On Tue, Jul 05, 2016 at 01:30:23PM +0200, Pali Rohár wrote:

On Tuesday 05 July 2016 12:14:55 Mika Westerberg wrote:

quoted

The whole point of this patch is that we expect that nobody never
uses that OpRegion. I'm 99% sure you don't find a single machine
where it is actually in use.

HP EliteBook 8460p uses it for sure! Here are DSDT snips:


                Method (\_SB.PCI0.LPCB.SMAB, 3, Serialized)
                {
                    If (LEqual (And (Arg0, 0x01), 0x00))
                    {
                        Store (0x01, Local0)
                        Store (\_SB.PCI0.SBUS.SWRB (Arg0, Arg1, Arg2), Local1)
                        If (Local1)
                        {
                            Store (0x00, Local0)
                        }
                    }
                    Else
                    {
                        Store (\_SB.PCI0.SBUS.SRDB (Arg0, Arg1), Local0)
                    }

                    Return (Local0)
                }

Crap, well that is in that 1% then ;-)

...

                Method (ALRD, 1, Serialized)
                {
                    Store (\_SB.PCI0.LPCB.SMAB (0x33, Arg0, 0x00), Local0)
                    Return (Local0)
                }

                Method (ALWR, 2, Serialized)
                {
                    Store (\_SB.PCI0.LPCB.SMAB (0x32, Arg0, Arg1), Local0)
                    Return (Local0)
                }


And ALRD and ALWR methods are used by hp_accel.ko kernel driver.

So are you able to test what happens when you unload the driver? I think
the safest thing to do is that we just pin the driver in the kernel once
we notice the OpRegion is being accessed. Does anyone else have better
ideas?
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html