Re: Get PCI resources from ACPI

dmesg diff, followed by full dmesg with 2nd patch applied. No panic, at least in the time it took to grab the dmesg and reboot. Devin 1c1 < OpenBSD 5.9-beta (GENERIC.MP) #10: Fri Jan 8 09:00:37 MST 2016 --- OpenBSD 5.9-beta (GENERIC.MP) #11: Fri Jan 8 10:37:00 MST 2016 18c18 < cpu0:

Re: Get PCI resources from ACPI

On Thu, Jan 07, 2016 at 06:28 +0100, Pablo Méndez Hernández wrote: > Hi team, > > On Thu, Dec 31, 2015 at 10:05 PM, Philip Guenther wrote: > > On Wed, 30 Dec 2015, Mark Kettenis wrote: > > ... > >> Updated diff. Once again the ACPI standard is ambiguous and/or violated > >>

Re: Get PCI resources from ACPI

Firstly diff, secondly all dmesg from unmodified and modified with yours patch kernel. OpenBSD-current amd64 with source code updated using cvs -d$CVSROOT up -D "2016-01-07 05:00" -Pd diff -u /katalogDmesg/nieZmodyfikowaneAcpiMark/dmesg.log /katalogDmesg/zmodyfikowaneAcpiMark/dmesg.log ---

Re: Get PCI resources from ACPI

Hi team, On Thu, Dec 31, 2015 at 10:05 PM, Philip Guenther wrote: > On Wed, 30 Dec 2015, Mark Kettenis wrote: > ... >> Updated diff. Once again the ACPI standard is ambiguous and/or violated >> by the hardware vendors. This should fix at least the ports Dell r620 >> that

Re: Get PCI resources from ACPI

On Wed, 30 Dec 2015, Mark Kettenis wrote: ... > Updated diff. Once again the ACPI standard is ambiguous and/or violated > by the hardware vendors. This should fix at least the ports Dell r620 > that naddy@ told me about. Here's the diff of dmesgs with-vs-without this diff on my yoga12: @@

Get PCI resources from ACPI

It seems more and more server machines ship in a "64-bit" configuration where the BIOS assigns addresses above the 4GB mark to 64-bit BARs. We have a hack in arch/amd64/pci_machdep.c:pci_init_extents() to deal with some of these configurations: /* * ...