powerpc: Fix undetected data corruption with P9N DD2.1 VSX CI load em…

…ulation

commit 1da4a02 upstream.

__get_user_atomic_128_aligned() stores to kaddr using stvx which is a
VMX store instruction, hence kaddr must be 16 byte aligned otherwise
the store won't occur as expected.

Unfortunately when we call __get_user_atomic_128_aligned() in
p9_hmi_special_emu(), the buffer we pass as kaddr (ie. vbuf) isn't
guaranteed to be 16B aligned. This means that the write to vbuf in
__get_user_atomic_128_aligned() has the bottom bits of the address
truncated. This results in other local variables being
overwritten. Also vbuf will not contain the correct data which results
in the userspace emulation being wrong and hence undetected user data
corruption.

In the past we've been mostly lucky as vbuf has ended up aligned but
this is fragile and isn't always true. CONFIG_STACKPROTECTOR in
particular can change the stack arrangement enough that our luck runs
out.

This issue only occurs on POWER9 Nimbus <= DD2.1 bare metal.

The fix is to align vbuf to a 16 byte boundary.

Fixes: 5080332 ("powerpc/64s: Add workaround for P9 vector CI load issue")
Cc: stable@vger.kernel.org # v4.15+
Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20201013043741.743413-1-mikey@neuling.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

arch/powerpc/kernel/traps.c

@@ -889,7 +889,7 @@ static void p9_hmi_special_emu(struct pt_regs *regs)
 {
 	unsigned int ra, rb, t, i, sel, instr, rc;
 	const void __user *addr;
-	u8 vbuf[16], *vdst;
+	u8 vbuf[16] __aligned(16), *vdst;
 	unsigned long ea, msr, msr_mask;
 	bool swap;