On Sat, Dec 5, 2020 at 7:41 PM Hans de Goede <hdegoede@xxxxxxxxxx> wrote:
>
> Hi,
>
> On 12/5/20 4:29 PM, Rafael J. Wysocki wrote:
> > From: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
> >
> > The PNP0D80 ("Windows-compatible System Power Management Controller")
> > device ID is used for identifying the special device object providing
> > the LPI (Low-power S0 Idle) _DSM interface [1]. That device object
> > does not supply any operation regions, but it appears in _DEP lists
> > for other devices in the ACPI tables on some systems to enforce
> > specific enumeration ordering that does not matter in Linux.
> >
> > For this reason, _DEP list entries pointing to the device object whose
> > _CID returns PNP0D80 need not be taken into account as real operation
> > region dependencies, so add that device ID to the list of device IDs
> > for which the matching _DEP list entries should be ignored.
> >
> > Accordingly, update the function used for matching device IDs in that
> > list to allow it to check _CID as well as _HID and rename it to
> > acpi_info_matches_ids().
> >
> > Link: https://www.uefi.org/sites/default/files/resources/Intel_ACPI_Low_Power_S0_Idle.pdf # [1]
> > Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
>
> Thank you for doing this, I contemplated doing the exact same
> thing but never got around to it.
>
> One small review remark inline:
>
> > ---
> > drivers/acpi/scan.c | 27 +++++++++++++++++++++------
> > 1 file changed, 21 insertions(+), 6 deletions(-)
> >
> > Index: linux-pm/drivers/acpi/scan.c
> > ===================================================================
> > --- linux-pm.orig/drivers/acpi/scan.c
> > +++ linux-pm/drivers/acpi/scan.c
> > @@ -719,25 +719,40 @@ int acpi_device_add(struct acpi_device *
> > /* --------------------------------------------------------------------------
> > Device Enumeration
> > -------------------------------------------------------------------------- */
> > -static bool acpi_info_matches_hids(struct acpi_device_info *info,
> > - const char * const hids[])
> > +static bool acpi_info_matches_ids(struct acpi_device_info *info,
> > + const char * const ids[])
> > {
> > + struct acpi_pnp_device_id_list *cid_list = NULL;
> > int i;
> >
> > if (!(info->valid & ACPI_VALID_HID))
> > return false;
> >
> > - for (i = 0; hids[i]; i++) {
> > - if (!strcmp(info->hardware_id.string, hids[i]))
> > + if (info->valid & ACPI_VALID_CID)
> > + cid_list = &info->compatible_id_list;
> > +
> > + for (i = 0; ids[i]; i++) {
> > + int j;
> > +
> > + if (!strcmp(info->hardware_id.string, ids[i]))
> > return true;
> > +
> > + if (!cid_list)
> > + continue;
> > +
> > + for (j = 0; j < cid_list->count; j++) {
> > + if (!strcmp(cid_list->ids[j].string, ids[i]))
> > + return true;
> > + }
> > }
> >
> > return false;
> > }
> >
> > /* List of HIDs for which we ignore matching ACPI devices, when checking _DEP lists. */
> > -static const char * const acpi_ignore_dep_hids[] = {
> > +static const char * const acpi_ignore_dep_ids[] = {
> > "INT3396", /* Windows System Power Management Controller */
>
> I think this one can be dropped now, I checked my acpidump / dsdt.dsl
> collection and 45/45 DSDTs declaring a _HID of INT3396 also added a _CID of
> PNP0D80 to this.
Sure, in a separate patch.
Thanks!
