[0/4] x86: ACPI and DMI table mapping fixes

Message

Jan Beulich Nov. 23, 2020, 12:38 p.m. UTC

The first three patches fix fallout from the re-work of
acpi_os_{,un}map_memory(): Direct uses of __acpi_map_table() are now
no longer valid once we've reached SYS_STATE_boot. This was originally
noticed by system shutdown no longer working (patch 1), but clearly
extends beyond of this (patches 2 and 3). The last patch relaxes
things such that entering S5 would still work even if there was a
problem with FACS (information collected from there is only needed for
entering S3 and, once we support it, S4 via S4BIOS_REQ).

1: x86/ACPI: fix mapping of FACS
2: x86/ACPI: fix S3 wakeup vector mapping
3: x86/DMI: fix table mapping when one lives above 1Mb
4: x86/ACPI: don't invalidate S5 data when S3 wakeup vector cannot be determined

Jan