-and-Tested-by: Jakub Kicinski <jakub.kicinski@netronome.com>
> > Reported-and-Tested-by: Patrick Bruenn <p.bruenn@beckhoff.com>
> > Reported-by: Chris Wilson <chris@chris-wilson.co.uk>
> > CC: <stable@vger.kernel.org>    [3.10+]

Note: 3.10+

> > Signed-off-by: Luis R. Rodriguez <mcgrof@kernel.org>
> > ---
> >  drivers/base/firmware_class.c | 5 +----
> >  1 file changed, 1 insertion(+), 4 deletions(-)
> 
> Why is this patch 7/7? 

Without the tests available on a development tree one cannot easily
reproduce.

> Shouldn't it go into 4.10-final now?  Why wait
> for 4.11-rc1?

Certainly, it should go into 4.10 now, sorry if it seemed otherwise.

  Luis