linux/drivers/base
Jiri Slaby 7e1b1fc4da base: make module_create_drivers_dir race-free
Modules which register drivers via standard path (driver_register) in
parallel can cause a warning:
WARNING: CPU: 2 PID: 3492 at ../fs/sysfs/dir.c:31 sysfs_warn_dup+0x62/0x80
sysfs: cannot create duplicate filename '/module/saa7146/drivers'
Modules linked in: hexium_gemini(+) mxb(+) ...
...
Call Trace:
...
 [<ffffffff812e63a2>] sysfs_warn_dup+0x62/0x80
 [<ffffffff812e6487>] sysfs_create_dir_ns+0x77/0x90
 [<ffffffff8140f2c4>] kobject_add_internal+0xb4/0x340
 [<ffffffff8140f5b8>] kobject_add+0x68/0xb0
 [<ffffffff8140f631>] kobject_create_and_add+0x31/0x70
 [<ffffffff8157a703>] module_add_driver+0xc3/0xd0
 [<ffffffff8155e5d4>] bus_add_driver+0x154/0x280
 [<ffffffff815604c0>] driver_register+0x60/0xe0
 [<ffffffff8145bed0>] __pci_register_driver+0x60/0x70
 [<ffffffffa0273e14>] saa7146_register_extension+0x64/0x90 [saa7146]
 [<ffffffffa0033011>] hexium_init_module+0x11/0x1000 [hexium_gemini]
...

As can be (mostly) seen, driver_register causes this call sequence:
  -> bus_add_driver
    -> module_add_driver
      -> module_create_drivers_dir
The last one creates "drivers" directory in /sys/module/<...>. When
this is done in parallel, the directory is attempted to be created
twice at the same time.

This can be easily reproduced by loading mxb and hexium_gemini in
parallel:
while :; do
  modprobe mxb &
  modprobe hexium_gemini
  wait
  rmmod mxb hexium_gemini saa7146_vv saa7146
done

saa7146 calls pci_register_driver for both mxb and hexium_gemini,
which means /sys/module/saa7146/drivers is to be created for both of
them.

Fix this by a new mutex in module_create_drivers_dir which makes the
test-and-create "drivers" dir atomic.

I inverted the condition and removed 'return' to avoid multiple
unlocks or a goto.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Fixes: fe480a2675 (Modules: only add drivers/ direcory if needed)
Cc: v2.6.21+ <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2016-06-15 19:21:31 -07:00
..
power Merge branches 'pm-cpufreq', 'pm-cpuidle' and 'pm-core' 2016-05-25 21:54:45 +02:00
regmap Merge remote-tracking branches 'regmap/topic/doc' and 'regmap/topic/flat' into regmap-next 2016-05-13 10:36:14 +01:00
attribute_container.c
base.h PM / sleep: prohibit devices probing during suspend/hibernation 2015-11-30 14:47:22 +01:00
bus.c driver core: bus: use to_subsys_private and to_device_private_bus 2016-02-09 17:25:27 -08:00
cacheinfo.c cpu/cacheinfo: Fix teardown path 2015-09-17 06:06:54 -07:00
class.c class_find_device: fix reference to argument "match" 2015-09-29 15:23:14 +02:00
component.c Merge 4.5-rc4 into driver-core-next 2016-02-14 14:29:55 -08:00
container.c
core.c driver core: Do not overwrite secondary fwnode with NULL if it is set 2015-12-07 02:29:23 +01:00
cpu.c drivers/base/cpu.c: use __cpu_*_mask directly 2016-01-20 17:09:18 -08:00
dd.c Revert "base: dd: don't remove driver_data in -EPROBE_DEFER case" 2016-05-03 08:06:06 -07:00
devcoredump.c devcoredump: add scatterlist support 2016-04-15 11:20:32 -07:00
devres.c devm: make allocations numa aware by default 2015-10-09 17:00:33 -04:00
devtmpfs.c wrappers for ->i_mutex access 2016-01-22 18:04:28 -05:00
dma-coherent.c drivers: dma-coherent: use memset_io for DMA_MEMORY_IO mappings 2016-03-22 15:36:02 -07:00
dma-contiguous.c driver core update for 4.4-rc1 2015-11-04 21:50:37 -08:00
dma-mapping.c dma-mapping: remove <asm-generic/dma-coherent.h> 2016-01-20 17:09:18 -08:00
driver.c
firmware_class.c Char/Misc patches for 4.6-rc1 2016-03-17 13:47:50 -07:00
firmware.c
hypervisor.c
init.c
isa.c
Kconfig
Makefile isa: Decouple X86_32 dependency from the ISA Kconfig option 2016-05-01 20:21:02 -07:00
map.c
memory.c memory-hotplug: add automatic onlining policy for the newly added memory 2016-03-15 16:55:16 -07:00
module.c base: make module_create_drivers_dir race-free 2016-06-15 19:21:31 -07:00
node.c Revert "mm: Check if section present during memory block (un)registering" 2015-10-13 10:57:25 -07:00
pinctrl.c drivers/pinctrl: Add the concept of an "init" state 2015-10-27 11:24:23 +01:00
platform-msi.c base: Export platform_msi_domain_[alloc,free]_irqs 2016-01-26 16:33:25 +01:00
platform.c device property: don't bother the drivers with struct property_set 2016-04-09 03:10:50 +02:00
property.c Merge back new device properties material for v4.7. 2016-05-06 22:07:33 +02:00
soc.c base: soc: siplify ida usage 2015-10-04 19:42:22 +01:00
syscore.c
topology.c
transport_class.c