Re: [PATCH] NFS regression in 2.6.26?, "task blocked for more than 120 seconds"

[Cyrill Gorcunov <gorcunov@gmail.com>]

[Glauber Costa - Mon, Oct 20, 2008 at 11:15:56AM -0200]
| On Mon, Oct 20, 2008 at 4:51 AM, Max Kellermann <mk@cm4all.com> wrote:
| > On 2008/10/17 16:33, Glauber Costa <glommer@redhat.com> wrote:
| >> That's probably something related to apic congestion.
| >> Does the problem go away if the only thing you change is this:
| >>
| >>
| >> > @@ -891,11 +897,6 @@ do_rest:
| >> >             store_NMI_vector(&nmi_high, &nmi_low);
| >> >
| >> >             smpboot_setup_warm_reset_vector(start_ip);
| >> > -           /*
| >> > -            * Be paranoid about clearing APIC errors.
| >> > -           */
| >> > -           apic_write(APIC_ESR, 0);
| >> > -           apic_read(APIC_ESR);
| >> >     }
| >>
| >>
| >> Please let me know.
| >
| > Hello Glauber,
| >
| > I have rebooted the server with 2.6.27.1 + this patchlet an hour ago.
| > No problems since.
| >
| > Hardware: Compaq P4 Xeon server, Broadcom CMIC-WS / CIOB-X2 board.
| > Tell me if you need more detailed information.
| >
| 
| There's a patch in flight from cyrill that probably fixes your problem:
| http://lkml.org/lkml/2008/9/15/93
| 
| The checks are obviously there for a reason, and we can't just wipe
| them out unconditionally ;-) So can you check please that you are also
| covered by the case provided?

Actually I'll wonder if it help. Do Xeon processors really not
have ESR register and not integrated?

...

		- Cyrill -