All of lore.kernel.org
 help / color / mirror / Atom feed
From: Gilles Chanteperdrix <gilles.chanteperdrix@xenomai.org>
To: Gajamohan Mohanarajah <gajamohan.m@domain.hid>
Cc: xenomai@xenomai.org, Jan Kiszka <jan.kiszka@domain.hid>
Subject: Re: [Xenomai-help] switchtest/xeon-test fails: Help needed (Ubuntu/Core i7)
Date: Mon, 27 Sep 2010 19:23:17 +0200	[thread overview]
Message-ID: <4CA0D305.6030306@domain.hid> (raw)
In-Reply-To: <AANLkTimYM+gKrWpJkRuPLVEBKy-E+mkhAjmj2tPhqA-e@domain.hid>

Gajamohan Mohanarajah wrote:
> Hi Gilles,
> 
>> Could you capture without truncating the lines? (option -w of minicom
>> itf it is indeed the culprit). And your configuration as NUMA enabled,
>> please disable it.
> With NUMA disabled, and minicom -w, I am attaching the log files. I
> spotted one more difference in the log files
> Without APIC (working)
> [    0.000000] Processors: 4
> [    0.000000] SMP: Allowing 4 CPUs, 0 hotplug CPUs
> With APIC (not working)
> [    0.000000] SMP: Allowing 8 CPUs, 0 hotplug CPUs
> 
> 
>> The only thing different between the two logs seems to be the IRQ routing:
>> working:
>> [    2.335156] ahci 0000:00:1f.2: PCI->APIC IRQ transform: INT B -> IRQ
>> 19
>> not working:
>> [    3.614523] ahci 0000:00:1f.2: PCI INT B -> GSI 19 (level, low) ->
>> IRQ 19
>>
>> On two core i7 I have access to, one gives PCI-MSI-dege, the other
>> IO-APIC-fasteoi, none is level. Could you show us the result of
>> grep ahci /proc/interrupts
>> in the working and non-working case?
> Without APIC (working)
>  26:       4613          0      16570          0   PCI-MSI-edge      ahci
> 
> With APIC (not working) - from initramfs
> 32:          0          0          0          0          0          0
>         0          0   PCI-MSI-edge      ahci

Please do not confuse APIC with ACPI. The non working configuration also
has only 4 cpus (I guess it does not handle hyperthreading).

> 
>> Do you have MSI enabled?
> yes, CONFIG_ARCH_SUPPORTS_MSI=y (I am attaching my config file with
> ACPI turned off and the diff file between the two configs for your
> reference)
> 
> Thank you very much for your help.

The irq line seems mis-configured.

Could you try progressively:
- removing acpi=force on the command line
- disabling CONFIG_X86_REROUTE_FOR_BROKEN_BOOT_IRQS
- disabling CONFIG_ATA_ACPI

I am a bit running out of steam here, maybe Philippe or Jan could help
better?

-- 
					    Gilles.


  reply	other threads:[~2010-09-27 17:23 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-09-24 20:05 [Xenomai-help] switchtest/xeon-test fails: Help needed (Ubuntu/Core i7) Gajamohan Mohanarajah
2010-09-24 20:31 ` Gilles Chanteperdrix
2010-09-24 22:15   ` Gajamohan Mohanarajah
2010-09-25  8:32     ` Gilles Chanteperdrix
2010-09-25 21:20       ` Gajamohan Mohanarajah
2010-09-25 21:39         ` Gilles Chanteperdrix
2010-09-26 11:37           ` Gajamohan Mohanarajah
2010-09-26 11:41             ` Gilles Chanteperdrix
2010-09-26 12:38               ` Gajamohan Mohanarajah
2010-09-26 12:42                 ` Gilles Chanteperdrix
2010-09-27  9:29                   ` Gajamohan Mohanarajah
2010-09-27 14:11                     ` Gilles Chanteperdrix
2010-09-27 17:06                       ` Gajamohan Mohanarajah
2010-09-27 17:23                         ` Gilles Chanteperdrix [this message]
2010-09-27 20:54                           ` Gajamohan Mohanarajah
2010-09-27 21:00                             ` Gilles Chanteperdrix
2010-09-28 10:39                               ` Gajamohan Mohanarajah
2010-09-28 11:09                                 ` Gilles Chanteperdrix
2010-09-26 12:43                 ` Gilles Chanteperdrix
2010-09-24 21:21 ` Stefan Kisdaroczi
2010-09-24 21:24   ` Gilles Chanteperdrix
2010-09-24 22:08     ` Stefan Kisdaroczi
2010-09-24 22:11       ` Gilles Chanteperdrix
2010-09-25 21:40       ` Gilles Chanteperdrix

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=4CA0D305.6030306@domain.hid \
    --to=gilles.chanteperdrix@xenomai.org \
    --cc=gajamohan.m@domain.hid \
    --cc=jan.kiszka@domain.hid \
    --cc=xenomai@xenomai.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.