Date: Wed, 20 Apr 2005 09:03:07 +0200 From: Guy Brand <gb@isis.u-strasbg.fr> To: freebsd-acpi@freebsd.org Cc: freebsd-current@freebsd.org Subject: Re: hw.acpi.battery.life always at 100 Message-ID: <20050420070307.GA928@chimie.u-strasbg.fr> In-Reply-To: <42640C2B.20100@root.org> References: <20050417205230.GC837@chimie.u-strasbg.fr> <4263E9BF.4050501@root.org> <20050418192235.GA867@chimie.u-strasbg.fr> <42640C2B.20100@root.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 18 avril at 12:36, Nate Lawson wrote: > I'm adding -current to the CC since a change elsewhere in the kernel may > have triggered this. Is anyone aware why it appears interrupts are now > disabled for a while just after launching init? The change appeared in > 5-STABLE between March 17 and April 17. > > Guy, can you try disabling usb? Also, can you cvsup to dates between > March 17 and April 17 to try to narrow down the exact date this started > occurring? Hi all, Nate, I followed your instructions. First of all disabling USB didn't change anything. I then suped RELENG_5 to several dates in between March 17 and April 17 and found that the problem appeared after Mar-31-05:00 and before Mar-31-11:00. cvsup between these two timestamps shows two changes: Edit src/release/doc/en_US.ISO8859-1/relnotes/common/new.sgml Add delta 1.761.2.38 2005.03.31.06.28.58 hrs Edit src/sys/dev/acpica/acpivar.h Add delta 1.79.2.7 2005.03.31.06.03.59 njl and diffing acpivar.h reveals: @@ -25,7 +25,7 @@ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. * - * $FreeBSD: src/sys/dev/acpica/acpivar.h,v 1.79.2.6 2005/03/02 09:18:41 obrien Exp $ + * $FreeBSD: src/sys/dev/acpica/acpivar.h,v 1.79.2.7 2005/03/31 06:03:59 njl Exp $ */ #ifndef _ACPIVAR_H_ @@ -406,7 +406,7 @@ ACPI_HANDLE acpi_GetReference(ACPI_HANDLE scope, ACPI_OBJECT *obj); #ifndef ACPI_MAX_THREADS -#define ACPI_MAX_THREADS 3 +#define ACPI_MAX_THREADS 1 #endif /* ACPI task kernel thread initialization. */ I suped back to 5-STABLE and changed ACPI_MAX_THREADS accordingly and the problem has now disappeared: # sysctl hw.acpi.battery hw.acpi.battery.life: 87 hw.acpi.battery.time: 167 hw.acpi.battery.state: 1 hw.acpi.battery.units: 1 hw.acpi.battery.info_expire: 5 # uname -a FreeBSD 5.4-STABLE FreeBSD 5.4-STABLE #7: Wed Apr 20 00:49:58 CEST 2005 root@6nq.u-strasbg.fr:/usr/obj/usr/src/sys/GENERIC i386 BTW -CURRENT has the same problem (MFC from 1.91) and the same change fixes it. gb
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050420070307.GA928>