Re: 2.6.13-rc3-mm3

From: Andrew Morton (akpm_at_osdl.org)
Date: 07/29/05

  • Next message: David S. Miller: "Re: iptables redirect is broken on bridged setup"
    Date:	Fri, 29 Jul 2005 12:33:30 -0700
    To: Michael Thonke <tk-shockwave@web.de>
    
    

    Michael Thonke <tk-shockwave@web.de> wrote:
    >
    > Andrew Morton schrieb:
    > > Michael Thonke <iogl64nx@gmail.com> wrote:
    > >> here again I have two problems. With 2.6.13-rc3-mm3 I have problems
    > >> using my SATA drives on Intel ICH6.
    > >> The kernel can't route there IRQs or can't discover them. the option
    > >> irqpoll got them to work now.
    > >> The problem is new because 2.6.13-rc3[-mm1,mm2] work without any problems.
    > >
    > > OK. Please generate the full dmesg output for -mm2 and for -mm3 and run
    > > `diff -u dmesg.mm2 dmesg.mm3' and send it? And keep those files because we
    > > may end up needing to add them to an acpi bugzilla entry ;)
    >
    > Well I did a little mistake..it only worked correctly up to
    > 2.6.13-rc3-mm1 but this dmesg output I have.
    >
    > Well as I save mm[2,3] are unable to setup the correct IRQs for the
    > devices..and please note that 2.6.13-rc3-mm3 only booted with irqpoll so
    > its in the dmesg output "dmesg.mm3"
    > Normaly the IRQ routed to something about 1xx now they are 1-21?! Caused
    > by irqpoll?
    >

    Are these problems only present in -mm kernels? Does 2.6.13-rc4 work OK?

    > > Odd trace. Do you have CONFIG_KALLSYMS enabled? If not, please turn it on.
    >
    > Mh I tried but my system freezes on boot then. And screen leaves blank.
    > >

    Oh geeze.

    @@ -53,10 +23,18 @@
     Enabling fast FPU save and restore... done.
     Enabling unmasked SIMD FPU exception support... done.
     Checking 'hlt' instruction... OK.
    + ACPI-0287: *** Error: Region SystemMemory(0) has no handler
    + ACPI-0127: *** Error: acpi_load_tables: Could not load namespace: AE_NOT_EXIST
    + ACPI-0136: *** Error: acpi_load_tables: Could not load tables: AE_NOT_EXIST
    +ACPI: Unable to load the System Description Tables
     ENABLING IO-APIC IRQs
    -..TIMER: vector=0x31 pin1=2 pin2=0
    +..TIMER: vector=0x31 pin1=2 pin2=-1
     NET: Registered protocol family 16
     PCI: Using configuration type 1
    +ACPI: Subsystem revision 20050708
    +ACPI: Interpreter disabled.

    Well it looks like ACPI committed suicide, so there's probably not much
    point looking at the other things until that gets addressed.

    Would you have time to raise a kernel bugzilla entry for this? Raise it
    against the ACPI AML interpreter, version 20050708 and mention the above
    failure. The output of acpidump (from
    ftp://ftp.kernel.org/pub/linux/kernel/people/lenb/acpi/utils/pmtools-20050727.tar.gz)
    will probably be asked for.

    Thanks.
    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/


  • Next message: David S. Miller: "Re: iptables redirect is broken on bridged setup"

    Relevant Pages

    • Only one processor detected in 8-Way opteron in 32-bit mode
      ... ACPI: IRQ0 used by override. ... Using 8 I/O APICs ... Enabling fast FPU save and restore... ...
      (Linux-Kernel)
    • FC5 Xen crash
      ... I tried to install the latest level on an available desktop and received the following kernel error. ... ACPI: IRQ0 used by override. ... Enabling APIC mode: Flat. ... Initializing CPU#0 ...
      (Fedora)
    • Re: Re: USB flash drive problem
      ... ACPI: IRQ0 used by override. ... Using 1 I/O APICs ... Enabling fast FPU save and restore... ... Using ACPI for IRQ routing ...
      (Debian-User)
    • Re: APIC vs PIC on Thinkpad r50p (Pentium-M)
      ... for IRQ routing it has to be able to use an IOAPIC, ... As far as the lack of power-off, I think the ACPI errors quite likely are ... Enabling interrupts during APM calls. ...
      (Linux-Kernel)
    • APIC vs PIC on Thinkpad r50p (Pentium-M)
      ... kernel compiled with local APIC for uniprocessors. ... Are these due to the ACPI errors indicated in the dmeag output? ... Enabling interrupts during APM calls. ... PCI: PCI BIOS revision 2.10 entry at 0xfd8d6, ...
      (Linux-Kernel)