At Thu, 4 Mar 2010 22:17:42 +0200, Jukka Ruohonen <jruohonen%iki.fi@localhost>
wrote:
Subject: Re: unable to build i386 ALL kernel, errors in smbus_acpi.c
(ACPI_SMBUS_COMPONENT undeclared)
>
> Fixed (except the acpitz(4) part; I am not quite sure what it was for...).
Do you mean the change below?
Mostly it's a cleanup of the poor and incomplete error message -- the
comment was noting some differences I had found in OpenBSD which might
improve operation on the machines I was having trouble with.
> > I made one more non-essential change to try to debug problems on some HP
> > hardware:
> >
> > Index: sys/dev/acpi/acpi_tz.c
> > ==================================================================
> > RCS file: /cvs/master/m-NetBSD/main/src/sys/dev/acpi/acpi_tz.c,v
> > retrieving revision 1.59
> > diff -u -r1.59 acpi_tz.c
> > --- sys/dev/acpi/acpi_tz.c 18 Feb 2010 14:10:15 -0000 1.59
> > +++ sys/dev/acpi/acpi_tz.c 25 Feb 2010 01:45:51 -0000
> > @@ -179,7 +179,7 @@
> > ACPI_STATUS rv;
> > ACPI_INTEGER v;
> >
> > -#if 0
> > +#if ACPIVERBOSE
> > sc->sc_flags = ATZ_F_VERBOSE;
> > #endif
> > sc->sc_devnode = aa->aa_node;
> > @@ -255,7 +255,7 @@
> > }
> >
> > if (acpitz_get_integer(dv, "_TMP", &tmp)) {
> > - aprint_error_dev(dv, "failed to evaluate _TMP\n");
> > + aprint_error_dev(dv, "failed to get ACPI _TMP value, no
> > acpitz(4) support?\n");
> > return;
> > }
> >
> > @@ -586,6 +586,11 @@
> > *tmp = ATZ_TMP_INVALID;
> > }
> >
> > +/*
> > + * XXX OpenBSD handles some values differently, "debouncing" invalid
> > readings
> > + * for some number of retries...
> > + */
> > +
> > static int
> > acpitz_get_integer(device_t dv, const char *cm, UINT32 *val)
> > {
--
Greg A. Woods
Planix, Inc.
<woods%planix.com@localhost> +1 416 218 0099 http://www.planix.com/
Attachment:
pgpLRzRmnDjK_.pgp
Description: PGP signature