| Message ID | 20201118133745.65628-1-mika.westerberg@linux.intel.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | thunderbolt: Fix use-after-free in remove_unplugged_switch() | expand |
On Wed, Nov 18, 2020 at 04:37:45PM +0300, Mika Westerberg wrote: > Paulian reported a crash that happens when a dock is unplugged during > hibernation: > > [78436.228217] thunderbolt 0-1: device disconnected > [78436.228365] BUG: kernel NULL pointer dereference, address: 00000000000001e0 > ... > [78436.228397] RIP: 0010:icm_free_unplugged_children+0x109/0x1a0 > ... > [78436.228432] Call Trace: > [78436.228439] icm_rescan_work+0x24/0x30 > [78436.228444] process_one_work+0x1a3/0x3a0 > [78436.228449] worker_thread+0x30/0x370 > [78436.228454] ? process_one_work+0x3a0/0x3a0 > [78436.228457] kthread+0x13d/0x160 > [78436.228461] ? kthread_park+0x90/0x90 > [78436.228465] ret_from_fork+0x1f/0x30 > > This happens because remove_unplugged_switch() calls tb_switch_remove() > that releases the memory pointed by sw so the following lines reference > to a memory that might be released already. > > Fix this by saving pointer to the parent device before calling > tb_switch_remove(). > > Reported-by: Paulian Bogdan Marinca <paulian@marinca.net> > Fixes: 4f7c2e0d8765 ("thunderbolt: Make sure device runtime resume completes before taking domain lock") > Cc: stable@vger.kernel.org > Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> > --- > drivers/thunderbolt/icm.c | 8 +++++--- > 1 file changed, 5 insertions(+), 3 deletions(-) > > diff --git a/drivers/thunderbolt/icm.c b/drivers/thunderbolt/icm.c > index b51fc3f62b1f..05323c442b56 100644 > --- a/drivers/thunderbolt/icm.c > +++ b/drivers/thunderbolt/icm.c > @@ -1976,7 +1976,9 @@ static int complete_rpm(struct device *dev, void *data) > > static void remove_unplugged_switch(struct tb_switch *sw) > { > - pm_runtime_get_sync(sw->dev.parent); > + struct device *parent = sw->dev.parent; > + > + pm_runtime_get_sync(parent); If you are saving a pointer to a structure, shouldn't you increment the reference count? struct device *parent = get_device(sw->dev.parent); Then you know it is valid when you use it later on. Just remember to call put_device() when you are done. thanks, greg k-h
On Wed, Nov 18, 2020 at 02:47:27PM +0100, Greg KH wrote: > On Wed, Nov 18, 2020 at 04:37:45PM +0300, Mika Westerberg wrote: > > Paulian reported a crash that happens when a dock is unplugged during > > hibernation: > > > > [78436.228217] thunderbolt 0-1: device disconnected > > [78436.228365] BUG: kernel NULL pointer dereference, address: 00000000000001e0 > > ... > > [78436.228397] RIP: 0010:icm_free_unplugged_children+0x109/0x1a0 > > ... > > [78436.228432] Call Trace: > > [78436.228439] icm_rescan_work+0x24/0x30 > > [78436.228444] process_one_work+0x1a3/0x3a0 > > [78436.228449] worker_thread+0x30/0x370 > > [78436.228454] ? process_one_work+0x3a0/0x3a0 > > [78436.228457] kthread+0x13d/0x160 > > [78436.228461] ? kthread_park+0x90/0x90 > > [78436.228465] ret_from_fork+0x1f/0x30 > > > > This happens because remove_unplugged_switch() calls tb_switch_remove() > > that releases the memory pointed by sw so the following lines reference > > to a memory that might be released already. > > > > Fix this by saving pointer to the parent device before calling > > tb_switch_remove(). > > > > Reported-by: Paulian Bogdan Marinca <paulian@marinca.net> > > Fixes: 4f7c2e0d8765 ("thunderbolt: Make sure device runtime resume completes before taking domain lock") > > Cc: stable@vger.kernel.org > > Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> > > --- > > drivers/thunderbolt/icm.c | 8 +++++--- > > 1 file changed, 5 insertions(+), 3 deletions(-) > > > > diff --git a/drivers/thunderbolt/icm.c b/drivers/thunderbolt/icm.c > > index b51fc3f62b1f..05323c442b56 100644 > > --- a/drivers/thunderbolt/icm.c > > +++ b/drivers/thunderbolt/icm.c > > @@ -1976,7 +1976,9 @@ static int complete_rpm(struct device *dev, void *data) > > > > static void remove_unplugged_switch(struct tb_switch *sw) > > { > > - pm_runtime_get_sync(sw->dev.parent); > > + struct device *parent = sw->dev.parent; > > + > > + pm_runtime_get_sync(parent); > > If you are saving a pointer to a structure, shouldn't you increment the > reference count? > > struct device *parent = get_device(sw->dev.parent); > > Then you know it is valid when you use it later on. Indeed, I missed that. > Just remember to call put_device() when you are done. Sure I'll do this in v2. Thanks!
diff --git a/drivers/thunderbolt/icm.c b/drivers/thunderbolt/icm.c index b51fc3f62b1f..05323c442b56 100644 --- a/drivers/thunderbolt/icm.c +++ b/drivers/thunderbolt/icm.c @@ -1976,7 +1976,9 @@ static int complete_rpm(struct device *dev, void *data) static void remove_unplugged_switch(struct tb_switch *sw) { - pm_runtime_get_sync(sw->dev.parent); + struct device *parent = sw->dev.parent; + + pm_runtime_get_sync(parent); /* * Signal this and switches below for rpm_complete because @@ -1987,8 +1989,8 @@ static void remove_unplugged_switch(struct tb_switch *sw) bus_for_each_dev(&tb_bus_type, &sw->dev, NULL, complete_rpm); tb_switch_remove(sw); - pm_runtime_mark_last_busy(sw->dev.parent); - pm_runtime_put_autosuspend(sw->dev.parent); + pm_runtime_mark_last_busy(parent); + pm_runtime_put_autosuspend(parent); } static void icm_free_unplugged_children(struct tb_switch *sw)
Paulian reported a crash that happens when a dock is unplugged during hibernation: [78436.228217] thunderbolt 0-1: device disconnected [78436.228365] BUG: kernel NULL pointer dereference, address: 00000000000001e0 ... [78436.228397] RIP: 0010:icm_free_unplugged_children+0x109/0x1a0 ... [78436.228432] Call Trace: [78436.228439] icm_rescan_work+0x24/0x30 [78436.228444] process_one_work+0x1a3/0x3a0 [78436.228449] worker_thread+0x30/0x370 [78436.228454] ? process_one_work+0x3a0/0x3a0 [78436.228457] kthread+0x13d/0x160 [78436.228461] ? kthread_park+0x90/0x90 [78436.228465] ret_from_fork+0x1f/0x30 This happens because remove_unplugged_switch() calls tb_switch_remove() that releases the memory pointed by sw so the following lines reference to a memory that might be released already. Fix this by saving pointer to the parent device before calling tb_switch_remove(). Reported-by: Paulian Bogdan Marinca <paulian@marinca.net> Fixes: 4f7c2e0d8765 ("thunderbolt: Make sure device runtime resume completes before taking domain lock") Cc: stable@vger.kernel.org Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> --- drivers/thunderbolt/icm.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)