linux-kernel - [PATCH] uio: ensure class is registered before devices

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 15 Jun 2018 17:52:49 +0200
From:   Alexandre Belloni <alexandre.belloni@...tlin.com>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     linux-kernel@...r.kernel.org,
        Alexandre Belloni <alexandre.belloni@...tlin.com>
Subject: [PATCH] uio: ensure class is registered before devices

When both uio and the uio drivers are built in the kernel, it is possible
for a driver to register devices before the uio class is registered.

This may result in a NULL pointer dereference later on in
get_device_parent() when accessing the class glue_dirs spinlock.

The trace looks like that:

Unable to handle kernel NULL pointer dereference at virtual address 00000140
[...]
Process swapper/0 (pid: 1, stack limit = 0xffff000008070000)
Call trace:
Exception stack(0xffff000008073810 to 0xffff000008073950)
3800:                                   0000000000000140 0000000000000001
3820: ffff800078f60000 ffffffffffffffff 0000000000000000 0000000000000000
3840: 0000000000000d39 07650764075f0774 0765076307690776 077207610770075f
3860: 07200774076e0765 0720072007200720 0720072007200720 0720072007200720
3880: 0720072007200720 ffffffffffffffff 0000000000000000 ffff0000089e88c0
38a0: 0000000000000010 ffff000008e0c000 ffff8000780990b0 ffff8000780f4ec0
38c0: ffff000008c41a48 ffff000008c41a30 ffff000008a59660 ffff000008c41a30
38e0: ffff000008a59660 ffff8000780f4c80 ffff000008c41000 ffff000008073950
3900: ffff0000084f3bb0 ffff000008073950 ffff0000089cc234 0000000040000045
3920: 00000000000005e7 ffff000008a59660 ffffffffffffffff 0000000000000000
3940: ffff000008073950 ffff0000089cc234
[<ffff0000089cc234>] _raw_spin_lock+0x14/0x48
[<ffff0000084f56bc>] device_add+0x154/0x6a0
[<ffff0000084f5e48>] device_create_groups_vargs+0x120/0x128
[<ffff0000084f5edc>] device_create+0x54/0x60
[<ffff0000086e72c0>] __uio_register_device+0x120/0x4a8
[<ffff000008528b7c>] jaguar2_pci_probe+0x2d4/0x558
[<ffff0000083fc18c>] local_pci_probe+0x3c/0xb8
[<ffff0000083fd81c>] pci_device_probe+0x11c/0x180
[<ffff0000084f88bc>] driver_probe_device+0x22c/0x2d8
[<ffff0000084f8a24>] __driver_attach+0xbc/0xc0
[<ffff0000084f69fc>] bus_for_each_dev+0x4c/0x98
[<ffff0000084f81b8>] driver_attach+0x20/0x28
[<ffff0000084f7d08>] bus_add_driver+0x1b8/0x228
[<ffff0000084f93c0>] driver_register+0x60/0xf8
[<ffff0000083fb918>] __pci_register_driver+0x40/0x48

Return EPROBE_DEFER in that case so the driver can register the device
later.

Signed-off-by: Alexandre Belloni <alexandre.belloni@...tlin.com>
---

Hi Greg,

I'm not sure using struct class::p to test for class registration is really
something we should do but this allows to have a small patch.

The other solutions to fix this are:
 - use a similar change but with a boolean to store whether the class has been
   registered.
 - or instead of EPROBE_DEFER, just register the class when the first uio
   device is registered.
 _ or stop allowing to compile uio as a module and use subsys_initcall instead
 of module_init

What do you think?

 drivers/uio/uio.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/uio/uio.c b/drivers/uio/uio.c
index e8f4ac9400ea..3fb84ab3dd99 100644
--- a/drivers/uio/uio.c
+++ b/drivers/uio/uio.c
@@ -853,6 +853,9 @@ int __uio_register_device(struct module *owner,
 	struct uio_device *idev;
 	int ret = 0;
 
+	if (!uio_class.p)
+		return -EPROBE_DEFER;
+
 	if (!parent || !info || !info->name || !info->version)
 		return -EINVAL;
 
-- 
2.17.1