linux-kernel - Re: Kernel Options: To use MSI-X or not?

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Wed, 04 Nov 2009 18:04:08 -0600
From:	Robert Hancock <hancockrwd@...il.com>
To:	Valdis.Kletnieks@...edu
CC:	Justin Piszcz <jpiszcz@...idpixels.com>,
	linux-kernel@...r.kernel.org
Subject: Re: Kernel Options: To use MSI-X or not?

On 11/04/2009 11:36 AM, Valdis.Kletnieks@...edu wrote:
> On Wed, 04 Nov 2009 11:53:18 EST, Justin Piszcz said:
>
>> Example, 3ware cards/Intel Audio Drivers you can use MSI-X instead of
>> having several devices use/share an interrupt.
>>
>> What are the pros/cons to using MSI-X vs. having devices share an IRQ?
>
> Cons: Sometimes, you'll find a device that claims to support MSI but in fact is
> to some extent borked...
>
> I'm the proud owner of a Dell Latitude D820, which includes this PCI:
>
> 00:1b.0 Audio device: Intel Corporation 82801G (ICH7 Family) High Definition Audio Controller (rev 01)
>          Subsystem: Dell Device 01cc
>          Flags: bus master, fast devsel, latency 0, IRQ 30
>          Memory at efffc000 (64-bit, non-prefetchable) [size=16K]
>          Capabilities: [50] Power Management version 2
>          Capabilities: [60] MSI: Enable+ Count=1/1 Maskable- 64bit+
>          Capabilities: [70] Express Root Complex Integrated Endpoint, MSI 00
>          Capabilities: [100] Virtual Channel<?>
>          Capabilities: [130] Root Complex Link<?>
>          Kernel driver in use: HDA Intel
>
> However, actually trying to enable MSI gets me this at some pseudo-random
> time after booting (one for each of 7 boots):
>
> [    4.519264] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x00df0900
> [   22.788015] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x008f0c00
> [    4.719263] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x00df0900
> [   12.497263] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x00df0900
> [  815.350265] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x00df0900
> [  595.171267] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x00171700
> [    4.968263] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x00df0900
>
> (Am slightly mystified about the resets at 4 and 12 seconds - at that point,
> we're not up far enough to actually try to use the audio. I suspect udev trying
> to initialize the card - except sometimes it works and runs just fine with
> MSI for quite some time:
>
> [14267.701267] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x00df0900
> [12841.892050] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x00df0900
> [10647.783274] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x001f1500
> [15091.102034] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x00171700
> [20085.955293] ALSA sound/pci/hda/hda_intel.c:683: No response from codec, disabling MSI: last cmd=0x00df0700

This may not be an actual MSI problem (in fact I'd say it's probably 
not). Currently hda_intel disables MSI after any codec response timeout, 
so it could be it would have worked if it retried while still using MSI. 
I believe Takashi was going to change the driver to be less sensitive 
about this.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/