> Date: Sat, 12 Sep 2020 12:48:48 +0200 > From: Marcus Glocker <[email protected]> > Cc: [email protected], [email protected], [email protected] > Content-Type: text/plain; charset=utf-8 > > On Sat, 12 Sep 2020 10:28:23 +0200 (CEST) > Mark Kettenis <[email protected]> wrote: > > > > Date: Sat, 12 Sep 2020 10:00:13 +0200 > > > From: Marcus Glocker <[email protected]> > > > > > > On Sat, 12 Sep 2020 09:34:18 +0200 (CEST) > > > Mark Kettenis <[email protected]> wrote: > > > > > > > > Date: Sat, 12 Sep 2020 09:20:26 +0200 > > > > > From: Marcus Glocker <[email protected]> > > > > > > > > > > On Fri, 11 Sep 2020 23:18:56 +0200 (CEST) > > > > > Mark Kettenis <[email protected]> wrote: > > > > > > > > > > > > Date: Fri, 11 Sep 2020 17:42:23 +0200 > > > > > > > From: Marcus Glocker <[email protected]> > > > > > > > > > > > > > > On Thu, 10 Sep 2020 23:44:38 +0200 > > > > > > > Joerg Jung <[email protected]> wrote: > > > > > > > > > > > > > > > Don’t give up so quickly ;) > > > > > > > > let’s try to make the driver work on your iMac, send me > > > > > > > > dmesg and sysctl hw output please. > > > > > > > > > > > > > > > > Your idea of converting it to ACPI is the right thing to > > > > > > > > do anyways, would be nice to get this working. > > > > > > > > > > > > > > Here we go: > > > > > > > > > > > > > > $ dmesg | grep smc > > > > > > > asmc0 at acpi0: SMC_ (smc-piketon) addr 0x300/0x20: rev > > > > > > > 1.64f564, 276 keys > > > > > > > > > > > > > > $ sysctl -a | grep smc > > > > > > > hw.sensors.asmc0.temp0=27.00 degC (TA0P ambient) > > > > > > > hw.sensors.asmc0.temp1=42.00 degC (TC0H cpu0 heatsink) > > > > > > > hw.sensors.asmc0.temp2=55.00 degC (TG0D gpu0 diode) > > > > > > > hw.sensors.asmc0.temp3=53.00 degC (TG0H gpu0 heatsink) > > > > > > > hw.sensors.asmc0.temp4=38.00 degC (TL0P lcd proximity) > > > > > > > hw.sensors.asmc0.temp5=41.00 degC (TO0P optical drive) > > > > > > > hw.sensors.asmc0.temp6=50.00 degC (Tm0P memory controller) > > > > > > > hw.sensors.asmc0.fan0=998 RPM (ODD, right mid rear) > > > > > > > hw.sensors.asmc0.fan1=1158 RPM (HDD, center mid rear) > > > > > > > hw.sensors.asmc0.fan2=1200 RPM (CPU, left lower rear) > > > > > > > > > > > > > > Does that work for you guys? > > > > > > > > > > > > $ dmesg | grep smc > > > > > > asmc0 at acpi0: SMC_ (smc-napa) addr 0x300/0x20: rev 1.3f503, > > > > > > 137 keys > > > > > > > > > > > > $ sysctl -a | grep smc > > > > > > hw.sensors.asmc0.temp0=63.00 degC (TC0D cpu0 die core) > > > > > > hw.sensors.asmc0.temp1=55.00 degC (TC0H cpu0 heatsink) > > > > > > hw.sensors.asmc0.temp2=58.00 degC (TC0P cpu0 proximity) > > > > > > hw.sensors.asmc0.temp3=52.00 degC (TN0P northbridge proximity) > > > > > > hw.sensors.asmc0.temp4=52.00 degC (TN1P northbridge 2) > > > > > > hw.sensors.asmc0.fan0=2077 RPM (Master, left upper front) > > > > > > > > > > > > So yes, this works for me. > > > > > > > > > > Cool. > > > > > > > > > > > You'll need to make changes to the i386 GENERIC kernel as > > > > > > well. > > > > > > > > > > Yep, done. > > > > > > > > > > > And I'd like to ask you to make one small change... > > > > > > > > > > > > +const char *acpiapplesmc_hids[] = { > > > > > > > > > > > > ...can you rename this variable to asmc_hids[]? > > > > > > > > > > Of course, copy/pasto, thanks for spotting. > > > > > > > > > > Also Joerg did suggest to hard code smc0 in GENERIC like it was > > > > > before, since it won't be possible that there is more than 1 SMC > > > > > available on a machine. > > > > > > > > Meh. We tend to only do that if there is a fundamental reason why > > > > there can only be one. But it doesn't hurt. > > > > > > Well, yeah, would be my initial preference as well. Maybe Joerg can > > > give a further explanation to underline why only one SMC can be > > > found. If we are unsure we still can change it to 'asmc*'. > > > > In practice there will only be one. But there is nothing in the ACPI > > driver to prevent it from supporting multiple SMC chips. > > > > For the ISA driver it made sense to have amsc0 at isa? since there can > > only be one at the specified address. > > Makes sense to me. I would finally also prefer to go with 'asmc*', and > it won't hurt either, right?
Yes. ok kettenis@ > Index: sys/arch/amd64/conf/GENERIC > =================================================================== > RCS file: /cvs/src/sys/arch/amd64/conf/GENERIC,v > retrieving revision 1.491 > diff -u -p -u -p -r1.491 GENERIC > --- sys/arch/amd64/conf/GENERIC 12 Sep 2020 07:47:26 -0000 1.491 > +++ sys/arch/amd64/conf/GENERIC 12 Sep 2020 10:46:53 -0000 > @@ -68,7 +68,7 @@ glkgpio* at acpi? > sdhc* at acpi? > acpicbkbd* at acpi? > acpials* at acpi? > -asmc0 at acpi? # Apple SMC > +asmc* at acpi? # Apple SMC > tpm* at acpi? > acpihve* at acpi? > acpisurface* at acpi? > Index: sys/arch/i386/conf/GENERIC > =================================================================== > RCS file: /cvs/src/sys/arch/i386/conf/GENERIC,v > retrieving revision 1.852 > diff -u -p -u -p -r1.852 GENERIC > --- sys/arch/i386/conf/GENERIC 12 Sep 2020 07:47:26 -0000 1.852 > +++ sys/arch/i386/conf/GENERIC 12 Sep 2020 10:46:54 -0000 > @@ -62,7 +62,7 @@ acpivideo* at acpi? > acpivout* at acpivideo? > acpipwrres* at acpi? > aibs* at acpi? > -asmc0 at acpi? # Apple SMC > +asmc* at acpi? # Apple SMC > > option PCIVERBOSE > option EISAVERBOSE > >
