ARM: 8692/1: mm: abort uaccess retries upon fatal signal
authorMark Rutland <mark.rutland@arm.com>
Tue, 22 Aug 2017 10:36:17 +0000 (11:36 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 13 Sep 2017 21:09:46 +0000 (14:09 -0700)
commit84478477d0b8572c6d267492aaaf49acd6fc4db5
tree6fd12371ec6024cf3191a514ffd8f88a9b57b79c
parentf7ec367c8ea7021517c9c04b0022c225d2d0785a
ARM: 8692/1: mm: abort uaccess retries upon fatal signal

commit 746a272e44141af24a02f6c9b0f65f4c4598ed42 upstream.

When there's a fatal signal pending, arm's do_page_fault()
implementation returns 0. The intent is that we'll return to the
faulting userspace instruction, delivering the signal on the way.

However, if we take a fatal signal during fixing up a uaccess, this
results in a return to the faulting kernel instruction, which will be
instantly retried, resulting in the same fault being taken forever. As
the task never reaches userspace, the signal is not delivered, and the
task is left unkillable. While the task is stuck in this state, it can
inhibit the forward progress of the system.

To avoid this, we must ensure that when a fatal signal is pending, we
apply any necessary fixup for a faulting kernel instruction. Thus we
will return to an error path, and it is up to that code to make forward
progress towards delivering the fatal signal.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Steve Capper <steve.capper@arm.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
arch/arm/mm/fault.c