From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754777Ab1JBXJp (ORCPT ); Sun, 2 Oct 2011 19:09:45 -0400 Received: from mail.klingt.org ([86.59.21.178]:47171 "EHLO klingt.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754807Ab1JBXJe (ORCPT ); Sun, 2 Oct 2011 19:09:34 -0400 X-Greylist: delayed 455 seconds by postgrey-1.27 at vger.kernel.org; Sun, 02 Oct 2011 19:09:34 EDT From: Tim Blechmann To: "linux-kernel@vger.kernel.org" Subject: threadirqs && irq nobody cared Date: Mon, 03 Oct 2011 01:01:57 +0200 Message-ID: <1515462.kJuO53sx2E@moka> X-KMail-UnExpanded-To: "linux-kernel@vger.kernel.org" User-Agent: KMail/4.7.1 (Linux/3.1.0-rc8+; KDE/4.7.1; x86_64; ; ) MIME-Version: 1.0 Content-Transfer-Encoding: 7Bit Content-Type: text/plain; charset="us-ascii" X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.3.9 (klingt.org [86.59.21.178]); Mon, 03 Oct 2011 01:01:58 +0200 (CEST) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org hi all, i've experienced problems with recent kernels that irqs are disabled. there was a similar report some time ago on lkml, suggesting that it might be an issue with threaded irqs [1]. my issue looked similar: [ 14.726028] irq 16: nobody cared (try booting with the "irqpoll" option) [ 14.726032] Pid: 46, comm: irq/16-uhci_hcd Tainted: P 3.0.1 #12 [ 14.726035] Call Trace: [ 14.726042] [] __report_bad_irq+0x36/0xd0 [ 14.726045] [] note_interrupt+0x136/0x1f0 [ 14.726049] [] ? local_bh_enable+0x57/0xc0 [ 14.726052] [] ? irq_thread_fn+0x50/0x50 [ 14.726054] [] irq_thread+0x1af/0x1e0 [ 14.726057] [] ? irq_finalize_oneshot+0x110/0x110 [ 14.726060] [] kthread+0x8c/0xa0 [ 14.726065] [] kernel_thread_helper+0x4/0x10 [ 14.726068] [] ? kthread_worker_fn+0x190/0x190 [ 14.726071] [] ? gs_change+0x13/0x13 [ 14.726072] handlers: [ 14.726074] [] irq_default_primary_handler threaded [] usb_hcd_irq [ 14.726080] [] irq_default_primary_handler threaded [] ahci_interrupt [ 14.726093] [] irq_default_primary_handler threaded [] snd_hdsp_interrupt [ 14.726098] Disabling IRQ #16 for me the issue mainly occurs under high cpu load. the issue in [1] is related to radeon_driver_irq_handler_kms and usb_hcd_irq. however i have mutually unloaded the ahci, uhci-hcd and snd-hdsp modules and the problem always occurred sooner or later, so it does not seem to be related to one of the drivers. however i haven't experienced it before 3.0. i've currently booted my kernel without the threadirqs options and it seems to resolve this issue for me, running for a few hours with a rather high cpu load. however the latency of the system becomes rather high, and i have quite a lot of audio dropouts (xruns of abour 30ms) ... how can i help to resolve this isse? thanks, tim [1] https://lkml.org/lkml/2011/7/11/119