ARC: Improve cmpxchg syscall implementation
authorPeter Zijlstra <peterz@infradead.org>
Tue, 19 Jun 2018 14:22:05 +0000 (17:22 +0300)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Fri, 24 Aug 2018 11:09:15 +0000 (13:09 +0200)
commit09af6c90819d9bdd9f8ce6233308cead86625965
tree6f16792a9d24f62e80aefb2cdddcd7ea73494baa
parente653e79ace50feb02030b5d4728e282c839c24fc
ARC: Improve cmpxchg syscall implementation

[ Upstream commit e8708786d4fe21c043d38d760f768949a3d71185 ]

This is used in configs lacking hardware atomics to emulate atomic r-m-w
for user space, implemented by disabling preemption in kernel.

However there are issues in current implementation:

1. Process not terminated if invalid user pointer passed:
   i.e. __get_user() failed.

2. The reason for this patch was __put_user() failure not being handled
   either, specifically for the COW break scenario.
   The zero page is initially wired up and read from __get_user()
   succeeds. A subsequent write by __put_user() induces a
   Protection Violation, but COW can't finish as Linux page fault
   handler is disabled due to preempt disable.
   And what's worse is we silently return the stale value to user space.
   Fix this specific case by re-enabling preemption and explicitly
   fixing up the fault and retrying the whole sequence over.

Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: linux-arch@vger.kernel.org
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
[vgupta: rewrote the changelog]
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
arch/arc/kernel/process.c