xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Bastian Blank <waldi@debian.org>
To: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>,
	"xen-devel@lists.xensource.com" <xen-devel@lists.xensource.com>,
	Keir Fraser <keir.fraser@eu.citrix.com>,
	"Zhang, Xiantao" <xiantao.zhang@intel.com>
Subject: Re: pvops-2.6.32 - Interrupt routing problem
Date: Fri, 19 Mar 2010 12:39:04 +0100	[thread overview]
Message-ID: <20100319113904.GA29200@wavehammer.waldi.eu.org> (raw)
In-Reply-To: <20100316182053.GA2258@wavehammer.waldi.eu.org>

On Tue, Mar 16, 2010 at 07:20:53PM +0100, Bastian Blank wrote:

Okay, lets go through the log again.

Initial readout before dom0 starts. Only 0 and 9 masked.

> (XEN)  NR Log Phy Mask Trig IRR Pol Stat Dest Deli Vect:   
> (XEN)  00 000 00  1    0    0   0   0    0    0    00
> (XEN)  01 000 00  0    0    0   0   0    0    0    28
> (XEN)  02 000 00  0    0    0   0   0    0    0    F0
> (XEN)  03 000 00  0    0    0   0   0    0    0    30
> (XEN)  04 000 00  0    0    0   0   0    0    0    F1
> (XEN)  05 000 00  0    0    0   0   0    0    0    38
> (XEN)  06 000 00  0    0    0   0   0    0    0    40
> (XEN)  07 000 00  0    0    0   0   0    0    0    48
> (XEN)  08 000 00  0    0    0   0   0    0    0    50
> (XEN)  09 000 00  1    1    0   1   0    0    0    58
> (XEN)  0a 000 00  0    0    0   0   0    0    0    60
> (XEN)  0b 000 00  0    0    0   0   0    0    0    68
> (XEN)  0c 000 00  0    0    0   0   0    0    0    70
> (XEN)  0d 000 00  0    0    0   0   0    0    0    78
> (XEN)  0e 000 00  0    0    0   0   0    0    0    88
> (XEN)  0f 000 00  0    0    0   0   0    0    0    90

Dom0 tries to set the first 16 IO-APIC pins to edge.

> (XEN) IOAPIC[0]: Set PCI routing entry (8-5 -> 0x38 -> IRQ 5 Mode:0 Active:0)
> (XEN) IOAPIC[0]: Set PCI routing entry (8-6 -> 0x40 -> IRQ 6 Mode:0 Active:0)
> (XEN) IOAPIC[0]: Set PCI routing entry (8-7 -> 0x48 -> IRQ 7 Mode:0 Active:0)
> (XEN) IOAPIC[0]: Set PCI routing entry (8-8 -> 0x50 -> IRQ 8 Mode:0 Active:0)
> (XEN) IOAPIC[0]: Set PCI routing entry (8-9 -> 0x58 -> IRQ 9 Mode:0 Active:0)
> (XEN) IOAPIC[0]: Set PCI routing entry (8-10 -> 0x60 -> IRQ 10 Mode:0 Active:0)
> (XEN) IOAPIC[0]: Set PCI routing entry (8-11 -> 0x68 -> IRQ 11 Mode:0 Active:0)
> (XEN) IOAPIC[0]: Set PCI routing entry (8-12 -> 0x70 -> IRQ 12 Mode:0 Active:0)
> (XEN) IOAPIC[0]: Set PCI routing entry (8-13 -> 0x78 -> IRQ 13 Mode:0 Active:0)
> (XEN) IOAPIC[0]: Set PCI routing entry (8-14 -> 0x88 -> IRQ 14 Mode:0 Active:0)
> (XEN) IOAPIC[0]: Set PCI routing entry (8-15 -> 0x90 -> IRQ 15 Mode:0 Active:0)
> (XEN) Pin 8-9 already programmed

Dom0 have a console.

> [    0.000000] Linux version 2.6.32-4-xen-amd64 (Debian 2.6.32-10~xen.1) (waldi@debian.org) (gcc version 4.3.4 (Debian 4.3.4-8) ) #1 SMP Tue Mar 16 07:46:55 UTC 2010

ACPI only reports overrides for 2 and 9.

> [    0.000000] Already setup the GSI :0
> [    0.000000] Already setup the GSI :1
> [    0.000000] Already setup the GSI :2
> [    0.000000] Already setup the GSI :3
> [    0.000000] Already setup the GSI :4
> [    0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 high edge)
> [    0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 low level)
> [    0.000000] xen: sci override: source_irq=9 global_irq=9 trigger=c polarity=3
> [    0.000000] xen_allocate_pirq: returning irq 9 for gsi 9
> [    0.000000] Already setup the GSI :9
> [    0.000000] xen: acpi sci 9

Readout after dom0 booted. Several other pins masked, all set to edge.

> (XEN)  NR Log Phy Mask Trig IRR Pol Stat Dest Deli Vect:   
> (XEN)  00 000 00  1    0    0   0   0    0    0    00
> (XEN)  01 009 09  0    0    0   0   0    0    0    81
> (XEN)  02 000 00  0    0    0   0   0    0    0    F0
> (XEN)  03 000 00  0    0    0   0   0    0    0    30
> (XEN)  04 000 00  0    0    0   0   0    0    0    F1
> (XEN)  05 000 00  1    0    0   0   0    0    0    38
> (XEN)  06 000 00  1    0    0   0   0    0    0    40
> (XEN)  07 000 00  1    0    0   0   0    0    0    48
> (XEN)  08 009 09  0    0    0   0   0    0    0    89
> (XEN)  09 000 00  0    0    0   0   0    0    0    58
> (XEN)  0a 000 00  1    0    0   0   0    0    0    60
> (XEN)  0b 000 00  1    0    0   0   0    0    0    68
> (XEN)  0c 009 09  0    0    0   0   0    0    0    79
> (XEN)  0d 000 00  1    0    0   0   0    0    0    78
> (XEN)  0e 000 00  1    0    0   0   0    0    0    88
> (XEN)  0f 000 00  1    0    0   0   0    0    0    90

Load of driver which drives interrupt 14. It tries to reprogram it as
level driven.

> [   43.620022] xen_allocate_pirq: returning irq 14 for gsi 14
> (XEN) Pin 8-14 already programmed
> (XEN) IOAPIC[0]: Set PCI routing entry (8-14 -> 0x88 -> IRQ 14 Mode:1 Active:1)

Readout after.

> (XEN)  NR Log Phy Mask Trig IRR Pol Stat Dest Deli Vect:   
> (XEN)  00 000 00  1    0    0   0   0    0    0    00
> (XEN)  01 009 09  0    0    0   0   0    0    0    81
> (XEN)  02 000 00  0    0    0   0   0    0    0    F0
> (XEN)  03 000 00  0    0    0   0   0    0    0    30
> (XEN)  04 000 00  0    0    0   0   0    0    0    F1
> (XEN)  05 000 00  1    0    0   0   0    0    0    38
> (XEN)  06 000 00  1    0    0   0   0    0    0    40
> (XEN)  07 000 00  1    0    0   0   0    0    0    48
> (XEN)  08 009 09  0    0    0   0   0    0    0    89
> (XEN)  09 000 00  0    0    0   0   0    0    0    58
> (XEN)  0a 000 00  1    0    0   0   0    0    0    60
> (XEN)  0b 000 00  1    0    0   0   0    0    0    68
> (XEN)  0c 009 09  0    0    0   0   0    0    0    79
> (XEN)  0d 000 00  1    0    0   0   0    0    0    78
> (XEN)  0e 008 08  0    1    0   1   0    0    0    B2
> (XEN)  0f 000 00  1    0    0   0   0    0    0    90

I did some digging and found similar code that disallows
reprogramming of IO-APIC pins in
arch/x86/kernel/apic/io_apic.c:io_apic_set_pci_routing
So this behaviour is correct.

So the question is: Is it wrong for the kernel to do register_gsi calls
early in bootup for interrupts it does not know anything about?

Bastian

-- 
Even historians fail to learn from history -- they repeat the same mistakes.
		-- John Gill, "Patterns of Force", stardate 2534.7

  parent reply	other threads:[~2010-03-19 11:39 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-03-10 11:19 pvops-2.6.32 - Interrupt routing problem Bastian Blank
2010-03-10 16:08 ` Ian Campbell
2010-03-10 16:19   ` Bastian Blank
2010-03-10 18:17     ` Jeremy Fitzhardinge
2010-03-10 20:32       ` Bastian Blank
2010-03-10 19:40     ` Ian Campbell
2010-03-10 20:00       ` Konrad Rzeszutek Wilk
2010-03-14 14:08 ` Bastian Blank
2010-03-14 15:23   ` Bastian Blank
2010-03-14 15:56     ` Jeremy Fitzhardinge
2010-03-14 16:17       ` Bastian Blank
2010-03-15  4:30         ` Zhang, Xiantao
2010-03-15 10:11           ` Bastian Blank
2010-03-15 10:36             ` Bastian Blank
2010-03-15 13:48             ` Zhang, Xiantao
2010-03-15 14:18               ` Bastian Blank
2010-03-15 14:38                 ` Zhang, Xiantao
2010-03-15 21:15               ` Bastian Blank
2010-03-16  1:31                 ` Konrad Rzeszutek Wilk
2010-03-16  8:18                   ` Bastian Blank
2010-03-16 15:32                     ` Konrad Rzeszutek Wilk
2010-03-16 18:20                       ` Bastian Blank
2010-03-17  2:05                         ` Konrad Rzeszutek Wilk
2010-03-19 11:39                         ` Bastian Blank [this message]
2010-03-19 12:13                           ` Bastian Blank
2010-03-21 21:47                             ` Bastian Blank
2010-03-22  6:24                               ` Jeremy Fitzhardinge
2010-03-23 11:37                               ` Jan Beulich
2010-03-23 12:37                                 ` Bastian Blank
2010-03-23 14:04                                   ` Jan Beulich
2010-03-23 15:49                                     ` Bastian Blank
2010-03-19 14:15                           ` Zhang, Xiantao
2010-03-19 14:36                             ` Bastian Blank
2010-03-19  2:55                     ` Zhang, Xiantao
2010-03-19 11:06                       ` Bastian Blank
2010-03-14 16:41       ` Bastian Blank

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20100319113904.GA29200@wavehammer.waldi.eu.org \
    --to=waldi@debian.org \
    --cc=jeremy@goop.org \
    --cc=keir.fraser@eu.citrix.com \
    --cc=konrad.wilk@oracle.com \
    --cc=xen-devel@lists.xensource.com \
    --cc=xiantao.zhang@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).