Hi, this breaks my ASUS E200HA because of missing acpi_attach_deps in dwiic_acpi_found_hid, see https://marc.info/?l=openbsd-bugs&m=149504791212937&w=2
- C. On Nov 9, 2017 14:07, "joshua stein" <j...@openbsd.org> wrote: On Thu, 09 Nov 2017 at 10:14:16 +0100, Remi Locherer wrote: > On Fri, Nov 03, 2017 at 12:01:15PM -0500, joshua stein wrote: > >> Intel 100 Series laptops have the DesignWare I2C controller >> attaching via PCI instead of ACPI, so move the guts of dwiic(4) into >> ic/ and add dwiic_acpi and dwiic_pci files. Unfortunately the PCI >> attachment still needs to reach into ACPI to discover client >> devices. >> >> There is still an issue with interrupt setup on these machines >> (unrelated to dwiic as far as I can tell), where ioapic interrupts >> registered for client devices like ihidev never fire, but those for >> dwiic (via PCI) do. This diff sets a flag for PCI-attached dwiic >> devices so that ihidev can do polling for client devices, which will >> come in a separate diff. >> >> I need some tests on machines where dwiic/ihidev attaches over ACPI. >> > > With your patch dwiic and also imt do attach. > > dwiic0 at pci0 dev 21 function 0 "Intel 100 Series I2C" rev 0x21: apic 2 > int 16 > iic0 at dwiic0 > dwiic1 at pci0 dev 21 function 1 "Intel 100 Series I2C" rev 0x21: apic 2 > int 17 > iic1 at dwiic1 > ihidev0 at iic1 addr 0x15, vendor 0x4f3 product 0x3035, ELAN1301 > ihidev0: 11 report ids > imt0 at ihidev0: clickpad, 5 contacts > wsmouse0 at imt0 mux 0 > > The touchpad does not work yet. > Let me know if you are interested in anything else from this machine. > Thanks. Has anyone tested this on currently-working machines with ACPI attachment? I am more concerned about not breaking anything right now.
