]> git.baikalelectronics.ru Git - kernel.git/commit
Revert "net: mdiobus: Fix memory leak in __mdiobus_register"
authorPavel Skripkin <paskripkin@gmail.com>
Thu, 30 Sep 2021 17:49:42 +0000 (20:49 +0300)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Tue, 2 Nov 2021 18:46:12 +0000 (19:46 +0100)
commitebe03d227f29ea1db8edf39f9b02034bbcf11ca6
treee3dff30d6cad27b7adf73366fc66d8a2728778dd
parent8e403b09cf441d432fb2157d676ead9ab4f3eb36
Revert "net: mdiobus: Fix memory leak in __mdiobus_register"

commit 10eff1f5788b6ffac212c254e2f3666219576889 upstream.

This reverts commit ab609f25d19858513919369ff3d9a63c02cd9e2e.

This patch is correct in the sense that we _should_ call device_put() in
case of device_register() failure, but the problem in this code is more
vast.

We need to set bus->state to UNMDIOBUS_REGISTERED before calling
device_register() to correctly release the device in mdiobus_free().
This patch prevents us from doing it, since in case of device_register()
failure put_device() will be called 2 times and it will cause UAF or
something else.

Also, Reported-by: tag in revered commit was wrong, since syzbot
reported different leak in same function.

Link: https://lore.kernel.org/netdev/20210928092657.GI2048@kadam/
Acked-by: Yanfei Xu <yanfei.xu@windriver.com>
Signed-off-by: Pavel Skripkin <paskripkin@gmail.com>
Link: https://lore.kernel.org/r/f12fb1faa4eccf0f355788225335eb4309ff2599.1633024062.git.paskripkin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/net/phy/mdio_bus.c