i2c: core: Fix double-free of fwnode in i2c_unregister_device()
Subsystems (3)
PATCH 1/1 i2c: core: Fix double-free of fwnode in i2c_unregister_device()
Before commit df6d7277e552 ("i2c: core: Do not dereference fwnode in struct device"), i2c_unregister_device() only called fwnode_handle_put() on of_node-s in the form of calling of_node_put(client->dev.of_node).
But after this commit the i2c_client's fwnode now unconditionally gets fwnode_handle_put() on it.
When the i2c_client has no primary (ACPI / OF) fwnode but it does have a software fwnode, the software-node will be the primary node and fwnode_handle_put() will put() it.
But for the software fwnode device_remove_software_node() will also put() it leading to a double free:
[ 82.665598] ------------[ cut here ]------------ [ 82.665609] refcount_t: underflow; use-after-free. [ 82.665808] WARNING: CPU: 3 PID: 1502 at lib/refcount.c:28 refcount_warn_saturate+0xba/0x11 ... [ 82.666830] RIP: 0010:refcount_warn_saturate+0xba/0x110 ... [ 82.666962] <TASK> [ 82.666971] i2c_unregister_device+0x60/0x90
Fix this by not calling fwnode_handle_put() when the primary fwnode is a software-node.
--- drivers/i2c/i2c-core-base.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/i2c/i2c-core-base.c +7 -1
@@ -1066,7 +1066,13 @@ void i2c_unregister_device(struct i2c_client *client)of_node_clear_flag (to_of_node (fwnode),OF_POPULATED ); else if (is_acpi_device_node(fwnode)) acpi_device_clear_enumerated(to_acpi_device_node(fwnode)); -fwnode_handle_put (fwnode ); + +/* +* If the primary fwnode is a software node it is free-ed by +* device_remove_software_node() below, avoid double-free. +*/ +if (!is_software_node (fwnode)) +fwnode_handle_put (fwnode);device_remove_software_node (&client->dev);device_unregister (&client->dev);
Signature · 1 line
2.49.0
Quoted 23 lines · Hans de Goede
Before commit
df6d7277e552("i2c: core: Do not dereference fwnode in struct device"),i2c_unregister_device()only calledfwnode_handle_put()on of_node-s in the form of callingof_node_put(client->dev.of_node).But after this commit the i2c_client's fwnode now unconditionally gets
fwnode_handle_put()on it.When the
i2c_clienthas no primary (ACPI / OF) fwnode but it does have a software fwnode, the software-node will be the primary node andfwnode_handle_put()willput()it.But for the software fwnode
device_remove_software_node()will alsoput()it leading to a double free:[ 82.665598] ------------[ cut here ]------------ [ 82.665609] refcount_t: underflow; use-after-free. [ 82.665808] WARNING: CPU: 3 PID: 1502 at lib/refcount.c:28 refcount_warn_saturate+0xba/0x11 ... [ 82.666830] RIP: 0010:refcount_warn_saturate+0xba/0x110 ... [ 82.666962] <TASK> [ 82.666971] i2c_unregister_device+0x60/0x90
Quoted 2 lines
Fix this by not calling
fwnode_handle_put()when the primary fwnode is a software-node.
Thanks for the fix!
It doesn't look elegant, but I think it's all due to the (current) design of the fwnode linked list.
Signature · 2 lines
With Best Regards, Andy Shevchenko
Wolfram Sang
AppliedQuoted 7 lines · Hans de Goede
Before commit
df6d7277e552("i2c: core: Do not dereference fwnode in struct device"),i2c_unregister_device()only calledfwnode_handle_put()on of_node-s in the form of callingof_node_put(client->dev.of_node).But after this commit the i2c_client's fwnode now unconditionally gets
fwnode_handle_put()on it.
Quoted 7 lines
When the
i2c_clienthas no primary (ACPI / OF) fwnode but it does have a software fwnode, the software-node will be the primary node andfwnode_handle_put()willput()it.But for the software fwnode
device_remove_software_node()will alsoput()it leading to a double free:
Applied to for-next, thanks!