On Thu, Jul 26, 2018 at 03:58:28PM +0200, Bastien Nocera wrote: > On Thu, 2018-07-26 at 16:30 +0300, Andy Shevchenko wrote: > > On Tue, 2018-07-24 at 11:19 -0700, dmitry.torokhov@xxxxxxxxx wrote: > > > On Tue, Jul 24, 2018 at 11:55:52AM +0300, Andy Shevchenko wrote: > > > > The comment was about the bit in FADT table which reflects the > > > > i8042 presence on platform in the question. > > > > Just to be clear, DMI quirks for Apple machines are _already_ in the > > kernel. As Lukas mentioned above it's just matter of adding one more > > parameter to the condition. > > The machine wouldn't boot without this patch, but as you can imagine, > this was 10 years ago, the machine is long gone, and I don't remember > whether I used EFI or BIOS booting, or anything else about the setup. On my MacBookPro9,1, the "8042 Present on ports 60/64" bit is set to 0 and I do see the "ACPI: i8042 controller is absent" message emitted by arch/x86/kernel/acpi/boot.c if I enable debug messages. I've tested what happens if I simulate that the bit is set to 1 by adding acpi_gbl_FADT.boot_flags |= ACPI_FADT_8042; in acpi_parse_fadt(). This results in a one second delay on boot, but otherwise no further issues: [ 2.484586] i8042: PNP: No PS/2 controller found. [ 2.484614] i8042: Probing ports directly. [ 3.521220] i8042: No controller found I grepped through my collection of Mac ACPI dumps and found one from 2013 for the MacBookAir5,1 which incorrectly sets the 8042 bit in the FACP to 1. However I then downloaded recent firmware updates from Apple and checked what the bit is set to nowadays: https://support.apple.com/en-us/HT201518 I checked the MacBookAir5,1 and MacBookAir1,1 firmware and they both correctly set the 8042 bit to 0. The bottom line is that Apple did set the bit incorrectly in the past, but this doesn't seem to have grave consequences with contemporary kernels (only a 1 second delay on boot) and users may fix it by updating the firmware. So it appears there's no need anymore for the patch. Andy, why did you dig this out now? Cleaning stale bugzilla/patchwork entries? :-) Thanks, Lukas -- To unsubscribe from this list: send the line "unsubscribe linux-input" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
